Wikipedia talk:WikiCup/Scoring

2014/2015 brainstorming

It's that time again! Every year, WikiCup rules are tweaked and adjusted, so: what works? What doesn't work? What do you want to see? What do you want to to stop seeing? Do points need to be adjusted? Do we need new rules/methods for running the competition? This initial thread is for throwing around some ideas, after which time I will create some straw polls with possible changes to gauge feelings. The polls can then be closed by the judges. Feel free to start subsections as appropriate, but they are not essential. J Milburn (talk) 17:37, 31 August 2014 (UTC)

• The spirit of the competition was broken in the past when one user would flood DYK with noms and I think he either won or was close to winning. This is happening again with FPs, and is reminiscent of what happened with Durova many years ago (although she seemed to have more variety). As a beginning, I think sets should be devalued. Nergaal (talk) 21:01, 31 August 2014 (UTC)

• Nergaal, I'm not clear why you feel that any user is causing a problem at the FPC project, which is what happened at the DYK project. This competition's FP specialists are very well-thought-of at FPC. There's something very, very perverse about the idea that someone who is producing a lot of high quality content must be a "spam"er or in some other way problematic. J Milburn (talk) 08:03, 1 September 2014 (UTC)

I think we should have points for PRs as was discussed earlier in the year because it is the one area of Wikipedia that seems to be left behind from all the others and often gets neglected. Adding it to the Wikicup should not only help the project, but also people who create PRs only having to wait for months before it gets a review. The C of E God Save the Queen! (talk) 21:18, 31 August 2014 (UTC)

You know what, Nergaal? Shut the fuck up. Because, currently, the logic is: "Someoone who specializes in FPs might win! But FPs suck! We must stomp this possibility down! Seriously, you have literally no other argument, and Godot might not even win: in no round has he even come close to what Cwmhiraeth tends to get from DYK in the last rounds of Wikicup competitions, for instance. If this competition devalues featured pictures - particularly in the absence of anyone winning with featured pictures - then please rename this the article cup, and stop pretending to value all content. This competition has turned from a friendly competition to an ugly one, where people who don't work in fields of content spend half their time attacking other fields.

If the judges wish to devalue FPs, I would petition them to instead stop all pretense and remove them from the competition completely. It would be far more intellectually honest than pretending all content is valued, but devaluing any non-article content that ever threatens to do well in the competition. Adam Cuerden ^(talk) 23:58, 31 August 2014 (UTC)

A reminder for civility czar ♔ 00:25, 1 September 2014 (UTC)

On a different topic, I wonder about benefit to project vs work involved - i.e. is it worth upping GA reviews to 6 or even 8 as there is such a backlog? Cas Liber (talk · contribs) 00:12, 1 September 2014 (UTC)

I would support that, however I feel that points for GARs must not exceed the points for base DYKs. The C of E God Save the Queen! (talk) 17:11, 1 September 2014 (UTC)

I also assume that for DYKs, the 5 year bonus will continue to roll over to be pre-2009 for next year? The C of E God Save the Queen! (talk) 08:57, 4 September 2014 (UTC)

Strong support with the caveat that reviews must be thorough - the 1500 character review (or whatever it is) should be adhered to, AND if met, equal points should be given for a pass OR a fail, lest reviewers be tempted to pass poor articles that don't improve even with a lot of time and effort put into a review. I had at least one GAN review where I think I wound up working harder than the creator did! (A lot of "why not say it this way?" - they'd copy and paste my suggestion verbatim - "How about putting this paragraph" there? -copy and paste- and so on). I have also seen GAN reviews abandoned. Montanabw^(talk) 19:08, 4 September 2014 (UTC)

Vital rebrand

• Saving any editorializing and in the spirit of openness, my honest suggestion would be to consider rebranding the Cup around vital articles (I know the scope is contested, but I also know it was good enough to matter in previous cups), say level three or above. This could still include the point ratios as they stand (and hopefully include peer and featured content reviews of a certain length), but it would do well to keep our benefit to the project clear. I don't plan to participate next year as is, mostly because I feel that my own content quality and article ambitions have decreased with the Cup's deadlines. I'm also discouraged by the air of antagonism and feel that working on BIG articles encourages (nay, almost requires) collaboration, perhaps even between Cup participants? czar ♔ 00:25, 1 September 2014 (UTC)

We discussed this a few years ago - problem is that there is always an issue of subjectivity with vital articles, and over time I am intrigued by the idea that we just go on numbers of how many language wikipedias the article appears in, so we do this after a fashion - make the multiplier bigger then? Cas Liber (talk · contribs) 02:46, 1 September 2014 (UTC)

I think the interwiki idea may even have been your idea in the first place, Cas. Czar, I understand your desire (and I agree with your thought about collaboration- that's not a bad thing!) but I wonder whether we are failing at this current time, in your view? Cwmhiraeth has won for the last two years mostly because she has played the bonus point game, and, in the process, has improved some very important articles. J Milburn (talk) 08:08, 1 September 2014 (UTC)

I worry that bonus points in their entirety become a metagame of optimization. Also I see how interwiki is used as an alternative to the Vital list, but it appears to have not sufficiently incentivized work on Vital-like articles. (For example, The arts, a vit-1 article of generally agreed importance, has a low interwiki count.) At least for one year, I would be more interested in the Cup actually prioritizing the articles that form the basic breadth of the encyclopedia. I can appreciate how, with the inclusion of individual works, the Vital list gets murky after level vit-3, but all the same, a work is not necessarily of greater priority because someone has labored to stub interwikis (or has had a bot do so). Perhaps this is less the Cup than the WP, but I'd like to see the content that reaches people most and the content most expected of an encyclopedia to be incentivized and completed before we discuss how many apples are equal to an orange, especially when both the apples and oranges are getting less than 50 hits a day (mostly bots, at that). czar ♔ 14:05, 1 September 2014 (UTC)

I have my reservations about the vital lists; even at level three it seems pretty arbitrary. Are Ibn Khaldun and Mary Wollstonecraft really more significant than Max Weber and Michel Foucault? I doubt it. Is Moses more important than Adam? Hard to say. Dog and horse are included, but invertebrates (other than insects) are entirely excluded? Seems a little odd. Nonetheless, I am interested in what you're suggesting- how do you feel we could incentivise work on these articles? J Milburn (talk) 16:35, 1 September 2014 (UTC)

You know, I'm going to have to question the level one vital articles. Is Culture actually the best top-level overview? Life has multiple meanings. The arts is, at least in English, a vague term I wouldn't imagine people would search for that often - and, indeed, they don't. Vital articles often seems like a vague attempt to make trees, then after coming to the top of the tree, declaring THIS ARTICLE IS MOST IMPORTANT! Is it? Honestly, I'd say level 2 was far more vital than level 1 in the main. Adam Cuerden ^(talk) 16:49, 1 September 2014 (UTC)

Personally I take the same view I take with DYK hooks "What may be uninteresting to some, could be very interesting to others" and I think this is the same where there is no definitive answer to what is vital unanimously. And since there is no definitive and non-arbitrary way to work out the vital articles, I would have to say that I do not believe that it should be used as a criteria for bonus points. The C of E God Save the Queen! (talk) 17:06, 1 September 2014 (UTC)

My point is chiefly that if the Cup is meant to incentivize content creation, I think the encyclopedia would be better served if that scope is narrowed to the articles agreed to be of core importance to the project, whatever they may be (not necessarily the current VItal list, as there are other metrics). I don't think there would be much of a need for bonus points in such a scenario. czar ♔ 18:24, 1 September 2014 (UTC)

How about a second scorecard just for vital articles, and another Cup awarded for winning? If the Cup is to incentivize content creation; that's great, but I think there will be a large loss of contributors if its scope is restricted. Part of the allure of Wikipedia is that people get to write what they like to write; most of my articles this year were about Sega because that's the project I wanted to complete, and sure enough I kept hammering away at it until I was eliminated in Round 3. That being said, vital articles are neat and important, so why not just do a separate scorecard for Vital articles that awards its own award at the end of the year? Red Phoenix ^{let's talk...} 18:29, 1 September 2014 (UTC)

I felt that the scope restriction would reduce the unhealthy, metagaming aspects of the Cup. From the responses so far, it seems like it's too far from what everyone else wants to continue next year. I don't have a particular interest in making it another institution so I'll let it languish here czar ♔ 18:59, 1 September 2014 (UTC)

Personallly, I think the bonus multiplier helps accomplish the same goal, albeit indirectly; I am considering working on broader articles next year because of the multiplier. That said, I also think the formula needs to be tweaked a bit so that points accumulate faster at the low end and less fast at the high end. Montanabw^(talk) 19:08, 4 September 2014 (UTC)

This is a very good suggestion. I don't think people have much of a problem with a 1.6x multiplier; its when it becomes possible to 500+ points for one article because of a very high multipler and the ability to count the same article up to 3x (DYK,GA,FA) that it gets out of hand. Is a 4x multipler article really twice as important/hard as a 2x multipler article? No! --ThaddeusB (talk) 19:40, 4 September 2014 (UTC)

I thought the multipliers have been a good objective way of encouraging broad articles. Pushing vital content was one of the reasons I resurrected the Core Contest to try and give us coverage of all different aspects of content improvement....Cas Liber (talk · contribs) 22:28, 2 November 2014 (UTC)

New way(s) of surviving to the next round

I would like to see a reworking of the advancement process to allow more people to survive longer into the competition. I feel that if more editors were in the Cup for longer people (both competitors and observers) have fun with the Cup. Some ideas:

1. The first round cutoff could be just getting any points at all. Looking back at the last couple of years, a large proportion (40%?) of people sign up, and then get zero points. Part of the trouble is they are intimidated by the big point scores of the early leaders and just give up. Also, if people know that they can advance by just getting any points, they might make the commitment to participate. Once they do, they might see it less daunting.
2. The next round(s) could have some method of weighting by content type. So, instead of getting 5 points for a DYK, 30 points for a GA and so forth, people could get some portion of a DYK "pie", a GA pie, and so one. What could be done is to count up how many DYKs were submitted in this year's second round, figure out how many points that amounted to, and assign something close to but less than that number to next year's "pie". So, if 1285 DYK points were received in round 2 this year, then a "pie" of 1000 DYK points would be allotted next year. Then points would be handed out proportionally. Thus, if 250 DYKs were submitted in round 2, people would get 4 points for each of them (1000/250). Conversely, if a "pie" is under-submitted, more points would be received for each submission. This would discourage people from flooding with one type of submission, and encourage diversity of submission types by each competitor, and generally even out the Cup.
3. By making the scores more difficult to figure out, true, it will increase the work for the judges, but it will also make it very difficult for people to game the system. It will also make it more difficult for people to consider themselves eliminated, and thus keep people competing in the Cup to the end of the round. Abductive (reasoning) 01:46, 1 September 2014 (UTC)

I like these ideas - just as someone can game the system by waiting to submit until the last minute, isn't the aim to submit material and not rest on one's laurels in the first place? Therefore I am intrigued by pie-slicing. Cas Liber (talk · contribs) 02:44, 1 September 2014 (UTC)

But if it's pie-slicing, doesn't that mean that the size of the pie could screw people over, or make it so one single contribution was amazingly high? For example, take featured portals. Few work on them. So either working on them suddenly becomes ridiculously disproportionately valuable, or there's never any point to doing more than one or two. Likewise, why is variety of contributions a good thing and what does that mean? Articles come in three different "pies" (four if you count lists) and all other contributions get one pie each; would this really serve to encourage variety? Similarly, pie size matters, and I suspect that it's far less fair to have arbitrary pie sizes to limit how much contributions one can get. If someone makes masses of featured articles, or pictures, or lists, that's a good thing, not something to devalue. Thirdly, this actually discourages variety: If people have already gotten, say, 100 GAs, it's not really that worth getting a GA at that point. Adam Cuerden ^(talk) 03:35, 1 September 2014 (UTC)

From what I have seen in the past few years—and even directly above this discussion—there is a lot of concern about people swamping the Cup with lots of one type of submission. As I see my pie system, limiting the total number of points that can be claimed for each submission type will not profoundly change the fairness of the scoring of the Cup, but it will "take the edge off" of the impact of these sort of mass submissions. So, if an editor submits 100s of DYKs, say 55% of all DYKs, s/he will still get 55% of the points of all DYKs submitted (or a bit more, if there are interwiki multipliers). What my pie system will do is reduce personal hard feelings against people who submit lots of one type of submission.

As for why a variety of contributions is good, it is good for the competitors, guiding them to consider making types of articles or portals or whatever that they might never have tried before. I for one have never worked on taking an article to GA status nor FA status. In fact, I've never done anything besides ITN and DYK. Abductive (reasoning) 05:14, 1 September 2014 (UTC)

If you want to encourage variety of content, give say, a bonus equal to the base points for that type of content for the first piece of content of a type submitted by the user in that round. As for the FP bashing, I consider it meritless. We don't attack (and shouldn't attack) Casliber for having over a hundred featured articles. We don't attack (and shouldn't attack) Cwmhiraeth for his massive amounts of work on vital article DYKs and GAs. We do however attack (but shouldn't attack) featured picture contributors. There is, however, no merit in these attacks.

Further, if the point is to encourage a variety of content, attempting to make sure featured pictures is unviable is not the way. Adam Cuerden ^(talk) 07:20, 1 September 2014 (UTC)

I think you are completely not getting my idea at all. Abductive (reasoning) 07:36, 1 September 2014 (UTC)

Abductive, I like your idea of having more people survive the first round, but I note that current numbers allow for a neat pooling system. I suspect if more survive the first round, we'd just have to have a bigger elimination in the second round so that we had a number which was 2-to-the-power-of-x going into the third round. How do you feel about that? Or is there a better way of working it? J Milburn (talk) 08:32, 1 September 2014 (UTC)

Thinking aloud- maybe we could call round 1 "qualification", and any points sufficed, but, so that people were still encouraged to do something in the first couple of months, people could carry 10% or so of their points over into round 2, giving them a head start against the bigger eliminations which were to come? This would mean that just about everyone would be able to have four months of WikiCup, but those first four months would also be fairly competitive, even for the higher scorers. J Milburn (talk) 08:35, 1 September 2014 (UTC)

Now, this has advantages and disadvantages, of course, but... You know, you don't have to halve each round. If you want to leave people in longer, you could have a big cull at, say, end of Round 3, which would also make round 3 more competitive, as a smaller portion would pass (although numerically the same).

Why do the pools exist? I was told that they exist to help people make it to the next round when otherwise they might not have, right? Abductive (reasoning) 15:39, 1 September 2014 (UTC)

It's just the way we've always done it- I'm open to alternatives! J Milburn (talk) 16:12, 1 September 2014 (UTC)

Underlying my thoughts is always, "how can more people keep competing longer"? Abductive (reasoning) 16:32, 1 September 2014 (UTC)

I like the idea of calling round 1 qualification where a set # of points is all that is required to continue and carrying some % of the points to the next round. --ThaddeusB (talk) 14:13, 2 September 2014 (UTC)

• Comment I had no problem getting to the second round in my very first attempt at the wikicup, and I even slid into the second round on the strength of a FAC that I (successfully) begged to be approved prior to the deadline. I'm OK with the way that part works, but here are my thoughts:

1. Maybe allow some points from one round to carry over to the next. If X amount of points is the cutoff, perhaps 50% of the points above that cutoff could carry over. This would also encourage people to put in FACs, which can take more than a month to get approved.
2. Perhaps do more bonuses for participants who contribute more than one category of work (or cap the points that count in each category of work) Kid of like the "pie" idea above, but with the stated goal of having people contribute different kinds of work - I did my very first ITN because of the wikicup, and I am inspired to think about whether to try to do a featured image next time.
3. I also favor more points for GAN reviews, so long as adequate quality controls are in place (see above) Montanabw^(talk) 19:08, 4 September 2014 (UTC)

The FP debate

The featured pictures debate is an attack on content that I have spent eight years developing skill in. My subfield, restoration, is a highly specialized skillset only a handful of Wikipedians possess, whereas I think most people, with a bit of mentoring and a lot of work, are at least theoretically capable of creating a good article, especially on a limited subject.

Featured pictures includes four broad categories of work:

1. Photography - Probably goes without saying
2. Restoration - The act of taking a damaged work, and removing the damage, while staying true to the original
3. Map and diagram making - (For example. Wikipedia:Featured_pictures/Diagrams,_drawings,_and_maps/Diagrams. These, of course, vary in complexity)
4. Researched content/released content [For example, Godot13 works with the Smithsonian Institute to prepare and release their holdings; Other users work to get screenshots from video games released under a free license, and so on]
5. "Found" works [e.g., "Isn't this image I found [on the internet/on commons] awesome?")

For obvious reasons, we don't allow #5 to claim Wikicup points.

Now, there may be some reasonable debate over the first four categories, and whether point values should be tweaked. But that's not what's happening. What's actually happening is vicious, often personal, attacks on the idea that featured pictures are at all valuable.

This goes directly against the spirit of the Wikicup, which is meant to celebrate all content on Wikipedia, not just articles. If you wish a purely article-based contest, there are such contests, such as WP:The Core Contest. Or you could start your own contest, keeping out those nasty picture and portal people. But the Wikicup is meant to be a friendly, inclusive contest.

I have seen very little friendliness, and quite a few attacks on inclusiveness.

And, just to be clear: Maybe Godot will win this year. However, there is every chance he will not, and if he doesn't, what does that say? That everyone's willing to blow up the featured pictures point system because there was a chance someone working with FPs might win? Frankly, to describe my views on what I would think should Godot13 lose and FP still be devalued would probably require far more cursing than I'd prefer. Also, I would lose all respect for the Wikicup organizers for devaluing something that already had little to no chance of being used to win. Adam Cuerden ^(talk) 07:51, 1 September 2014 (UTC)

I'm split on this. Obviously things such as photos are easy to take if you have a quality camera and don't really require an awful lot of work however older images require serious restoration work in order to make the FP criteria. I think some sort of subcategories for FPs might be beneficial to try and stop allegations of easy points being claimed. The C of E God Save the Queen! (talk) 08:00, 1 September 2014 (UTC)

"Obviously things such as photos are easy to take if you have a quality camera and don't really require an awful lot of work"- I think you are seriously underestimating the amount of legwork, artistic flair and technical ability required to take professional-quality photographs. J Milburn (talk) 08:39, 1 September 2014 (UTC)

If you think taking a high quality photo is merely a function of camera quality, then you "obviously" have no actual experience with high-level photography. --ThaddeusB (talk) 14:17, 2 September 2014 (UTC)

Adam, thanks for these helpful categories. There are at least a few more- Scanners fall somewhere between photographers and releasers, editors are a kind of restorer-light, and writers find suitable article placement or even write articles. I agree that differentiating between different kinds of FPs may be profitable, in a kind of "extra base points" way. Maybe it could be three-tiered system: photographers and drawers are offered x+5 or x+10 points, restorers, scanners, significant editors and releasers are offered x points, while finders, writers and non-significant editors are offered 0 points. Researchers are hard to place- probably 0 points. How does that sound to you? J Milburn (talk) 09:04, 1 September 2014 (UTC)

I do think you underestimate just how much work can go into a restoration, but that's rather the point of this discussion, isn't it? To actually discuss what's happening. Adam Cuerden ^(talk) 09:55, 1 September 2014 (UTC)

And I think that's really the problem with the featured picture debate - the actual work involved for the various types is easy to underestimate. For photographs, remember that not every subject is featureable. One might think one could photograph common objects, but common objects are held to vastly higher standards, to the point that simply planning, setting up lighting for, and getting a good background for the shot add huge amounts of work. And if you aren't photographing common objects, then what subject are you doing? Monuments? You're going to run out of ones near you soon, and need to start travelling. Perhaps this might be a small distance at first, but it's going to spread. And if the lighting isn't quite good, people will oppose. Buildings? Only a limited number are independently notable, and you still have all the lighting issues. Fine then, interiors? That limits you to buildings that, firstly, have interesting interiors that you can access. So, maybe major hotels, churches, museums, and theatres, say. Okay. Those are spread out too, and access for some of them will cost - and that's ignoring problems such as the fact a lot of the bigger English churches block photography without a lot of permission-gathering. There's a reason we don't have Westminster Cathedral featured pictures despite it being stuffed full of monuments to notable people, for instance. Okay, so how about objects in a museum? Well, presuming it allows photography in the first place, you still have to deal with crowds of people blocking the view, and glass cases, which can put really awful reflections - and that's presuming there's not greasy fingerprints.

Another problem is that, within any subject, there's easier and harder work to do. With restoration, if the image isn't all that damaged in the first place, it's pretty easy; if it has small little spots of damage, but they're literally everywhere, settle in for a few weeks of work. [There's a reason the Mahan images I've been working on are taking so long, for example]. Even with things you'd think might be similar, say, two Mathew Brady American Civil War photos from the same era? They range from something I could do in 8 hours work, to things I've been working on, off and on, for months and am still not anywhere close to finishing.

This is, of course, true of all content. I'm pretty sure a GA on a subject with only three or four major sources is pretty easy. I should know. But there we have a bonus system that can compensate. However, Featured Pictures do not have this quality: For example, most of the images in User:Adam_Cuerden#Civil_War_Images_that_are_awful_.28in_queue.2C_not_started.29 - are tiny thumbnail versions of images available on the Library of Congress at high-resolution, but which will need some unknown amount of restoration to complete. [ACW images are particularly variable in how damaged they are]. Most of them are used in dozens, if not hundreds of Wikipedias. I will not get a single point more for restoring them, though.

Now, if you're not aware, I do not claim images I don't think I've done enough work on. For example File:Magdalene College Dining Hall, Cambridge, UK - Diliff.jpg has actually had a about.. I don't know, probably about two to four hours put into it by me to remove lens flares that went over a painting. I talked it over with the ed17, and decided it wasn't quite enough work to claim. And trivial amounts of work I don't even consider.

But let's look at another case. For those that don't know, I'm considered disabled. Getting better, but still disabled and thus unemployed. Last September, I was spending all the spare time I had on featured pictures, which is quite a lot if you're unemployed. The results of this was a single set of 20 images, the Puck of Pook's Hill set.

It exhausted me. Soon after that, I left Wikipedia for a three months, burned out. It was a huge amount of work, done in far too little time. That's 700 points. Let's just put that into comparison with a few of the things other people did in that last round

• In the same round, Hawkeye7 got 620 points each for his featured articles of Enrico Fermi and Niels Bohr.
• Cwmhiraeth got 820 points for his featured article Sea

The score to beat that round was over 3000. I managed 23 featured pictures, and came in 6th place. I literally physically could not have done more work than I did, let alone the four times over that work I would have needed. FP is not actually a viable way to win the Wikicup, as far as I can tell. Indeed, even Godot13, so far as I can tell, probably won't win with the scores he's been getting. Adam Cuerden ^(talk) 09:55, 1 September 2014 (UTC)

Adam, there seems to be little progress in understanding each-other's point of view, so I propose a solution: please, pick any single article you want from the almost 5 mil around here, or start a new one, and make it a GA and/or an FA. Once you do that, please come back and tell us how do yo see the amount of work for a GA/FA compared to one of your regular FPs. Nergaal (talk) 10:46, 1 September 2014 (UTC)

Do me a favour and search for Adam Cuerden on Wikipedia:List_of_Wikipedians_by_featured_article_nominations. Or maybe just look at my history in previous rounds and note Urania's Mirror (GA, moving towards FA)? I thought Urania's Mirror was pretty easy, actually. Particularly given I should be able to claim it three times [DYK, GA, FA]. I think last year I had two, Fatinitza and something else - Fatinitza mainly didn't make FA because I can't read German, and suspected there were sources in German I should consult. It certainly does vary - W. S. Gilbert took a hell of a lot more work than Urania's Mirror - but then, the ones that are most difficult are also likely the ones that have large bonuses, and if that isn't the case, we should adjust the rules to fix that. Adam Cuerden ^(talk) 12:01, 1 September 2014 (UTC)

And of all those articles, how does the "easiest" one compare to sets of 10+ FPs? And how do the points for the two cases compare to the amount of work? Nergaal (talk) 13:11, 1 September 2014 (UTC)

Oh, Urania's Mirror was ridiculously easy compared to the images in it (with one or two exceptions: Cancer was exceptionally clean, for instance). It was most of the way there after a day [Seriously: I checked, there's half a paragraph of new material added, and some formatting fixes after that.], and it never required the intensive, unending focus FPs do. The Urania's Mirror images tended towards the easier side as FP restorations go, but even given that, each one was probably harder work. Admittedly, I am good at research, and this helps; and the subject is a very limited one. But I believe that was my point: The GAs likely to be difficult should be the ones eligible for more points. This was not one of the difficult ones. There's a limited number of sources. I actually discovered some new information in the research for it (and told the author of one of the sources, who updated the source, which is always gratifying. =) ) Adam Cuerden ^(talk) 13:32, 1 September 2014 (UTC)

Should probably finish raising it to FA, actually. It doesn't need much if the Peer Review is accurate. I want to get a photograph at the local museum, where they have one of the cards with a light behind it; and I need to create an article on Richard Rouse Bloxam, to cover some additional material in the sources that is a little too off-topic for here. And, of course, I'll probably get at least a DYK off of that article, probably GA.

See, you work on articles on big subjects. Noble gas, Ceres (dwarf planet) and the like are all really major topics, with literally thousands, if not millions of sources. Urania's Mirror is not that sort of subject. It's a limited topic with a few key sources. That's a lot simpler; you can pretty much write the article by planning out the structure, then going through with each source in turn. I think it's also generally easier to get a good article out of something you can write from scratch - bad pre-existing structures tend to be much more difficult to work around or fix. Adam Cuerden ^(talk) 13:46, 1 September 2014 (UTC)

@Adam Cuerden: I was not thinking of major articles, since those get many bonuses. I am thinking of simpler articles like ununoctium: even though that would get interwiki bonuses, it is a relatively simple article, which when I did I had to access all the 20-or-so articles published on it. 20 sources is not much, but believe me it didn't take just a day (excluding the nomination). I am willing to agree that taking 3 random FPs would probably need about a similar amount of work. However, you can't tell me that 3 pics in the same set take more than a day to take/cleanup/etc to FP-level. Maybe even that does, but I am really not willing to accept that a 10-set FPs takes more than 3 days worth of work. Nergaal (talk) 11:36, 5 September 2014 (UTC)

@Nergaal: You seem to be under the impression that restoration is easier in sets. It's not. This is one of those cases where we probably need to break down cases: For a release, the number of images you get could vary for the same amount of work, and, if you're lucky to get a large release, you might save time. For things like @Godot13: does, ask him. For photographs, I suppose large sets could represent some savings - the location scouting and travel and so on would compact. But for restorations, the amount of effort to do one image in the set is multiplied by the number of images. There really is no savings. If the set is made up of relatively easy nominations, you might be able to plough through it somewhat quickly, but, when, for example, I finish the USS Mahan set (4 images), I'll have likely put about 60 hours work into it. They're proving fiendishly difficult so far. Adam Cuerden ^(talk) 01:04, 9 September 2014 (UTC)

[unindent] Perhaps an abstraction will help. Imagine you need to turn all "0"a [the letter] in this table to "0"s [the number], manually, without using search and replace.

7	6	2	2	O	5	2	7	1	1
5	3	1	6	4	1	4	1	7	7
O	7	8	8	6	1	1	9	5	O
O	9	4	8	1	4	3	3	O	6
8	8	5	2	6	5	7	4	8	5
8	8	9	7	6	9	4	0	4	5
8	6	4	9	9	8	4	8	3	5
2	9	9	6	7	5	1	3	9	5
2	5	9	8	4	6	8	9	8	7
6	6	7	O	5	4	4	7	5	3

Would having done that substantially save you time when you then had to do it again on this table, which also has an "I" that needs replaced with a "1" (since different images often have different types of damage, even in a set)?

5	O	7	6	2	8	1	4	9	9
3	6	6	9	7	8	1	2	7	6
5	9	6	9	9	8	3	7	5	2
1	5	I	4	3	9	9	3	8	9
2	3	O	9	5	6	O	1	9	8
8	3	5	1	4	7	8	3	5	3
8	O	4	4	5	7	7	5	6	O
2	7	3	O	3	2	5	8	1	9
5	6	8	1	5	O	7	4	1	8
8	O	9	5	5	8	5	9	2	8

And that's presume you're not dealing with really weird damage, but I think I'll actually deal with that below.

Adam Cuerden ^(talk) 02:05, 9 September 2014 (UTC)

• Suggestion - Here's a thought on how to handle FP's. It can take a lot of work to make one, that's for sure. I know I certainly don't have any for that reason, and because I'm not terribly into photography. I'm not sure the issue is so much that FP's shouldn't count, but that FP's shouldn't be spammed in lieu of written content creation, which is the core part of this project. So here's a thought: why not do a staggered point system for FP's? The first three to five can still earn the full 40-45 or however many it is points, then the next 5-10 can earn up to 20 apiece, then any further earn 5. In essence, FP's can still earn points that way and there's credit given for being featured, but not to the point to where it can be excessive. Thoughts? Red Phoenix ^{let's talk...} 18:16, 1 September 2014 (UTC)

I have 72 FPs so far this year. I'm not going to work on article content if I'm not working on images; I use the Wikicup as a spur to keep myself constantly productive. If later images are devalued, literally the only result is you'll probably get fewer restorations from me next year, as I won't have the motivation to pace myself, as I won't be doing the Wikicup. Adam Cuerden ^(talk) 18:38, 1 September 2014 (UTC)

You know, under my "pie" system, later contributions could be worth more than earlier ones. It could be done by making the pie bigger, or just by the loss of competitors in later rounds. Abductive (reasoning) 19:40, 1 September 2014 (UTC)

That's only true if there's competition for the slot. Once you have, say, 90& of the pie, it's probably not worth doing ten times the work you've just done over again to get 99%. [Sample math: If you have 90x, and others have 10x. to get to 99% you need to get to 990x.]</snakk> The pie system has rapidly diminishing returns. Adam Cuerden ^(talk) 19:52, 1 September 2014 (UTC)

If the pie system is adopted it does not need to be for all the rounds. My hope is to keep a lot more people in the Cup in the first few rounds. The last few rounds will have to eliminate competitors all the more mercilessly, so it would not be feasible to have the pie system in place in those rounds. Abductive (reasoning) 21:38, 1 September 2014 (UTC)

I started off as a Wikipedian many moons ago, and until this year I have been barely involved in Commons. My interest was and still is articles about places, which probably make up two thirds of the articles I have ever edited. If you'd asked me about photos and image restoration last year I wouldn't have given you the time of day because I felt sourced text was more important than pictures. However, this year I have made an effort to do more editing on Commons, took a few photos (not very good ones), uploaded some Flickr photos and even tried my hand at things like airbrushing out watermarks (not very successfully). I now see myself as a WikiMedian rather than a WikiPedian and I would highly recommend all Wikipedians to expand their contributions beyond sourced text to include images. This dispute about articles versus pictures is frankly a banal distraction because we shouldn't lose sight of the fact that this is an online encyclopedia and most of us voluntarily edit here. Adam shouldn't really have to tell us his personal circumstances because we should judge his contribution by what he actually does here rather than belittling his efforts. I think an ideal solution would be to modify the scoring system so that the four major content items below featured articles should get the same scores, i.e. good articles, featured lists, featured portals AND featured pictures should all get 40 points. In line with this, good topics should go up to 4 points per article and good article reviews should go up to 5 points per review. Featured articles and DYKs should stay the same but I think ITN should be dropped to 5 points because the criteria are so much lower than anything else on the main page - a five-sentence update (with at minimum three references). I don't think that that criteria matches up to strictures like DYKs needing at least 1500 characters or five fold expansion for existing articles. Giving ITN 10 points puts it on a par with a 10 point DYK, which may have need more effort involving 5KB of prose. Another way of making things more interesting is if we could have bonus points for the effort involved in getting articles to GA status e.g. bonuses for writing a good article from scratch or for taking an article from stub to GA within the time period of the round. Anyway, chill out everybody, stop sniping at each other, eat more vegetables and enjoy the WikiCup and that is my tuppence worth of Tyke opinion. Green Giant (talk) 22:18, 1 September 2014 (UTC)

Re: ITN - while a minimal update may qualify for ITN at times (usually not), it doesn't qualify for the WikiCup under the "significant contribution" criteria. There used to be a line in the rules saying judges would scrutinize such submissions extra closely specifically to prevent such submissions. Additionally, ITN has time constraints and the significant risk of your work updating an article receiving 0 points because it was decided the story was insufficiently significant or was too outdated by the time the content work was done. If that happens, there is nothing one can do about it - unlike other areas where one can fix it up and resubmit. As someone who does a lot of ITN work and a lot of DYK work, I must say the criteria certainly are not "so much lower than anything else on the main page". Indeed, most ITN regulars believe our criteria are stricter than DYK. And there are no multipliers either. If my goal wa sto maximize points, I would focus on expanding stubs with heavy multipliers to 5k DYK articles. That is a far easier way to rack up quick points than ITN (or FP or just about anything else). --ThaddeusB (talk) 14:39, 2 September 2014 (UTC)

Apologies if it seemed as though I was insulting ITN, it was just that that particular criteria seems quite low compared to 5KB for DYK which also gets 10 points. Perhaps if we had a similar split to DYK so that larger ITN items would get 10 points and smaller ones get 5 points?

• The multi-tier point system for FPs is a good idea. It would also allow the opportunity to give a few points for "merely" finding and uploading high quality images (which is not actually zero work). I would propose something like:

Photography, Restoration, and other creative efforts (e.g. drawing) - 35 pts

Scanning, Securing relicensing/permission - 10 pts

Finding already properly licensed photo (online) & uploading - 3 pts

--ThaddeusB (talk) 14:46, 2 September 2014 (UTC)

• Frankly, this whole kerfuffle is overblown; impressive as the FP editors have been thus far, people have been holding back and not unleashing the true Cup-winning strategy of a flood of DYKs with large bonuses. It's won the last two cups in a rout and none of the rule changes made in the last two years have done much to address the issue. So I hate to tell y'all that this has all been, to quote Shakespeare, "a lot of sound and fury, signifying nothing".--Sturmvogel 66 (talk) 00:32, 3 September 2014 (UTC)

• This is correct - high multipler stubs remain by far the most efficient ("easiest") way to earn points. --ThaddeusB (talk) 16:20, 3 September 2014 (UTC)

• I think we either need to lower the FP points or make it so someone can't run the table with them. FPs are good, but way overrated. I'm only getting 36 points for an article like this one. Why should an FP be almost the same amount? Don't forget, you can't have multiple GANs or FACs in a single nomination. Honestly, I have no problem with those articles that get ridiculous amounts of bonus points. No offense, but those are way more work than an FP.--12george1 (talk) 15:48, 3 September 2014 (UTC)

• Actually, a typical species article or city article (both which generally get good multipliers) is not "way more work than an FP". Indeed they are probably less work on average. If the points are supposed to indicate greater importance of such articles, that's one thing, but to claim they are more work on average is flat out wrong. By far the most efficient way to earn points is to bring an old stub with a high multipler to DYK and then possibly GA. Any claim that any type of content gets too many points based on work required can not be taken seriously by me if it claimed that 2x,3x,etc., and old stub DYKs are substantially harder than new article DYKs and thus points reflect work required.

The perception of work required to make a FP does not meet with the reality - creating or restoring a high quality image involves a lot of effort and skill. For people who are not photographers, it may seem like "push a button and you are done" but taking a high quality image is not at all like causual photography. Similiarly, a quality restoration is not like pushing a few button in Photoshop - it is a like of time consuming work. I say all this as someone who does a ton of article work and very little image work. --ThaddeusB (talk) 16:20, 3 September 2014 (UTC)

While I agree, I do think that the bonus system for articles could be improved a bit, to accurately reflect the work and importance put in. I still think we should lower the base GA points, but have bonus base points available for length that could raise the base above what it is now. Adam Cuerden ^(talk) 19:41, 3 September 2014 (UTC)

What sort of numbers were you imagining? J Milburn (talk) 19:49, 3 September 2014 (UTC)

I'd need to check lengths to get it right, but maybe set up to work out to between 20 and 50 base points? Of course, one probably should list int in the score table as 20 to 50; listing the lower number alone would put people off. The thing is, if an incredibly difficult, 40,000 word GA is worth the same as a simple GA of a few thousand, it's not really surprising that people who work exclusively in the former should complain, is it? But surely the response is not to attack FPs, but to deal with the things being undervalued. We should probably discuss DYK bonus multipliers again as well, since they're probably a bit high. Adam Cuerden ^(talk) 19:52, 3 September 2014 (UTC)

• I just saw this page. Not planning to follow it right now (just a bit demoralizing). I'll make comments after the cup is over. Thanks--Godot13 (talk) 05:03, 4 September 2014 (UTC)

• I tire greatly of this constant bickering, yet am thankful for the reminder to never enter the Wikicup. Speaking as an editor active in pretty much every area covered by the Cup (I have credits in DYK, GA, FA, FP, FL, FT, and FPo), I am shocked that people without much experience in one or more areas seem to automatically consider said areas to be less valuable then what they do themselves. DYK has long been claimed to be mere hat collecting, GAs have been said to be less strenuous than FA (and even, sometimes, easier than DYK)... I've heard FL called "just compiling data" and seen users try to sabotage FL nominations because they included a lead (really now...). FPo? "Copying and pasting articles by others and then making it look pretty." And... the countless bytes killed in this pointless discussion of FP.

The value of an image... we can all agree, I think, that TLDR applies to both posts (like this one) and articles. If I have to read a 40k article like Departures without even a single image, I might just give up. Images break up the flow of the text, attracting the eyes, and letting one measure the progress down the page. They also help expand on ideas discussed through the text, or lend a face to a subject, which in turn expands readers' understanding of the article. Writing articles without any images is like cooking without any spices: sure, you can do it, but is the result really the best you can do? And the better the image, the more useful it is for the reader (and we, the article writers). Even paper encyclopedias have images.

Taking a featured photograph is a pain in the ass, especially to start. You need thousands of dollars of equipment (unless you get lucky), access to objects / people which need a good quality image, and a familiarity with different aspects of optics to get the right effect. You also need to time your shoot to get good lighting (not very easy if you live far from what you're shooting), and time to do postprocessing work (lighting adjustments, warmth, etc.) It took me three trips to Taman Sari to get the shots I wanted for my FP set, each trip at least an hour and a half in length. As I was doing stitched panoramas, there was another 4 or 5 hours each shot of stitching and restitching, to fix errors caused by parallax or masking issues. Restorations are likewise time consuming, and require great patience and understanding of the different tools available (as well as a fair investment, for those who use Photoshop).

But what people seem to be complaining about most this round is scans. I've only started doing scans of banknotes (example), and even then with notes I had reasonably ready access to. How long do you think it takes to make a single scan like what I've linked to? If you guessed anything under an hour, you're way off. We need to take time to clean off the scanner (careful not to use something that will smear and make the scan blurry), clean the note, run the scanner (at high resolutions, with multiple passes, this can easily be half an hour on its own), and then through postprocessing make the note presentable (black background for attractiveness, digital removal of any gunk [hairs, for instance] which passed our cleaning), etc. Godot has the added challenge of actually getting access to to the notes: travelling to the Smithsonian, getting permission to access these (sometimes hundreds of years old) notes, and following the rules they set when giving him permission. This is not a simple venture, as anyone who's tried to negotiate the release of free media can attest. I'd be surprised if the average time for each note was under three hours (for comparison, some of my shorter FAs were written in less time).

Rather than complaining that a certain process is too easy, and thus the results of the Cup (whatever they may be) are unfair, why not try to learn about the process and what it involves, and once an understanding is reached, civilly discuss whether there should be a change to the points system? I mean, it guarantees we won't lose much needed contributions, broadens our own horizons, and builds a spirit of camaraderie that the Cup is supposed to promote. Hell, some may be interested in trying their hand at scanning notes - and that can only benefit the encyclopedia. — Crisco 1492 (talk) 05:44, 5 September 2014 (UTC)

Samples of damage

Here are examples of some of the challenges restorationists might face:

Before	After	Notes
		Philippe Pétain portrait with odd "scribbly" damage in most darker areas of the picture. Note that there's more of it than just on the obvious dark bit - look a bit closer. Part shown: The chin and neck. Restoration incomplete, but forthcoming, probably this round.
		From File:Sébah & Joaillier - Sultan Mehmed VI.jpg, lower right corner. Bits of blue paper got stuck to the carte de visit.
More coming shortly

But, honestly, and I'm not going to link anything for reasons that will become clear - I recently did a Urania's Mirror series. It took quite some time, but the restorations were relatively easy as such things go, with a few notable exceptions. Now, the thing is, though... A lot of them had restorations attempted by other people. And by and large, they were... really bad. And a lot of people had tried things. Quite simply, for a lot of these things, it does appear that, if one of maybe four people don't do it, noone else can. Stains seem particularly hard for people to figure out what to do with (Though, to be fair, they can take hours.) The LoC is particularly bad with uneven tone to scans; Most of my savings on things I scan myself are down to not having to deal with as many tone issues. Adam Cuerden ^(talk) 05:54, 9 September 2014 (UTC)

I think 30 or 35 pts for an image is relatively fair. There are FPs that require far less effort than 30pts, but I am fine with that. What I am personally not OK with is having a set of 10+ somewhat similar images that get put into a set and then are worth 3 whole FAs. I am not ok with having those 10+ images being isnerted as a gallery in the bottom of a relatively limited-scope article, and then asking for 350+ pts for that. I genuinely believe that those 10+ pic require only the amount of research of a single image and I am pretty sure that the amount of overlap between the restoration is large at best. I wonder, how would the current edition of the wikicup look if every FP would keep the same amount of pts but the sets would only get 35+10(n-1) pts? (A 10-pic set would be worth 125pts not 350.) Could someone find out? If anyone wokring on FPs wants the 350pts at least they should word on 10 different pics that get spread across 10 articles instead of throughout a single section of an article. Nergaal (talk) 08:16, 9 September 2014 (UTC)

Nergall, it's literally impossible to have work on one image apply to another. You seem to think there's some magic way that if I work on removing damage from one image, it will magically make damage go away from the next image in the set, or something else that I can't possibly figure out what you mean. Removing things like the damage shown is a whole ton of work, and there are no automated processes that can fix this sort of damage. I honestly cannot possibly figure out what your logic is here. If you're presuming that research is the rate-determining step of an image restoration; the need to do extensive actual research on images is rare (And it's generally things like finding the death date of the authors, etc). Finding something to work on is not something I will ever have problems with; admittedly, it does take a certain amount of time, but that's far dwarfed by the time spent actually restoring images.

Obviously, this depends on how you work. I have a wide range of subjects I'm willing to work on, as well as living in Edinburgh, where acquiring old books fairly cheaply is pretty easy. So there's that. As such, it seems like your argument is the same saying that I don't deserve credit for Creatures of Impulse as most of the sources are the same ones I used on W. S. Gilbert.

I'm honestly trying to understand your point, but I can't figure out how to make your argument make sense without presuming you're fundamentally misunderstanding something. Adam Cuerden ^(talk) 15:06, 9 September 2014 (UTC)

• A basic question should be asked: Are you working on these articles for your own reasons or to win or do well in the Wikicup? The Wikicup is pretty easy to game, a picture set that gets reviewed and done at once can be speedily done whereas my 50 Good Articles require months of waiting and need to be reviewed individually. Simply from a processing point, picture sets blow through the processes where other works are bottlenecked and delayed for longer periods than the entire round itself. You can make all sorts of criteria for defining points, but work is work and Wikipedia does not need a textbook for rules for something like the Wikicup. 30 images of rare bills in a gallery may be of lower "value" to general readers than 30 famous restored portraits of historical figures and articles on those figures themselves. Those out gunning for points to "win" and make a fuss rather than enjoying their Wikipedia work need to re-evaluate their motives. ChrisGualtieri (talk) 15:28, 9 September 2014 (UTC)
  • Chris, I think you've hit the nail on the head with the reviewing point. For various reasons, the FP process is extremely efficient in the way that GAC, FAC and FLC aren't. This is likely a big part of what makes the FP points seem "unfair" to some people. Deciding how we as a group respond to that, or if we should respond at all, is difficult. J Milburn (talk) 16:26, 9 September 2014 (UTC)

It's hard to see how you could justify responding to it without it being massively unfair. What are you going to do? I can't see anything you could possibly do that doesn't come down to telling productive members of the community to stop being so productive. It's probably impossible to win the Wikicup with FPs. Adam Cuerden ^(talk) 08:00, 10 September 2014 (UTC)

I think the point is this: We have previously talked mostly about balancing points in terms of difficulty of producing content of the requisite quality (FA is worth more points than FL, say) and value (bonus points, say). However, there's at least one other issue which is potentially pertinent- the efficiency of the system. Ideally, what we'd be able to do is balance the points so that they are fair all things considered, which will certainly mean that they are unfair when considering only one dimension (say, a highly important GA may be easier to write than an obscure but technical FA, but, all things considered, we may be justified in giving more points for the GA). J Milburn (talk) 16:25, 10 September 2014 (UTC)

Last year, I promised to review any GA I was asked to review. Almost noone took me up on the offer, repeatedly made. I can't promise that this year, I'm much busier, but, in any case, for people who actually use the Reviews part of the Wikicup, things do seem to be dealt with fairly quickly. If people are complaining, but not actually using the resources offered to fix the issue, that's their problem, not FP's. Adam Cuerden ^(talk) 16:50, 10 September 2014 (UTC)

• The issue with "balancing" FP points is that you'd have to treat set and individual nominations separately (if you did want to discourage set nominations in the first place, which I strongly disagree with: nominating as a set may take less time to review, but you still have to deal with 10, 20, 30 pictures and bring them up to snuff). Personally, I'd try and improve the efficiency of other processes: introduce points for Featured content reviews (FA, FL) as an incentive for Wikicup participants to review, not just allow but recommend that Cup participants request reviews from fellow Cup participants (beneficial for both parties, as both get points, assuming its all done neutrally and the article passes review). As an FL delegate it pains me to have to fail so many nominations because of a lack of reviews; if we have an incentive, and Cup participants work towards reviewing, a surge of reviewers will work to improve the Featured Content processes for everybody ... maybe even in a sustainable manner. — Crisco 1492 (talk) 01:48, 14 September 2014 (UTC)

An FP question

• For a FA you are limited by only being allowed one nomination + one joint nomination at a time. Content must have been worked on during the course of the WikiCup.
• For a GA you are limited by the willingness of a reviewer to take up the nomination. Content must have been worked on during the course of the WikiCup.
• For a DYK you are limited by certain date constraints, the willingness of a reviewer to review and the selection of the nomination once approved to join the queue. Content must have been worked on during the course of the WikiCup.
• What limits are there for a FP? Cwmhiraeth (talk) 07:54, 14 September 2014 (UTC)

• Oversimplification on FA: with leave from an FA delegate, you can nominate more than one at the same time. This is usually as one nomination is wrapping up, however.

• Limits on FPC: a smaller reviewer base (check how many regular reviewers each process has), a greater number of supports required (five for FP vs. three for FA), the fact that for many images if you screw up you can't fix it (check out this nom for an example; at least with prose you can polish it further), and (I believe) content still has to have been worked on during the course of the Wikicup. Simultaneous noms are allowed (as at GA and DYK), and there is a firm deadline of 11 per nom (which may work for or against a nom). — Crisco 1492 (talk) 08:21, 14 September 2014 (UTC)

• I think I have more successful FP noms than all other FCont together. I don't think I had to do any work any of those FPCs, while all my FLCs and FACs, plenty of which unsuccessful, took a very non-negligible amount of work. Nergaal (talk) 16:42, 14 September 2014 (UTC)

• @Nergaal- For some reason you feel that it is necessary to demean the work of others. Is that what this forum is for?--Godot13 (talk) 23:36, 14 September 2014 (UTC)

• Sorry if the impression you got was me trying to trivialize FPs. None of those FPs I had would have been eligible for points in the WC because I did not do any work to the pics. They were simple drive-by noms, which did not require any actual work from me. The only thing I wanted to point out about the "C" part (the nomination part) is that it is impossible to get drive-bys at FLC and FAC while at FPC that is not a limitation in any way. It does not imply that any WC-eligible FPs require a small amount of work; just that during/for the nomination part of 90% of FPCs the work is very limited. Nergaal (talk) 00:50, 15 September 2014 (UTC)

@Nergaal: This forum is discussing FPs in the context of the WikiCup, meaning that the issue of "drive-bys" (which quite rightfully do not receive any points) is completely irrelevant. Introducing this just serves to divert discussion from the actual issue. It doesn't even show that "during/for the nomination part of 90% of FPCs the work is very limited", as even if the images required adjustments, "drive-by" nominators would almost certainly be unable to adjust properly. Those that do require adjustments (such as my example above, or the recent image of Peter Carey) are sometimes irredeemable without starting from scratch, a situation rarely (if ever) faced by experienced editors at the other featured processes.

(On a further point: you are incorrect about it being "impossible" to get "drive-bys" at FLC, FAC, and/or GAC. Nominations of articles one hasn't written oneself can succeed, but it is considerably more difficult, such that in my experience it is almost never done. I only know of one that succeeded, an IP GAN nom for "Langit Makin Mendung", and even then I was there to deal with any issues) — Crisco 1492 (talk) 02:24, 15 September 2014 (UTC)

Actinide also succeeded, if you want another example (this was from me in 2011, still kinda not getting the point that you should have actually done work on articles you nominated!). Double sharp (talk) 14:16, 17 September 2014 (UTC)

For that matter, I don't think I did nearly enough on Fanny Bullock Workman to claim it here, but it's just passed. Adam Cuerden ^(talk) 19:45, 21 September 2014 (UTC)

By the way, if FP is supposedly so easy to get votes for, I'd just like to note that half my current FPs are failing for lack of sufficient votes, because I did what some here apparently want to insist on: Opened a lot of nominations at around the same time. Adam Cuerden ^(talk) 19:45, 21 September 2014 (UTC)

So something that can have unlimited submissions, pass in only 10 days, and have the nominator be a supporter is way harder than an FAC? In other words, nominating 40 pics for FP simultaneously is fair? All I am seeing is you guys crying about having to score the same way everyone else does. Have fun next year, because I won't be in the WikiCup.--12george1 (talk) 02:47, 1 October 2014 (UTC)

Personally I was sitting on the fence on this due to being out for the final but just seeing the scores, I have to say that to get that much for pictures is bordering on obscene. I do not wish to belittle others work but to get over 3,000 points in a month for a bit of image work I think is not really fair and reflective of scores. Things like that I know would discourage many people from participating fully if it happens next year. The C of E God Save the Queen! (talk) 08:02, 1 October 2014 (UTC)

If you don't want to "belittle others [sic] work", maybe you shouldn't refer to their efforts as merely "a bit of image work". It's ridiculous that we can't have a discussion about this (and I'm looking at certain people on both sides...) without being aggressive or dismissive. J Milburn (talk) 16:35, 1 October 2014 (UTC)

I doubt anyone who works with FPs will join the Wikicup for a long time, after the abuse that began being hurled around the moment it looked like someone who worked in them might win. I like that, after months of people abusing FPs, you decide to blame both sides. Because the people upset at their content being belittled are just as at fault, of course. Adam Cuerden ^(talk) 10:59, 2 October 2014 (UTC)

Please do not put words into my mouth. I very clearly did not say that. I've been one of the few people on this talk page defending you, and it gets harder for me to do so every time you post. J Milburn (talk) 17:46, 2 October 2014 (UTC)

Then what, praytell, was the "on both sides" meant to mean? I realize you may want to be diplomatic, but blaming the victims isn't on.

“ "The most important rule is that the WikiCup is just a bit of fun — at the end of the day, we're all here to improve Wikipedia. The second most important rule is to treat Wikipedia and other users with respect. If through the WikiCup any participants are hurting the encyclopedia (whether through abusing the rules/systems, creating a negative atmosphere, or whatever else), they will be removed from the Cup. " - WP:Wikicup/Scoring ”

But that isn't true, is it? This competition, for the last six months, has been toxic, and I quit. Adam Cuerden ^(talk) 19:08, 2 October 2014 (UTC)

Peer Reviews

I made this suggestion earlier but the discussion got overshadowed a bit. Having experienced lengthy delays personally, I feel it would be beneficial to Wikipedia and to WikiCup participants if Peer Reviews are made part of the WikiCup. At the moment, people have to wait for months to get PRs done, if they are part of the Cup and eligible for points then it is likely that people won't have to wait long as people will want to do PRs for points. This would help the people requesting the reviews and also newer Cup participants. The scoring would be on a par with GANs, except slightly lower and with a minimum requirement of tangible comments made for improvement, for example 600 characters with a minimum of 3 improvement suggestions. This is just a theory and I hope it can be refined through discussion here. The C of E God Save the Queen! (talk) 08:13, 1 October 2014 (UTC)

• I am in for having PR having same points as GAN and also having the same requirements. Nergaal (talk) 12:41, 1 October 2014 (UTC)
  • Nergaal, my worry with that is that a GAC is necessarily "full"- you have to review all parts of an article. With a peer review, you could easily add 1500kb of text just talking about the images or reference formatting or something. Maybe we could specify that the review has to be relatively "full"- that is, not exclusively looking at a small area of the article. An image review or source spotcheck is not sufficient for points, for instance. I think the feeling is generally in favour of points for peer reviews- we just need to get it right. J Milburn (talk) 16:31, 1 October 2014 (UTC)
    • As I've said, put the same requirements that GANs have had this year, and give it the same points. In principle if an user gives a complete PR and then the article gets GANed, then the article can be quick-passed (and without additional points for GAN the PR would be in essence the same thing as a normal GAN). Nergaal (talk) 06:03, 2 October 2014 (UTC)

Fairer groups

Right now, the groups/pools are, as far as I know, completely random. I think it should be like the following instead.

• Round 1 is still one group.
• Round 2 is seeded based on round 1. So for example if we have 8 groups again, the top eight scorers from round 1 are put into pot A, 9th–16th make up pot B, etc., and a random editor from each pot goes into each group.
• Round 3 is done the same way with round 2's scores.

Personally I would still have wildcards with this, but if it ruined the point of wildcards to some people then it's not like we have to have them. –Ugncreative Usergname (talk) 04:57, 22 October 2014 (UTC)

A FIFA-like or NFL draft-like system would be good, and could be a decent incentive to prevent some people from stacking/delaying entries for next rounds. Nergaal (talk) 09:37, 22 October 2014 (UTC)

Yeah, the short version of my thought was that it should be like UEFA World Cup qualifying, but obviously not everyone knows what that means. –Ugncreative Usergname (talk) 15:10, 22 October 2014 (UTC)

UU: We have some discussions about the way pools will work next year at Wikipedia talk:WikiCup/Scoring. The proposal there is not necessarily in support of fairer pools, but in support of keeping more people in the competition longer. I'm not sure if the two are compatible, but they may be. J Milburn (talk) 16:58, 22 October 2014 (UTC)

I take "scoring" to be a separate thing from "format", so I just put it here. I took a look and saw the "more people" discussion you're talking about, and no one else seems to be pedantic about the word, so I'll copy this section to that page since it seems to get more attention and the discussion can continue there.

And they are compatible, you just put an arbitrary number of editors in each pot to make the same number of groups. Or you can raise the number of pots if you want more people in each group instead. –Ugncreative Usergname (talk) 21:49, 22 October 2014 (UTC)

Straw polls: 2014-2015

Ok, based on the discussions above, I am now opening some straw polls to help work out what changes, if any, we need to make for next year. J Milburn (talk) 17:10, 2 November 2014 (UTC)

Peer reviews

Should peer reviews be awarded points, provided there are certain guidelines to prevent incomplete/shorter reviews are not eligible for points? If yes, how many points should reviews be worth?

Yes- peer reviews should be awarded points

1. tentatively yes. Anything that encourages collaborative editing should be encouraged, and Peer Review should be a vital cog in the wheel of article improvement. Harder to quantify than others though. Cas Liber (talk · contribs) 05:18, 3 November 2014 (UTC)
2. As nominator, of course since it gets lots of backlog with people waiting months for reviews because there is little incentive to work there. Including it in the WikiCup would change that and also give newer editors a chance to pick up a few points. The C of E God Save the Queen! (talk) 10:37, 3 November 2014 (UTC)
3. Strong Support - The backlog at peer review is huge. Plus, peer reviews are just as helpful as GA reviews. GA reviews is just a general check to see if it meets the criteria, point out some parts that need fixing. Peer reviews are basically the same, it just points out the parts that need fixing on the criteria set by the nominator. Technically, any article could request a peer review. Peer reviews could help an article come from a C-class to a B-class or help prepare for GA or FA. It's a really important part in an article, especially if the article is written mostly by one editor.  ΤheQ Editor  ^Talk? 20:46, 5 November 2014 (UTC)

Peer reviews should be worth fewer points than GA reviews

1. Support as the nominator. It is the one section of user related contributions that is untouched by the WikiCup and should be part. My opinion on point numbers are on the lower side but I don't mind so long as PR gets into the WikiCup. The C of E God Save the Queen! (talk) 23:16, 2 November 2014 (UTC)
2. Yes, it is unstructured, and even a few passing comments can be extremely helpful - would rather just give a base 2 point bonus and the minimum requirement of two pieces of advice in a Peer Review section of a given article. That way, if someone makes two driveby comments each to several articles at PR, I think it is in some ways more helpful than one in-depth analysis. There are plenty of subjects I know ziup about but could make a couple of comments for. Cas Liber (talk · contribs) 05:18, 3 November 2014 (UTC)
3. My preference is for "lighter" requirements for a PR review to count - as Cas Liber notes minor comments can be very helpful - and a lower point top match for PR in comparison to GAR. I would also accept the same requirements and same point numbers as a second choice. --ThaddeusB (talk) 22:48, 3 November 2014 (UTC)
4. Agree, as per Cas Liber.  ΤheQ Editor  ^Talk? 20:50, 6 November 2014 (UTC)
5. Support this would be really useful to the PR process and would prevent articles languishing, however I agree with Cas Liber above that, as many useful reviews are not very long (and often an encouragement to try for GA), it should not be worth many points. --Tom (LT) (talk) 22:23, 6 November 2014 (UTC)

Good article reviews

Should good article reviews be worth more points? Indicate the amount you think would be appropriate in your vote.

No- Good article reviews should stay at 4 points per review

1. I feel that GARs should never overtake DYKs in relation to points. DYKs require hard work, individual research and continuous improvement while GARs require going along the GA and copyedit checklist and making sure that the article meets the list. The C of E God Save the Queen! (talk) 10:41, 3 November 2014 (UTC)
2. Four points, i.e. justa bit less than a DYK, seems right to me. --ThaddeusB (talk) 22:49, 3 November 2014 (UTC)

Yes- Good article reviews should be worth more points

1. torn on this. we really need to push for content. I could live with the same but maybe a bit more would be good. Cas Liber (talk · contribs) 22:36, 2 November 2014 (UTC)
2. A proper GA review takes considerably more time than writing the most basic DYK. My review of Kate Millett took about 7 or 8 hours of active reviewing. — Crisco 1492 (talk) 00:32, 5 November 2014 (UTC)

Good article review discussion

• I guess it is a question of what is worth more to the 'pedia..one good article reviewed (and passed/failed) or one DYK. If I think of it like this, I don't think a GAR is 40% (or less) of a DYK nomination in value, so maybe increasing it to 8 or 10 points maybe (?) Cas Liber (talk · contribs) 22:36, 2 November 2014 (UTC)
• A GA review could take an hour to a week to finish reviewing. It depends on the article, sometimes it's really easy and you get it done really fast while maintaining quality. But sometimes, longer articles take longer to review, especially if a lot of the sources are book sources. It also depends on how well you are reviewing it. Unlike FA reviews, it's not that strict. Some people might review the major aspects thoroughly but may miss some minor mistakes and sometimes people review everything thoroughly so that it strictly meets the criteria. What I'm trying to say is that how the points awarded should be looked at the length of the review and complexity of the article.  ΤheQ Editor  ^Talk? 20:54, 5 November 2014 (UTC)

Featured pictures

Does something need to be done about featured pictures? There has been concern that they are currently inappropriately handled. If yes, what?

Featured pictures should be worth fewer points

1. I suggest that FPs should score 20 points each. If that score were applied to this year's competition, the winner would still be Godot13, but he would win by a much smaller margin. Cwmhiraeth (talk) 19:32, 2 November 2014 (UTC)
2. The point numbers we were seeing were obscene, I think we maybe need to narrow the field on this a bit. The C of E God Save the Queen! (talk) 23:17, 2 November 2014 (UTC)
3. 20 or 15 15-25 per FP. YE ^{Pacific Hurricane} 16:13, 3 November 2014 (UTC)
   • If you're going that low, Just remove FPs, FPOs, and FT's from the stystem, and stop pretending this is about anything but article work. Seriously. That's ridiculous, insulting, and wrong. Just have and article club, and stop pretending to care about any other content. Adam Cuerden ^(talk) 18:42, 3 November 2014 (UTC)

   Having different types of FP worth different values seems tricky and a source of potential problems down the road. Based on the speed of the various review processes, concisely illustrated by Sturmvogel 66 below (Review Speed), some kind of FP point reduction sounds appropriate, until such time as other review processes become more efficient (I’m not holding my breath). Below 20 points is insulting (and to me reflects a devaluing mentality as opposed to fairness in scoring). At 20-25 points, I would be looking to try and develop an FP set bonus structure. Since nobody appears to appreciate the difficulty of sets over individual FPs that’s highly unlikely to happen. At a minimum score of 25 points (a 30%+ reduction) I would hope there is the possibility of a lively (productive?) discussion.--Godot13 (talk) 23:53, 3 November 2014 (UTC)

   On second thought, why? Because people don't understand (or want to) the work that goes into the process? The fact that much of the FP discussion started in April/May and continued throughout the competition reflects a serious lack of gamesmanship. In a perfect world this would actually be a larger concern to the community than how FPs are scored. Realistically, I was just hoping that someone actually noticed.--Godot13 (talk) 01:20, 5 November 2014 (UTC)

   • Eh, on a second thought, I think 25 per FP is fair. YE ^{Pacific Hurricane} 06:13, 4 November 2014 (UTC)

4. 20 points. Gloss 05:12, 8 November 2014 (UTC)

Featured pictures should not be worth fewer points

1. There hasn't been a single argument for devaluing that hasn't reduced to "how dare someone who works in a field I don't do well in this competition." This has been explicitly made at times, with people complaining that it was somehow wrong for people working in pictures to be doing better than people working in articles. Given people working in pictures aren't going to touch this toxic, hostile competition again anytime soon, making sure that we never return would be a terrible idea. Adam Cuerden ^(talk) 17:55, 2 November 2014 (UTC)
2. The maximum number of points from a FP shoudl stay unchanged. However, I am in favor of a tiered structure. See comments below in the "Different kinds" section and in the "Discussion" section. --ThaddeusB (talk) 00:25, 4 November 2014 (UTC)
3. ThaddeusB and especially Adam Cuerden's postings of how much work is required convinced me. If it takes more time, then of course it gets more points, that's the only thing that makes sense. And if Godot wins by a large margin because he put tons of effort into the category that takes the most work... then duh? –Ugncreative Usergname (talk) 04:39, 4 November 2014 (UTC)
4. I honestly think that if some of the nay sayers tried to take or restore FPs, they'd very quickly figure out that it's no instant points, by any means. There's serious effort. — Crisco 1492 (talk) 00:27, 5 November 2014 (UTC)
5. Per Crisco (and per above generally).--Godot13 (talk) 20:57, 5 November 2014 (UTC)

Featured picture sets should be worth fewer points

1. FPs should be worth the same amount of points, but I think with sets of more than 3 images, any image past the 3 threshold should be worth half the points. It is still possible to runnaway with FP sets and slaughter everybody, but at least this discuorages sets in the same nom, and encourages separate FP noms. Nergaal (talk) 13:29, 4 November 2014 (UTC)

Feature picture sets should not be worth fewer points

1. There hasn't been a single argument for devaluing that hasn't reduced to "how dare someone who works in a field I don't do well in this competition." This has been explicitly made at times, with people complaining that it was somehow wrong for people working in pictures to be doing better than people working in articles. Given people working in pictures aren't going to touch this toxic, hostile competition again anytime soon, making sure that we never return would be a terrible idea. Adam Cuerden ^(talk) 17:55, 2 November 2014 (UTC)
2. Sets should not be treated any differently than multiple individual nominations. --ThaddeusB (talk) 00:23, 4 November 2014 (UTC)
3. Per Adam. — Crisco 1492 (talk) 00:30, 5 November 2014 (UTC)
4. Per above. It would essentially be saying that the whole is worth less than the sum of the parts which is incorrect (IMO).--Godot13 (talk) 20:51, 5 November 2014 (UTC)

Different kinds of featured pictures (photos, scans, restorations, etc) should be worth different amounts

1. The main "problem" with FPs is one of perception - non-image people do not understand the amount of work that a good image requires. However, there is also a real problem - vastly different amounts of time/effort are required for some types of image work than for others. Therefore, I am in support of a tiered scale. In the original discussion above, I suggested 35pt for photographing and (extensive) restoration work, 10pt for scanning, licensing, and minor cleanup, and 3pts for obtaining the image from a third party and uploading it. Those numbers can easily be adjusted, but I feel a scale is needed. See more extensive comments below in discussion. --ThaddeusB (talk) 00:25, 4 November 2014 (UTC)

   Interesting idea, but it would need further discussion. Personally, I don't think "3pts for obtaining the image from a third party and uploading it." is really necessary, but some would rather have it. — Crisco 1492 (talk) 08:30, 21 November 2014 (UTC)

   This is one of those good ideas that's very hard to implement. Adam Cuerden ^(talk) 08:56, 21 November 2014 (UTC)

Featured picture discussion

I've been keeping an eye on this year's competition and some of the discussions about featured pictures. I personally feel the best way to move forward from here is to eliminate points for featured pictures all together and keep this competition focused on articles. It would be a shame to have to do this, of course. But pictures and articles are just two very different items to work on and it's near impossible to compare the two and come up with a "correct" answer on which is harder. The nomination processes and the preparation for the nominations are very different for FACs/GAC and FPCs. It's great how many featured pictures have been produced over the past few years, likely as a result of this competition, however with only one-two editors utilizing the featured picture points and the remaining editors focusing on articles, I feel the best way to move forward is focus the competition on articles, though I'm expecting a lot of you will disagree with that. Gloss 17:37, 2 November 2014 (UTC)

So, you're basically saying "this produced a lot of great images - but bugger improving the encyclopedia! Let's make this article-only." It's not like people who specialize in pictures are going to jump over to articles. Adam Cuerden ^(talk) 17:52, 2 November 2014 (UTC)

I have nothing against improving the encyclopedia, you can clearly see that you've put those words in my mouth. I'm talking about this competition and this competition only. For the sake of this competition, only 6 editors claimed a featured picture out of 137 competitors this year, and it's been the same every year so far. It's not a widely used category, and it's a whole different ball game from what the other 131 editors are working on to stay alive in this competition. So maybe it's best to keep the two worlds apart and focus on the majority (131 to 6). Gloss 18:12, 2 November 2014 (UTC)

If the number of editors claiming points in a given field becomes a criterion for inclusion in the Wikicup, then we would also need to drop Featured Portals and both Featured and Good Topics.--Godot13 (talk) 18:37, 2 November 2014 (UTC)

It is unfortunate that the initial public discussion regarding the “over-valuation” of FPs was launched by two protest withdrawals from the competition (which seems to have gone missing from the archived talk page material). I understand the desire for competitors to feel there is a level playing field among the various categories in the Wikicup. What I do not understand, evidenced immediately above (and elsewhere on this and other talk pages) by several editors repeatedly, is the need to put down or minimize both the work involved in Featured Pictures, and their importance as content in an encyclopedic reference.

This is not said with any condescension or provocation intended – unless you have spent multiple hours at a time working on a single image at 400% size (give or take) using applications like Photoshop, Lightroom, or equivalent software, respectfully, you do not understand what goes into making the editing and restoration of Featured Pictures, or how maintaining (or creating) uniformity between images make FP sets more difficult than single image nominations.

Should FP and FP Sets be “devalued”? Never. Should they be worth fewer points? Perhaps, depending on the type of work involved (e.g., “merely scanning” versus scanning and restoration). Some sets required new articles to be researched and written (as did some single images). The large sets planned for the final round (coats of arms, and the BEP portraits) took months of restoration, not to mention researching and writing original articles to legitimately use them (in addition to placing over 90% in separate individual articles). The Cuban silver certificates took the better part of two multiple-day trips to the Smithsonian.

During the course of this year alone I made at least five trips (possibly more) from the New York area to Washington DC and spent 4-5 days per trip digitizing, cataloging, and doing condition assessments of currency. I have never done any notable restoration per se on a museum object (despite the often tedious work of alignment, color correction, and image preparation). Why is this relevant? I cringe to think that under some of the proposed scoring revisions offered during the year my work would be classified as “merely scanning”.

Like many of you, I worked my ass off on the Wikicup this year. But I worked too hard on the final round. It was not enjoyable producing a massive number of (unnecessary) points, and it provides little satisfaction winning (still unofficial) by such a large margin. Unfortunately, knowing the opinions of the FP critics throughout the year, it seemed necessary to produce a score that would “hold up” even when judged by the revisions/devaluations almost certain to take place moving forward.

What has given me great pleasure is furthering the advancement of numismatics, in particular U.S. banknotes. I would have done this with or without the Wikicup, but participating certainly drove me to work much harder than I otherwise would have thought possible.

I'm happy to engage in cooperative and constructive (and respectful) dialogue on how to best achieve some kind of harmony moving forward. I hope we can all keep in mind that this is a competition (game) ultimately for the benefit of Wikipedia...--Godot13 (talk) 06:54, 3 November 2014 (UTC)

Thanks for your thoughts. Point taken on "devalued"- I have rephrased. J Milburn (talk) 08:51, 3 November 2014 (UTC)

Adam C put it best in one of the original discussions above: It is easy to underestimate the amount of work required to do all types of image work. Photography might feel like "go to a location and push a button" to a casual photographer, and indeed that may be all that is required for casual photography, but professional quality photography is much more involved. And no, it is not a matter of havign better equipment - skill trumps equipment by a large factor. A high quality image involves carefully picking prospective, angle, and lighting (which may come down to good luck) to get things just right. And it normally involves post-processing to clear up minor, barely perspectable flaws. A professional photographer will normally be happy to get one truely good image out of a session of photographing the same subject for an hour or more. The kind of restoration work Adam does is even more involved, and nothing at all like knowing the right buttons to push in Photoshop. I can easily understand why he has been upset greatly by people claiming pictures don't take much effort (or worse aren't very important).

Obtaining permission to access/relicense a professional photo of someone else's can also be quite involved. And scanning an image, while less work than restoring or creating one, is not trivial. Obtaining a great image on the first try is virtually unheard of, and even a great image will normally require "minor" (perhaps an hour of) postprocessing to get it looking as good as possible.

Even just uploading, categorizing, formatting, etc. an existing free image can be a fair bit of work. I often do this for my articles, and it usually takes a half hour to do it right.

Regardless of the amount fo work required, there is the other issue raised by some people who said or implied image work is not important. Those people are utterly wrong. A well-written article with an amateur (or no) picture looks just as amateurish as a poorly written article. Fighting the perception that Wikipedia is unreliable requires effort on all fronts, and image work is no less valuable. So any point adjustment should be based purely on work required, not importance.

As someone who has extensive article experience and a good amount of image work (mostly outside Wikipedia), I can say that "technical work" (scanning, obtaining re-licensing, minor restoration such as rebalancing the colors, etc.) is usually about as much work as an average DYK. Ten points seems fair, given that a DYK can easily earn lots of bonus points. "Creative work" (detailed restoration, photography, drawings, etc.) takes a good deal more effort. I do not think it is vastly different than the amount of work required for an average GA. Now, some GAs require a lot more work than others and some pictures require a lot more work than others. GA has the multipler to partially compensate for extra work, while FP does not. The current 35 points seems reasonable. Finally, I would like to see a small amount fo points awarded to those who "find" existing high-quality images suitable for Wikipedia and go through the efforts of properly uploading, describing, and categorizing them. In theory this will open FP up to more people since no special skills are required, while giving non-image people an idea of how much work is required in even the easiest FP cases. Perhaps 3 points could be given for such work.

The only issue with a tiered system is that some judgement calls will have to be made, but I trust the vast majority of participants to fairly assess their own work and the judges to weed out an bad apple that tries to exploit the system. --ThaddeusB (talk) 00:23, 4 November 2014 (UTC)

Here's what I think everyone's real beef about FP's. They can score points so quickly, and that from a non-FP standpoint, makes competing against users who do FP's for lack of a better word, frustrating (don't get me wrong, this is nothing personal). Aside from lowering the FP points that I have agreed with aove, I'm afraid there isn't much that can be done to prevent such scoring outbursts and lulls (I don't think it is right to devalue sets). YE ^{Pacific Hurricane} 06:42, 4 November 2014 (UTC)

My 2¢: In the past round, Godot13 produced 181 FP's in 61 days. Assuming an average of 6 hours per day doing nothing but FP's, that's one FP, and 35 points, in 2 hours. That seems like a much higher point accrual rate than possible with GA's, FA's, FL's, FPO's, DYK's, or ITN's. Even though FP's are undoubtedly important, the effort involved is significantly less than with other types of recognized content. - Ypnypn (talk) 21:08, 4 November 2014 (UTC)

That rate would be impossible with the type of restoration work done by Adam C, for example, which is why I'm in favor of a tiered schedule. While equally valuable to the encyclopdia, scanning + the associated minor fixes is certainly less time consuming than photography or full-scale restoration (which do take a similar amount of time/work as a typical no-bonus-point GA). --ThaddeusB (talk) 22:20, 4 November 2014 (UTC)

Ypnypn & ThaddeusB- Correction - I had 181 FPs promoted. Some of those restorations (the majority of the final round submissions) were the product of months of work.--Godot13 (talk) 22:48, 4 November 2014 (UTC)

I wonder if these two editors have ever contributed FPs, in any format, to be honest. I've discussed issues related to investment of time, money, and energy above, so I shouldn't need to repeat it here. In short, it's easily possible for an FP to take up a whole day of non-stop work, and to have the best chance at photographing an FP one often needs at least $1000 (likely more) of equipment and software. Lightroom and Photoshop aren't cheap, and lenses and bodies are even more expensive. Scanning presents its own challenges - including getting access to the material. — Crisco 1492 (talk) 00:24, 5 November 2014 (UTC)

@Crisco 1492: Indeed, I have barely done any significant photography or restoration at all. And I realize that FP's are quite time-consuming, difficult, and expensive to produce. In terms of time, however, "for an FP to take up a whole day of non-stop work" is clearly not always the case, as one user managed to produce over 250 of them in ten months. No one managed to produce other types of content at nearly that rate. Should we reward expertise and monetary expenses in addition to time spent? Maybe. I don't know. -- Ypnypn (talk) 01:58, 5 November 2014 (UTC)

• I never said its always the case, and indeed there are some which are relatively easy. However, even the individual scans can take several hours, particularly if we're using a high DPI, or the notes are dirty or wrinkled, or we need to treat them carefully to avoid ripping them (a lot of the notes Godot scanned were from the 1800s and earlier). — Crisco 1492 (talk) 02:11, 5 November 2014 (UTC)

@Crisco 1492: Perhaps you should actually read what I wrote before making statements like "I wonder if these two editors have ever contributed FPs". If you did, you would see that I have experience in both text & image work and have defended image work as taking a comparable amount of time on average. However, while text has a tiered point system (i.e. bonus points) to partial compensate for harder work, images do not. I am not in favor of decreasing the max points for FP, just for recognizing that while some pictures may indeed "take up a whole day", others do not. It is unfair for images that take an hour (e.g. an average scxan with simple restoration) to earn the same points as images (and articles) that take many hours or days to complete. --ThaddeusB (talk) 15:23, 6 November 2014 (UTC)

• Links? I've been unable to find any featured picture credits on your talk page or its archives. — Crisco 1492 (talk) 07:54, 8 November 2014 (UTC)

Again, try reading what I wrote... "As someone who has extensive article experience and a good amount of image work (mostly outside Wikipedia)". You can see what I've done here by vistingmy userpage on Commons, but most of my image work is unrelated to Wikimedia in any way. Instead of implying I am unqualified to judge image work effort, perhaps you can respond to my actual argument that certain types of image work take longer than other types? --ThaddeusB (talk) 15:29, 10 November 2014 (UTC)

Vital articles

Should the WikiCup take greater account of vital articles? If so, how?

No- The WikiCup should not include vital articles

1. I think the points multiplier strikes the right balance. I should add that I agree the list of vital articles is somewhat arbitrary and is covered well by current bonus points system. Cas Liber (talk · contribs) 22:36, 2 November 2014 (UTC)
2. I stand by what I said last year, vital is flawed and arbitrary and shouldn't be relied on for extra points. The C of E God Save the Queen! (talk) 07:26, 3 November 2014 (UTC)
3. I agree with The C of E, the vital article lists are too inconsistent, with lots of things that are just as famous or basic as what makes it on being left out. If the lists were actually good, it would be good to award extra points, but they're not. –Ugncreative Usergname (talk) 04:13, 4 November 2014 (UTC)

Bonus points

Should bonus points be adjusted "so that points accumulate faster at the low end and less fast at the high end"?

No- linear progression is fine

1. Cas Liber (talk · contribs) 22:36, 2 November 2014 (UTC)
2. Fine as it is with the multi-wiki and rolling 5 year rule. The C of E God Save the Queen! (talk) 23:26, 2 November 2014 (UTC)
3. If anything, articles like Jesus that are on every single wikipedia should get a bonus because it there were more languages then they would have for sure more links. I think from a fairness pow, any article above 150 (or 100) should be worth even more. Nergaal (talk) 13:39, 4 November 2014 (UTC)

Yes- bonus points should be changed

1. I think the multipliers can get out of hand for very well-covered subjects (which aren't necessarily very important topics - cities and species are low hanging fruit that often get made early in a Wiki's history, sometime by a bot, regardless of actual importance). Additionally, while some articles are certainly harder to write because they are more well covered and/or broad in nature, there is a limit to this. A broad subject (say "birds") might be 5x harder to write than a narrow one ("Yellow-throated sparrow"), but is not much easier to write than a very broad one ("animals"). Thus, I would prefer less accumulation of additional points at the upper end. --ThaddeusB (talk) 00:44, 4 November 2014 (UTC)
2. I agree, the current linear scale needs to be changed so that an article in 100 languages isn't 5x as valuable as one in 20 languages. Thaddeus makes a very good point about common topics that are going to be a lot of different wikis that don't necessarily have any more value and topics that aren't.--Sturmvogel 66 (talk) 15:36, 17 November 2014 (UTC)

Bonus points discussion

I'm not sure of the merits of receiving bonus points in DYK for expanding older stubs. Cwmhiraeth (talk) 19:36, 2 November 2014 (UTC)

I'm a strong advocate of anything encourages the expansion (rather than the creation) of articles, especially ones that have been in poor shape for so long. I believe that the bonus points helped foster this expansion (I know it encouraged me to pursue older stubs). I got strong satisfaction out of seeing people earn the extra points, knowing that they helped improve something older. That said, I haven't heard a good reason to scrap the bonus points but am willing to listen if someone feels strongly. Regards, Ruby 2010/2013 01:40, 4 November 2014 (UTC)

I agree with Ruby2010 - so much content is left to stagnate on wikipedia that any incentive to improve old content I feel is very worthwhile...Cas Liber (talk · contribs) 00:51, 5 November 2014 (UTC)

I would also like to raise a discussion on something that was raised last year but got buried under the text. Nergaal made a good suggestion for DYK that for every year that the article existed 2009 and before, an extra bonus point could be awarded on top of the 5 for the 5 year rule. The C of E God Save the Queen! (talk) 13:53, 4 November 2014 (UTC)

I would be in favor of this. --ThaddeusB (talk) 15:02, 4 November 2014 (UTC)

I would also support this. So if an article was created in, say, 2006, it would earn 8 points (5 + 1+ 1+ 1)? Ruby 2010/2013 16:33, 4 November 2014 (UTC)

That is how I envision it would work, Ruby. The C of E God Save the Queen! (talk) 13:16, 5 November 2014 (UTC)

me too. Cas Liber (talk · contribs) 00:56, 5 November 2014 (UTC)

Just to formalize, I endorse this proposal. The C of E God Save the Queen! (talk) 13:16, 5 November 2014 (UTC)

I'm not going to bother opening a poll for this- it seems uncontroversial and sensible. Unless anyone raises an objection, we can make this change. J Milburn (talk) 22:40, 6 November 2014 (UTC)

Rounds discussion

Should rounds be adjusted so that more people stay in the competition for longer? Should some points carry over from round-to-round? Should we have a "draft" system to ensure fairer pools?

Rounds should keep people in the competition for longer

1. Somehow, but not sure how. Maybe fewer and longer rounds? Cas Liber (talk · contribs) 22:36, 2 November 2014 (UTC)
2. Add more users to the rounds. YE ^{Pacific Hurricane} 16:10, 3 November 2014 (UTC)

Points should not carry from round-to-round

1. The rounds should be completely separate. I guess this is to stop people from delaying entries or to make an epic performance in a previous round not worthless, but if we really want to do that then we should abandon elimination entirely and have one long round, and that would be really boring. Non-entirely random groups will make more points than most other advancees in a particular round worth something, and keep the rounds from weirdly flowing into each other. Also, early contributions would be worth more in that they would help you through rounds they weren't in—people nearly idling for a couple months then getting through on a previous round's points would be really annoying. (Having just some points carry over might fix that, but not the other problems. Also few would be happy with the exact fraction, and it's just not as simple as discrete rounds.) –Ugncreative Usergname (talk) 20:14, 2 November 2014 (UTC)
2. No, I think the current system is fine. Cas Liber (talk · contribs) 22:36, 2 November 2014 (UTC)
3. Agree with the above. The C of E God Save the Queen! (talk) 07:28, 3 November 2014 (UTC)

We should have a draft for fairer pools

1. Support as nom, in a weird way. I'll put more detailed thoughts in the discussion section. –Ugncreative Usergname (talk) 18:02, 2 November 2014 (UTC)

Pools discussion

• Right, this proposal also involves keeping slightly more people in the competition for longer, which is why it doesn't strictly belong in one category, but it's mostly incidental to make the round progression more symmetrical. You'll see.

Round 1 would be one group. The top 81 advance to 9 groups of 9.

The best 9 performers each go to into a different group, then 10th-18th, etc.. It could be by random selection, which is simpler, or a "zigzag" where 1st-9th go into groups 1-9, 10th-18th go in 9-1, and so on.

Best two in each group and nine wildcards advance to 3 groups of 9. Seeding is the same.

Then best two in each group and three wildcards go into the final round. –Ugncreative Usergname (talk) 18:15, 2 November 2014 (UTC)

There's nothing stopping you from voting in three different sections, here- as you say, there are three different questions. 19:01, 2 November 2014 (UTC)

• I am in favor of the idea proposed in the original discussion of making round 1 "qualifying" with a small number of points required to get through. Thus, all have a chance to advance and a clear idea of what they need to do. From there, pool as normal. It's not a big deal if pools are slightly uneven for 1 round - iot is very rare for someone to get through a group who wouldn't have been a wild card on points anyway. --ThaddeusB (talk) 22:18, 10 November 2014 (UTC)

Recent deaths

Should recent deaths be eligible for ITN points?

Yes- recent deaths should count

1. There's no reason not to treat recent deaths like other ITN posts. –Ugncreative Usergname (talk) 23:10, 6 November 2014 (UTC)
2. Per my reasoning in closed discussion below. Preference is for same points, with usual "significant work" rule blocking simple RDs from receivign any points. --ThaddeusB (talk) 15:14, 7 November 2014 (UTC)
3. Yes, but for trial give them 1/2 of the ITN points then increase it the cup after. Nergaal (talk) 01:09, 10 November 2014 (UTC)

Featured articles

Should featured articles be worth more points? If so, how many?

No- featured articles should stay at 100 points

1. While I do appreciate FAs, the reasons for the point increase being given aren't about FAs, they're about FPs. Given an FA will likely also get GA points in the Wikicup, a slight increase might be justified, but anything much larger is rather questionable given the reasoning for the votes. If you want this to be about featured articles talk about featured articles. But if people are only talking about FPs, it makes this look like a way to work around losing the vote up above. Does everything have to be people attacking FPs? Adam Cuerden ^(talk) 06:12, 8 November 2014 (UTC)

Yes- featured articles should be worth more points

1. Normally I would have said no but after seeing how easy it can be that FAs can get out-pointed then I think to redress the balance, the points should be increased. The C of E God Save the Queen! (talk) 22:45, 6 November 2014 (UTC)
2. Absolutely - The debate can be made on how much time it takes to do a featured article versus a featured picture, and that's not my debate to pick up, but it's clearly much quicker to go through the FPC process than it is the FAC process. On time constraints, I think FA's almost have to be worth more. Red Phoenix ^{let's talk...} 01:28, 8 November 2014 (UTC)
3. Agree with Red -- if nothing changes on how FPs are scored, then FAs need to have their value increased. It's simply just much much harder to get an article featured than it is a picture, and thus it should be recognized as such. Anyone comparing the admirable work done by both Godot and Cwm in the last round can clearly see the point imbalance. Ruby 2010/2013 04:45, 8 November 2014 (UTC)
4. Absolutely. Tons of work goes into these, and the nominations take forever. Up it to 150. Gloss 05:14, 8 November 2014 (UTC)
5. 150 works. I would not mind going higher. YE ^{Pacific Hurricane} 15:52, 8 November 2014 (UTC)
6. If the future winner will have only FA points seems fine by me. Until now people have won the cup with predominantly working only one one of the following categories: FP, GA or DYK; somehow FAs would seem a more legit pathway to winning. Nergaal (talk) 01:11, 10 November 2014 (UTC)
7. My initial reaction was "stay at 100" - its a nice round number and traditionally all other areas have been adjusted to balance things while FA stated the same. Additionally, I do not think FA is more than 3X harder than GA on average (which is the relevant comparison, not FP). However, I looked back and it looks like GAs outscored FAs on the last round every single year. That strongly implies GA is a more efficient way to earn points (likely due solely to FA;s one at a time rule). Since I think the other areas are already fairly balanced, upping FA is the most logical choice. Due to the extreme bonus multipliers possible, I can't support a very large increase though. I say up it to 125. If bonus multiplers are reformed to increaser slower on the upper end (seems unlikley at this point), I could go higher, but under the current system I am afraid of a single article being too large a % of the total points needed to win. --ThaddeusB (talk) 15:43, 10 November 2014 (UTC)
8. Yes, the bottleneck in reviewers needs to be compensated for. I don't think that anyone's done more than 4 in a single round, ever, so things should be scaled that that number should be competitive in the final round. Getting blown out by easily reviewed DYKs or FPs just demonstrates that the Cup is unbalanced.--Sturmvogel 66 (talk) 15:55, 17 November 2014 (UTC)

Featured articles discussion

• "It's simply much much harder"... I welcome people to actually create featured content in both areas before making such statements. Otherwise there's nothing to back up the claim. — Crisco 1492 (talk) 07:48, 8 November 2014 (UTC)

Agreed. Particularly for small-scale topics - the kind that would get merely the base points - if I start an article from scratch, and sufficient resources exist, I can probably get it featured. Only reason Urania's Mirror hasn't entered the process yet is that I wanted to get one more source.

Further, one FP is worth about 1/3 of an FA; much, much less if any bonuses apply, and only the work between GA and FA is relevant, as points are given for GAs as well. Adam Cuerden ^(talk) 12:17, 8 November 2014 (UTC)

I guess the other thing to consider is I'd expect that an FA that is only on en-wiki would be pretty esoteric, and that many would be on some other languages too, resulting in bonus points. do we have an idea what the average points awarded for FAs was? Cas Liber (talk · contribs) 12:25, 8 November 2014 (UTC)

For instance, Banksia lemanniana, which is a sitter for a quick FA buff for me at some point (and pretty esoteric), is on 7 other wikis, so would be 120 pts. Australian raven (currently at FAC) is on 17 wikis, so is 160 pts...something else I'll buff soonish....Telescopium is a small constellation that is on 58 other wikipedias...so 320 points (probably similar to other constellations)...elements might be similar..... I'll find some others....Cas Liber (talk · contribs) 12:32, 8 November 2014 (UTC)

I have some experience in both as a contributor to multiple featured articles, and though I don't have any featured pictures because it's not my particular area of interest and because the content I choose to work on mostly deals with copyrighted works, I have done quite a bit of Photoshop work and image creation before. It's not a new concept to me at all, but it's not something I choose to work with on Wikipedia. I know what it's like to take hours of painstaking work to a picture restoration to get it just perfect. That being said, several points I'd like to make here.

• Having experience with both, featured articles can certainly take longer than featured pictures. Conversely, some pictures can take longer than articles. That's normal; varying works will take varying degrees of effort.
• GA work can take quite a while as well due to the backlog at WP:GAN. but as GAs are worth less than FP's but work on the same idea that you can nominate as many as you like at once, I don't think there's an issue with the point balance between GAs and FP's.
• The bump to FA from GA is quite significant in my personal experience, and is not so immediate that you can immediately milk every GA into an FA without significant work.
• The main problem with FA's is that you can only usually collect one, sometimes two, in a two-month round span. Effectively, that's a point limiter on the highest quality of article work.
• We have effectively been demonstrated during the WikiCup this year that FPC has no point limiter because you can run through as many as you like.

So here's the main question: Why is it fair to place a limit on the highest quality article contribution such that being a picture contributor is so much more prolific? If this contest is designed to drum up Wikipedia activity, why can't we have a points system where being a fantastic picture contributor and a fantastic article writer are in balance and can be fairly weighted as such? I quit trying this summer because I saw the ridiculous imbalance and thought there was no way I could catch up. Giving more points to featured articles is the first step in correcting this imbalance, as it is by far the most time-consuming in terms of how much time it takes to accumulate points in that fashion, which is what is supposed to lead to the highest quality articles. Red Phoenix ^{let's talk...} 02:56, 9 November 2014 (UTC)

@Red Phoenix, The C of E, Ruby2010, and Sturmvogel 66:If we can shut up about pictures, I'm actually inclined to support, but is it really necessary to insult other types of content in the process? Seriously! I agree FAs, due to their slowness and amount of work, could reasonably be raised to 125 or so. I've certainly had several articles stall at GA over just needing one more source, but I really wish that at least one thread here didn't have to get turned into being about devaluing FPs. Adam Cuerden ^(talk) 01:52, 21 November 2014 (UTC)

What I'd like to see is that a maximal effort devoted to FAs (like the 4 that Cas got this year) has a decent shot at winning, excluding bonuses, so the maximums achieved in each category should roughly balance. Those are 4 FAs, 181 FPs, 121 DYKs and my 83 GAs back in '10. The FAs are obviously an outlier in comparison to the others and reflects the slow review process. If a FA-centric strategy is to have a chance of winning, then the points for them need a massive boost, far beyond the paltry 125 points suggested earlier. This could be an increase in base value, additional bonus points for length (CasLiber's four entries in the final round averaged 13K apiece, while those of Cwmhiraeth and myself ranged from 13 to 49K), a combination of both, or some other possibility. I'm not really interested in discussion about the amount of work thought to be required for each type, but only in how to balance them. And the potential for DYK→GA→FA points for the same article all in the same round is a possibility, although the last stage will be exceedingly hard to do. Thoughts, comments?--Sturmvogel 66 (talk) 17:52, 23 November 2014 (UTC)

Anything else?

Are there any other changes you would like to see for next year's competition? J Milburn (talk) 17:10, 2 November 2014 (UTC)

Delete text from WP:WikiCup/Scoring

The constant attacks on FPs and the management's failure to do a thing about them shows this should be deleted:

“ The most important rule is that the WikiCup is just a bit of fun — at the end of the day, we're all here to improve Wikipedia. The second most important rule is to treat Wikipedia and other users with respect. If through the WikiCup any participants are hurting the encyclopedia (whether through abusing the rules/systems, creating a negative atmosphere, or whatever else), they will be removed from the Cup. To quote Durova, winner of the 2009 Cup, "Wikipedia is the real winner". Let's hope we can keep it that way. ”

That's a lie. Noone will be removed, they'll be allowed to drive people out, then vote to declare their contributions worthless. Adam Cuerden ^(talk) 18:46, 3 November 2014 (UTC)

Of the several people who have been in favor of having a status quo for FPs, I have failed to see anyone that has moved their own opinion anywhere closer to the middle (as in agreeing at least partially to change the status quo). If people are of the way "my way or I am out" then I don't feel inclined to empathize with them. Nergaal (talk) 13:44, 4 November 2014 (UTC)

Review speed

My only major concern about the Cup scoring process is that the winners of the last few years have been able to exploit processes that review submissions far more quickly than any of the article review processes. The last couple of years it's been DYKs with their QPQ policy that enormously speeds reviews and this year FP. In contrast it can take a month or more to get a GA or FAC reviewed. I was impressed that one finalist managed to get four FACs reviewed this round as I only managed a pair myself and that is likely the record over the entire history of the Cup. I'd never make the argument that FPs are easier to do than articles, but it's undeniably easier to get them reviewed.

So how do we restore some sort of parity? If I have everything all to hand and don't have any distractions, I can write a 6 to 8K GA-quality article from scratch in 3-4 hours or less, depending. For subjects with more available information or detail, that roughly doubles for a 20K article, etc. My 50K article on HMS Illustrious (87) took me about a solid week, writing 3-4 hours each day. So I could have written something like 7-8 GAs in that time and gotten more points, if I'd been able to get them reviewed in a timely manner. It happens on rare occasions, but it does happen, so you can't discount the possibility.

In general I think that people have been more concerned about the amount of work involved when assigning values for the Cup than about the speeds through the review process. We need to address the issue directly as we have no way of speeding up reviews unless we want to massively increase the points awarded for reviews (which I'd discourage, although a modest increase would probably be beneficial). Awarding bonus points for longer DYKs was a good start, although the bar may be set too low at 4500 words, and I think we might want to consider doing the same for GA and FA articles. I'd suggest that the minimum bonus for those be set at 10K as my average GA-quality article is about two-thirds of that size. I'd also like to see the bonus scale for longer articles so that a 50K article has some sort of multiplier so that very long article are rewarded commensurately. We can discuss how to the scale the multiplier and how the multiplier ties into the existing bonus system as I think both should apply, although I think that the current linear bonus scale might need to be reworked so that a very long article like User:Cwmhiraeth's 86K Sea, that's on a hundred or so other wikis, doesn't win the Cup single handed.

The more competitive people among us, myself included, have long exploited the points available through the review processes that work quickly and I think that it's past time to rebalance the Cup's points to account for the work involved and how fast it gets reviewed. Having won once I'm not as hung up about winning as I once was, but I'd like to have a real chance to win without being overwhelmed by a flood of bonus-point heavy DYKs or speedily reviewed FPs.--Sturmvogel 66 (talk) 20:42, 3 November 2014 (UTC)

This all sounds very sensible (to someone who is essentially a WikiCupOutsider) but I might comment that the amount of work on Sea was very considerable. Not only was the article long, but being a major topic, it more or less inevitably attracted a substantial "audience" of editors, both during the construction phase and during FAC, who expressed enough opinions to consume many hours of our time. Getting an article of this type through to FA is therefore both slow and risky. To a non-competitor, the chances of success often seem slim; to a competitor, I imagine the delays and amount of input required must be gut-wrenching. As a rough rule of thumb, I'd cheerfully predict that the amount of hassle would be exactly proportional to the length of the list of other wikis ... and hence to the resulting score: i.e. the existing scoring system is exactly fair, because risk, effort and delay (all critical in a competition) are well and truly factored in. Just my tuppence-worth. By the way I think the Cup admirable in its aims and methods. All the best, Chiswick Chap (talk) 21:23, 3 November 2014 (UTC)

My 2 cents. In the past, the DYK flood was dealt in two ways: decrease the value of DYKs, and give bonuses for working on the higher traffic articles. The people who wanted to work had their attention grabbed from DYKs. Currently, FPs are not a problem, but FP sets. I truly believe that most people attending the cup are happy to see FPs, so FP contributors are actually welcome here (although that might not be visible here). I believe these contributors should be attracted to to cup and encouraged to participate in non flood-nominations (in other words tweak the discrepancy between the rewards of regular FPs and those for FP sets). Nergaal (talk) 13:51, 4 November 2014 (UTC)

Nergaal- You've been putting your two-cents in for months now and have been a very outspoken critic of FP set scoring while revealing you have no clue about the work involved. You do realize this was a competition? "Flooding" has fairly negative connotations, no? You are free to have your opinions, you are free to express them wherever you want, but I do think you may take some enjoyment from being disruptive...--Godot13 (talk) 08:24, 5 November 2014 (UTC)

@Godot13: I have been around the cup for around 5 years now, and I've seen these sort of issues happen almost every year. A much bigger debate happened a few years ago when two of the last 8 editors in the final withdrew from the negative attitude raised around similar issues, and I think even the person who would have ended up winning for sure withdrew. What I am trying to point out repeatedly, but people seem to mss that, is that the attention of contributors would best be directed away from such controversial topics. Wether you want to agree or not, if you truly think that sets of n pictures takes n amount the time necessary for a single picture, then you can see how unwelcoming other people are who are trying to encourage you to work on n different FPC than a single FPC set of n picture. From your point of view, the two options require the same amount of time, and if you want to disagree with other preferring you to work on one of the two options, then you are either stubborn, or are not willing make public your true opinions. Nergaal (talk) 18:37, 5 November 2014 (UTC)

see Wikipedia:WikiCup/History/2010. Nergaal (talk) 18:40, 5 November 2014 (UTC)

Nergaal, unless I'm very much mistaken, one person withdrew during that upset, not two. There was also a clear difference, because, in that case, there was concern about the quality of the user's contributions. In this case, no one seems to be in doubt that Godot is doing good work and that his images should have been promoted to featured status, the only disagreement revolves around how many points the images should be worth. Further, you are being unduly confrontational. Accusing people of refusing to "make public [their] true opinions" seems intended to put them in a bad light, and certainly looks like a failure to assume good faith. Finally, to be quite frank, I have no idea what it is that you're claiming, other than that it's something to do with FP sets. Perhaps you could rephrase your concern in clearer language, and try to do it without any mudslinging? J Milburn (talk) 20:15, 6 November 2014 (UTC)

I genuinely think that whoever states that a set made of 30 pictures takes 30x the amount of time necessary for a random FP is genuinely disingenuous. I anticipated the FP sets problem very early in the competition, long before people decided to pull out and start the whole controversy. Like then, nobody seems to rally behind my personal opinion now, so I won't bother to prolong the discussion and let the others have the fun of getting out of this mess themselves. Nergaal (talk) 09:58, 8 November 2014 (UTC)

How? Seriously, How? Restoring one image does not magically restore the second image. There may be savings in some sets in overhead - e.g. you don't need to travel to multiple locations - and documentation may be able to be combined, but the time spent on the first image and the time spent on the last image when restoring is usually quite similar. I've asked you several times to state where you think the savings in time are and you have, rather disingenuously, not answered to my knowledge. Adam Cuerden ^(talk) 00:16, 10 November 2014 (UTC)

@Adam Cuerden: Anything that is part of a set tends to have a similar set of issues that needs to be fixed, be it tweak lighting, erase some watermark, remove creases in a similar color palette range, etc). Similarly a few years ago, a user was submitting a flood of extremely similar articles such as al the basketball players at a specific (not really top level) team throughout the years: they looked very similar and probably used very similar sources. The solution for that was to disincetivize similar articles by lowerng their points, and increasing the relative points for other type of content-writing (that is why there is a somewhat evolved scoring system for DYK). Nergaal (talk) 01:01, 10 November 2014 (UTC)

First of all, basic colour adjustments are the quickest part of restoration - 5 to 15 minutes, perhaps. Sets can have similar adjustments on that account, but it's a small part. These are very rough figures, but, for an average restoration:

A huge proportion of the time is spent on grunt work - removing spots of dust and the like. And, no, this can't be automated: there's a certain amount of skill and knowledge of the medium required to avoid removing things that should be in the image. This may sound easy, and, in a way, it is, but it's a grind-down thing. There can be literally thousands of specks removed in a perfectly average restoration. And, of course, care needs to be taken that the fix blends in with surrounding elements. That's not to say there aren't going to be challenges, but even when you've worked out what to do, that doesn't mean that applying it won't be a bit gruntworky. For example, File:William_Simpson_-_George_Zobel_-_England_and_America._The_visit_of_her_majesty_Queen_Victoria_to_the_Arctic_ship_Resolute_-_December_16th,_1856_-_Original_scan.tif has two orangish vertical stains going from top to bottom of the image. As the two stains differ in saturation, fixing one didn't really help much with the other, and since they're analogue gradients, removing them is a long, multi-stage process, with a bit of reconstruction along the edges, and especially in the sky area where any difference in colour would be very visible I was reconstructing a LOT. I succeeded in fixing that, but fixing two took pretty near twice as long as fixing one, although, of course, each being a challenge in their own right - they go through different parts of the image, and one is a bit less visible due to passing through a lot of dark colours, so perhaps saying "took as long as fixing each individually de novo would have" would be more accurate. Adam Cuerden ^(talk) 03:57, 10 November 2014 (UTC)

Secondly, fixing, say, a rip on one won't affect another. The skill in fixing such things is making the content work across the divide, unless the rips are in exactly the same place, with the exact same content going across it, the fixes will be just as time-consuming and difficult.

Thirdly, Creases are hell. If they distort the image at all, that's a very, very nasty fix - c.f. File:Joseph_Ferdinand_Keppler_-_The_Pirate_Publisher_-_Puck_Magazine_-_Original_LoC_scan.png for how bad they can get. even a simple crease will likely have some level of paper distortion, and fixing that is actually a highly-skilled job.

I hate to say this again, Nergaal, but you really need to actually do a restoration if you want to have any idea what's involved in them. Your view of what it involves

@Adam Cuerden: I never doubted that plenty of FPs take time, but my issue is with sets ONLY. If you noticed my vote, I am against decreasing points for FPs in general, I am only pro capping the number awarded in sets. Do you have some specific sets in mind that show that an x picture set takes roughly x times amount of time than for a single picture? The examples I listed below (this, this, this, and this) don't IMHO take a linear amount of time than for a single pic. Do you agree that these sets include a large portion of automation that decreases the amount of total time to work on them, rather than it would have been necessary to do for a similar number of pics spread over a 1-pic nominations? Nergaal (talk) 10:56, 10 November 2014 (UTC)

I don't know Godot's prep methods, but I'd imagine that, short of time saved finding where to go to get the rare and fragile media, it's going to take about the same amount of time. There's no way to automate restorations. Certainly I know that State Arms of the Union had some tricky restorations. Due to some confusion, I did a couple unnecessarily.

You'd be better off asking Godot, since those aren't my sets. If you'd like to ask about my sets, I'd be happy to give you all details I remember. Adam Cuerden ^(talk) 22:29, 11 November 2014 (UTC)

Actually, this does come across as a legitimate concern to me. This is not intended as a blast on featured pictures, but the rule structure does make article-writers undervalued. I had several featured articles this year before I was knocked out mid-summer, but it took nearly six weeks for me to get each one through, and at WP:FAC you're not allowed more than one article nomination at a time—I had some ready for nomination while I was waiting on others to pass. Contrast this with Godot13's nearly 6400 FP points in the last two months alone... that's the equivalent of passing 64 featured articles in two months! I think what has to be the concern here isn't whether or not anyone's contribution is more valuable than another, but how efficient is it to score in the systems we work in? I will note, aside, though, that WP:GAN doesn't really have this problem given you're allowed any number of nominations, but what's the incentive to write the highest quality of articles, then? You couldn't be an article writer and be competitive just because the systems at WP:FAC and WP:FPC differ so much on how quickly it is to keep up. Maybe the solution, then, is to increase the scores a featured article will give because it takes weeks just to have a review completed. Featured pictures would not need to decrease if FA's were increased by, say, a factor of 5 or so. Red Phoenix ^{let's talk...} 23:25, 5 November 2014 (UTC)

One thing you haven't factored in is that FAs can gather bonus points. On several occasions, we've seen people gain more than 500 points per article; if the base points were up to 500, they'd be scoring 2,500 points per article. Maybe this is what you'd like to see, maybe not- just something to bear in mind. J Milburn (talk) 20:07, 6 November 2014 (UTC)

I did factor that in, and if you're getting that many bonus points, you've clearly had your work cut out for you on the article you wrote. What's really wrong here is that the point balance is entirely wrong; we're finding with the FA-FP debate this year that the two systems of WP:FAC and WP:FPC have very different limitations, and it's much harder to score as an article writer of featured content than if you're a picture producer/restorer with a lot of base content. Is that still not a bit silly that even saying with the max amount of bonus points (which is in itself ridiculously difficult because of their ties to broad subjects and "vital articles"), you still in two months couldn't match what an editor could do with featured pictures alone? Credit to Godot13 for his hard work on all of those pictures, but if he were trying to score that many points through articles in two months, he would find it downright impossible because the systems in place limit what can be achieved in article writing. Red Phoenix ^{let's talk...} 21:49, 6 November 2014 (UTC)

I have opened a poll above. J Milburn (talk) 22:40, 6 November 2014 (UTC)

There is another aspect to the FP/FA discussion. With a featured article there is much preparatory work as the article is brought up to standard, and then ongoing work in responding to reviewers as the review proceeds. With a featured picture, there is much preparatory work as the image is brought up to standard, but after it is nominated, the nominator mostly sits back hopefully and waits for sufficient support to roll in. The timing is also easier with the featured pictures, ten days concludes the process. With featured articles, timing is less precise, takes much longer and cannot be predicted with precision (a FAC I nominated on September 11th is not yet concluded). Cwmhiraeth (talk) 11:43, 8 November 2014 (UTC)

And you need three FPs for every base-point FA, and several dozen for anything with a lot of bonus points. Adam Cuerden ^(talk) 12:43, 8 November 2014 (UTC)

If I am not mistaken, the highest single-entry in the competition was Sea with 7.2x multiplier, so worth maybe 130*7.2, or almost 950 points. Using the current point distribution, that comes to a set made of 27 FPs. I am genuinely curious, what 27-picture set you think took the same amount of work as taking the sea article though both GAN and FAC (or 50-picture set for double the time). At a quick glance, this, this, this, and this are examples of sets that got about as many or much more points than sea did altogether. Nergaal (talk) 01:21, 10 November 2014 (UTC)

It's impossible for me to talk about someone else's work. Wikipedia:Featured picture candidates/Set: Puck of Pook's Hill took almost all of a month of working in all my free time, to the point that I pretty much burned out of Wikipedia afterwards. The constellations split over several sets were easier restorations, but there was certainly nothing about them that made later ones easier than the first. Indeed, some of the last ones, like Canis Major, were actually the most difficult, as the level of damage across the set varied a lot. Adam Cuerden ^(talk) 15:43, 10 November 2014 (UTC)

Ok, those are fair arguments. 1) Do you think that sea took about the same amount of work to make it a FA? 2) Do you think that restorations are a category of FPs which take an appreciable larger amount of work to do than any other categories of FPs? Nergaal (talk) 20:00, 12 November 2014 (UTC)

It's hard to judge with FAs. In some ways, if you can start off without too much content you need to salvage, it can be a lot easier. For example, Creatures of Impulse, or my forthcoming FAC, Urania's Mirror. But I didn't work on Sea. So, let's use W. S. Gilbert, an FA I did some time ago. 23 Wikipedias, so, theoretically around 560 points, I think? Hm. I was really passionate about the subject, so it didn't really feel like work. But I did read four or five books for it, plus bits elsewhere... but I read quite fast... Hmm. Given that I didn't burn out after Gilbert, but did after Puck, I'd say compareable.

I'm not sure if FPs are gameable. Things that I think will be easy have often turned out much more difficult; things I thought would be hard have unexpectedly proven to be near-pristine copies. I suspect the time sink of trying to find only easy restorations would eat up the time they saved. Hell, the simplest FP I did all this year, File:Red Skelton 1960.jpg, was one I only did because the article reached FA.

Let's see, for other types of things - Photography probably isn't gameable. Getting sufficiently good lighting and subject material is going to be tricky. We don't give points for things without effort, so one couldn't just grab paintings.

I suppose, if I'm going to boil it down: If we ever get a competitor who has really, really easy access to some grand archive of near-pristine material - say, they work in an art gallery - yeah, they could game the system. But for people who actually have to research and find their own things, then I think it'd be hard. But I don't know.

If I really had to speculate as to what kind of content might be easiest to gain points in? I'd probably go Featured Lists, if the review system was fixed a bit. Not that you can't do excellent work in them - and people often do - but because there's classes of lists that have readily-available pre-done research (say, sports statistics), and thus mainly just need the opening paragraphs. Plus, unlike FPs, they're eligible for bonus points. If one could find lists with bonus point spreads like Sea in such categories, you could game the Wikicup easily.

On the other hand, I could mine article points. I brought Creatures of Impulse to FA level. It wasn't that hard. If I so chose, using the same sources, I could write articles on the other hundred or so plays by W. S. Gilbert. Outside of a few particularly notable ones, I wouldn't need to change sources much. I probably would be slowed at FA, but, especially if I reused background paragraphs (dodgy, but probably acceptable) I could probably get a bunch of GAs.

But we should be presuming that the goals of competitors is to improve the encyclopedia, not to win. Everything I've just said has been about ways to game the system. If people are acting in good faith, that won't happen. Adam Cuerden ^(talk) 00:45, 13 November 2014 (UTC)

In theory, the FL gaming technique you say could work. In practice, most lists are not eligible for bonus points since nobody outside English-speaking countries care about low-hanging lists in whatever sports you want. In addition, my last 4 FLCs I've had did not pass simply because of no reviewers (I guess with a buddy system you can skip this hurdle as one could with GANs). Now coming back to sea, the point I was trying to ask is about really broad-audience articles such as sea. I believe that regardless of the level the article starts at, there is a significant burden one has to pass in order to make sure the article covers well all the sides relevant to that topic. For example, take iron or gold: they are both decent articles, but I am not looking forward to working on either of them because I know for sure that they need a very, very large amount of work just to make sure they have a featured-level structure and coverage. Nergaal (talk) 00:59, 13 November 2014 (UTC)

I guess this is a topic where each of us has a hard time truly understanding the work necessary in the two discussed cases: I presume you don't have much experience on working on really high impact articles (in my "young" career here I had the energy to work on planet and oxygen, but since then I preferred aiming for 10 low-hanging FAs instead of another one of these), while I have no actual experience on getting restorations, scans, or even acquiring FP-level photographs. Unless we both accept to try to actually attempt to work on these two areas where we don't have experience. Nergaal (talk) 01:06, 13 November 2014 (UTC)

There is a reason I try to avoid guessing work done by others (and why I refocused on one I had actually done.) I'd presume that working the really generalist ones are basically a specialized skill in themselves, but, as they're so valuable, we should probably heavily reward them.

I don't think it's ever possible to judge something like this completely fairly. I'm going to unindent for this.

[Unindent] So, another example based on my work. Here's a quick table:

File	Number of Wikipedias	Number of articles on en-wiki	Number of wikis featured in	Difficulty?
	13	4	2 (En and commons)	Relatively straightforwards. A few hours without too many tricky bits
	1	4	2 (En and Commons)	Very difficult: Many days' work with many difficult challenges, several unique
	1	2	1 (En)	Difficult: Some challenging sections, but nothing too different to things I've seen before.
	3	3	3 (En, Commons, Tr)	Medium difficulty. Lots of detail, and I think I had to start over once when I started editing out a chandelier wire. Oops.
1000x200px	1 (can't be on others)	2	1 (En)	Fairly difficult - some very odd damage.
	2	3	1 (en)	Moderately easy: More time consuming than hard.

Can you think of any good way to weight these so that the most deserving - and what does that even mean? Most Wikipedias? Most featured? Most work? - get more points, that isn't arbitrary, and which doesn't require me to rate myself? Note: For images that replace an old, crappy copy of the same image, distributing to many wikis is easy. For completely new images, it could take a year or more to see how far they spread. The Puck of Pook's Hill images appear to be used on four Wikipedias, and featured on three, but almost certainly weren't at the time of last Wikicup. (That said: I'm really glad they're used.) Adam Cuerden ^(talk) 02:32, 13 November 2014 (UTC)

All of these seem to be really nice work. The only thing I can say they have in commons is that they are restorations, which from my grasp seem to generally take more time on average than the average FP. In my opinion, none are really the main focus/problem of all the discussions lately. I think only the first is part of a set. I think the most contentious part until now hasn't been giving out points, but how many points can be gained in the upper limit from FPs. My impressions is that all the negative comments have been about the massive amount of points potentially gained. For example, I don't remember many complains for the amount of points gained by sea, while some FP sets did. I think if we focus on discussing how many points some of the high-scoring sets did get, and how many points they are truly worth in terms of amount of work, people will be much more calm an lenient with awarding whatever amount of points FPs might be worth or not. (1) Do you think there is a major difference between sets of restoration FPs, and sets of other FPs (say scans)? (2) Do you think it would be unfair to you to have a rule that discourages you from working on FP sets, but say, you would same, or even more points on single-pic FPCs? Specifically, if there was a rule that kept everything the same, but halved the points for FPs, do you think you would really have a genuine problem on being dis-incentivized on working on sets? The reason why I am asking (2) is because I do genuinely think people outside FP "feel" the current system is somehow unfair, even if they don't know why, and if this potential is partially muted, some of the same people might think that at least some categories of FPs are actually worth more points. As a FP contributor yourself, do you really think that the current system where one cant potentially scan 300 pages of the same book in decent condition, do some minimal work, and then claim all the points necessary to win the competition for 3 years straight? Nergaal (talk) 10:05, 13 November 2014 (UTC)

If that was really all that was necessary, I'd have finished that Gustave Doré Dante's Inferno set years ago, instead of only the first few. Adam Cuerden ^(talk) 16:29, 14 November 2014 (UTC)

---

The most important thing is that sets are far more valuable to the encyclopædia. One of the constellations had been featured years ago by User:Durova, and I basically had to replace that, and a lot of tiny little thumbnails (about 200px by 300px or so) - but all of these images were very, very widely in use over dozens of Wikipedias. By doing that set, I improved literally hundreds of articles in many, many languages. Had I done one, then flittered off, as your proposed changes to scoring would have actively encouraged, this big important project would not have happened. Rules should never be made to encourage worst practice. Adam Cuerden ^(talk) 15:57, 10 November 2014 (UTC)

Nergaal I’m really making an effort to stay out of discussions here. Any effort to educate you as to the workings of FP sets (seeing as you have absolutely no basis of experience for any of your claims, and lack the necessary insight to understand this), have been futile.

Now… your comment about being “genuinely disingenuous”. What you describe as the criterion for this is generally speaking (or on average), what I have been trying to say. If you know of a magic application where I only need to restore a single image in a set and then it automatically applies the magic restoration everywhere else, please let me know. Now, am I to understand that you are accusing me of being “genuinely disingenuous” (oxymoron), which I would roughly interpret as deceptive and/or deceitful? Perhaps you have said your peace in the FP set arena and it’s time to stop?--Godot13 (talk) 08:14, 11 November 2014 (UTC)

Bonus for medical articles?

Given as I am a doctor I am hardly neutral in this, yet I have found improving medical articles alot more arduous and time-consuming than most of the stuff I work on. Would anyone think some sort of bonus or multiplier is feasible here? If so, are there other categories of articles that are hard...those under arbcom sanctions come to mind..... Cas Liber (talk · contribs) 12:30, 8 November 2014 (UTC)

I'm potentially open to the idea, but there seem to be at least a few issues which would have to be dealt with. Problems of scope are the most obvious: what counts as a medical article? My Phellinus ellipsoideus has some discussion of possible medical application, but I wouldn't call it a medical article. Is finger a medical article? Happiness? Suicide? Problems of fairness spring to mind: Why do medical articles (which are definitely difficult) get favoured over other kinds of technical articles? Difficult philosophical concepts, ideas in theoretical physics, legal complexities, economics... (Smaller questions: How do bonus points interact with each other? How can the bot deal with them, if at all?) J Milburn (talk) 17:00, 8 November 2014 (UTC)

Yeah I know, I came up with the same issues when I thought about it in more detail. I don't think it is insurmountable as such...but it is very fiddly. Cas Liber (talk · contribs) 05:17, 9 November 2014 (UTC)

Surely this would be the same as vital articles, an arbitrary system that favours some articles over others. Yes medical articles may be hard but so are military or historical articles. 08:55, 9 November 2014 (UTC)

Adding rule concerning "delayed updates"

Back in June, J Milburn and I discussed adding a new rule with regards to delayed updates (something along the lines of):

“ Content must be claimed within a week of promotion. Failing to update your submissions page in a timely fashion when you do not have a legitimate reason is considered an abuse of the system. If you are regularly failing to update your submission page in a timely fashion [alternatively, "within x number of days"] without good reason, your points may not be awarded, at the judges' discretion. This is to ensure transparency, not to deny you your points. ”

—Bloom6132 (talk) 08:58, 11 November 2014 (UTC)

Considering this was motivated solely by me claiming some content late due to various factors, but never for gamemanship reasons, I think making it a rule is overkill. Efforts to game the system (which would include purposely holding back reporting of points) have always been highly frowned upon, and are already ground for removal is problems persist... I will make sur eto claim points mroe promptly next year. --ThaddeusB (talk) 15:13, 12 November 2014 (UTC)

Agreed. I know I was late in putting things in for points a couple of times because I was away. Such a rule does indeed insinuate that people are all out to game the system while the truth is that there are several perfectly reasonable real life reasons why someone may be slow in updating their scorecard. The C of E God Save the Queen! (talk) 17:14, 12 November 2014 (UTC)

@ThaddeusB – Actually, J said earlier in the same thread that "Ed brought [the issue] up" earlier, so it wasn't "motivated solely by [you] claiming some content late". This has been a persistent problem, and though you point out that this has "always been highly frowned upon", no one has ever had points deducted or been kicked out of the competition because of it. Unless I'm mistaken (and feel free to provide me diffs showing otherwise), this has gone unpunished, solely because there is no rule in place to deal with these "delayed updates". And since the rule I proposed above does not punish those who "have a legitimate reason", why are you and The C of E so worried? —Bloom6132 (talk) 19:51, 12 November 2014 (UTC)

You do realize anyone wishing to actually game the system could make up a reason right? As to why I'm "afraid" of the rule, I am against it for the reason I already stated: it is unnecessary instruction creep. If someone is actually withholding points to gain an edge, that is a pretty sad reflection on them, but not a reason to punish people who simply forget to report points in a prompt fashion. If the judges feel this is a "persistent problem", they already have the authority to stop it (but I welcome input from J Milburn, The ed17, &b Miyagawa as to the actual frequency of issues). --ThaddeusB (talk) 20:32, 12 November 2014 (UTC)

@ThaddeusB – I don't see how this new rule would be an unnecessary instruction creep. Time and time again, judges (well, J specifically) have been unwilling to use their so-called "authority to stop" these delayed updates from happening. Statements like "the user in question did not break any particular rule … so we will not be docking any points" and "I'm certainly not going to start enforcing rules that weren't in place at the start of the competition" show how impotent beyond belief the enforcement of the primary rule – not abusing the system – has become. Hence, it is of utmost importance that this rule be put in place, and anyone who believes in transparency, fair play and integrity would also agree. On the other hand, your advocating the status quo will result in nothing being done, but I'm not surprise you continue to support this position given that you directly benefited from there not being a rule in place last year. —Bloom6132 (talk) 11:44, 13 November 2014 (UTC)

OK, I see J Milburn has said he won't disallow points because of late reporting. So, I will retract the comment about the judges being able to do so under current rules (while they can, apparently they won't). However, I stand by the part where I said someone could eaisly make up an excuse if they wanted to game the system, which means the rule would only punish honest contestants who forgot to report. I will also stand by my assertion that it would be pretty pathetic for someone to purposely withhold reporting just to slightly increase their chances. (There are probably a dozen more effective ways to game the system than reporting content late, for what it's worth). This is supposed to be a just for fun contest; if people are taking it that seriously (yourself included), they are doing something wrong.

I will again point out that I did not actually benefit from reporting some of my points late. I would have advanced based on the prompt points alone in the third round and didn't advance out of the fourth round. To deny me (or anyone else) points based on what I would deem a technicality is offensive. The contest should be about having some fun and doing good work - not some regid adherance to a set of rules, and not gamemanship. Again, if someone is truly trying to gain an advantage through sly behavior (and I don't think anyone has, other than one guy I know who actually did get points removed a couple years ago and subsequently was banned from Wikipedia for further disruption in other "badge gaining" venues), that is pathetic and they should be removed from the contest. However, I find it quite insulting that you continue to imply I did something wrong. --ThaddeusB (talk) 15:13, 13 November 2014 (UTC)

Bloom, I certainly will (and have...) use my "authority", as you put it, to stop people abusing the system. Here, we do not have a situation where I am "impotent beyond belief" and incapable of stopping abuse, we have a situation where you see ill-intent everywhere, and I don't. (And of course I'm not going to enforce rules that weren't in place at the start of the competition. Do you actually want me to? Think about what you're saying...) Thaddeus, I certainly have not said that I "won't disallow points because of late reporting". I have said that, in a particular case, I did not see any abuse. They are completely different claims. Please do not put words into my mouth, that is completely unfair. I don't think this discussion is going anywhere productive- I advise we stop it now. J Milburn (talk) 00:05, 14 November 2014 (UTC)

I'm not sure how you reconcile "won't enforce rules that weren't in place at the start of the competition" with "would consider removing points for late reporting (under the current rules)." That is all. I (obviously) have no problem with the decisions you made. --ThaddeusB (talk) 15:17, 14 November 2014 (UTC)

@J Milburn – "I certainly will (and have) use my "authority" … to stop people abusing the system." Diffs please. Correct me if I'm wrong, but you haven't expelled a single competitor for abusing the system in the two years I've participated in the comp, even though there were plenty of grounds to do so. I'm not asking you to "enforce rules that weren't in place at the start of the competition"; the one core rule that was already in place of "not abusing the system" is the rule that should be enforced but isn't. It's not just this year's Cup, but in 2013, you failed to dock any points off Bonkers The Clown for his abusive DYKs. Instead, you defended him tirelessly time after time – thankfully Ed took a different approach and removed those DYKs (i.e. actually "stop[ping] people abusing the system"). And your accusation that I "see ill-intent everywhere, and [you] don't is just a poor attempt at victim blaming. I'll use your own words in reminding you, "Please do not put words into my mouth, that is completely unfair". —Bloom6132 (talk) 03:32, 15 November 2014 (UTC)

The fact you insist on calling yourself a victim is telling. I am not providing diffs. If you don't believe me, don't believe me- I'm not going to lose sleep over it. J Milburn (talk) 09:38, 15 November 2014 (UTC)

And Thaddeus: The judges have always been willing to do what they have to do what's needed to stop people abusing the system, even if it's abusing the system in a way that isn't explicitly mentioned in the rules. However, at the same time, once the rules are set for a year, they're set. It would not be fair to introduce a "you must claim within two weeks" rule in the middle of the competition, even if it would be reasonable to remove some user's deliberately delayed update as a calculated, deceptive move designed to cheat another user out of their place in the next round. I assume this is what I was saying- if so, I stand by it. To clarify: I will always be willing to do something about abuse (even if we might have legitimate or not-so-legitimate disagreements about what constitutes abuse, or how much evidence of ill-intent would be necessary) but I am not willing to introduce new rules mid-competition. J Milburn (talk) 09:45, 15 November 2014 (UTC)

Thanks for explaining what you meant. I understand now. --ThaddeusB (talk) 15:35, 17 November 2014 (UTC)

The fact you insist on putting words into my mouth is telling, as I've never insisted on anything, let alone insist on calling myself a victim. I'm simply calling a spade a spade. And since you're not willing to back up your claim that you actively "stop people abusing the system", then stop making such an empty and baseless claim. To address your unjustified accusation – what is telling is your reluctance to enforce already-existing rules. And it's not just me who has been frustrated by this – look at what happened between you and Adam Cuerden during the FP fiasco. Sure, feel free to dismiss the concerns of one person as "seeing ill-intent everywhere". But when there's more than one person expressing their legitimate concerns about how the competition is run, it's still always going to be someone else's fault, isn't it? —Bloom6132 (talk) 15:56, 15 November 2014 (UTC)

Of course, there needs to be a rule to prevent unduly delayed submissions. Snowman (talk) 21:53, 19 November 2014 (UTC)

@J Milburn, The ed17, & Miyagawa – with just over a month to go before next year's Cup begins, what is it going to be then? Are you going to finally introduce a new rule that prevents competitors from gaming the system with delayed updates by deceitfully hiding points from other competitors. Or are we going to continue having the same ineffective system currently in place where J grants a general amnesty to everyone citing "good faith", despite damning evidence to the contrary? —Bloom6132 (talk) 11:42, 25 November 2014 (UTC)

Bit of a loaded question there. I think we should rely on the integrity of the judges to ascertain if someone is gaming the system. Which frankly isn't too hard to notice where there is a pattern of subs being nominated late regularly. The C of E God Save the Queen! (talk) 13:09, 25 November 2014 (UTC)

"I think we should rely on the integrity of the judges to ascertain …" – sorry, we've tried this for all the previous Cups, and in the last two years the status quo has proven ineffective and impotent. It's interesting to note how the two people who oppose my call for necessary reform prohibiting abusive "delayed updates" have directly benefited from there being no rule. —Bloom6132 (talk) 15:40, 25 November 2014 (UTC)

Excuse me, but I have not benefitted from there being no rule. I have always made my nominations promptly either on the day or within a couple of days of promotion. If it would help, I would suggest a system where people can inform judges if they are going to be away for a long time (holidays and the like) so there can be no allegations of deliberate gaming. The C of E God Save the Queen! (talk) 15:44, 25 November 2014 (UTC)

I like The C of E's idea about informing the judges about being away. But what do you think would be a suitable time period for which to submit points by? DYK goes with a 5 day nomination, so while this isn't the same thing I think that sort of deadline is an intrinsic one? Does that sound sensible? Miyagawa (talk) 17:50, 25 November 2014 (UTC)

@The C of E – But you said in your first reply, "I know I was late in putting things in for points a couple of times because I was away". Now you're saying you've never benefitted from there being no rule. Care to explain the discrepancy? —Bloom6132 (talk) 22:27, 25 November 2014 (UTC)

You are nitpicking here. By that comment, I was referring to a few days before I submitted as I was on holiday at the time not weeks and months as you seem to infer that is widespread and requires a rule regarding it. The C of E God Save the Queen! (talk) 22:30, 25 November 2014 (UTC)

Once again, I did not gain any actual benefit from the handful of items I reported late. Even if my legitimate work was removed on a technicality, as you proposed, I would have still advanced to the next round. I will remind you that assume good faith is a policy, which you appear to be in violation of. If you take the cup so seriously that you see no problem libeling other competitors over a simple mistake (which I have promised not to repeat) and insulting judges because they didn't do what you wanted, maybe you should not participate. BTW, Bonkers did have DYKs removed and was later banned from Wikipedia for inappropriate behavior at various "reward venues", among other things.--ThaddeusB (talk) 18:14, 25 November 2014 (UTC)

ThaddeusB – I will remind you that no legal threats is a policy, which you appear to be in violation of when you accuse me of "libeling other competitors". Whether or not you would've advance to the next round with or without those submissions is completely irrelevant to the fact that delayed submissions made without a valid reason should not be accepted in the comp. I also fail to see where I have not assumed good faith – I haven't accused you of doing anything wrong, because you haven't done anything wrong under the current rules. —Bloom6132 (talk) 22:44, 25 November 2014 (UTC)

All I can say to that is wow, WTF dude. --ThaddeusB (talk) 15:44, 26 November 2014 (UTC)

"WTF dude" – now that's an intelligent argument. —Bloom6132 (talk) 16:50, 26 November 2014 (UTC)

It's not supposed to be. Your rediculous assertion that I was making a legal threat doesn't warrant a serious reply. --ThaddeusB (talk) 16:55, 26 November 2014 (UTC)

Thaddeus has done nothing in violation of NLT, and your failure to AGF is clear for everyone to see. Drop the stick; you're fooling no one. J Milburn (talk) 23:18, 25 November 2014 (UTC)

You clearly haven't read the NLT page then. Perceived legal threats include "assert[ing] that another editor's comments are "defamatory" or "libelous", which can be "interpret[ed] as a threat to sue for defamation". Thaddeus is fooling no one, and neither are you with your "soft-on-abuse-and-gaming" attitude. —Bloom6132 (talk) 23:35, 25 November 2014 (UTC)

I am perfectly aware of what that page says. Do you think Thaddeus is threatening to sue you? I'll give you a tiny, weenie little clue: If you think he is, you're a moron. (And, for what it's worth, if I was firmer on abuse, step one would be blocking you. WikiCup or no WikiCup, your conduct has been way beyond the Pale for a long time, and remains completely unacceptable. I've not exactly been unambiguous about that.) J Milburn (talk) 23:49, 25 November 2014 (UTC)

First off, WP:INVOLVED applies if you do try to block me, given all the times you've ruled in favour of everyone else even in the face of legitimate evidence, simply because it was me who was raising and inquiring about those concerns. Secondly, your block threat to me while letting Bonkers The Clown (who violated countless Cup rules and was eventually indeffed) off the hook scot-free poignantly shows the hypocrisy in your "enforcement" of rules. Do you really think I've been more disruptive than BTC? To quote you – "I'll give you a tiny, weenie little hint: If you think I have, you're a moron". —Bloom6132 (talk) 01:05, 26 November 2014 (UTC)

Bloom: do you understand that being a judge is a completely voluntary act that yields nothing to the editor than having to deal with disruptive editors like you? Also, if you don't like what is happening on this page you are free to leave; I highly doubt your contributions here will be missed. Nergaal (talk) 04:10, 26 November 2014 (UTC)

Even ignoring the judges, this is meant to be a friendly competition. Over ten months, people are going to not edit Wikipedia, or have a limited time for Wikipedia that may not include documenting, for at least one contiguous week. Adam Cuerden ^(talk) 07:01, 26 November 2014 (UTC)

Nergaal is, of course, correct. Bloom, your failure to get the point is astonishing, and your comment dated 01:05, 26 November 2014 is a typical example of the reason you're a nightmare to work with. The constant insistence that you're some kind of victim (the claim that I have "ruled in favour of everyone else ... simply because it was me who was raising and inquiring about those concerns" is nonsense) and the endless wikilawyering (saying you're no worse than someone who got indefinitely blocked isn't necessarily a convincing argument against your being blocked) and demands for enforcement (along with accusations of bad faith- I'm a hypocrite, I'm biased against you for some reason, and so forth) are precisely the issue. J Milburn (talk) 09:49, 26 November 2014 (UTC)

Not that this comes as any surprise, but this isn't the first time you've been called out for victim-blaming before. Disappointing, but not surprising at all. —Bloom6132 (talk) 16:50, 26 November 2014 (UTC)

Nergaal – after the constant barrage of abuse you've given both Adam Cuerden and Godot13 for their Cup contributions in featured pictures, you're in absolutely no position to lecture me about what constitutes a "disruptive editor". You've managed to piss off and drive more people away from the Cup in half a year by denigrating and belittling the work of others than I have in my full two years of participating. If there's anyone whose "contributions here will not be missed", it's yours (without a doubt). —Bloom6132 (talk) 12:29, 26 November 2014 (UTC)

Can I please ask that I not be used as a stick to beat people with? I'd rather rescind all comments I've made about Milburn - despite still thinking he handled the situation I was actually talking about very badly - than have my words be used in some petty squabble to support a view I don't endorse. Milburn is an ass. Bloom, you're a much bigger ass. Adam Cuerden ^(talk) 17:10, 26 November 2014 (UTC)

Bloom, generally, if you're trying to encourage others to go along with your ideas, throwing your weight around like this is not going to be helpful. This shouldn't be a revelation. I've personally got absolutely no interest in engaging with you any further on this matter. If the other judges do, that's their call. J Milburn (talk) 19:59, 25 November 2014 (UTC)

Recent deaths

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Postings to the recent deaths section of ITN are not current eligible for points. This is not intentional, but rather an artifact of the fact such postings didn't exist when the Cup started. From an ITN perspective, such posts have teh exact same requirements as any other post - a well written, throughly updated article. However, by nature, it is sometimes easier to get an RD posting because there isn't a lot to say about someone dying. If the article, was already in great shape, there isn't much to do (unlike other current events, which enatil a lot of new information). On the other hand, sometime articles are in very poor shape and take more work than would typically be required.

Recent deaths should logically be worth some number of points. The simpliest thing to do is just make them worth 10 like normal ITN posts, but with teh caveat (which applies to all ITN material), that substantial work is required to claim Cup points. If an article was already in great shape, it simply wouldn't earn any points here. The alternative, is to make RD worth less (say 5 points). I am fine with eitehr option, but feel RD should definitely count for something. --ThaddeusB (talk) 00:38, 4 November 2014 (UTC)

I think they should be the same as any other ITN post. As you say, we already have a substantial work rule, so there's no need to worry about that. –Ugncreative Usergname (talk) 04:16, 4 November 2014 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Bonus scaling

Right now an article gets 20% bonus pts for each 5 interwiki links. Would people think an increase from 20% to 25% would be fine? An article like sea instead of getting a 7.2x multiplier would get a 8.75x one. Nergaal (talk) 23:16, 20 November 2014 (UTC)

Seriously? DYK's are already scoring very high with bonus points, sometimes more than a base GA, and you want to up the multipliers? Oppose: the bonus system isn't a bad idea, but it certainly doesn't need more of a boost. Adam Cuerden ^(talk) 01:14, 21 November 2014 (UTC)

Oppose First, the 20% works out nicely because 20% of 5 is 1 - no need to introduce fraction points. More importantly, the current multiplier is plenty (I would argue too much on the upper end of the spectrum). --ThaddeusB (talk) 04:45, 21 November 2014 (UTC)

"Value" and FPs - should there be a bonus system?

An outside proposal

I've never participated in the Cup, and I've not been an active editor for some time, but I've watched this discussion and the project from afar. I think a lot of proposals here risk hurting the encyclopedia (capping points, excluding images, generally promoting bad faith). Both articles and media are important to the encyclopedia. The main dispute seems to be that some content can earn points much more quickly than others.It's easy to see that both take a lot of work, and that it's difficult if not impossible to make a direct comparison between them. No one scoring system is likely to be perfect.

Articles and media work in different ways for the project. Articles are self contained, but can exist on many projects. Bonus points aim to compensate editors who work on more difficult and fundamental articles. This makes sense because they are, broadly, more important and of larger benefit for the project. Images by contrast can be used in multiple articles (and encyclopedias). Those images are likely more valuable for the same reasons that articles that earn bonus points are.

I would propose the following solution (numbers assuming 100 points per FA):

FPs would receive 25 base points. For each additional article they are used in, they would receive 2 bonus points, to a maximum of 10 points. The maximum score would thus remain 35 points.

This way, sets which are used only in one article space are no longer able to score massive amounts of points at once, but images which are used in many topics (pictures, art works, illustrations and diagrams nominated on their own are more likely to fit this criteria) aren't under valued. There's obviously room for gaming the system, but that will be true in any kind of scoring. We need to have some faith that editors make their contributions honestly. I don't think that's too much to expect.

The important thing to remember, I think, is what we're trying to measure: benefit to the project, and the work required to make that happen. It's also critical that different content is used and is valued for different reasons. That seems to be lost in the above. TownCows (talk) 01:16, 14 November 2014 (UTC)

• • As a followup. There's a lot of different ways that we can approach this problem. Scores for FPs could also be scaled by resolution (on the presumption, probably a good one) that the larger an image, the more time it takes to do (i.e. panoramas, massive scan restorations etc. Diagrams done as svgs would be given a set number). Bonuses could be given for article placement (Lead images, in body versus listed). The point is that looking at this just in terms of content type (be it articles vs. images, restorations vs. photographs etc.) is not solving anything. TownCows (talk) 02:47, 14 November 2014 (UTC)
    • Re: resolution: that's nowhere near the be all and end all some would make it out to be. Focus stacking takes some considerable time (especially on slower computers) but with really small subjects may only give a 1.5k pixel image. This is, of course, excluding the time it takes to set up such a shoot (very difficult with moving subjects). — Crisco 1492 (talk) 02:55, 14 November 2014 (UTC)
      • That is true, but I don't think it's insurmountable. One solution that comes to mind is simply counting the total resolution of base images. TownCows (talk) 18:58, 14 November 2014 (UTC)
• I could support the 25 point base w/up to 10 bonus point proposal. --ThaddeusB (talk) 15:21, 14 November 2014 (UTC)
  • That idea is probably going to cause things to start drifting from actual value rapidly. For example, an image used in an infobox at the bottom of pages will get lots of bonus points, but the lead image on a featured article might not. Plus, with every other bonus point system used here, there isn't a hard cap, this can only reduce points, never increase: why? Adam Cuerden ^(talk) 16:13, 14 November 2014 (UTC)
    • I think the requirements for use would have to be more stringent than just infobox use at the bottom (situations like this is why there would be a cap - to prevent abuse. Maybe it's worth thinking about putting a cap on the other bonus systems too). But that's not the issue, really. Another way this can be looked at is in terms of the FA project. During later rounds where higher scores occur the FAC process should maybe consider allowing competitors to submit more than 1 article at a time. I doubt that it will lead to an overwhelming flood, and it would certainly mitigate the review speed issue to some degree (This would certainly be the least disruptive solution). I think there's also a case to be made that while sets are collectively more valuable, the individual images may not. I also think it's fair to state, as some have above, that sets do to some degree mitigate the amount of work required at certain steps in the process (in the case of scanning and restoring, at the acquistion, not the restoration stage). TownCows (talk) 18:58, 14 November 2014 (UTC)
• I'll make one last proposal. This is based on the idea that 1) all images require a certain amount of effort as a baseline and 2) producing sets reduces work to some degree at some stage in the process. Here, images in sets would be progressively less valuable, up to a certain point when they would all receive the base points. I think that having all images after the first 5 in the set receiving equal points seems like a reasonable enough line to draw. My suspicion is that at 35 points FPs may actually be underscored individually and overscored, so I did some experimenting with Godot's scores to see what I could come up with that seemed to work well without changing the scoring dynamic. In effect, most FP nominations would be unaffected, but gargantuan sets (which appear to be the main grief) would be somewhat slowed. I did this with three potential valuations: 35, 33, 31, 29, 27, 25, 25... (Average value of the first 5 images: 34 pts); 40, 37, 34, 31, 28, 25, 25... (First 5 average: 31 points); 40, 36, 32, 28, 24, 20, 20... (Average value of the first 5: 32 pts). I thought the third option seemed most equitable: it benefits small sets and individually nominated images, but keeps scoring in line after a certain point. To be truly fair, this would have to go hand in hand, in my opinion, with capping or rescaling article bonuses and allowing competitors more than one FAC at a time during later rounds. The table below illustrates the results for Godot's final round scores (sorry for putting you in the spotlight)

Set	Total Images	Current Score	35, 33, 31...	40, 37, 34...	40, 36, 32...
Individually Nominated	8	280	280	320	320
Secretaries	40	1400	1030	1045	860
High Denominations	8	280	230	245	220
Coat of Arms	46	1610	1180	1195	980
Philippines	13	455	355	370	320
Presidents	25	875	655	670	560
1880 Notes	9	315	255	270	240
1928 Notes	8	280	230	245	220
Cuba	24	840	630	645	540
Total	181	6335	4845	5005	4260
Final Competition Score	---	6400	4910	5070	4325

The numbers are fungible, obviously. But I think there's something in this for all contributors to like. The final score, especially in the 3rd case is no longer insurmountable. FPs are still given large amounts of points (even more in some options and circumstances), but massive sets would no longer skew the competition. TownCows (talk) 20:19, 14 November 2014 (UTC)

Again, though, while it may be true in some cases, such as multiple photos of one location, I don't think it's generally true that there's any significant savings when doing a set. If nothing else, a set means that, no matter how bad an image is, one must still work on it. And one or two incredibly damaged images can easily raise the average amount of work hugely. It also encourages breaking up sets, which damages the FP process. Adam Cuerden ^(talk) 05:12, 15 November 2014 (UTC)

I think the reality is that some projects will always take more time than others. It's true of articles as well as for images. I also doubt that people will try to split sets to earn more points. It would be rather transparent, and I think the judges would be able to prevent it from being abused. TownCows (talk) 12:40, 15 November 2014 (UTC)

Just to point out, I divided up what could have been a set (Urania's Mirror) into smaller sets as I did the work over several months, and to help encourage full reviews. There are sometimes good reasons for not just nominating everything in one big set, indeed, the size of a set can be arbitrary at times, as there's hierarchies. Adam Cuerden ^(talk) 04:39, 16 November 2014 (UTC)

It's an interesting example. It also depends on how we think about FP sets. I would look at those as one actual set. It happened over a longer period of time, but once they're featured, they're for all intents and purposes a single group. My thought would be that in those cases, the newly added images would be scored as a continuation of the set. TownCows (talk) 05:03, 16 November 2014 (UTC)

Even though the ponts were split over three rounds? Adam Cuerden ^(talk) 16:50, 16 November 2014 (UTC)

Yes. It doesn't make a real difference, IMO. There is only one set; it's not too hard to make the argument that the time savings for working on similar images aren't muted by the fact that the work is spread out. TownCows (talk) 18:04, 16 November 2014 (UTC)

But, as I keep saying, there's only a few cases where those time savings exist, and this set isn't one of them. Adam Cuerden ^(talk) 23:42, 16 November 2014 (UTC)

There are still going to be some savings in most cases, even if it is just part of a streamlined workflow. Also, as I mentioned earlier, the individual EV of the pictures diminishes as the set grows, which is another reason why sets should be looked at as one entity and scored as if nominated en masse. TownCows (talk) 15:25, 17 November 2014 (UTC)

I would challenge whether that's true about savings. Vigorously. For some types, sure, but for restorations, the content matters less than amount of damage, size, and type(s) of damage. Plus, consider the Urania's Mirror pseudo-set: I believe every single image in that set is used in one or more individual articles on constellations, because it's a really valuable set of images. Adam Cuerden ^(talk) 18:29, 17 November 2014 (UTC)

I agree that you're right about images used in distinct articles are just as valuable as standalone FPs (the map projections set is another good example). They should be treated distinctly. I would dispute though that it's true of all sets (like large amounts of banknotes, for example. Or the images of Puck of Pook's Hill). TownCows (talk) 01:51, 18 November 2014 (UTC)

Quite a number of the banknotes are used in articles, because they depict people and other such things. "Was featured on a banknote" is usually going to be highly relevant to any discussion of commemoration or legacy. Further, Puck of Pook's Hill is featured on three different wikis (including Commons), and the set, in full, is used in three languages (English, Farsi, Japanese). Further, if we get a wikisource of Puck, which is to be hoped for, they will gain exponentially in value. Adam Cuerden ^(talk) 09:54, 18 November 2014 (UTC)

Basically as I said in my original proposal, wider usage means greater value. Cross-wiki usage is more difficult to measure for FPs, which is why I proposed multi-article use as a base for points. It's not all that hard to imagine a hybrid system of varied intra-set values combined with multi-article bonuses. There's been discussion below about EV as the measure, but I think that's too subjective to judge readily. I think as long as the picture clearly adds value to the article it is in (i.e. no flags, portal images etc.) it should count. TownCows (talk) 06:29, 19 November 2014 (UTC)

1) I kind of like the idea of 25 + bonus points, but I think the cap in that case can be raised to more than 35, possibly up to 50. 2) As for the second one, what comes in to my mind is a simpler rule: first 5 pics in a FP set get 35, all above that get 25pts? Nergaal (talk) 20:57, 14 November 2014 (UTC)

I would be fine with a higher cap than 35 if this system was adopted. --ThaddeusB (talk) 15:33, 17 November 2014 (UTC)

Tiering FPs according to their EV

How about trying to quantify the encyclopedic value of each FP? In the past we had a flood of low-notoriety DYKs, and this was solved by decreasing the base value of DYKs to 5 and compensating more important EV ones by keeping the 10 pts. What if the same is done for ANY FP, be it a set or not.

1. FPs that are "infoboxeable" (i.e. if there is an actual article on that picture, and that picture is representative of the topic of the article (stub or not)) should get 40 pts.
2. All other pictures should get like 20 pts. That way, if there is a set of 100 pics, and there is only an article on the whole set, then one pic gets 40 pts, and all the other pics get 20 pts.

If of all the other 99 pics, some of them have their own article, then those would also get 40 pts. For example, if there is a set on some famous pictures, all the pictures that are worthwhile enough to deserve an article shoudl get 40 pts, while all of the lesser-importance ones got 20 pts. If there is a banknote set, there will be an article on the set probably, but only the more important banknotes should have their own articles. Or if there is a set on pictures of presidents, if there is no existing, more worthwhile picture of that president, all the pics in the set should get the full (increased) 40 pts, while all the extra "neat" additions (that tend to become a gallery at the bottom, or as a part of a table) get only 20 pts.

I think this way, the impact of the work is taken into account, same way more impactful articles do currently get some sort of bonuses. Lastly, this encourages FP users to focus on pictures that have a higher impact on the whole wikipedia. I do think is in the spirit of the cup, AND would try to quantify the FP? criteria that relates to encyclopedic value of a picture. Nergaal (talk) 20:57, 14 November 2014 (UTC)

That sounds a little problematic in practice. Let me give you eight cases: The Puck of Pook's Hill set, a forthcoming The Diverting History of John Gilpin set (where the article is primarily on a set of artworks), Beer Street and Gin Lane, File:Gaspare Fossati - Louis Haghe - Vue générale de la grande nef, en regardant l'occident (Hagia Sophia - Ayasofya Mosque nave).jpg - Not the lead image, but certainly valuable, being the best historical image in the article, File:Rudyard_Kipling_-_The_Fringes_of_the_Fleet_(cover)_-_cleaned.jpg - lead image of an article but relatively boring as an image, File:C.M._Gilbert._-_John_Hay,_c._1904.jpg Not a lead image, but a great illustration, File:Levin C. Handy - General Robert E. Lee in May 1869.jpg - lead image, but may eventually be swapped with the only other good image in the article, File:Ivan Logginovitch Goremykin, c. 1906.jpg Lead image and only good image of him.

Outside of paintings, and a few 18th-century engravings, there are vanishingly few articles about images, as opposed to having images that represent the subject. Other than Beer Street and Gin Lane, Join or Die is probably my only FP with an article specifically on the image. Not all of us go to art galleries to take photographs, after all, and that's certainly not the only way to get EV in an image. The thing is, EV is a requirement of FPs being promoted. Adam Cuerden ^(talk) 03:25, 15 November 2014 (UTC)

I am not doubting the value of the images in these sets. Far from it. What I am trying is to come up with a scoring system which encourages FP contributors like you to work on images with the highest possible impact. As nice as these sets would be, I would honestly prefer you work on images that would be representative in some articles where images don't exist at all. Similarly, I strongly prefer people working on sea or melon, rather than the article on some obscure, high school basketball player. Nergaal (talk) 10:52, 15 November 2014 (UTC)

My point is that the rules currently under discussion would devalue all of these except Join or Die, because they're far too narrow. Adam Cuerden ^(talk) 04:41, 16 November 2014 (UTC)

Not necessarily, if you can prove (say to the FP reviewers) that that picture in particular deserves to be the lead pic of an article, then you can get more points than now. The 2013 system awarded 35 pts, the system I propose would be for a 20/25 and a 40/50 pts choice, if some reviewers think the picture is of some extra value to the wikipedia. Nergaal (talk) 21:51, 16 November 2014 (UTC)

I think the problem is that you're presuming FPC determines article content. It does not. Have a look at my current Robert E. Lee image. As I said in the FPC, this may move to a different section of the article, as there's one other possibility for the lead, but it's the best we have by far until then. I was quite open about that. However, what we don't want is someone getting lead image points for the same article every round of the competition. Adam Cuerden ^(talk) 18:23, 17 November 2014 (UTC)

• The lack of understanding of how FP sets work inherent to the statement here is quite concerning. It is little wonder why FP sets have been so divisive when well-meaning editors are under the impression that "a set on some famous pictures" would have a chance of passing. The other suggestions get closer, but still miss the mark. FP sets are not just "all presidents", but rather "all presidents as depicted by a particular publication at a certain time" (i.e. actually printed and published as a set). Compiling images of Washington through Obama, by different artists/photographers, would have vanishingly small chances of being promoted - unless they were all the "Presidential portraits", which are a series which have their own article. — Crisco 1492 (talk) 03:41, 15 November 2014 (UTC)

I think you misunderstood my point. If there IS a president without an existing image, or if the existing image is not particularly good, an image from a set would be the lead image. I am trying to think of a system where FP people work on images with the highest possible impact. In principle, there are pics which are FP on commons, but have no EV value here, so the cup technically values them with 0 pts. On the other hand, we have stubs with no info, where just a single image conveys much more info than the article text itself does. I do think the latter can be given more points than FPs which end up in galleries, or in tables somewhere. Nergaal (talk) 10:52, 15 November 2014 (UTC)

Just to clarify my point: I genuinely do not think people have problems with some contestants getting lots of points for FPs, but with the fact that there IS an impression that a seemingly very large amount of points are awarded to pictures that do not greatly enhance the value of existing articles, or if they do, they do that only for low-importance articles. I genuinely think if the FP work would seem as impactful as say working on big articles like sea, nobody would really mind. Nobody expressed significant concerns with somebody getting almost 1k pts for working on sea, because the perception was that it was a deservedly important articles. If FPs people would ged same points for seemingly similarly important work, we wouldn't have these endless discussions now. Nergaal (talk) 11:07, 15 November 2014 (UTC)

• Alright, but it would be much better for discussion purposes if the example "sets" would actually fit the set criteria. Otherwise we just confuse the point. — Crisco 1492 (talk) 04:44, 16 November 2014 (UTC)

• Nergaal - Are you now suggesting what content should or should not be worked on? I choose to work on numismatic and other related articles. I do this accepting that I forfeit any chance of earning bonus points generally reserved for existing more popular articles. I choose to significantly re-write some articles while starting several more. For this I am already “penalized” by the bonus point structure. It seems you suggesting that FPs are only helpful if they are attached to existing “important” work. Perhaps these images will bring attention to “low-importance” articles? Regarding “in galleries or in tables”… Are you aware that in the final round where 181 FPs were promoted, over 100 appear in an individual article in addition to the main set? Of the 24 presidential portraits, 23 appear in individual articles; of the 40 treasury portraits, 35 appear in individual articles. For what you might consider “low-importance” articles, 44 out of 45 state coats of arms appear in individual articles, 8 of the 9 Series 1880 United States Notes appear in individual articles, and the 8 Series 1928 Gold Certificates appear in 3 articles. By my calculations I have contributed FPs to at least 110 articles – in the final round. Do you like choosing what you get to work on? Well, so do I.--Godot13 (talk) 06:33, 16 November 2014 (UTC)

• 1) Yes, currently articles that have a higher notoriety get a bonus (currently being decently quantified as being featured on other wikipedias too). I see a similar tiering system for FPs being beneficial to the whole spirit of the cup. 2) I see absolutely no problem having pictures in a same set being awarded different points. I truly believe if there are 1% or 100% in a set of pictures of people that we do have an article of but they do not have pictures, then IMO they should get more points than those pictures in a set where wikipedia already has a picture on. 3) Here is a numismatics example: I know that there are some individual articles on some specific banknotes or coins; if you manage to get a FP on that article which is representative of the subject such that it deserves being the "infobox picture" then I do think you deserve to get more points for it, wether it is a set or not. However, if you work on a set of pictures where there is an article on that set, then only one of those pics should get that bonus points. 4) The only truly debatable issue about this system is wether this picture deserves to be featured in the infobox or the intro, but this can be decided at FPC by reviewers. For example, of the 24 presidential portraits, (not sure what happened to the 24th one) the 23 pictures that appear in separate articles DO deserve more points if those presidents do not have any other pics available here, OR if all the other pics that are available, are of less EV such that your scans "deserve" to be featured in the intro. 5) Don't you think a picture of a banknote that is notable enough to have its own article, should deserve more points? Nergaal (talk) 21:46, 16 November 2014 (UTC)

Another good point: Gently encouraging people to find content where it's lacking is one thing, but we shouldn't penalize people for having different interests. In any case, outside of people finding images and nominating them - which is irrelevant as they can't get Wikicup points for that - FPCs are generally created/restored relatively soon before the nomination. While sometimes they can replace a crappy image - See User:Adam Cuerden#Civil War Images that are awful (in queue, not started) for examples - but I'd suspect the majority, or, at least, a quite substantial minority of FPs are created de novo. We want to encourage both the replacement of crappy images like I just linked to and creation of new content. And let's not forget a third purpose I do FPs for: To celebrate a new FA or other high-quality content by giving it another go on the main page at some point. Adam Cuerden ^(talk) 10:29, 16 November 2014 (UTC)

I am not saying we should penalize anybody! I am saying to incentivize them to work on content where there is a higher impact to the readers. As I've said, FPs do deserve points, REGARDLESS of where they are featured. However, the same way we incentivize people to work on articles like sea rather than some obscure college basketball player, I do believe we should incentivize people to work on important drawings of Van Gogh, and give them less points if they work on Van Gogh drawings that do not deserve a separate article (i.e. don't pass wp:GNG). In addition, if a FP contributor thinks that a FP deserves a separate article, they can even get the bonus DYK points too, and the FP bonus that they would get for this. In this way at least, a random reader has the chance to find more about that scan or picture, if it is notable enough. Also, what is wrong with working on images that already exist but are of low quality? In the past DYK allowed only new articles, then it became evident that expanding stubby articles is probably even more important. Getting the best possible state for an existing picture, and incentivizing that is perfectly fine IMO. Nergaal (talk) 21:46, 16 November 2014 (UTC)

• I am in favor of incentivizing more important FP work, just as is done with articles. Perhaps a simplier proxy for importance would be to use to other Wikipedia article count. It is not unreasonable to think pictures in high importance articles are also more important to the encyclopedia on average. Something like 20 base points X article multipler could work. Or if the multipler is seen as too great that way, then maybe have the multipler accumulate at half the article rate (10% bonus per 5 wikis instead of 20%) and have 30 base points. An additional rule of only the first image per article gets the multiplier.

The only difficulty is some pictures are used in multiple articles. In such a case, I would say use the highest multipler on any individual article it is in. An image used in a navbox on many pages would only get to count articles where it is also used in the body of the article. --ThaddeusB (talk) 15:23, 17 November 2014 (UTC)

The problem is the that people would create galleries such that their FPs get more points. What is wrong with simply separating infobox pics from random pics? Nergaal (talk) 19:13, 17 November 2014 (UTC)

Like I said, only the first FP from a given article would get bonus points, so creating galleries would be less effective than it is now since th base points would be lower (and I'm not convinced it is a particular good strategy now). The infobox idea is also fine, but more complex to track. --ThaddeusB (talk) 20:54, 17 November 2014 (UTC)

Articles without pictures

Another variation: DYKs for new articles get 5 pts, while expanding existing articles gets 10 pts. Why not have an analogous rule for FPs? Pics that go into existing articles that do not have an image yet, they get 2x pts (therefore capped to one pic per article), while pics that go into articles with existing pics only get 1x pts. Nergaal (talk) 11:02, 15 November 2014 (UTC)

It's a good idea, but a couple things we need to work out first:

1. What counts as not having an image yet? Last year, I created the article Fatinitza and raised it to GA. It also has a featured picture I created for it. Would this count?
2. William Russell Flint had a replaceable fair use image, which was duly removed, I think well, well before the free-use images I found for it were ready. But, had I done this in a different timescale, and removed a fair-use image and put in a free image, should that count as well?
3. If the article has an Infobox at the bottom with an image, does that count? What about flag icons? If, as I think is sensible, these are ignored, can we actually get the bot to recognise this sense?
4. I've linked to my list of unnecessarily really terrible American Civil War images earlier. If I replace the only image in an article with a much, much better one, should I get bonus points?
5. Can this be done by bot? If not, will we have to request points manually?

I think this is a decent idea, but - as you can see - the practice is a bit messier. Adam Cuerden ^(talk) 10:47, 16 November 2014 (UTC)

Firstly, I do believe that the status quo is already messy. Trying to find solutions to the status quo is something worthwhile aiming for if we don't want the drama from this year.

1. Why not? If you can start an article from scratch, and get a representative pic for it you deserve both the DKY points and the FP bonus points. This encourages FP editors not only to work on non-pic content, but also advertise their own FPs with articles. You get your own pic seen by people who read articles on that image.
2. Yes (according to my expanded rule discussed above), if an article has only one pic, and you manage to get a pic of a better quality that is more representative of the subject, then you deserve the bonus. The old pic may be removed, but it could also remain in the bottom part of the article. The priority can be established say at FPC.
3. Obviously those extra unrelated pics can be ignored for this count. My optimal solution is to discuss the priority of the image with any other existing image at FPC.
4. YES (see #3)! My point is that the old pic can be kept in the article. The only important part is to see if the FP deserves to be the "top" pic.
5. I think the priority issue should be discussed at FPC (reviewers decide if pic deserves to be at the top of the article). Should be done manually, and the appropriate vote would be contained in the liked FPC.

I still prefer the "infobox" or "top" picture rule as you can see. Nergaal (talk) 22:11, 16 November 2014 (UTC)

Would flags count as pictures? Are coats of arms to be excluded? I don't quite see the point of this. The C of E God Save the Queen! (talk) 22:31, 16 November 2014 (UTC)

A picture is something that has a caption. Anything else is a sign. Nergaal (talk) 23:01, 16 November 2014 (UTC)

That's an interesting way of splitting it up. What about captionless infobox/taxobox pictures? What about sound files? J Milburn (talk) 08:54, 17 November 2014 (UTC)

I am not sure what is confusing about my proposal. I believe that adding pictures to articles that do not have one, is a good EV boost, therefore shoudl be worth more points than regular FPs. As an advanced idea, I think if an article has pics, if the FP deserves to be the "intro" or "infobox" picture on top, then it should be worth more points also. Nergaal (talk) 11:31, 17 November 2014 (UTC)

You defined "picture" as "something with a caption". I am challenging that account, because it seems reasonable to call a captionless taxobox/infobox image a "picture". Further, it is unclear where sound files and video files fit into this. Do you now see my concern? J Milburn (talk) 17:33, 17 November 2014 (UTC)

Milburn, please remember that the featured sound process is currently dormant. If you want to talk about it in the context of the Wikicup, you may want to work on reviving it first. Adam Cuerden ^(talk) 18:25, 17 November 2014 (UTC)

Adam, I am not talking about the featured sound project. I am asking Nergaal about his definition of "picture". I'm not really sure how I could be any clearer, here. J Milburn (talk) 23:01, 17 November 2014 (UTC)

I am confused as to what are you thinking. Do you have an example of FP-able or high quality picture that does not have a caption or is in the top infobox? I am pretty sure that the FP process pretty much implies that the picture in discussion have a caption OR be in the main infobox (the one that sits to the right of the intro). Nergaal (talk) 00:40, 19 November 2014 (UTC)

Milburn isn't very clear. There are occasional images that are in weirder templates - {{Wide image}}, {{Double image}}, and so on. But I don't know why Milburn mentioned sounds even after asking him.

As for examples of FPs without captions, besides things in tables, where the description is next to them... maybe some animated diagrams, where the text is in the animation, and they need to be displayed fairly large? Adam Cuerden ^(talk) 04:35, 19 November 2014 (UTC)

You're right that sounds aren't relevant. I misunderstood this conversation- I'm trying to follow everything going on on this page, and I really don't have the time (or the motivation, as I'm facing snippy comments from unpleasant people left, right and centre). In any case: there are FP nominations up at FPC right now of images which don't have captions (see Leotia lubrica), and videos are occasionally nominated at FPC. I don't buy your "images have captions, things without captions aren't images" for these reasons and others that Adam mentions- panoramas, images in tables, maps, diagrams- these are often presented without captions as well. J Milburn (talk) 10:22, 19 November 2014 (UTC)

I should have more faith in myself. Of course my comments about sounds are relevant. "Pics that go into existing articles that do not have an image yet, they get 2x pts (therefore capped to one pic per article), while pics that go into articles with existing pics only get 1x pts." Do sounds count as "images" in this context? They can certainly have captions. If sounds don't, do videos? If free media does, does non-free media? J Milburn (talk) 10:26, 19 November 2014 (UTC)

@J Milburn: Not many articles have sounds, and not much more have vids. I believe that even with that, a picture has a quite different role from a sound, and even a video, therefore if an article has a sound or a video, they would still get the 2x bonus. As for the fungi example, I am proposing two options: 1) if the article has no pictures, then the pictureless case gets 2x points (in this case the FP would still be worth 1x since there are other pics); 2) if the picture is "infoboxable" or "top image" (i.e. it sits in the intro) then it does deserve the 2x points (in this case the FP is of noticeably higher EV, therefore it deserves being in the top infobox, therefore should be 2x; all navigation infoboxes and such would be excluded from this rule). I prefer #2, but it is more difficult to enforce that unless the "infoboxable" or "top" status is voted and accepted at FPC. Nergaal (talk) 10:50, 19 November 2014 (UTC)

• Personally, I think this is rule creep to the worst degree; way too complicated to be enforceable. Furthermore, it wouldn't even have prevented much of the conflict from this year, as the sets that were nominated were (in several instances) included in brand new articles (and thus, if all added at the same time, they would all have been the "first" picture in an otherwise unillustrated article). — Crisco 1492 (talk) 10:32, 19 November 2014 (UTC)

@Crisco 1492: Not really, what I am proposing is have regular FPs be worth 20 pts, and the pics that go into pictureless articles be worth 40 pts. As for sets, only one of the pictures in the set would normally be worth the 40 pts. So large sets like say banknotes would average around 22pts per picture rather than 35pts like now, while sets of presidents would probably be worth closer to 40pts depending on wether there are other pics available in the article. Nergaal (talk) 10:50, 19 November 2014 (UTC)

• Your original wording was "Pics that go into existing articles that do not have an image yet, they get 2x pts" (then you summarize that this means a maximum of one image per article, which is not necessarily true based on that wording, as I pointed out). Adding "except if" and "only when" and other similar wordings is rule creep, pure and simple, in my opinion. — Crisco 1492 (talk) 12:41, 19 November 2014 (UTC)

• @Crisco 1492: How about this

A picture that goes into a new article or an existing article which on 31-12-2014 did not have any images, gets 2x points, with a limit of one such bonus per article.

This is in no way near any more complex than any of the current DYK rules, AND would allow for a decrease in base points for FPs (to say 20 or 25) since such a bonus exists. Nergaal (talk) 14:24, 19 November 2014 (UTC)

Value and FPs - more complicated than assumed

Okay, let's talk about value. First of all, let's start by pointing out high-value images that are NOT in the start of an article, but greatly add to it.

• Scallop leads with a picture of a living organism, but Scallop#Anatomy uses File:Scallop Diagram2.svg, a clear, detailed diagram that guides the rest of the discussion.

Diagrams seem to me to be a legit category to get 2x points, as per the bonus points I proposed for videos and animations below. Nergaal (talk) 15:42, 19 November 2014 (UTC)

• Due to WP:OPERA standards, most opera pages start with a picture of the composer in an infobox that interlinks articles. As such, for example, images of the opera itself don't appear straight away in Le Cid (opera).

I think in such non-typical cases you can ask for the bonus if it was say File:Jules Massenet - Le Cid 1885.jpg, and let the judges decide on that. No system is perfect, but the 2014 system seems to be further from ideal than the EV-based one. Nergaal (talk) 15:44, 19 November 2014 (UTC)

More examples could be found, but I think those show typical examples. Further, several of Godot13's sets form the active content of a featured list - without the images, the list simply wouldn't have sufficient details to be useful.

Those lists get the bonus of having multiple images in the same FPC. Nergaal (talk) 15:42, 19 November 2014 (UTC)

You've not really listened to anything I've said in the last month, have you? Adam Cuerden ^(talk) 17:01, 19 November 2014 (UTC)

Why do I think you are not assuming good faith? I only want to find a system with which non-FP contributors are more comfortable with. Nergaal (talk) 19:22, 19 November 2014 (UTC)

Those lists get the 'bonus' of having multiple images in the same FPC - Bonus = more work as a set. Nope, he's not listening...--Godot13 (talk) 21:07, 19 November 2014 (UTC)

So you guys are 100% not interesting on modifying the status quo and make non-FP contributors more comfortable with the FP scoring? Nergaal (talk) 23:28, 19 November 2014 (UTC)

• They (and I) are interested in having people understand FPs better. Bonus multipliers are just that: multipliers. They are awarded whether or not there was extra work necessary to reach them, so long as other criteria are met. An FP set, meanwhile, requires the same amount of work for each image included. Would you say that writing 4 featured articles is the same as writing one featured article with a 4x multiplier? — Crisco 1492 (talk) 08:46, 20 November 2014 (UTC)

• @Crisco 1492: This shows to me that you clearly do not understand the bonus multiplier system. A 4x FA could, but likely is not the same amount of work as 4 1x FAs. However, there is more hassle with the former, AND wikipedia gains (together with its readers) more form it from the 4x FA than the four different 1x FA. Wikipedia has plenty of articles now but what it lacks is good quality content in areas where people care. Generally there are more FAs on obscure topics that on high-impact/importance ones. Why should this competition favor writing another 100x articles on obscure high-school basketball players instead of high-impact articles like sea? Nergaal (talk) 15:07, 20 November 2014 (UTC)

Secondly, let's look at the idea that number of articles should be a guide. This is problematic in many ways:

• Overcounting: Take the (non-featured) File:Lt. Gen. John B. Hood.jpg. It's used in eight articles. However, in every article but John B. Hood, it's a tiny little thumbnail shoved in with lots of other tiny thumbnails. However, none are in galleries. If this image was raised to featured status, it could easily be overcounted.
• Overusing: At the moment, there's no real reason to try and stuff images into articles that they don't add much to. If the Wikicup was actively encouraging it, it's not inconceivable that people will overuse images, probably not to the point of abusiveness, but beyond what they actually deserve. For example, the Puck of Pook's Hill set: It's not uncommon for articles to have an "in popular culture" section. With this set, I decided to put it out there, and let it spread organically over years. But, if I wanted more points, it wouldn't be hard to justify lots of uses - the first Puck image into Puck (mythology) and Puck (A Midsummer Night's Dream), the image about Wayland could go into Wayland the Smith - it even mentions Puck of Pook's Hill already! Viking stuff might be useful in Popular culture sections... you get the point. They might be justified, but instead of spreading organically, we're now pushing things hard, and that leads to...
• The points, they are a changing: Let's say that, with the best of motives, I've put my image in... five articles. Indeed, this isn't a particularly strange thing to do with an image. The thing is, when I do it outside the new Wikicup rules being proposed, it's a sort of WP:BRD process - Bold edit proposing the use of the image; maybe people will Revert its addition in some of the more marginal articles, but there's no real need to Discuss, I'm just here to propose its use, if they don't like it, I have my big main article that noone's going to remove it from. And it might well spread to other articles - that happens too. I did not add this image to Turkey Travel Centre. (One could argue there's good reasons for my wise decision in this case, but let's ignore that; point is, others will use FPs.) The system works. Images spread and are pruned on their merit over time. But we want to score points based on how many articles it's in. At what time do we do the count? Do we update? What if I lose a usage in article X, but gain one in Y and Z? Should I gain points? Should I gain points in round four because my image from round one is now lead image of a new article?
• Undercounting: Sometimes, a slightly different version of an FP will be preferred for different uses. For example, I was asked to provide an uncropped constellation card for the Urania's Mirror article, just to show the context. Similarly, sometimes people will crop an FP for certain uses. (Sometimes these people will even be justified in cropping it, though that's rarer.) This can mean that an FP has four apparent uses, but minor variant nfP is used in six other articles. We don't like to feature two variants.
• Undercounting, Part II: Usage on other Wikipedias can take time, unless you're replacing an existing, low quality image. Maybe we want to give extra points to FPs that replace old, crappy images - but, if the idea is most valuable, that's harder. We DON'T want people shoving captionless images into languages they don't understand, but they will spread... over time. Probably too much time for any WikiCup scoring system. Still, though, many of my FPs from a few years ago are featured on three different Wikis or more. (This led to amusing things when I told User:Hahc21 about this, as he gives awards for that kind of thing. Time consuming, difficult to check awards for that sort of thing. Throw him a barnstar, he's great.)
• Counting is tricky: There's probably some usages that shouldn't count. But they're not ones that can be defined by bot easily. Footer infoboxes probably shouldn't count, but the infoboxes at the top should... except when they shouldn't; I don't think that I should get loads and loads of points for File:Bain_News_Service_-_Franz_Lehár.jpg because WP:OPERA uses composer images in their infoboxes for operas. And heaven help us if some complicated flag SVG gets featured, then put into the templates that mark nationality with flags or the like.

Adam Cuerden ^(talk) 09:38, 19 November 2014 (UTC) Adam Cuerden ^(talk) 09:38, 19 November 2014 (UTC)

• I notice you didn't comment on my proposal which is to use the same multiplier (number of interwikis) as the article the picture appears in would get. I realize it isn't perfect (it isn't for articles either), but it at least gives some sort of proxy for the importance of the subject. --ThaddeusB (talk) 15:09, 19 November 2014 (UTC)

I don't think you've said it quite so clearly before, or if you have, I missed it. Eh... It's not bad, but we'd need to ask for a main article if we want the scoring to be done by bot, because bots aren't that clever, and a really triviial usage would be enough to throw them. Adam Cuerden ^(talk) 17:06, 19 November 2014 (UTC)

Upper limits for scores

I think this is a worthwhile point to consider for discussion some of the issues above. The highest single-entry in the competition that I am aware of was Sea with 7.2x multiplier, so worth 130*7.2, or almost 950 points (FAC+GAN). Using the 2014 point distribution, that comes to a set made of 27 FPs. On the other hand, picture sets like this, this, this, and this were awarded about as many or way more points. Any thoughts? Nergaal (talk) 11:06, 10 November 2014 (UTC)

There is a discussion about changing the bonus point system above... The comparison to sets of pictures is pretty irrelevant, though. A set of 27 pics could have just as easily be 27 separate nominations. Making sets worth less than individual nominations would be akin to making additional GAs or DYKs after a certain number worth less. --ThaddeusB (talk) 15:48, 10 November 2014 (UTC)

We shouldn't be doing things that encourage worst practice. It is entirely possible to flit around, sipping at one thing here, another thing there, and getting a wide variety of featured picture credits, but without any depth. This leads to a lot of problems later, for example, I've actually cost Durova a couple featured picture credits this year because I did the Urania's Mirror set, and consistency and some minor issues called for replacement of a couple of her restorations. However, in the interim, noone with sufficient talent to finish the set had actually stepped in to do so, so a whole ton of really crappy images - things like File:CetusUrania.jpg and File:Draco and Ursa Minor.jpg - were widely in use, because it turns out that having good depictions of the constellations is important, and Urania's Mirror is amongst the clearest, easiest-to-use settings out of such constellations. They're in use in literally hundreds of articles, and on dozens of Wikipedias. That's a massive improvement to not just en-wiki, but far farther spreading than that.

Any attempts to discourage this would literally encourage worst practice: do one image from a set, then flit over to another type of image to avoid reviewer fatigue, and never come back. We never become a good resource on any one topic, and we ruin reuse.

Another example: Puck of Pook's Hill. I could have probably got that passed with only a few images, or only did one. However, for our reusers: teachers, children wanting to learn about the book, and even other Wikimedia projects like Wikisource, that would have ruined the reusability. My arbitrary selection would limit them, and mean that only bits of the book could be talked about with illustration. Adam Cuerden ^(talk) 16:19, 10 November 2014 (UTC)

I've got to agree with Adam Cuerden on this one. Point-capping does encourage worst practice and doing math to game the system. The idea here should be to make contribution fruitful in all areas (not everyone's an expert in everything; I have several FAs, FLs, and GAs to my name, but believe it or not, no DYK's because I'm not interested in them) and place a balance so that contributors who put in the same amount of work in the same timespan can be scored equally, regardless of what content area they work in and specialize in. Because caps will impede work in some areas, I think this is a bad idea. Red Phoenix ^{let's talk...} 03:14, 14 November 2014 (UTC)

Proposal for a new rule

I would like to propose that there be a new rule in next year's competition, applicable specifically to the final round. I propose that contestants' points for each scoring category should be capped at the total they received in that category in all the previous rounds. In most instances this extra rule would have little or no effect but it could be significant in some. Applying that rule to this year's competition, Casliber would be permitted his 4 FAs in the final round because he had previously claimed for 5 FAs in previous rounds, but I would not be allowed to claim for my 2 FAs because I had only claimed for a single one in previous rounds.

Turning now to featured pictures, under the new rule, Godot13's FP submissions in the final round would be limited to 111 and Adam Cuerdon's to 72 (Godot actually claimed for 190 in the final round). Adopting this rule would encourage contestants to score more heavily in the early rounds in order to build up a sufficient reserve in any scoring category on which they hoped to rely if they were still competing in the final round. Cwmhiraeth (talk) 06:28, 12 November 2014 (UTC)

I would not be in favour of that, it does seem to limit creativity particularly if you work hard throughout the year to bring an article up to FA standards but can't nominate it in the final round. The C of E God Save the Queen! (talk) 08:47, 12 November 2014 (UTC)

Also, would Wikipedia really be better if we took the most productive 8 people, and told them "you have to stop now" after a certain point? Adam Cuerden ^(talk) 11:06, 12 November 2014 (UTC)

Absolutely not. Why encourage people to stop editing? YE ^{Pacific Hurricane} 14:16, 12 November 2014 (UTC)

I see problems - I might be away or have limited time for an early section which then hamstrings me for the rest of the competition. I might have a slow FAC and some delays and only get 1 FA in one round. Or somesuch. Cas Liber (talk · contribs) 19:22, 13 November 2014 (UTC)

I was thinking specifically of the final round, – if you haven't done any FA's earlier in the competition (you've got a whole 8 months), you can't score for one in the final round. The purpose of the proposal is to try to resolve the mismatch between FPs and other point-scoring means. It would prevent users from working on images and amassing potential pictures over the course of the whole year (or longer) and then submitting them all for FP in the final round. Cwmhiraeth (talk) 12:01, 14 November 2014 (UTC)

As a general comment, I think working up towards a very large good or featured topic, which is then nominated in the final round, would be a perfectly reasonable (and pleasantly neat) way to take part. Topics (and something similar is true of portals) are few and far between, and require much preliminary work, and so this system would seem to very much go against them. J Milburn (talk) 12:58, 14 November 2014 (UTC)

Yes, well the idea has its problems, and as several people don't like it, its not likely to get anywhere, anyway. Cwmhiraeth (talk) 13:18, 14 November 2014 (UTC)

Long DYKs

Currently long DYKs (over 4.5K) get twice as many points for short ones (minimum 1.5K). Let's balance the value with the work required and triple the points for long ones because two short DYKs are generally easier to do than a single long one. 4.5K is long enough that it will almost certainly be B-class in quality and could be a contender for GA. I've written many GA-class articles around that length and I think that it would help to encourage editors to write longer articles that cover their subject more in depth. Something that I think that the Cup should encourage in general. Thoughts?--Sturmvogel 66 (talk) 15:46, 17 November 2014 (UTC)

That could play oddly with bonus points. It's already easy to get a DYK that's worth more than a GA, I think we may be incentivising DYK above all other content if we're not careful. Adam Cuerden ^(talk) 17:50, 17 November 2014 (UTC)

True, maybe the bonus points ought to apply to the base value rather than the total value. That would go a pretty fair way to cutting the massive points for DYKs on many wikis.

I think the status quo makes a correct balance of points. I don't think we should tamper with it. The C of E God Save the Queen! (talk) 23:28, 17 November 2014 (UTC)

So you don't see the blowout victories over the last three years by the exploitation of speedily reviewed processes like DYK and FP as a problem?--Sturmvogel 66 (talk) 23:44, 17 November 2014 (UTC)

There is no inherent reason why GAC or FAC (or any of the processes, to be honest) shouldn't be speedily reviewed. — Crisco 1492 (talk) 00:01, 18 November 2014 (UTC)

True, and the shortest review period that one of my FACs has ever had was 8 days, but that's far from the norm as they average about a month. Generally the only way that a GA will get speedily reviewed is during a backlog reduction drive or something like the GA Cup. Most importantly, you can submit an unlimited number of GANs, unlike FAC, so you have a better chance of reviewers picking one of yours. But reviewers in general tend to review things like DYKs or FP that involve a lot less work or work under a QPQ system.--Sturmvogel 66 (talk) 04:05, 18 November 2014 (UTC)

• I've had a couple which went only four days (Hiram Wesley Evans comes to mind), though that was with a couple days of PRing beforehand. But yes, many people do prefer to do DYK reviews. I recall Adam mentioned having one left a standing offer to review any GA if requested, but nobody took him up on it. Was that on this page? — Crisco 1492 (talk) 04:27, 18 November 2014 (UTC)

• Points are given for GA reviews. Why isn't the same done with FA reviews? They're clearly more involved, and also obviously valuable. Might speed things up (and allow for more than 1 FAC at a time)TownCows (talk) 06:26, 19 November 2014 (UTC)

• A reasonable point for discussion, I think (though what counts as a "point-worthy FA review" would need discussion). — Crisco 1492 (talk) 10:33, 19 November 2014 (UTC)

Yeah, last year basically offered to review anything anyone asked me to, and... think I was asked about ten times total over the whole Wikicup. Had I been a lot more swamped then, I'd be a lot more sympathetic now. Adam Cuerden ^(talk) 07:09, 18 November 2014 (UTC)

FP repoll

Since there doesn't seem to be a consensus above, I took the liberty to start a new poll with some improved ideas. Nergaal (talk) 15:45, 19 November 2014 (UTC)

All FPs should continue to be worth the same

All FPs require a lot of work so their value should remain 35 points!

All FPs should be worth fewer points

Some FPs take less time therefore they should be worth less points for sure. Until a better rule is agreed upon, the value of all FPs should be lowered. Should they be worth 25 or 30 points?

This is already an open vote above. Adam Cuerden ^(talk) 18:20, 20 November 2014 (UTC)

FPs should be worth less points, except for those going in articles without existing images

A picture that goes into a new article or an existing article without any images on 31-12-2014 will get 2x points, with a limit of one such bonus per article. Should the base value be 20, 25, or 30 points?

Any flags, navigation boxes, videos, sounds would NOT count towards this rule. Nergaal (talk) 15:06, 19 November 2014 (UTC)

FPs should be worth less points, but those featured in the lead of an article should get a bonus

Any picture that is agreed upon at FPC to be representative enough to be featured in the introduction or in the infobox of an article should get a bonus. Should this bonus be 2x or linear to the importance of the parent article as per WP:Cup/Scoring#Bonus? Should the base points for FPs then be 20, 25, or 30 points?

Only one bonus per picture is allowed. Linear bonus would mean the article where the FP is featured in the lead or infobox would need to be present in 25+ other wikipedias to get 2x points, while the top infobox in Jesus would be worth about 11x base points. This incentivizes people to work on pictures representative for higher-importance articles. Nergaal (talk) 14:55, 19 November 2014 (UTC)

I would support this with the linear scale (maybe only 10% per interwiuki instead of 20% given articles though) as an alternative to my less stringent suggestion below. As Adam C has pointed out, the lead image is not always the most important one, so I am hesitant to judge based on infoboxing, but it is an acceptable trade off to prevent excessive counts from merely appearing in a super general article. --ThaddeusB (talk) 16:08, 19 November 2014 (UTC)

IMO, for clear examples of "not-infoboxed but for sure most important one" an user can claim the bonus points as long as a judge (explicitly?) agrees to it (a similar criteria as we ask editors to submit only those where they have done a "significant" amount of work). Nergaal (talk) 18:55, 19 November 2014 (UTC)

After some thought, and good points by Godot and Adam, I would like to refine my position. I think the bonus idea is a good one, but it should be a fixed number, not an article-style multipler. I would suggest 5 points for 10+ interwikis and 10 points for 50+ interwikis (based off host article). I do not think "infobox" is a necessity, but galleries should be excluded.--ThaddeusB (talk) 21:20, 20 November 2014 (UTC)

I could support the basic concept, but think we need to tweak the numbers a fair bit: even very important, well-known people like Franz Lehar come under 50 interwikis (30-something, if I remember right. Iconic general Robert E. Lee only has 57; The big Shakespeare plays like Romeo and Juliet and Hamlet look to hover in the 70s, but the still quite iconic Much Ado About Nothing only has 29. As such, I'd change the proposal to a base of 30, with 5 bonus points for 5+ interwikis, 20+ interwikis, and 50+ interwikis. This would be a reasonable proxy of value, although the 5+ might be better at 10+. Adam Cuerden ^(talk) 21:26, 20 November 2014 (UTC)

That could work- three fixed tiers, perhaps at 5, 20, and 50+ interwikis, (with values to be determined based on some research/discussion), and a base score of 30. Also, if an image is not in a formal infobox, but it is clearly the sole lead image, it should (I think) be bonus-eligible.--Godot13 (talk) 21:50, 20 November 2014 (UTC)

I'd say any non-gallery image. Adam Cuerden ^(talk) 22:31, 20 November 2014 (UTC)

I like that we are getting somewhere! I think starting next year with a smallish range (30-45) would be a fine trial period. IMO, I would prefer this: base value 25, and a bonus system capped at 50pts. For example every 10 interwikis it gets 5pts. So if it is on an article with 50+ interwikis, the FP gets 50 pts; only 40 pts for 30-39 interwikis. Adam, how do you define galleries? I believe if you add a 11th picture to a well-developed article isn't significantly different from a truly <gallery> section. I believe restricting bonuses to (1) infoboxes, (2) lead images, (3) anything else of similar status (like the snail diagram, or the actual scan of an important opera, or a single FP from a set in the table of a list article, or any article/list that doesn't make sense have put a pic in the intro/infobox but where the FP is representative of the subject) would be most ideal. I believe (3) should also include a "pending judges approval" clause to prevent abuses. I believe this system would prevent FP versions of File:Georg Ernst Stahl.png, File:Wisoff on the Arm - GPN-2000-001069.jpg, or File:Rust screw.jpg from getting bonuses for being used in oxygen. Nergaal (talk) 23:33, 20 November 2014 (UTC)

I am quite surprised (and pleased) we might actually be moving towards a consensus. I could support either the proposed 30-45 (Adam) or 25-50 point scheme (Nergaal). How do you guys feel about the 5 per 10 wiki up to 50 system @Adam Cuerden: and @Godot13:?

As to the infobox/importance issue, I think it is probably tricky to judge. Placement alone is not sufficient, as (other than the infobox/lead) the article may well be chronological, so an image being higher up may just be for that reason. Perhaps a no gallary/gallary like section rule plus a strong caution that adding images to articles solely to gain bonus points won't be tolerated work work. (Note, the FP process requires an image to be stably placed in an article. This in of itself rules out someone tacking a non-relevance image on to, say Jesus, and getting it to FP - if the image is unimportant to a high-profile article it is quite likely to be removed fairly quickly.) --ThaddeusB (talk) 04:58, 21 November 2014 (UTC)

I think Nergaal's has far too low of base. This isn't meant to be a massive drop in points for sets by a back door. Adam Cuerden ^(talk) 07:06, 21 November 2014 (UTC)

As I've said, I am fine with either, and as a trial period it is probably better to start with a small variance in points. However, how do you decide for example File:Rust screw.jpg not to get the bonus for being used in oxygen (it would almost for sure get the max bonus for being present in rust but that is not the point here)? Nergaal (talk) 09:38, 21 November 2014 (UTC)

If you guys feel strongly about "all pics outside galleries" rule here are two additional criteria that I think would be sensible and limit snowballing:

1. One bonus per article, or bonus only in articles without FPs

2. Bonus is eligible only for the most possibly representative article

No 1 could encourage people to work on articles with less pics/without FPs, while no 2 would imply that an image of a rusty screw would only be eligible to get the bonus on the rust article. Nergaal (talk) 09:55, 21 November 2014 (UTC)

In the end, this has to be run by a bot. Let's keep it simple. Adam Cuerden ^(talk)

I don't know how the bot works. You put the name of your entry and then the bot comes and automatically checks for bonuses? If yes, for FPs one should put the name of the image, the FPC, and then the article which one would be claiming the bonus for. Nergaal (talk) 10:41, 21 November 2014 (UTC)

I'd say lose both 1 and 2. They're a nightmare waiting to happen. Adam Cuerden ^(talk) 13:16, 21 November 2014 (UTC)

I'm not sure there is a huge difference between the two scales, but I will gladly endorse the 30/35 (5+ wikis)/40 (20+)/45 (50+) scale. I do agree with Nergaal that one should only be able to claim the bonus once per article - and it is simple enough to enforce. The "most representative article" idea is probably not worth the trouble given the smallish bonuses possible. --ThaddeusB (talk) 15:58, 21 November 2014 (UTC)

I'm not willing for this to be a backdoor way to reduce points in highly valuable sets. I've already tacitly agreed to reducing images in galleries to 30, that agreement, however, is conditional on valuable images getting more. If there's any other restrictions, I withdraw my support. Adam Cuerden ^(talk) 23:25, 21 November 2014 (UTC)

Looking at your FPs, I am pretty sure that 90%+ of your pics will get more than 35pts next year. I am fine with that. What I don't understand is if you think that all the pictures used right now on oxygen should get 45pts if they were to be FPs, and if yes why. Nergaal (talk) 23:58, 21 November 2014 (UTC)

My understanding is that getting a second, third, etc. image featured on a given article is considerably harder than a first one. The idea being that another image doesn't add as much EV. However, I have rather limited experience, so I'm not quite sure. Is this correct Adam Cuerden? --ThaddeusB (talk) 00:21, 22 November 2014 (UTC)

For larger articles, a number of distinct "topics" arise. For example, Orion (constellation) can reasonably have both an FP of the stars and nebulae in the constellation, and a depiction of the traditional mythological depiction. Ulysses S. Grant served in the Mexican War, the American Civil War, and as United States President. A ship article might have a photo of the ship, and a depiction of the wreck of the ship. An article on a city might need a number of photos to illustrate different concepts - History, culture, major sites, and so on. An article on a theatrical work might show several of the key scenes, and each will add value. You get the idea. Adam Cuerden ^(talk) 00:30, 22 November 2014 (UTC)

Right, but each has to convey unique information, yes? That is, a high-quality picture used merely to add more illustration to an article with other FPs wouldn't be featured? The reason I ask, is that would make FP's criteria already a fine filter to prevent someone getting many bonuses for the same article. (This would necessitate the claimant adding an bonus point claimed article to their nomination rather than letting the bot decide, but that's not a big deal.) --ThaddeusB (talk) 00:56, 22 November 2014 (UTC)

Pretty much, yes. I can't think of any case where redundant images would both be featured and used. I think the only way you'd get redundant works is if it was something along the lines of (ignoring copyright for the example) "Andrew Warhol did a lot of notable silkscreen artworks that were very similar. Here's five representative samples of his $BLAH paintings. But if you did that, they'd almost certainly be in a gallery. Adam Cuerden ^(talk) 03:16, 22 November 2014 (UTC)

Here is a possible case that comes in to my mind: is it possible for somebody to add to the Civil War article the image of a less important general or character? I agree that Grant deserves its place on that article, but would anybody at FPC check that a forgotten character is also present in that article (that character would have its own article, so the FP is clearly stable in at least a low-traffic article). How do you prevent such abuses? I think the bonus should be limited to articles where the picture has a notable impact. This rule would be something that judges can double check in cases of abuses (i.e. if somebody consistently gets bonuses for semi-dubious uses). What do you think? Nergaal (talk) 08:52, 22 November 2014 (UTC)

I don't think we can have a perfect rule system. What we can do is set it up in such way to limit exploits. So, here's what I see as the major exploits and what protects against them.

• High-value article, low-value artwork. File:Columbus Taking Possession.jpg is in Sea, but has problems for use in any Columbus article - it's neither a famous depiction, not particularly historically accurate. Luckily, since encyclopedic value is considered at FPC, it would probably fail on encyclopedic value grounds: It's only tangentally connected to one article.
• Putting things into high-value articles for the points. High-value articles tend to be well-watched. Adding an illustration without good justification will likely be reverted. We may want to have the bot check and flag for manual review any FP whose value score has changed every so often.
• Image great for low-value article, meh in higher-value one: One of the reasons I didn't want to have too many tiers is that it helps avoid how many points you could get from these sorts of things. The trouble is summary style and spinoffs - consider New York City#History (extensive discussion in an article on 195 wikis) and History of New York (even more extensive discussion, only 17 wikis.) - I think this hits lists worst, actually, as I've rarely seen a list eligible for bonus points. Hmmm. Might want to tweak a little. Adam Cuerden ^(talk) 09:57, 22 November 2014 (UTC)

Ok, a sensible rule that would cover part of the abuses is in order to claim a bonus, the pic needs to be used in the article claiming bonus for at least one month (or two?). FPC only checks to have it present for a period on any single article, but for bonus, the rule should apply to the bonus article. Nergaal (talk) 10:15, 22 November 2014 (UTC)

I think the spirit of your idea is good, but there's two things to work through. As a kind of "Zeroth" thing though: Note that any image that hasn't been in an article seven days at the start of the ten-day FPC review (without good reason to skip the seven days) is generally rejected out of hand. Good reasons are things like "The article lacked an image before this", "It replaced a very similar, but much lower-quality image", or (less helpfully) "The article hadn't been edited in two months" (but then, in that case, that image probably isn't getting many bonus points anyway). If there's any doubt it'll stick, the image will almost certainly be opposed (though there are some weird articles with traditions that screw everyone over) That said:

First, it would tend to overvalue cases where you improve an existing image as opposed to getting new ones, and I'm not entirely sure the former is the more valuable of the two). Waiting one month is a good way to kill the excitement at your work before getting it to review. Also, before I set up my user page the way it is now, waiting too long was a good way for me to forget I had done a restoration.

Second, consider the case where File:X.jpg gets replaced by its restored version, File:X restored.jpg - to humans, it's trivial to tell there's continuity, to machines, not so much. Likewise, during some FPCs, minor tweaks will be made to an image, these are meant to be uploaded separately, under a new filename. If the new version is successful, all uses will be replaced with it on closing of the FPC.

It's not a perfect solution, as one could have a lot of stuff promoted at the end of a round, but what if the bot checked the bonus points awarded to FPs again at the end of the round, and flagged up any cases where the point value had changed for manual review by judges?

I'm good with the 30-45/5-10-15 structure. Just to clarify- if a single FP is used in more than one article with different bonus levels, the highest bonus can be claimed, but not the bonus for each article... correct? Also, if this will require spot decisions from the judges regarding non-infobox but lead image placement, shouldn't we hear from them too?--Godot13 (talk) 19:48, 21 November 2014 (UTC)

Yes, the one highest bonus would be earned. If judges @J Milburn, The ed17, and Miyagawa: or bot programmer @Jarry1250: wishes to provide some input on the mechicanical side of this, I would certainly welcome their input. --ThaddeusB (talk) 21:03, 21 November 2014 (UTC)

Bot query?

Would it be possible to have a bot take at look at all the FPs that scored in 2014 and see on average how would they score bonuses? Specifically, check how many of them are used on articles with 10+/20+/30+/50+ interwikis? Nergaal (talk) 10:15, 22 November 2014 (UTC)

FPs should have a lower base score, but earn a multipler like articles do

Any picture should receive a multiplier based on importance of the subject. That is, use the interwiki bonus (or a fraction there of) applicable to the article it appears in. If a picture appears in multiple articles, the highest bonus applies (exlcuding articles where the picture is only part of a footer navbox.) Only one picture per article can earn a bonus. EDIT: alternatively the multipler is given based solely on the article the picture most closely associates with, not the highest one.

I suggest a 20 point base with a 10% per 5 wiki bonus, but am flexible on both numbers. Thus, a picture illustrating really important subject can earn substantially more points than it does now, while a set of low interest pictures wouldn't get any bonuses. This is not a reflection on the amount of work required - which is probably about the same regardless - but rather a reflection of what most improves Wikipedia. This is primary justification offered for the large multipler possible on articles, and I see no reason it shouldn't apply to pictures too. --ThaddeusB (talk) 15:49, 19 November 2014 (UTC)

This is a more loose rule to the one I proposed above. How do you exclude say some wacky priest making some ridiculous remarks about Jesus, then the FP of that priest appearing in Jesus? Or more likely, an image of Einstein appearing in physics or science (I am pretty sure that these 2 have more interwikis than Einstein itself, but if not other similar cases could apply)? Nergaal (talk) 15:53, 19 November 2014 (UTC)

(edit conflict)Well that example wouldn't actually happen because there is no way his comments would rise to the level of notability required to appear in the Jesus article, let alone his picture. That said, a common sense exception could be applied if some less extreme example occurred that seemed wrong. A better question is are we comfortable with say File:Georgia_Aquarium_-_Giant_Grouper_edit.jpg getting the fish multiplier (it is the lead image of that article, but is more directly applicable to Giant grouper). [After the ec, you have added examples similiar to my fish one]. --ThaddeusB (talk) 16:01, 19 November 2014 (UTC)

If the "main article" is used instead of the highest one, a larger multipler (e.g. same as article) might be reasonable. The lower multipler was suggested mainly to prevent excessive scores from, for example, featuring pictures in broad animal articles. --ThaddeusB (talk) 17:34, 19 November 2014 (UTC)

Issues like the fish example above should be (and can be easily) discussed at FPC. If people at FPC feel that the File:Georgia_Aquarium_-_Giant_Grouper_edit.jpg is the best example for the fish article I see no issues with getting this multiplier. IMO, if we stick to the infobox rule, I think exceptions from it could be given by the judges here. Nergaal (talk) 18:30, 19 November 2014 (UTC)

I have abandoned this idea in favor of a smaller, fixed number of bonus points. See my comment in above section. --ThaddeusB (talk) 21:20, 20 November 2014 (UTC)

Diagrams, animations and videos should be worth double points

Diagrams, animations and videos are much rarer and more difficult to get, so they should get 2x the base value of regular FPs!

The EV of such ideas is especially high and the work is probably more substantial on average. Thus I support a 2x multipler for these regardless of whatever else is decided about FPs. --ThaddeusB (talk) 15:51, 19 November 2014 (UTC)

I'm not convinced that the standard Wikipedia User-created Video is as difficult as claimed here. They tend to be things like "Falkirk Wheel making a full rotation", not any studio setup, and, as such, could be massively overvalued. And you're undervaluing restorations so very much... Adam Cuerden ^(talk) 17:10, 19 November 2014 (UTC)

I was the only person who supported giving restorations (and photography) more points than other types of FP work in the original discussion, so I'm not quite sure where this idea that I personally am devaluing restorations comes from... I support extra points for these works primarily because of the high EV. --ThaddeusB (talk) 17:18, 19 November 2014 (UTC)

More of a generic "you". I'd be careful with all of these, actually. Take Conway's Game of Life, which has a lot of awesome animations, several featured, but, as it's showing a well-supported simulation, would be relatively trivial to remake. Adam Cuerden ^(talk) 17:23, 19 November 2014 (UTC)

I have a hard time imagining anyone abuse this bonus with animations like Falkirk Wheel or Conway's Game of Life. I can't say I remember seeing any animations submitted at WP:CUP in the years I've been around. As for the restorations split, if you think there is ON AVERAGE a clear difference in time taken to get a restoration from getting a scan, then I fully 100% support splitting them into 1x and 2x pots. My only worry is for someone not to abuse this. Also, the "infobox bonus" shouldn't stack witht his 2x. Nergaal (talk) 18:45, 19 November 2014 (UTC)

Bonuses are meant to be an automatic system. If you've got to the point where the only way you can get a bonus to work is to say "well, people won't claim it", then your plan's broken. Adam Cuerden ^(talk) 18:18, 20 November 2014 (UTC)

We don't seem to speak the same language. The spirit of a bonus system is IMO to encourage people to work on content that is either underrepresented, or undervalued. For example, we have points for FPo but nobody does them, so "giving bonuses" to that is in the spirit of the competition IMO. Nergaal (talk) 19:12, 20 November 2014 (UTC)

FP sets should be scored differently

As sets grow, FPs should be given reduced points up to a certain baseline

Already a poll for this up above. Please don't open duplicate polls. Adam Cuerden ^(talk) 08:12, 21 November 2014 (UTC)

This whole poll is a duplication. Don't discourage discussion. Unstruck. TownCows (talk) 17:28, 21 November 2014 (UTC)

If the whole poll was really a duplication, the whole poll should be deleted - the other poll is still open. Restruck: having two polls, with exactly the same subject, both open at the same time, harms discussion. Adam Cuerden ^(talk) 23:23, 21 November 2014 (UTC)

Take FPs out of the competition completely

This area is quite distinct from all the other parts featured in the competition, and there has been so much energy spent (without getting anywhere) of this subject (which in reality is a relatively small part of the whole Wikipedia) that the whole WP:CUP would be a better place for almost everybody involved if FPs would be taken out of competition completely until at some point in the future a clear consensus emerges with how to score them.

• Oppose - having difficult agreeing to a new scoring system is insufficient reason to remove FPs from the cup. --ThaddeusB (talk) 18:35, 20 November 2014 (UTC)

Discussion

@Adam Cuerden, Cwmhiraeth, The C of E, Godot13, Yellow Evan, ThaddeusB, Ugncreative Usergname, Crisco 1492, Gloss, J Milburn, Ypnypn, Sturmvogel 66, Chiswick Chap, Red Phoenix, and TownCows: Notice that the bottom 2 choices could award more points than the existing 35, but at least would incentivize some specific areas (hopefully higher-impact ones). Nergaal (talk) 15:47, 19 November 2014 (UTC)

The ping didn't work, probably because the comment is (accidentally) unsigned. Perhaps try again with a new comment (I would do it, but then I wouldn't know if it worked or not.) --ThaddeusB (talk) 15:41, 19 November 2014 (UTC)

Resigning the existing comment doesn't work. Here I will ping for you: @Adam Cuerden, Cwmhiraeth, The C of E, Godot13, Yellow Evan, ThaddeusB, Ugncreative Usergname, Crisco 1492, Gloss, J Milburn, Ypnypn, Sturmvogel 66, Chiswick Chap, Red Phoenix, and TownCows:. Hopefully that works. --ThaddeusB (talk) 15:53, 19 November 2014 (UTC)

How come you've coupled a lot of things to lowering points overall? That's rathe r deciding in advaance. Adam Cuerden ^(talk) 16:58, 19 November 2014 (UTC)

I do not think that is true. Sure the minimum points possible is lowered, but the maximum is raised at the same time. If you want a tiered system (I do), it is pretty much a given that the minimum should be lowered. Otherwise, you are saying FPs need more points than they get now. I mean, that is a valid stance, but not one that will get anywhere as to resolving the disconnect between FP points and people's perception of their difficulty/value.

Honestly, after all this discussion I do not really know what you would like to see happen. Are you completely happy with the status quo where every picture regardless of effort or importance gets the same # of points? You spend a ton of time explaining how difficult restorations are (and I agree), yet you don't seem to be willing to admit scans, for example, are easier than restorations. Given that articles are worth vastly different number of points depending on the subject, I find it a bit hard to justify a position that all pictures should be worth the same number of points. I do not think FPs earn too many points on average, but I do think it would be more fair if they had a tiered system of some sort. (Originally, I supported one based solely on type of work. That idea received zero comments other than my own. The idea of tiering on encylopedic value is also reasonable if a sutible system can be found.) --ThaddeusB (talk) 17:28, 19 November 2014 (UTC)

I don't think we've come up with robust ways to describe importance, and, as such, think this is highly premature. Further, for the ways we can mark, lowering the value of all FPs isn't necessarily a requirement - we could give 5 points extra to lead images, say, while not lowering the value of other FPs. If we have robust ways of determining value, lowering base points makes sense, as most of the valuable stuff will be benefiting anyway, but if those ways of determining value aren't very good, then we really aren't helping things, just inserting ways to game the system. Adam Cuerden ^(talk) 18:06, 19 November 2014 (UTC)

Here is my general worry: since we have examples of 2 different winners doing predominantly FPs, I cannot be supporting keeping the same system in place AND on top of that give bonuses. I am perfectly fine keeping the average value the same, by decreaseing some categories and increasing the value of others (say have 30 pts for normal FPs and 40 for infobox FPs). The question I have for you: at the end of 2015 would you rather: a) have 100 random FPs; b) 50 random FPs and 50 infobox FPs; c) 100 infobox FPs? Nergaal (talk) 18:48, 19 November 2014 (UTC)

Also, Adam, would you describe your stance as: "Some FPs take less time therefore they probably should be worth less points for sure. Until some rule is unanimously approved, the value for ALL FPs should be kept exactly the same"? Nergaal (talk) 18:52, 19 November 2014 (UTC)

The previous winner with FPs was half a decade ago, before bonus points, before a whole host of revaluing of article content.

And I don't think that you're fairly representing me. My point is, making a bad change that hasn't been well-thought through, well-discussed, and informed by facts is far, far worse than the status quo. I'm not against change, but I don't think we're anywhere near reaching a good solution.

Something that makes things worse is not worth doing just to have made a change. Adam Cuerden ^(talk) 23:52, 19 November 2014 (UTC)

Thanks Nergaal, you've just described the majority of my last 10 months of work as "random" because it may not have fit neatly into an infobox. Like Adam, I'm open to change but not just because an idea sounds good or might serve to pacify a very small group of competitors. We need concrete criteria and a logical rationale. “Given that articles are worth vastly different number of points depending on the subject, I find it a bit hard to justify a position that all pictures should be worth the same number of points.” There is no existing method to assess the EV of an image and to think we can apply the methods for assessing the EV of written text to an image leaves too much unaccounted variance. It is important to know that if someone else is willing to put in the massive work to mount an almost exclusive FP wikicup in the future, they can still win. Proposals to raise the value of FAs seems to have been a direct result of the perceived disparity between FA-FP values. I understand that people want some kind of change, but raising FA and reducing FP starts to seem (for lack of a better word) like gerrymandering.--Godot13 (talk) 00:46, 20 November 2014 (UTC)

@Godot13: Right now the FP score is like the US Senate: both Rhode Island and California get two senators. I propose that if we have a hard time doing a census (i.e. accounting for how much time an FP takes) at least look at the exports of those two states (how much comes out of those FPs) and give them a number of representatives (points) somewhat proportional to that. There is no existing method to assess the EV of an image. This exact rationale was used before the current interwiki system was put into place (and that bonus system was actually taken from wp:TFAR not copied from here). And you might want to victimize yourself as much as you want for me using the adjective random, but I genuinely do think that a picture that describes the whole subject (infoboxable) is almost always much more valuable than a picture that describes only part of the subject of the article. Also, taking a banknote example, if there is a banknote that deserves its own article (like $100) then that picture does deserve more points than a single banknote in a series of banknotes that has an article only on the series. I am not saying that all the n banknotes in the series should be worth only as much as the FP of the $100 bill, but it shouldn't be worth *n points either (JUST from an EV point of view). Also, I bet everyone here can agree that File:New100front.jpg has more EV than File:US-$100-LT-1863-Fr-167.jpg, and therefore should be worth more points. Nergaal (talk) 14:41, 20 November 2014 (UTC)

Perhaps you or Adam can make some suggestions to guide us then? Just saying "you guys are wrong" doesn't really help move forward, now does it?

Please explain how using the number of wikis (a proxy for importance of the subject, not the text) causes "too much unaccounted variance". I have no clue what that is supposed to mean. And again, a tiered system is not about reducing the points given FP (which I have explicitly !voted against), but rather keeping them roughly the same overall, while giving some pics more and some less. Nergaal's proposal, for example, would give a FP on Jesus 220 points, even if the lowest of 3 suggested base points was chosen. I am pretty sure that is considerably more than 35! --ThaddeusB (talk) 01:43, 20 November 2014 (UTC)

I do wish I didn't have to explain things in three different threads every time. I think the problem is that EV of images doesn't necessarily tie well with article value. To use your own example, the lead image at Jesus at the moment is File:StJohnsAshfield_StainedGlass_GoodShepherd-frame_crop.jpg. A stained glass window at a random Australian church from ~1932. It's not bad stained glass, but, at the same time, I hardly think it merits massive bonus points; it's one of a thousand similar images that could have been used and is not discussed at all. It probably doesn't add a huge amount of EV. Now, of course, this could be a coincidence, but before we go giving massive points out, perhaps we should actually find out what the results of such things would be, by having a look at some high-value articles, and seeing if the images generally live up to the value. I'll start. Isopoda's lead image would probably be featureable if of higher resolution. However, I wouldn't consider it particularly uniquely valuable. Sea's lead image is also potentially featureable if it was higher-resolution, but it's also hard to call it uniquely valuable.

This is me speculating, but I think the articles that would be considered most valuable by these rules are also the ones that are most general. With articles above a certain level of abstraction, there's so many possible images that could be used that what is used generally comes down to a crap shoot.

This has... consequences for gaming the proposed system. If I could get a very good image of the Sea - and, trust me, I can get a very good image of the sea, I own Cassell's The Sea - under these rules, I would get massive points just by getting it into Sea. But what if you don't own Cassell's The Sea? In that case, you may be interested in noting that lithograph of Columbus used in Sea is available from the Library of Congress. Adam Cuerden ^(talk) 07:53, 20 November 2014 (UTC)

I did offer as an option that the picture bonus would only be earned for the most directly relevant article. Thus, it would be difficult to claim one for an article like sea. Any photograph would be more relevant for Caspian Sea, Atlantic Ocean, or whatever. You would probably need to make a illustrative diagram of some important concept to qualify. Is this idea perfect? Of course not, but its a starting point for debate that hopefully will create a good system. --ThaddeusB (talk) 15:17, 20 November 2014 (UTC)

[Unindent] If I might make a suggestion for a general rule? If you want a range, try for a range of about 30-45 for FPs. Because you can't change the rules mid-competition, so don't want to discover in February that you've either massively undervalued valuable FPs, or massively overvalued not-that-valuable ones. It's a lot harder to mess things up if you try a small change first. I'll be honest, though, I'm not signing up next year. I might give the Wikicup another chance in 2016, but there were some pretty ugly moments this year. Adam Cuerden ^(talk) 08:23, 20 November 2014 (UTC)

I am not of the same opinion. IF you could convince editors that whatever image you have should be the infobox image of Jesus rather than the current one, THAT image increases the clarity of a very important article; you improve the quality of a very important article such that very many readers get more out of that article. And in my opinion, if such an image were to get even 1k points, I would be more happy than now when a random set of some images sitting somewhere in some semi-forgotten articles gets that 1k points. Unless you are mega lucky to live right near a church with a very good representation of Jesus, you will REALLY have to put quite some energy into getting such an image. Nergaal (talk) 14:23, 20 November 2014 (UTC)

Also, I would be curious how many people would be against you or somebody else finding good images for sea or Jesus that are actually deserving of the infobox spot, and then being awarded 200+ points. I think if you can get 30*bonus pts for a GA, you should also be able to ask for say 15*bonus (or 10*bonus) points for a real FP. If people are really worried about this we can have say 25 or 30 pts for a FP, OR a 10*bonus pts for that FP, whichever is larger (so if you get a FP in some obscure article, then you can still get a minimum 25/30 pts, but if you get at least around 25 (or 50) interwikis for that picture, then you should be able to claim more pts. This is much better than the current system where it is quite rare to get more than 150 pts from a GAN, but is rather common to get it from a FPC of pictures that are mostly featured in obscure articles. Nergaal (talk) 14:34, 20 November 2014 (UTC)

I will echo this point. I will not try to imply that a super great image of Jesus is that much harder to do, but I do think people will be a lot more accepting of large image scores if the images are going in high profile articles. --ThaddeusB (talk) 15:17, 20 November 2014 (UTC)

Lastly, could you guys try to focus on how to improve some of the proposed solutions rather than point out why they should not be implemented? This page is for WP:CUP, not WP:FPCup. Please try to consider solutions that would make happy ALSO the 95% of people taking part in this project which wouldn't flinch much if this were to become WP:CUP−FP. Nergaal (talk) 14:47, 20 November 2014 (UTC)

I give up. You're having the same conversation six places, you're not listening to a word said, and you seem to think that giving hundreds of points for a single FP will make people more accepting of people getting FP points than if people do the work on a massive set, which you just keep acting, over and over and fucking over again: "This is much better than the current system where it is quite rare to get more than 150 pts from a GAN, but is rather common to get it from a FPC of pictures that are mostly featured in obscure articles." [Three comments up]. No, you're wrong. NO FPC IS WORTH MORE THAN 35 POINTS AT PRESENT. END OF STORY. Having multiple related pictures is not having one picture. You're like talking to a brick wall, your start conversations in six different places and make the same damn arguments in every one, and now you've opened a thread to throw FPs out of the competition. I'm fucking done here, this is a waste of my damn time. Adam Cuerden ^(talk) 18:26, 20 November 2014 (UTC)

I would not have started this section, as its been clear for some time that most everyone has left the conversation. That said, it is a shame Nergaal feels the need to continue to attack sets - this shouldn't be about sets. It is also unfortunate that you seem to be unwilling to offer any sort of suggestion as how to make image scores more fair, as you are precisely the person who is in the best position to do so... Maybe next year we can revist the issue and have a more productive discussion. --ThaddeusB (talk) 18:40, 20 November 2014 (UTC)

@Adam Cuerden: I've thrown multiple ideas at you "FP guys" and I fail to sense any of you bulge from your positions of looking down on anybody not working on FPs. If you think the status quo is what the Cup needs, then you probably believe that the points system (excluding FPs) in 2010 was better than now. Nergaal (talk) 19:08, 20 November 2014 (UTC)

You seem to have the idea that pointing out problems that should be discussed and dealt with is the same as not moving. Here's a hint: You can't change rules during the round, so half-assing this isn't an option. We should be tearing ideas apart, then seeing if any parts of them still hold up, or can be fixed, because we won't get a chance later. But what's been happening is you suggest something, a discussion starts, you start a second thread repeating the initial idea, taking in none of the discussion, repeat ad nauseum. The whole discussion is taking place in about ten parallel threads, with no structure whatsoever, and with a huge amount of redundancy, because you keep insisting on starting votes before the idea's even been discussed for a day. You keep putting words into my mouth: "If you think the status quo is what the Cup needs, then you probably believe that the points system (excluding FPs) in 2010 was better than now." can hardly be considered helpful, or even an attempt to describe my points.

I can't work with you. Noone could. Any good ideas you have are scuttled by your unwillingness to discuss things seriously, but instead just wanting your ideas to pass, instantly, without any attempt to improve them first. The general idea of trying to encourage highest-value content in FPs is admirable. But we need good ways of doing this, not ways that mean I could win the competition with maybe 30 FPs over ten months and no other content, like an easily exploited 200+ point bonus system.

And that's another thing. Seriously, the thing that made you finally decide to blow up and try and throw FPs out of the competition was me saying that 200-point single FPs was a terrible idea, and showing how exploitable it is? Because these are my posts from immediately before you opened a thread asking for FPs to be kicked out of the Wikicup, in which I'm arguing against giving them as many points as a proposed suggestion. Would you rather I went along with this thing that would almost certainly skew things massively in my favour? You realise I'm mainly focusing on American Civil War battles and generals and World War I at the moment, right? Things that would have gotten massive, ridiculous bonus points under the scheme I objected to. Clearly, I'm trying to game the system. Adam Cuerden ^(talk) 19:23, 20 November 2014 (UTC)

Nope. I've had plenty of problems in the past with run-of-the-mill DYKs on obscure topics winning, or coming close to blowing the competition outright. What I have problems with is continuing to encourage people to work endlessly on obscure topics and then praise them for doing great, lots of work on topics that nobody really cares about outside a very narrow circle. Also, I think I made it pretty clear somewhere that I would be happier to see a FP of Jesus getting 1k points rather than continue with the status quo. Same with a representative picture of sea, that truly deserves to be in the infobox. Also, if you think something like this is easy to game, then cap the possible bonus points to either 50 or 100pts per FP for a trial period (until next cup). Oh, and I don't recall implying that you are trying to game the system. Nergaal (talk) 19:45, 20 November 2014 (UTC)

I said I could game the system, and part of winning the Wikicup is gaming the system slightly, after all. It's not like anyone has continuous cup-winning points in every round of the entire ten-month competition. But your proposals make it trivial to win the cup with FPs, without really trying. Hell, in the last four months, without trying to get things that would give me massive bonus points, I restored and brought to FP the lead images on Franklin Pierce, William H. Seward, Robert E. Lee, Franz Lehar, the Battle of Franklin, John Jellicoe, 1st Earl Jellicoe and Ivan Goremykin. I was arguing against my own benefit because I didn't think your proposals would be fair on other contestants than me. If anyone said "Don't you think you're getting too many points for that image?" I don't want to be in the position of having to say "Yes, massively". But I'm done here. Adam Cuerden ^(talk) 19:59, 20 November 2014 (UTC)

(ec)I'm fairly certain that you have not considered the ramifications of potential 200 point single FP images and will just start complaining about it a few months into next year's cup. Personally, I think 200 points for a single image is absolutely insane. And to avoid gaming, we then use a fairly non-objective means of determining whether it should qualify for bonus or not. Should we evaluate, on a case by case basis, whether a GA should receive bonus points based on how much work/text was added? I know I've seen some with 90% of the text written by the contestant and others where less than 10% (even 5%) was written by the contestant. If we come up with new rules they can not be subject to interpretation mid-game. ThaddeusB, my issues above are not with you, I want you to know that. @Nergaal I find your comments often tactless, and somewhat disorganized and tangential. Right now the FP score is like the US Senate? How about it more resembles a negotiation where all parties (except you) seem to have some grasp of what’s involved (or at least exhibit an openness to understand)?

Victimizing myself? Okay, if you say so. I thought I was just responding to your persistent lack of tact. I won working on material that interests me and (at least according to TFL one-day main page hits) a few others. I started out discussion willing to take some kind of value hit on FP. For the right kind of bonus structure we may be able to make that happen. But, your involvement in the discussion (purely in my own opinion) has been and continues to be disruptive to progress.--Godot13 (talk) 20:31, 20 November 2014 (UTC)

I don't have an issue with a 200pts FP on Jesus. I would rather have 200pts truly impactful FPs than massive sets of 1k pts in galleries of low-traffic articles. This is similar to having a Jesus-like FA worth 1k pts than have 10 FAs on some low-traffic [insert here obscure categories overrepresented at wp:FA]. Nergaal (talk) 23:02, 20 November 2014 (UTC)

[Unindent] And, may I point out, Godot, that I did the lead image for Albert Einstein in the last round of this Wikicup? He's in 172 Wikipedias. I also had Franklin Pierce (93 Wikipedias), and Robert E. Lee (57), amongst others. There's other high-multiplier leads in there, but I think that's the top three. Had multipliers been in effect, we all know that the current upset would be that I used massive multipliers on FPs to win the competition, which I could have, easily, because I almost did it without trying. Adam Cuerden ^(talk) 21:08, 20 November 2014 (UTC)

Again, I have absolutely no problem with all these 3 pictures receiving significantly more points than the average FP. If someone doing FPs wins the Cup by working on high impact images I have absolutely no problems with that. And again, if you believe gaming can be an issue we can have say a 3x or 5x cap for bonuses on FPs. Nergaal (talk) 22:58, 20 November 2014 (UTC)

Since everyone seems worried about having 200pts FPs, why not do this: all FPs have a 20pts base value, to which bonus multipliers can be added capped to a total of 50pts. Nobody will be prevented on doing whatever FPs you want, same way nobody prevents people from working on Epacris impressa. However, if someone wants to get most value for their bucks then they will be 2.5x more inclined to work on infobox pictures on the likes of Einstein, R. Lee, or F. Pierce. Nergaal (talk) 23:08, 20 November 2014 (UTC)

• Wow, I'm frankly amazed that everyone is still arguing this. There's clearly no consensus for 200 point FPs, and (as shown by the above discussion) there may be a consensus for lower points for FPs, but a fixed number has not been determined. I feel sorry for the Cup coordinators who have to close this.

However, I can guarantee that, after the welcome that FPs have received this year, whatever changes are implemented will have little effect on the 2015 WikiCup. No FP-oriented people will be participating.

(I still think people should be asked to try and take or restore an FP before judging how much value they should be given. As one of the few people who has done work in almost all areas of the cup which earn points [I've never had a GT], I can promise that you will be surprised). — Crisco 1492 (talk) 08:27, 21 November 2014 (UTC)

• You'll note the large multipler bonus idea has been abandoned above after good points by Adam & Godot. We actually are now making good progress towards implementing a slight (5 point) reduction in the base score for FPs while also allowing for a small amount of bonus points. See the "FPs should be worth less points, but those featured in the lead of an article should get a bonus" section. --ThaddeusB (talk) 16:02, 21 November 2014 (UTC)

  Glad to see that, while I was sleeping, progress has been made. Yeah, the current wording of the proposal would probably get support from me too. — Crisco 1492 (talk) 01:15, 22 November 2014 (UTC)

  I'm afraid I've lost track of these various FP discussions (and I suspect I'm not alone in this). What is this sort-of-conclusion that has been reached? Perhaps I could put this up for a vote above. J Milburn (talk) 12:49, 23 November 2014 (UTC)

Re: the main discussion now seems to be favoring dropping FPs to 30 pts, and giving a 5/10/15 bonus depending on the article where the pic goes into. At this point, what is 100% agreed is to not include galleries, and the max bonus going to articles with at least 50 interwikis. Some people want only one bonus per article, some think otherwise. Also, the thresholds where to put the bonuses aren't very clear. Personally, I would favor 10/30/50 interwikis. Nergaal (talk) 14:17, 23 November 2014 (UTC)

The scaling on interwikis is a little funny, 5/20/50 probably works better, since you can be very notable and have less than 30 (Much Ado About Nothing (29), W. S. Gilbert (24), Charles Cornwallis (28), Battle of Stirling Bridge (29), Roanoke Colony (21), James Longstreet (26), The Bacchae (24)) - all of these seem like things we should be giving the second tier to, so 20 seems a better point above which fairly important things start appearing, not 30.

I'm willing to be convinced on the 5 vs 10 issue, but think that list is fairly good evidence of why 20 is better. I prefer 5 as that's a good cutoff for "It's probably at least somewhat notable", which is likely what we're going for at the first tier. I'd be happy enough with 10/20/50. Adam Cuerden ^(talk) 16:49, 23 November 2014 (UTC)

FP bonus issues

While I'm not fundamentally opposed to reworking FPs to include bonuses, I have grave reservations about just how practical it would be. Unlike an entire article, images are easily deleted from an article and people can differ drastically how on appropriate a given image is for an article. I've seen edit wars over images in certain volatile articles, so how can we (the judges) be sure that it really is appropriate for the infobox or whatever? The thing is that the FP reviewers assess the image in a fairly objective way, but use of the image is entirely objective and editors will differ in their opinions so that the very next edit could remove it from the article on one basis or another. Forex, somebody adds a FP of a German tank to the Battle of Kursk article and then the next editor substitutes a picture actually taken during the battle that is kinda fuzzy. So which is more "important" or "valuable"? My answer would probably be the latter, but that reflects my bias towards for candid pictures. And if it were the article on the tank, I'd prefer a FP-quality picture in the infobox. Both editors would have valid points, so I'm not sure that we can realistically award points for use in articles as they may not remain there for very long.

And in more general terms, I'm doubtful that the bot can be programmed to figure out how to award the bonuses under discussion. Which would mean that the judges themselves would have to do so. Something that I'd much prefer to avoid on their behalf as assessing Godot13's output could be very time consuming next year, <BEG>.--Sturmvogel 66 (talk) 22:37, 21 November 2014 (UTC)

FPs not used on articles are delisted. If somebody comes and removes a FP from the only article where it is used, then it loses its featured status. Plus, I think there has been a rule for some time requiring the picture be used on that article (therefore stable) for some period before if can be put up for FP. Nergaal (talk) 22:55, 21 November 2014 (UTC)

Nergall is correct. In order to achieve (and keep) FP status, an image's use must to stable. As to the bot issue, as a programmer I can say it is trivial to assign a bonus based on the article(s) it appears in. Whether the infobox criteria is desirable or necessary is still be discussed, but if it were to be adopted a bot could detect it properly with a bit of work. The only case that would require judge approval is "important, but not in lead/infobox". --ThaddeusB (talk) 00:15, 22 November 2014 (UTC)

I didn't know that FPs already have to be in an article to be an FP; thanks for educating me on the requirements and that a bot can award the points appropriately. So strike all of my above comments.--Sturmvogel 66 (talk) 16:24, 23 November 2014 (UTC)

Penalties for delisting

I don't think this would have a large impact, but if some editor gets points for say a GA or FP and that item loses its GA or FP status, should the user get some penalty (say equal to the initial score he got)? Nergaal (talk) 08:56, 22 November 2014 (UTC)

This already happens, at least in the same round. Adam Cuerden ^(talk) 09:19, 22 November 2014 (UTC)

I know I had a DYK that ran but then got deleted and I lost all the points for it. So it makes sense that if GA or FP status is lost during the competition (not likely though I must admit) then the points get lost too. The C of E God Save the Queen! (talk) 09:43, 22 November 2014 (UTC)

Well, in the most trivial possible sense it happened to me this year: Just after my restoration of the lead image of Franklin Pierce passed, I was talking to the author of the featured article about it, started saying how it was a shame there was the huge black area at the top... realised that, for once, a crop was really justified, and so delisted and replaced about two days after closure. Though, that said, had I tried to claim both the delisted original AND my replacement, I'd expect to be slapped upside the head, and quite rightly too.

A more likely case is late-identified copyright issues. Those are terrible. It's why, if you've ever got an FAC image review from me, you'll find I am very thorough: I just apply the same things I do before putting all the work into a restoration.

I think that, once a round has closed, though, we should probably draw a line under it, except in cases of outright fraud. If a GA from round 1 gets delisted in round 3, you probably shouldn't take a negative thirty point penalty, and it's probably too late to reconsider whether you should have been promoted in round 1 after all. Adam Cuerden ^(talk) 10:09, 22 November 2014 (UTC)

Of course if an item is delisted\deleted before a round deadline, it shouldn't earn points. I would also agree that if it happens after the round is over, it shouldn't cause negative points for the round it was demoted. I would assume that is the status quo. --ThaddeusB (talk)

Think the only other case that could be problematic is something like the Franklin Pierce thing I identified above (say, promoted in round 3, then replaced in 4), but I think that said replacement in round 4 would so clearly NOT be worth points that any attempt to claim them would be outright cheating. (Not that Delist and replace can't be a valid way to gain points: Consider Wikipedia:Featured picture candidates/delist/Walt Whitman, where a low-res 2005 FP was replaced with a high-res restoration. But if you're delisting and replacing things you got points for already in the same Wikicup....) - Anyway, I don't think we need a rule to forbid something so obviously, fundamentally cheating. Adam Cuerden ^(talk) 11:48, 23 November 2014 (UTC)

I can confirm that this does already happen when something is delisted in the same round it is promoted. As the C of E says, deleted DYKs are also removed (I think, though I can't remember for sure, that I have also removed GA reviews after the result was challenged). Naturally enough, it doesn't come up too often; I don't think it belongs in the rules (once something is written down, people get lawyer-y about it...) but these kinds of cases are why we have judges. J Milburn (talk) 12:46, 23 November 2014 (UTC)