Wikipedia

Wikipedia talk:WikiCup/Archive/2022/1

< Wikipedia talk:WikiCup
Latest comment: 2 years ago by SL93 in topic Rebalancing Featured content


On Featured Article points

Latest comment: 2 years ago36 comments13 people in discussion

First of all, I'm going to be using the last couple years' Wikicup final rounds for this, but I don't want to deny the legitimacy of the achievements documented there. None of this is personal, it's simply a follow-up to the peer review comments above.

So, Featured articles are worth 200 base points, and this can be multiplied up to three times. In addition, an extra 35 points (plus the same bonus) is available part-way through this process for GA, and additionally 5 or 10 points at DYK, again times bonuses. This gives articles a maximum score of (200+35+10)*3 = 735 points. Plus potentially another 20 points from Good/Featured topics, though I'm not sure if anyone has ever claimed for both types of topic in the same round. Theoretically more given overlapping topics. They may also generate review points.

This can be awarded to multiple participants. Last year, the gold and silver medallists shared 3 FAs, all with large bonuses, and each received full points.

Let's compare Featured lists. Lists rarely work well with Wikidata, so bonuses, if any, tend to be low. Lists are, as far as I can tell, ineligible for GA. If they're generally accepted for DYK, I see little evidence of this. So a Featured list is probably capped at pretty close to the base 45 points, plus review points.

Featured pictures get 30 points. I believe I'm at least in the top five highest number of FPs on Wikipedia, and my best year was just short of 100. There are no bonuses or review points available. Also, I don't talk about it much, but I'm disabled, so have a lot more time to dedicate to this. I think the only time FP points reached the level FA ones regularly do is the year Godot13 was collaborating with the US Treasury and releasing gigantic (30-100 item) sets of bills and coins.

I think we're kind of getting to the point of needing a rebalance. FAs are certainly right to be worth a lot of points, but multiple people getting 400+ points for a single FA (not even counting GA points) seems to devalue everything else, and while FA should be an accomplishment, well worth points, maybe not that many?

Adam Cuerden (talk)Has about 7.5% of all FPs 11:02, 20 January 2022 (UTC)

While it's a lot of work to get an article to FA, in my experience after doing several FAs it's more on the order of twice as much work as a GA article rather than 5.7 times as much work. The high number of points to FA in my opinion does not necessarily benefit either Wikicup (as outlined above) or FAC. As stated above, I'd support reducing the number of points for FAs down to 100 or 150. (t · c) buidhe 00:10, 21 January 2022 (UTC)
How much work it takes to get an article to FA status relative to GA depends on the topic, the content area, and the GA reviewer (medical GAs are pretty much across the board meaningless, while an FA can be years of work), but I generally agree with the gist of Buidhe’s post. The high number of points assigned by WikiCup to FAs has affected FAC in not always optimal ways.
I am just thinking out loud on the following proposal … not a fan of WP:VITAL (how is it possible that a condition like autism isn’t even level 5?), but it does give us some relative measure of articles, and there are articles that reflect years of effort.
What if the FA points given are scaled, with 200 retained for Level 1 Vital or WP:CORE or WP:MILLION articles, scaling down from there (eg 175 level 2, 150 level 3, 125 level 4, 100 level 5) to 75 for non-vital? And perhaps consider points for a featured article save award, also scaled accordingly, since a lot of “core” article work is happening at FAR and a FAR “save” is becoming the same as a FAC promotion. SandyGeorgia (Talk) 14:36, 21 January 2022 (UTC)
I like this proposal. A. C. Santacruz ⁂ Please ping me! 14:44, 21 January 2022 (UTC)
Still thinking out loud, by adding WP:MILLION to the scale, you pick up articles like autism that the vital people have overlooked. Writing an FA on the level of autism is a lifetime of work (as indicated by its loss of the star when its main watcher stopped editing). SandyGeorgia (Talk) 14:51, 21 January 2022 (UTC)
I'd support reducing points for FA from 200 to something less, but not 75. Quite honestly, 75 is less. – Kavyansh.Singh (talk) 16:42, 21 January 2022 (UTC)
One issue is that you can have 50 GA nominations open at the same time, while you can't have several FA nominations open, and FAC usually takes longer than GA once somebody has decided to start reviewing. Dropping to just double or triple GA points for FA could make FAC rather unattractive for WikiCup participants. SandyGeorgia's idea of more points for core is already somehow encoded in the bonus system, although it is not equivalent. Personally, I have invested more time in my most recent FA (first in over a decade) than in a handful of GAs.
Another point is that Adam's computation is a bit exaggerated. I don't believe that there are any articles with a 3x multiplier that are currently available to create or 5x expand. For such articles, getting them through GA and FAC within one round of the Cup also is quite difficult. —Kusma (talk) 14:48, 21 January 2022 (UTC)
I think I’ve figured out where bonus points are covered, but I may have missed it; I’m not finding anything similar? SandyGeorgia (Talk) 14:55, 21 January 2022 (UTC)
I agree with this assessment. I never went through FA before so I didn't realize that it wasn't unusual for my FA so far to have lasted around two weeks with three out of three supports. I do have another article I want to work on soon to nominate for FA, but I can't nominate it until the first one passes or fails. SL93 (talk) 14:58, 21 January 2022 (UTC)
@Kusma: While I'll agree the maximum is unlikely to be reached regularly, last year's Wikicup final round had a lot of sports articles in the FAs leading to several ones in the 2 to 2.6 range, with the three FAs shared between Bronze and Gold I mentioned being 2.2, 2.4, and 1.2 multipliers. 2020's silver place participant had two 3x multiplier FAs on major wars, and also claimed one of those for GA in the same round, and it was part of a featured topic, which is 720 points right there, and pretty close to the maximum I mentioned. Again, I'm not bringing this up to criticise anyone, I just want to use actual data. Adam Cuerden (talk)Has about 7.5% of all FPs 16:25, 21 January 2022 (UTC)
I wouldn't mind seeing the maximum multiplier reduced a little bit, but 400 points for bringing an article about a country or a world city to FA status does not strike me as being widely off the mark. Giving 600 points to each of two joint nominators may be a bit much, though.
(You should note that my comments are perhaps not extremely well informed, as this is the first time I am participating in the WikiCup, and I expect my real life won't allow me to play to win). —Kusma (talk) 16:42, 21 January 2022 (UTC)
@Kusma: The thing about articles in the Wikicup is that you can claim the same article for points over and over, and that's not entirely a bad thing, but it's also not true of, say, Featured lists or Featured pictures. A featured picture is worth less than an unmultiplied GA, but that GA might also be getting DYK, FA, GT, or FT points. Likewise, a featured list is only worth slightly more than a GA, but I'm not seeing any lists at GA, and, though I honestly can't find confirmation, I suspect they're ineligible for GA. They might be possible to add to a topic, though, so there's that. Basically, while I support celebrating FAs, we've seen that people can get 8 FAs in a round, and regularly do in final rounds. That... kind of makes for one winning strategy to the exclusion of other content, though. Adam Cuerden (talk)Has about 7.5% of all FPs 16:47, 21 January 2022 (UTC)
@Adam Cuerden, do you think there could be a reasonable "multiplier" rule for FPs to make the game more even between different types of content? —Kusma (talk) 16:51, 21 January 2022 (UTC)
@Adam Cuerden: The unwritten rule here is that lists which have less than nine entries, or have a significant amount of prose, it can be a GA (Example: List of Puerto Rican flags). That being said, I sort of agree that FLs should have more points that 45. – Kavyansh.Singh (talk) 16:56, 21 January 2022 (UTC)
I also agree that FLs being only 45 points is too low. A few examples of lists that I think are just fantastic and I'd feel discouraged to create if I only got 45 points relative to an article:
There's definitely more, but thought to include just a few. A. C. Santacruz ⁂ Please ping me! 17:10, 21 January 2022 (UTC)
I've made a mockup of the current ideas we have cooking here (I marked what's new in green). Basically, the featured article rules, the peer review addition, and something else for featured pictures. For now I put the WP:MILLION rule (10 points bonus if the article gets one million views annually, excluding the other three WP:MILLION awards), but this can be removed if decided otherwise. Panini!🥪 16:15, 21 January 2022 (UTC)
@Panini!: I'm not sure "Image restoration and passing the featured picture process are not one in the same; this means, if you were to restore an image, but it is nominated and becomes a featured picture two years later, you will still be awarded points no matter the length of time in between." should be in the rulebook. It might be true that that's the judge ruling, but it feels like it should still be outwith the spirit of the Wikicup for me to not nominate anything this year, and then win the 2023 Wikicup with all the images I restored. Adam Cuerden (talk)Has about 7.5% of all FPs 16:38, 21 January 2022 (UTC)
I've strikes it for now, mainly because your reasoning could possibly be enforced. Like GAR, if this exact idea becomes an issue we can simply disqualify, but I'd like to hear what the judges think about this. Panini!🥪 17:12, 21 January 2022 (UTC)

I think as a baseline, 200 points for an FA is fine. An FA is probably 5-10x the work of a GA (maybe more, if you consider some of the shorter GAs.) The problem, I'd say, is with the multipliers. The most you can get for a GA is 105 points, just over half a standard FA. The most you can get for an FA though is 600 points, which is a lot. Maybe total bonus points earned per nomination should be capped in the interest of fairness. Guettarda (talk) 16:29, 21 January 2022 (UTC)

Well, there is actually no accurate way to say that a FA take either 2 times or 10 times the work in comparison to a GA. And if you feel that 200-35 ratio is fine, then 600-105 should also be fine. – Kavyansh.Singh (talk) 16:41, 21 January 2022 (UTC)
Most of the FAs that garnered points last year and IIRC the year before were not on articles that are "5 to 10 times" the work of a GA relative to most of the project. Anyway, I may regret suggesting that these issues extend to FAR saves, for the same problems that have occurred at FAC. Unwatching now, thanks Panini, and please ping me if anything advances to a point where I might want to weigh in again. SandyGeorgia (Talk) 16:53, 21 January 2022 (UTC)

I've made this chart:

Year Number of finalists with this many FAs
0 1 2 3 4 5 6 7 8
2010 4 1 1 1 1
2011 5 1 2
2012 3 1 3 1
2013 3 1 1 4
2014 4 1 2 1
2015 4 1 1 2
2016 5 2 1
2017 4 1 3
2018 6 1 1
2019 4 3 1
2020 2 2 1 1 2
2021 1 1 2 1 1 1 1

As you can see, the number of FAs in the final round has gone up drastically. And I'm not sure how or why. @SandyGeorgia:, if you'll forgive the resummoning, has anything changed since, say, end of 2019? Adam Cuerden (talk)Has about 7.5% of all FPs 17:14, 21 January 2022 (UTC)

Made a similar table at User:A. C. Santacruz/Wikicup stats. IIRC 2013 or so is when multipliers were added, but idk how that affects the data. I'll try and expand it to earlier rounds as well, see how the table changes. A. C. Santacruz ⁂ Please ping me! 17:34, 21 January 2022 (UTC)
I think it's gone up because finalists plan to nominate FAs during those rounds to maximize their chances to win the Wikicup. I know this contributes to the higher number of nominations at FAC from June-October. I think getting articles to FA status is a net positive for Wikipedia, but the high amount of points for an FA makes it very difficult for an editor to reach the final round of the cup without having an FA, let alone having a chance to win the cup. Z1720 (talk) 17:40, 21 January 2022 (UTC)
Z1720 I agree on it being positive to Wikipedia to have more FAs. However I think that that benefit is not equivalent to 6 GAs. On the final round bias, I agree so I'll expand the table soon once the LEC finishes today. A. C. Santacruz ⁂ Please ping me! 17:46, 21 January 2022 (UTC)
Evolution of the ratio between FA and FL submissions over time, by year
Made table based on results, for full data see my user subpage linked above. Data from summary tool.A. C. Santacruz ⁂ Please ping me! 19:26, 21 January 2022 (UTC)
I think the reason why we doubled up FA to 400 was because of that year where Godot smashed it with the FPs (and fully deserved too, he did very well that year!) and a lot of people complained (rightly or wrongly). But bringing it back down might be a good idea. The C of E God Save the Queen! (talk) 19:41, 21 January 2022 (UTC)
Perhaps then the criteria should be changed on the character of the FP? I'm unaware if WP has Featured picture sets but I know they exist in Commons. Perhaps awarding a maximum amount to a Featured picture set (e.g. 120) divided between submissions that are part of that set would be one way to do so. A. C. Santacruz ⁂ Please ping me! 19:48, 21 January 2022 (UTC)
I agree that the points may need to be adjusted slightly (potentially the multiplier is a bit broad), and that FLs in particular are too short on points. They don't seem too viable to write, unless you do them on mass. Last year I managed to complete a career wikipedia ambition, and got snooker to FA. This did get the 600 points, but the vast majority of the articles I work on don't have much, if any multiplier, such as the 2020 World Snooker Championship. I don't think these articles are any less viable, nor that they are somehow worth less. It is, however, easier to write about a tournament that took a couple weeks, than the whole history, rules, format and lineage of a sport. I do think the wikidata rules, however, do give a good tilt towards things being bigger priority, but the discussion above suggests that we are trying to get users to write and update certain types of articles, which I'm not going to agree with. I've written 21 FAs, and I wouldn't have written a single one if it weren't for the cup. In my eyes, the cup each year gives the encyclopedia an improvement, so the points are kind of irrelevant. Yes, there is a winner at the end, but it is purely for the good of the site.
In terms of numbers, 150 would be better, or 200, but instead of the multipliers being x1.2,x1.4, etc, we would go up by 20 points each time. A current 3x multiplier would instead get 400 points. I'm not sure how the vital list would work, as giving out 200 points for working one of the ten articles to FA is ridiculous. I would also suggest that, as FLs don't get much in the way of interlanguage links, that they would be more suitable for say 70 points. I have no opinion on the FP list, other than in 2019, the winner only posted FPs, with zero prose, so I don't see how it is under valued.
Even if the points were changed so that I was no longer competitive, I think I would still do the same content I do now. Best Wishes, Lee Vilenski (talk • contribs) 20:03, 21 January 2022 (UTC)
Coming late to this discussion, there is one thing that happened in the 2021 WikiCup which I really didn't like. Some contestants were asking the FAC coordinators for permission to post a new nomination before the previous one was formally passed. This potentially gave an advantage to "old hands" over less experienced FAC nominators, giving them the possibility of scoring higher marks than would otherwise have been possible. Cwmhiraeth (talk) 19:51, 22 January 2022 (UTC)
Hearing that saddens me a bit. I don't like to hear that editors are trying to skirt rules to gain internet points. We're all here to increase the free global access to quality information, not just to beat others in a contest. The contest should be a way to motivate editors to make more quality content; a means and not an end. Are these the kinds of issues SandyGeorgia and buidhe have been experiencing at FAC mentioned above? A. C. Santacruz ⁂ Please ping me! 22:25, 22 January 2022 (UTC)
I observed worse than what Cwm describes, but that’s not the only problem. SandyGeorgia (Talk) 23:20, 22 January 2022 (UTC)
Hardly skirting the rules, as this is very much the norm and allowed as Buidhe says. Best Wishes, Lee Vilenski (talk • contribs) 23:24, 22 January 2022 (UTC)
Such requests are allowed and routine for all FAC nominators regardless of participation in the wikicup. The requests are only granted if the nomination is close to promotion, although I agree that some newer participants may not be aware of this rule. I don't see this as the problem, rather I'm more concerned about a risk of quid pro quo favorable reviews and consequent slippage in quality standards. (t · c) buidhe 22:35, 22 January 2022 (UTC)
That, and WikiCup participants pressuring Coords for contest reasons (not that the Coords succumbed to those pressures, but they still should not be exerted). SandyGeorgia (Talk) 23:22, 22 January 2022 (UTC)
Where exactly have coords been pressured? Best Wishes, Lee Vilenski (talk • contribs) 23:24, 22 January 2022 (UTC)

Reminder

Latest comment: 2 years ago6 comments5 people in discussion

Currently, 35 people have 0 points in this round. As we know, anyone who gets points in this round will likely move on, and one of the easiest ways to guarantee points is to do a good article review. There also just so happen to be ten days remaining in the January 2022 GAN Backlog Drive. Anyone have an objection to me dropping a note to the aforesaid 35 participants, suggesting the backlog drive as a way to grab some WikiCup points? --Usernameunique (talk) 02:25, 22 January 2022 (UTC)

Nope. No objections from my side. The only minor request that I would have is that we emphasize the level of attention to detail and other expectations from a WP:GAN. I have recently seen less than our best there. But, I am sure participants here will already be attuned to the expectations. Thanks for checking here. Cheers and good luck. Ktin (talk) 03:49, 22 January 2022 (UTC)
You are welcome to leave a message on people's talk pages, although a few participants are novices and unlikely to be competent to review anything. Cwmhiraeth (talk) 19:34, 22 January 2022 (UTC)
Sounds good, just dropped a note. Example here. --Usernameunique (talk) 19:48, 22 January 2022 (UTC)
More than happy to chip in once my submissions page is up and running! :) Kingoflettuce (talk) 23:17, 22 January 2022 (UTC)

Might've been a bit premature. Haven't even got the end of January, after all, and FPC has a ten day turnaround. Otherwise, reasonable enough. Just weird to get a note asking me to claim points on the same day the lengthy process I went through for points have them to me. But, y'know, harmless and enthusiastic is good. Adam Cuerden (talk)Has about 7.5% of all FPs 23:49, 22 January 2022 (UTC)

Avril de Saint-Croix

Latest comment: 2 years ago5 comments3 people in discussion

Talked about this up here, but never got a reply. As it's going to be passing soon, would really like some sort of yes or no reaction. Adam Cuerden (talk)Has about 7.5% of all FPs 16:17, 19 January 2022 (UTC)

I once queried this when I was a competitor. Now that I am a judge, I would allow it, the criterion being the date it becomes a featured picture and not the date on which the restoration was done. How would anyone except the restorer know when the restoration was done? Cwmhiraeth (talk) 20:27, 19 January 2022 (UTC)
Well, um... The fact it was uploaded in 2019 and unchanged since is a big clue. Adam Cuerden (talk)Has about 7.5% of all FPs 04:19, 20 January 2022 (UTC)
A mystery indeed. We'll never know... Panini!🥪 13:04, 20 January 2022 (UTC)

Well, I've claimed it. I'd feel more conflicted if two more definitely from this year wouldn't be joining it shortly. Adam Cuerden (talk)Has about 7.5% of all FPs 23:50, 22 January 2022 (UTC)

Rebalancing Featured content

Latest comment: 2 years ago52 comments17 people in discussion

If you are struggling to follow along at home, please look here for a strawpoll.


Since we're discussing early this year, how about an early proposal?

I think when we vote on changes, they'll need to be done as a package, as we don't want to reduce FAs, while in a separate proposal we raise FLs and FPs until it's a mess.

So. I'm going to be trying to go off of the GA/DYK points. An article can get, before bonuses, 35 points from GA, with a bonus from DYK if you started the article. FLs are, from what I can tell, largely outwith the bonus process, and FPs completely so.

So. Here's my proposal:

FLa are increased to 60 points. FPs go up to 40 points. I shouldn't put any restrictions on set moms because the only time they've mattered they've either been fairly small or related to a Wikipedian personally getting archive access for Wikipedia and preparing pristine quality images for upload from that archive. And, frankly, if you manage to do that, just... have your Wikicup win, okay? Being too harsh with sets encourages incompleteness. Consider the complete set of illustrations from, say, The Hunting of the Snark vs the single image we have from The Wizard of Oz. And yet, it would have been a near equal amount of work to jump around to several different books for illustrations.

Correspondingly to these raises, FAs drop to 150 points, and we see where that gets us. If we think they should go lower, we can reduce further later. I

Thoughts?

Adam Cuerden (talk)Has about 7.5% of all FPs 00:17, 23 January 2022 (UTC)

Just to confirm I understand, the new points would be:


Featured article Featured list Featured picture Featured topic Featured article review Featured list review Good article Good topic Good article review Peer review Did you know? In the news
200->150 45->60 30->40 15 per article 5 5 35 5 per article 5 10 5 or 10 12
I'd be ok with that, although I will say that you don't just get a DYK for creating a GA, there's time and size issues with that. Best Wishes, Lee Vilenski (talk • contribs) 00:26, 23 January 2022 (UTC)
Considering the time constraints and restrictions with ITN, I think it would be fair to bump it up to 15 points (and also eliminate non-bonus points that end in 0 or 5). Also, I neglected to participate in the earlier discussion, but I don't think the FAs should be reduced too much in light of the amount of effort needed; a FP is also less accessible to editors who don't have good camera equipment or editing software that taxes most PCs. SounderBruce 01:41, 23 January 2022 (UTC)
Off topic, but you know I'm using a five year old laptop without a dedicated graphics card, right? And was previously using a 15-year-old XP desktop. I just use GIMP. But on the main point, I don't think 150 is too precipitous of a drop - I'm trying to be a bit conservative in point changes, and, of course, and GAs and FAs may well get bonuses atop thisAdam Cuerden (talk)Has about 7.5% of all FPs 03:13, 23 January 2022 (UTC)
I think we should implement some kind of guardrail so we don't have a Godot situation again. Perhaps a maximum number of FP points of the same kind per round? Something like, Featured pictures submitted by the same user within one topic (such as Currency, Prints, or Portraits) may receive a maximum combined points of 1000 per round. or Featured pictures submitted in collaboration with a single entity (such as the Bureau of Engraving and Printing, the Smithsonian, or the U.S. Treasury) may receive a maximum combined points of 1000 per round. This would be a nice way of preventing systematic uploads from gaining too many points (while still being valued, as they're great work), such as in the Godot case, without penalizing more piece-by-piece work done individually, such as in the Adam Cuerden case. What are y'all's thoughts? A. C. Santacruz ⁂ Please ping me! 08:59, 23 January 2022 (UTC)
That... Just feels like a way to punish people who focus on one topic. In the end, the point of the Wikicup is to encourage content creation, not to tell people to stop. That's the equivalent of saying "No more than three sports FAs per round." The thing is, Godot didn't win by that much, despite resources I'm not expecting to see here again. If the actual point here is "FP should never be a viable way of winning the Wikicup" then just remove it from the Wikicup, because it does feel like anytime someone does well in the Wikicup with FPs, this gets pointed to as a "problem", and punishing "solutions" are implemented. Adam Cuerden (talk)Has about 7.5% of all FPs 16:11, 23 January 2022 (UTC)
I'm not saying that FP should not be a viable way of winning the WikiCup, I just don't believe a single partnership (e.g. the BEP) should give multiple thousands of points. Also, Godot did win by much (in my opinion) in 2014, when he got 2.55x the amount of points as the next closest in the final round. In 2015 he was 588 points ahead of the runner-up who had over 4000 bonus points from their 4 FAs and 15 GAs. 3rd place that year had ~4000 less points than either of them.
Yes, FPs take a serious amount of work and convincing such an institution like the BEP to provide the resources to add those images to Commons is a massive undertaking I greatly admire. FPs should be valued more than they currently are in the Wikicup is something I strongly believe. But I think FPs that are part of large series of pictures should not be valued as much as isolated FPs in terms of points. There surely is a better way to regulate that than what I suggested, but it's just an idea. A. C. Santacruz ⁂ Please ping me! 21:08, 23 January 2022 (UTC)
  • Take my words with a pinch of salt since a) I do not really 'compete' in this event; I just pop-in my regular entries into this project to see how far it goes, and b) I probably am amongst the newest to this project and hence might lack any historical context that has resulted in where we are right now. With those caveats out of the door, here goes:
  1. I think FA reduction from 200 to 150 is alright. But, the real deal I think is in the multipliers. A 2.5x multiplier on 200 points is quite a lot. I am assuming the multiplier counts across multiple wikipedias and is a proxy for an article's significance. But, I think a better proxy might be in order. Assuming you want to reward for effort, multipliers do not really reflect effort as it stands right now. Bonus points for size perhaps does and I think you guys already have that.
  2. Yes, FP has a few entry barriers viz. a) access to a good camera, and b) good image editing software (GIMP, PhotoShop, etc.). But, I do not understand the effort that goes in. So, I shall not comment on the points being awarded to FPs.
  3. In a minor point, I recently learnt that WP:DYK is not eligible for 'points' if it is based on a WP:GA article. At 5 points, I assumed the points were reflective of the effort that went into championing the article through the WP:DYK process as opposed to work done for expanding the article. But, it seems like it is actually for the latter. So, I would think more points might be due there.
More later. Good luck. Ktin (talk) 19:26, 23 January 2022 (UTC)
I'd appreciate if you could you expand on Assuming you want to reward for effort, multipliers do not really reflect effort as it stands right now., Ktin. A. C. Santacruz ⁂ Please ping me! 20:45, 23 January 2022 (UTC)
Sure. Why does FA have 200/150 points while GA has 35 points? The answer I think is because, on the input side, FA requires more effort in a) fleshing out the article b) championing the article through the review process. And, this effort results in a much more valuable article (FA>GA from a measure of value) on the output side. Now, I am assuming that at the time that y'all worked on the points, someone determined that the effort in building out an FA (completing both actions a and b) required somewhere between 5-6x the effort of getting the article to a GA status. Hence the 200/150 points for FA vs 35 for GA. This is good. Now, there is a notion of multipliers, that gets you up to 3x the base-points based on the number of Wikipedias that an article is present in. Wikipedia:WikiCup/Scoring. The multiplier is not a proxy for effort, but, is a multiple that gets decided based on the number of wikipedias that the article is present in. So, assuming an article is present in 50+ wikipedias, one gets 3x the points. i.e. 600 points in this case. The base 200 is for effort whereas the 400 is not linked to effort but is based on significance of the article. Does that make sense? Good luck. Ktin (talk) 21:23, 23 January 2022 (UTC)
My memory is that the high points also reflected an assumption that, thanks to the number of nominations permitted at a time, 2 FAs would be the maximum someone could get under normal circumstances. Adam Cuerden (talk)Has about 7.5% of all FPs 21:26, 23 January 2022 (UTC)

Maybe someone can answer this, but I have been wondering why being on more Wikipedia websites are worth extra points. I don't think that necessarily means that those articles are more valuable. SL93 (talk) 21:36, 23 January 2022 (UTC)

I believe it's meant to be a proxy for how quote unquote "important" an article topic is. But that's a very imperfect metric at best. Something extremely important to, say, Indonesia might be totally ignored on other language Wikipedias, while something we might consider far less important, such as some random celebrity, might have articles on a dozen different Wikipedias. But of course that begs the question, how do we define importance? Having the core articles is a start but those are by necessity arbitrary and subject to bias. I'm not sure if other language Wikipedias have an equivalent to our vital articles system or not. Essentially any interpretation we use to determine importance will be flawed in some way. My guess is that the "number of Wikipedias an article is present on" was decided as the method with the most upsides and fewest downsides. It does work to some extent, for instance essentially every single Wikipedia is going to have articles on Earth or Gravity or Human. I don't claim to have the answers, but I wonder if those more experienced might have further thoughts about this. Trainsandotherthings (talk) 21:54, 23 January 2022 (UTC)
I agree with the above post from Trainsandotherthings. Another metric (and I know this is going to be a polarizing metric) that could be used is WP:PAGEVIEWS. Assuming that an article is important because more people are looking for that article. Alternately, another view of importance is using WP:VITAL designation. Irrespective, I think the upto 3x multiple for FAs might be worth re-examining, particularly if the base points are as high as they are right now. Ktin (talk) 22:10, 23 January 2022 (UTC)
This, Ktin, is similar to a proposal SandyGeorgia made a little bit back. I've struck this from the draft of the current proposals because it's fallen out of the discussion's scope, but I'm all for introducing it back. Panini!🥪 12:35, 24 January 2022 (UTC)
There are only a few heuristics we can use if we wish them to apply very broadly to a large number of articles and retain the heuristic's accuracy over time. I think interwiki articles is the best metric of those, and so we should keep it as is. Length is the next-strongest contender, but I think it is not a great indicator of how much work has been put in, and it's possible to be manipulated artificially. I would expect en.wiki pageviews to have more geographic systemic bias, while WP:VITAL applies only to a very small fraction of articles.
The whole points system, including those given for FAs, FPs and GAs, is also a heuristic, and in many cases scores will not represent how much work someone has put into WikiCup submissions, nor how much value it has added to Wikipedia. This is fine, and the rules may be improved here, but we should remember that we are not aiming to create a situation where whoever works hardest will win.
I agree with reducing the points for FA, and I have no objections to increasing them for FP or FL. Increasing the points for DYK by 5 (to 10 or 15) would fit well with my experience of DYK vs GA effort. Ktin makes a good argument for this. — Bilorv (talk) 18:49, 24 January 2022 (UTC)
  • To make scores more reflective of their articles value, I do like using pageviews. I think then we would take the average monthly page view over the last year, divide it by some factor to make it a manageable number, round it off so we get nice even numbers, multiply the score by it, and bob's your uncle. It would entice people to quickly look at at articles XTools and think "what is the average impact of this article". CaptainEek Edits Ho Cap'n!⚓ 19:33, 24 January 2022 (UTC)
Page views has the issue that it doesn't work for brand new articles, and does tend to promote WP:RECENTISM, and popular articles of today, but not necessarily items that have the biggest impact. Best Wishes, Lee Vilenski (talk • contribs) 20:54, 24 January 2022 (UTC)
I agree with Lee Vilenski. However, I also think that the number of wikipedias might tend to benefit topics like sports or movies rather than other types of articles, so it encourages competitors to work in these areas for maximum points (see for example Rambling Man's multipliers last year). Not that I think it has too big of an effect on the competition (competitors will tend to work in areas they're most passionate about anyways), but it is something to note. Another possibility would be the age of the article, perhaps? But I think that brings issues of its own. A. C. Santacruz ⁂ Please ping me! 21:28, 24 January 2022 (UTC)
A perhaps naive comment from my side -- why even consider a multiplier? If the thinking is that we need to reward effort (in a. expanding the article and b. championing it thru the review process), just using the base points should suffice. If needed add a bonus for size (rather than multiply) to ensure that efforts commensurate with volume are adequately rewarded. Yes, this approach will not reward for 'significance' in whatever form we eventually think of measuring. But, I think that is alright. Definitely, can see why it need not be a 'multiplier'. More later. Cheers. Ktin (talk) 22:39, 24 January 2022 (UTC)
I agree with using a base increase rather than a multiplier. A. C. Santacruz ⁂ Please ping me! 22:40, 24 January 2022 (UTC)
Multiplier is easier to apply to all kinds of articles (FA, GA, DYK, ITN). But multipliers above 2 are perhaps a bit exaggerated. —Kusma (talk) 14:13, 25 January 2022 (UTC)
  • I've been wanting to propose a similar decrease in the points for FA. The points should reflect editor fun, reader benefit, and effort imo, and especially the reader benefit is much more for a poorly written article to go to GA than FA. The current points are also not great for getting a level playing field for core articles. They are often more than 3x as much work, and it would be great if the multiplier could go up to 4x or 5x for articles above 50 languages. Reaching 50 languages is quite easy now, but there is a difference between articles that exist is 50 languages or a hundred. The fact that the 3x multiplier is used so rarely is an indication it's not enough of a boost. Femke (talk) 18:59, 25 January 2022 (UTC)

Arbitrary ease of reading break

Reading all this makes me wonder about GAs. Because they require "broad by not comprehensive" coverage of a topic, one GA may be less than 1,000 words while another that's over 5,000 words might fall short of this criterion. Would there be benefit in making longer GAs worth more point, in the way that longer DYKs are worth more points? Guettarda (talk) 16:03, 25 January 2022 (UTC)

That's a very interesting idea. I certainly think an article like Jean Batten should be worth more points than Nashua, Acton and Boston Railroad (the latter is one of mine, there's not a ton that can be said on the subject so it's fairly short). It's not that short GAs aren't worthwhile, but longer ones certainly take more effort and provide more to the reader, so I think some kind of bonus would be helpful, within reason - we don't want to encourage people to artificially inflate article length solely for more points. Trainsandotherthings (talk) 16:27, 25 January 2022 (UTC)
Long GAs are sort of their own punishment, since they provide reviewers more opportunities to nitpick ;) Guettarda (talk) 16:32, 25 January 2022 (UTC)
You make a valid point and I definitely agree. Not only are smaller articles much easier to do they're far more attractive to good article reviewers (would you rather review 5550 words or 700?), meaning they're faster to complete and quicker to claim points. I don't feel this needs a complicated multiplier system and can simply boil down to "more words get more points". This brings up another question, then; should good article reviews get more points, considering they take longer to review as well? It looks like my simple peer review proposal has opened a pandora's box of rebalancing... and re Trainsandotherthings, while articificial fluffing could be a problem, that stuff is usually caught in a good or featured review. Panini!🥪 17:54, 25 January 2022 (UTC)
I don't think longer article reviews should get more points, given that FA reviews currently also only get 5 points. I think rebalancing those points would be too complicated. Guettarda (talk) 17:59, 25 January 2022 (UTC)
Some sort of rebalancing is needed. I don't agree that 1 FA should be worth the same as forty FAC reviews! Trainsandotherthings (talk) 18:01, 25 January 2022 (UTC)
I was just talking about points for reviews. Though I'm not sure I agree with you that the ratio of FA reviews to FAs is wildly out of balance. Guettarda (talk) 18:12, 25 January 2022 (UTC)
I agree the some sort of re-balancing is needed. I know this might have been raised before. Currently, both a GAN review and FAC/FLC review are of 5 points, when, in my opinion, it is much easier and faster to review GANs than FACs. Quite honestly, GA criteria is the simplest (in few cases, even simpler that DYK criteria). There are very limited parts of MOS that are being considered while reviewing GANs. While, in FAC, all parts of MOS are being concerned. This is a normal GA review, very correctly passed when the GA criteria is met. This is the FAC review for the very same article; notice how almost every review is substantially longer than the GA one. If we give 200 to FA and 35 to GA, then why same for reviewing both? – Kavyansh.Singh (talk) 18:14, 25 January 2022 (UTC)
What about 10 points, same as PR? A. C. Santacruz ⁂ Please ping me! 18:26, 25 January 2022 (UTC)
@A. C. Santacruz – Ten sounds good for FAC. But, I assume you are saying that PR should also have 10 points. The problem with PR is that it can be about anything. Often, it is pre-FAC, but there are many cases in which user demands a PR pre-GAN. In that case, it would be more or less gaming the system by giving a pre-GAN peer review, and getting 10 points, when the GAN itself is getting 5 points. I have also seen cases where an user lists an article for peer review just because they need help with copy-editing/improving NPOV. Giving 10 points for that, is quite-honestly a lot. Thoughts?– Kavyansh.Singh (talk) 18:37, 25 January 2022 (UTC)
That makes sense. I think perhaps peer reviews should only give as many points as the goal of the article, e.g. if its pre-GAN give 5 points and pre-FA give 10. Thing is I think peer-reviews like the Milhist A-Class reviews fit in a weird in-between, so I don't know how those would be scored. I think 10 would be possible but leaving it at the judges' discretion to award less. A. C. Santacruz ⁂ Please ping me! 18:41, 25 January 2022 (UTC)
That makes the system a bit more complicated, doesn't it? On another note, Cwmhiraeth, do you judges check each GAN and FAC review entries, if they are long/comprehensive enough? Because, I assume it'll take a lot of time ... – Kavyansh.Singh (talk) 18:55, 25 January 2022 (UTC)
Kavyansh.Singh yeah, I agree it makes it more complicated, but it's just an idea. A. C. Santacruz ⁂ Please ping me! 19:02, 25 January 2022 (UTC)
I'm probably doing my GA reviews wrong, but I don't think all FA reviews are typically twice as much work as GA reviews are. The current system is nice and simple. Anything "fair" would require "Good 'featured article review' reviews" so let's not go there. —Kusma (talk) 19:25, 25 January 2022 (UTC)
Also see this GA and this FAC, which is confusingly larger than the review of a literal presidential campaign. Panini!🥪 18:20, 25 January 2022 (UTC)
Panini! I see your Paper Mario and raise you Nonmetal's GA nom and its level 3 FAC >:) A. C. Santacruz ⁂ Please ping me! 18:29, 25 January 2022 (UTC)
I've thought about this with regards to GAN length as well. My smallest GA is "Uncomfortable", at 845 words. My longest, Brad Marchand, is 5,613. While they both went through the same expansion, review, revision cycle, one clearly took much more time and effort. While I won't stop working on longer articles just because of the cost/benefit factors, some point balance might be in order. — GhostRiver 19:24, 25 January 2022 (UTC)
See Israel the Grammarian for a short FA, for example. A. C. Santacruz ⁂ Please ping me! 19:38, 25 January 2022 (UTC)
Ever seen Miss Meyers – Kavyansh.Singh (talk) 19:40, 25 January 2022 (UTC)
Holy cow I never thought I'd see a FA with less than 1000 words. SandyGeorgia gave it a look so I assume it meets FA but that length's surprising. A. C. Santacruz ⁂ Please ping me! 19:47, 25 January 2022 (UTC)
A._C._Santacruz The criteria for FA does not include length, but rather comprehensiveness for the individual topic based on reliable coverage that is available. SL93 (talk) 21:04, 28 January 2022 (UTC)
SL93, yes, this was pointed out to me at URFA a month ago or so. However, it's still striking to me when an article the length of a middle school essay is considered "comprehensive". I tend to come out from reading those articles feeling like I haven't been given enough context about the article's topic. That's just me, of course. A. C. Santacruz ⁂ Please ping me! 21:18, 28 January 2022 (UTC)
A._C._Santacruz I understand. I was told about length for FA almost a month ago. SL93 (talk) 21:21, 28 January 2022 (UTC)
  • Here's my summary based on a distillation of above points:
  1. FA - consider reducing base points
  2. FA Multiplier - consider doing away with this one in favor of bonus points (add rather than multiply)
  3. GA - consider revising points based on length
  4. FP - consider increasing points
  5. DYK - consider increasing points
  6. Bonus points - currently focuses on length. If alternate measures of significance needs to be considered other than interwiki appearances, consider either of WP:PAGEVIEWS or WP:VITAL (this idea is not fully fleshed out).
  7. GAR / FAR - have not heard any issues with these.

Good luck. May the odds always be with you. Ktin (talk) 20:26, 25 January 2022 (UTC)

To be more specific, see here. Panini!🥪 23:17, 25 January 2022 (UTC)
I think bonus points for longer quality-assessed articles might be reasonable. It could be based on the structure used in the July 2021 GAN backlog drive, where you get X additional points per 5000 words cumulative (between all articles), or it could be per article. I'd like to see it applied to both GA and FA level -- bringing a huge topic to FA is even bigger than bringing it to GA (let me tell you...). I'm not enthusiastic about the plan to reduce FA points -- FAC is both fiendishly difficult and much more consistent. If it were to happen, my suggestion would be that second and later FACs listed for consideration (in a single round? in a set of two rounds? in the whole year?) receive a slightly reduced number of points, as a very rough reflection of the fact FAC gets easier the more you have. Vaticidalprophet 00:33, 26 January 2022 (UTC)
  • REALISTICALLY, as it states on the scoring page, "The rules will not change mid-competition." If there is a consensus for change, @Jarry1250: will need to be consulted on feasibility, and we are looking at the 2023 cup for any changes to be implemented. Cwmhiraeth (talk) 12:13, 26 January 2022 (UTC)
I kind of like that we're talking about it now, though: It always feels like the November/December discussions are really poorly attended, and, honestly, a little too reactionary against the specific results of that year's WikiCup. Obviously, though, no changes until next year. Adam Cuerden (talk)Has about 7.5% of all FPs 12:39, 26 January 2022 (UTC)
Hopefully everyone gets that this is a discussion for the 2023 WikiCup. And I agree, it's probably good to discuss this early, while more people are still engaged. Guettarda (talk) 15:48, 26 January 2022 (UTC)
Retrieved from "https://en.wiki.x.io/w/index.php?title=Wikipedia_talk:WikiCup/Archive/2022/1&oldid=1073329286"