Jump to content

Wikipedia talk:Featured article candidates/archive87

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


Becoming a FAC mentor

Can I encourage editors who have five or more successful FACs behind them to sign up as FAC mentors by adding their names here. I would especially like to encourage editors with single figure numbers of FAs to sign up - they are the ones most likely to relate to the issues first time nominators experience. Mentoring involves as much or as little work as you want it to, but can be extremely rewarding. It can also hone your own article-writing skills. Having a mentor roughly triples the chances of a first time nominator's article being promoted, so consider helping out. Thanks. Gog the Mild (talk) 17:51, 28 October 2021 (UTC)

Another example of diagnosing the problem to be a lack of throughput. Please tell me if a first-time nominator brought a GA such as Pythagoras theorem, Black hole, Desert or Czech language to FAC, who is this vaunted FA-writing community will help them? There is no expertise. I suspect that there are many WPians who take their articles to DYK or GA, but don't bother with FAC because what would be the value of affirmation by people who know nothing about the topic? Fowler&fowler«Talk» 09:01, 29 October 2021 (UTC)
It's actually Pythagoras' theorem. Detail. The Rambling Man (Keep wearing the mask...) 09:09, 29 October 2021 (UTC)
A case in point. Would you like to discuss the spherical geometry section of that page with me? I could evaluate your fitness to mentor the hypothetical nominator. Fowler&fowler«Talk» 09:21, 29 October 2021 (UTC)
I don't believe you are a suitable individual to assess anyone's suitability, but thanks for the generous offer. I'm sure your approach is really engendering a spirit of collaboration and community, I really do. The Rambling Man (Keep wearing the mask...) 09:33, 29 October 2021 (UTC)

Looking for a mentor

I recently(-ish) overhauled Immortality in fiction completely and brought it to WP:Good article status. I was curious as to what it would take to get it up to WP:Featured article quality, so I looked for similar articles that had reached featured status, but I couldn't find any (though maybe I was just looking in the wrong place?). I thought it might be worth giving it a shot, if nothing else because it would be nice to have something to point to and say "this is how we want our X in fiction articles to be written". First-time nominators are encouraged to get help from a mentor, so I'm looking for someone to take on that role (or else tell me that it's DOA and my time would be better spent elsewhere). TompaDompa (talk) 17:59, 28 October 2021 (UTC)

On the face of it, is this even an encyclopedia article? It reads like a term paper, i.e. has too much original research. Fowler&fowler«Talk» 20:50, 28 October 2021 (UTC)
And there it is. Again. I wonder if this will ever be stopped. The Rambling Man (Keep wearing the mask...)
(edit conflict) It's more or less what other encyclopedias such as The Greenwood Encyclopedia of Science Fiction and Fantasy, Science Fact and Science Fiction: An Encyclopedia, and The Encyclopedia of Science Fiction (all cited in the article) focus on, though with less focus on enumerating examples. It's all cited to secondary and tertiary sources (the version before I started editing the article relied almost entirely on the works themselves). TompaDompa (talk) 21:11, 28 October 2021 (UTC)
@TompaDompa: You asked, "tell me that it's DOA." My reply was not a put-down, just a note that we can't choose any topic we want. One step further, can we write an article on "The concept of immortality in the poetry of Seamus Heany?" We can't really. It is a term paper topic, a thesis topic, a monograph topic. When we enumerate examples that we find (albeit cited to secondary sources), we really are engaging in WP:SYNTHESIS. I think. Fowler&fowler«Talk» 21:23, 28 October 2021 (UTC)
Sources such as this, which exactly matches the topic of the article, are perfectly acceptable. I haven't looked at the article, but if the sources are of that quality, there's no reason there can't be a featured article on the topic. Mike Christie (talk - contribs - library) 21:29, 28 October 2021 (UTC)
Without having looked at the article, but just the sources, I am quite sure that an FAC on this topic could be crafted from them. It is a serious issue which has attracted a lot of scholarly attention. The sources used seem to cover a reasonable proportion of those I am aware of. Gog the Mild (talk) 21:39, 28 October 2021 (UTC)
(edit conflict) Agreed—tonally I think it needs work, but what first-time FAC article doesn't. This isn't some TV Tropes nonsense where someone has coat-racked a few things together, the sources do cover the device itself ("Life Without End: A Thought Experiment in Literature from Swift to Houellebecq", "Immortality: The Search for Everlasting Life", "Immortality and Longevity. The Greenwood Encyclopedia of Science Fiction and Fantasy: Themes, Works, and Wonders", "How Cyberspace Signifies: Taking Immortality Literally", &c &c) so there isn't, by the seems of it, an issue of synthesis. We do have featured articles that cover a non-discrete theme based on sources that discuss that theme (Walt Whitman and Abraham Lincoln, for example) so this isn't breaching new ground. I see Mike Christie has already responded but he and PresN have done a lot of work with science-fiction literature at featured level and may be good people to speak to about further sources. 𝄠ʀᴀᴘᴘʟᴇ 22:00, 28 October 2021 (UTC)
I can’t commit to mentoring, I’m afraid, but I’ll see if I can check out the sources over the next couple of days and try to come with more if possible. Mike Christie (talk - contribs - library) 22:12, 28 October 2021 (UTC)
@TompaDompa: I've taken a look at Gary Westfahl's three-pager "Immortality and Life Extension" (2021) that you reference, I think. It is not really a tertiary source on which to base your article's (initial) organization. It is too skimpy, not even bothering, for example, to mention "A Voyage to Laputa ... Luggnagg ..." in Swift's GT. When the tertiary references are skimpy, they need to be replenished and therein the problem lies. Rest assured that I won't appear at FAC and give you a hard time. Just trying to communicate something to you. Fowler&fowler«Talk» 22:15, 28 October 2021 (UTC)
@TompaDompa: The problem with Whitman and Lincoln, which is offered as a counter-example, is not the content, but the title which is very poorly chosen. The article is about Whitman's poetry on Lincoln, his lectures, even the hero-worship of Lincoln. But there is no real evidence that Lincoln knew him. Whitman claimed that Lincoln nodded to him on the street. But others thought Lincoln nodded to everyone he passed on the street; it was an aspect perhaps of his Hoosier friendliness (or neighborliness). The article should have been titled "Whitman on Lincoln," as Whitman's own lectures were, or "Whitman's Lincoln poetry." Much has been written about that and that is what the article is mostly about. Fowler&fowler«Talk» 06:20, 29 October 2021 (UTC)
@TompaDompa: You might consider changing the article name to "Immortality in Science Fiction," or "Immortality in Science Fiction and Fantasy." You might also consider removing references to epic poetry as fiction. (Epics, such as Gilgamesh, or Mahabharata, Ramayana, Illiad, and Odyssey are not generally considered fiction.) If you narrow down the scope (and the title) you'll have a better article. Bowing out now. Fowler&fowler«Talk» 07:04, 29 October 2021 (UTC)
Epics, such as Gilgamesh, or Mahabharata, Ramayana, Illiad, and Odyssey are not generally considered fiction—They most certainly aren't considered non-fiction, I don't know where this comes from. Especially as the article clearly attributes their discussion to sources treating them (rightly) as fiction. Much has been written about that and that is what the article is mostly about. Funny, because that's exactly why it's a good comparison, they're both articles discussing what much has been written about. TompaDompa I would suggest the best next step is to open a peer review for the article for additional feedback; if you want to ping me when you do so I will give the article a going over. After that I would also suggest running it past a copyeditor—WP:GOCE can help but they can't always guarantee FA-level prose, but another pass over the article will always help. 𝄠ʀᴀᴘᴘʟᴇ 10:01, 29 October 2021 (UTC)
I posted some comments on the article talk page. Hopefully it's helpful. (t · c) buidhe 13:49, 29 October 2021 (UTC)
Taking the cue from Buidhe, I've posted some further thoughts on the article's talk page. Fowler&fowler«Talk» 15:59, 29 October 2021 (UTC)

TompaDompa I'm sorry your request has been somewhat derailed by an unhelpful user. It's not my bailiwick at all, but I could certainly help with aspects like MOS if that'd be useful. The Rambling Man (Keep wearing the mask...) 21:54, 28 October 2021 (UTC)

Off topic (slightly..) Take the Lead!

Posting this here as this seems to be the most active talk page of content editors. Now is the time to make some leads in articles with crappy leads for a chance to win prizes and help all mobile users! See Wikipedia:Take_the_lead! - starting in a matter of hours....Cas Liber (talk · contribs) 18:17, 31 October 2021 (UTC)

In appreciation

The Tireless Contributor Barnstar
By the authority vested in me by myself it gives me great pleasure to present this barnstar jointly and severally to every editor who reviewed an article or had an article at FAC during October. They contributed to a frankly amazing 46 articles being promoted, the highest monthly total for more than 11 years. You should all be proud of yourselves. These promotions represent a staggering aggregate of hard work. Please keep it up, it is much appreciated. Gog the Mild (talk) 11:36, 1 November 2021 (UTC)

Updated FA stats table

Oct 25, 2021

Percentage Growth in FA Categories, 2008–2021, Legend:

Considerably above average, Above average, Average

Below average , Considerably below average

Featured Article Category as of Feb 23,
2008
Sep 16,
2008
Sep 16,
2010
Dec 1,
2011
Jan 1,
2015
Jan 1,
2020
Oct 25,
2021
Pct chg
Jan 2020
to Oct 2021
Pct chg
Feb 2008
to 2021
Pct of total 10-year trend
Art, architecture and archaeology 65 72 117 128 175 271 290 7% 346% 4.8% 127%
Awards, decorations and vexillology 24 26 28 27 26 24 17 -29% -29% 0.3% -37%
Biology 130 155 261 326 456 625 659 5% 407% 10.9% 102%
Business, economics and finance 16 19 22 44 73 116 119 3% 644% 2.0% 170%
Chemistry and mineralogy 29 31 34 37 40 46 46 0% 59% 0.8% 24%
Computing 17 17 17 18 16 14 10 -29% −42% 0.2% -44%
Culture and society 40 48 61 65 77 104B 109 5% 173% 1.8% 68%
Education 30 34 36 38 40 40 32 -20% 7% 0.5% -16%
Engineering and technology 35 37 38 40 43 49 56 14% 60% 0.9% 40%
Food and drink 11 11 9 13 17 21B 21 0% 91% 0.3% 62%
Geography and places 148 158 181 185 213 232 217 -6% 47% 3.6% 17%
Geology and geophysics 9 12 18 20 23 29 34 17% 278% 0.6% 70%
Health and medicine 31 36 42 43 51 52 48 -8% 55% 0.8% 12%
History 146 154 189 201 239 308 321 4% 120% 5.3% 60%
Language and linguistics 17 15 13 13 12 15 14 −7% −18% 0.2% 8%
Law 29 34 41 49 65 72 76 6% 162% 1.3% 55%
Literature and theatre 108 134 161 191 258 316B 323 2% 199% 5.4% 69%
Mathematics 13 14 19 17 18 18 17 -6% 31% 0.3% 0%
Media 159 171 221 231 324 424 439 4% 176% 7.3% 90%
Meteorology 61 78 111 126 147 168 177 5% 190% 2.9% 40%
Music 153 182 232 254 331 398 424 7% 177% 7.0% 67%
Philosophy and psychology 12 13 12 12 12 14 12 -14% 0% 0.2% 0%
Physics and astronomy 67 82 98 101 127 153 154 1% 130% 2.6% 52%
Politics and government 62 67 98 117 166 217 217 0% 250% 3.6% 85%
Religion, mysticism and mythology 36 44 73 84 105 121 130 7% 261% 2.2% 55%
Royalty, nobility and heraldry 75 90 94 108 124 173 187 8% 149% 3.1% 73%
Sport and recreation 119 162 268 298 365 449 528 18% 344% 8.8% 77%
Transport 47 74 107 128 171 213 215 1% 357% 3.6% 68%
Video gaming 72 96 127 137 180 222 239 8% 232% 4.0% 74%
Warfare 145 173 318 366 537 729 891 22% 514% 14.8% 143%
Total 1,906 2,239 3,046 3,417 4,431 5,695 A 6,022 5.7% 215.9% 100% 76%
  • Note A: Total is off by one; not worth looking for the error.
  • Note B Three food biographies moved [1] per discussion at WT:FAC
  • Note C - 10-year trendline is calculated by (2021# - 2011#)/2011#

A couple notes here.

  • 1) I'm off by two on the count. About half of these I manually counted before I realized I could copy the contents of the section into excel, delete all the headers, and see how many rows there were. Both ways were tedious. Feel free anyone to track down the two errors, but I frankly am tired, and as an auditor, I have the right to pull the materiality card here
  • 2) The previous version(s) had six colors; this one only has 5 because of the excel function I used to calculate the colors (more below)
  • 3) Many of y'all are probably wondering why the average change in the bottom doesn't really fit with the average range in the color bars. This is because the average change is a mean, which is skewed by outliers (looking at you, computing and warfare). The color bars were done with an excel function that essentially took the range (for the 2020-2021, -36 to 22, or 58) and then divided it into quintiles (58/5 = 11.6). So the "considerably below average" for 2020-2021 is from about -36.0 to -24.4; "below average" is from about -24.4 to -12.8, "average" is -12.8 to -1.2, "above average" is about -1.2 to 10.4, and "considerably above average" is about 10.4 to 22.0. Yes, it's a bit clunky (for the 2020-2021 the 40-60 quintile is entirely negative!), but it's the best I could come up with on short notice.

It took me about 2.5 hours to do this; I'm sure with all the brainpower we have at FAC someone can probably come up with something easier and quicker. Hog Farm Talk 03:24, 26 October 2021 (UTC)

Thanks for doing this - interesting. It would also be good to have a column with the current % of the total, if someone is feeling keen. Johnbod (talk) 03:47, 26 October 2021 (UTC)
@Johnbod: - I have added this. Hog Farm Talk 04:39, 26 October 2021 (UTC)
@Hog Farm: and as an auditor 🤔. Am at Uni for Accounting currently. Iazyges Consermonor Opus meum 03:54, 26 October 2021 (UTC)
@Iazyges: - I just survived that myself about 5 months ago. Hog Farm Talk 04:21, 26 October 2021 (UTC)
@Hog Farm: A great time; on a related note I went through the FAC page and searched for the  • 's, took off the one at the end and the the two in the edit description, to get numbers. For whatever reason my final count was 6021, so I'll dive through it again. I had 177 for metereology (compared to your 179) and 239 in video games (compared to your 240). I'm gonna duplicate rule match the contents of the featured article category to the page to see if I can hunt down the missing one. Iazyges Consermonor Opus meum 04:24, 26 October 2021 (UTC)
@Iazyges: - I think it's gonna just be the two meteorology. David Fuchs moved Power Mac G4 Cube from video games to computing after I created the table, which will be the -1 in the video games. As computing which saw an increase, was the bottom end of the range for both color coding, I will need to redo the color scales, which shouldn't be too hard now that I know what function to use. Hog Farm Talk 04:39, 26 October 2021 (UTC) Hog Farm Talk 04:37, 26 October 2021 (UTC)
@Hog Farm: You tell me this after I make my computer process 12,000 hyperlinked cells. Iazyges Consermonor Opus meum 04:41, 26 October 2021 (UTC)
Sorry to make more work for people, but if it weren't for these stats I would never have looked and seen it got added to the wrong category! Progress! Der Wohltemperierte Fuchs talk 12:57, 26 October 2021 (UTC)
Awesome, Hog Farm, thanks so much for doing this; this is the very kind of "take the bull by the horns" leadership that you are known for and a good example of why you are good Coord material :) A couple of things:
  • Could you add the chart to where I historically kept this data, at Wikipedia talk:Featured article statistics#FA category tallies? It will be much easier to find it there in the future then to have to search through much more lengthy WT:FAC archives.
  • I see you have already discovered how time consuming it was for me to generate this data, and are taking steps to make the work easier. Again, kudos for taking the leadership initiative. Should this be helpful, I used to use a combination of Word and Excel to generate the numbers because I found it easier to strip some text out via Word before moving to Excel for sorting and counting. But I'm pretty sure you will find better ways to skin this cat!
  • It is grand that someone found the mistakes, but for the purposes of this work, if finding discrepancies proves too hard, I believe it acceptable to simply footnote the differences.
  • A note of explanation: the reason the dates used are so odd just has to do with when I happened to generate the chart, which was usually a matter of writing an article for {{FCDW}}; sorry the data doesn't correspond to exact years.
The data highlights quite well exactly the concerns I have about the things we might focus on going forward, but it will take me some time to write that up. It is encouraging to see some of the upward trends, but basically, the downward trends are still a pretty big concern ... but more on that once I get caught up. Thanks again for taking this on ! Yep, there is lots of FA process work that anyone can do, but it's the natural born leaders who actually get it done, and that's what makes you a Coord :) Best, SandyGeorgia (Talk)

Everything should be fully updated to reflect the error correction (thanks Iazyges!) and the move of one article. Hog Farm Talk 05:00, 26 October 2021 (UTC)

So, in short, FAC is flourishing, and a few areas would benefit from more contributions but as this is a volunteer project, it will be down to volunteers to help with that, if indeed it really does need help. "Outreach" for this kind of thing is definitely not part of the role of (or even tangentially related to) FAC co-ordinators. The Rambling Man (Keep wearing the mask...) 06:10, 26 October 2021 (UTC)

Steady TRM. Let us concentrate our comments on the positive aspects of all of this, and what positive lessons it might hav for us, going forward. Gog the Mild (talk) 06:25, 26 October 2021 (UTC)
'Flourishing' would imply increasing growth rate. Only warfare, business, and biology appear to be growing. Meanwhile 11 areas are collapsing with 7 of those having shrunk in total, a further 9 are falling behind, and 7 are as stagnant as FAC overall. If it says anything, it is that SandyGeorgia has been correct in her concerns. Mr rnddude (talk) 07:34, 26 October 2021 (UTC)
Not, flourishing means many subjects are continuing to have increased numbers of FAs in their categories. Arbitrary delimiting of what is good and bad (or "collapsing"!!!) isn't helpful. And as for "concerns", well once you can find a load of volunteers who aren't between 18 to 40, white, male, to contribute to Wikipedia (that's more than half the editorship last time I looked) then you're going to struggle. And no, this isn't a FAC problem, it's a Wikipedia "problem" but really just a reflection of the kind of demographic that is interested in contributing here. And I'm still failing to see where, why or how we expected FAC co-ords to perform outreach. And all I can see is three categories shrinking since 2008, not seven. Maybe I can't read stats any longer. Ah, no, I see you're taking a micro-scopic view, over the last 20 months or so where FAR has been totally overwhelming in its taking down existing FAs, whereas before then it almost didn't really exist. We should run separate stats for FACs promoted and FARs demoted, for clarity, because of course that will put a very different spin on things. The Rambling Man (Keep wearing the mask...) 08:19, 26 October 2021 (UTC)
Growing 'considerably below average' is what I am describing as collapsing. The three you mention – that have shrunk since 2008 – have safely 'collapsed', particularly as two are ~ 2/3 of their original size, and awards and computing may not have had a single new FA in a decade [I say may because it's possible that more were demoted than promoted in that period, rather than purely demoted; I don't know how I'd go about checking]. I am aware that FAR has been diligently working to address old, dated FACs. I doubt this is the whole picture, but separate stats for comparison would be useful. FAC is not dying, but with three* areas growing ['above' and 'considerably above'] and twenty* areas declining ['below' and 'considerably below'] it just isn't 'flourishing'.
*That's according to the macroscopic column, although it is a bit skewed given that transport, sport, and art which are 120% above the global average are labelled as 'average' while literature and meteorology at 20% below are labelled as 'below average'. So, I'd also include those as 'above average' for a total of six growing areas, and the other two as 'average' for eighteen declining areas. Unless I've misunderstood that column. Mr rnddude (talk) 09:15, 26 October 2021 (UTC)
@Mr rnddude: I do think you've misunderstood the column, the percentage is growth, not percent of previous, so a 90% means 190% of the 2008 articles, not 90% of 2008 articles. So although three niche categories have lost some, and another stalled, on the whole most categories are gaining articles, but certain categories are growing much faster. As TRM says, it's largely a demographic constraint. A huge body of young White men pretty much guarantees dominance of Warfare and other history-related articles. Those three categories are harder to write about and less interesting, on the whole (no offense to those who find it interesting, but you are clearly not in the majority). The number of wrestling good articles is more than the combined FA and GA computing articles for a reason. Pretty firmly agree with TRM that that isn't going to change based on outreach; people develop to FAC, they aren't brought up to it. I'd even argue that it could bring about an increase in noms that clearly aren't going to make it. I'm certain if someone advertised FAC to me when I was new I would confidently bring my bare URL ref articles with half the text uncited. Iazyges Consermonor Opus meum 09:27, 26 October 2021 (UTC)
See my first comment, I am well aware that is percent growth. That is why I said decline, not loss. As in 'declining growth'. No, what I may have misunderstood is whether the ratings ['below'/'above' average] are being tied to the global average, or whether each category is being compared against growth in previous periods in that category only. Addendum: but that's infeasible, since that column covers 2008-2021, so the whole period, which also means not declining growth, but low growth. I do see though how writing 'eighteen declining areas' might have been interpreted as shrinking, rather than what I had meant. Mr rnddude (talk) 09:38, 26 October 2021 (UTC)
I think you're conflating FAC with FA. My conjecture is that FAC is doing just fine. FAR is counteracting that by removing several FAs from the total, especially older ones where principal editors have retired from the project. This has been particularly acute since last year since FAR has been cutting swathes through the total of FAs we have. This table does not (cannot) demonstrate that FAC is "declining" or "collapsing" or "shrinking", just that the summation of FAC and FAR is showing a negative trend in a handful of categories. None of this is surprising. The Rambling Man (Keep wearing the mask...) 12:16, 26 October 2021 (UTC)
I don't think we can really split things up that way. While yes, FAR and FAC are clearly distinct bodies with different processes, I think it would help if their purposes were looked at as more connected. The hope is to not only produce new FAs, but to also take the time to maintain the prior FAs, even if that means less time can be given to nominating at FAC. If we just dismiss FAR as "counteracting", then we forget that it's counterproductive to let our older FAs fall to the wayside and just cycle through FAs instead of building upon a corpus of them. Yes, I agree that the purpose of this talk page is mainly for how the FAC process itself should be run, but there also needs to be some thought into how we're maintaining what we have. There's currently 134 article listed at WP:FARGIVEN; some have been improved and won't need FAR after all, while others will go there (will be sending one there tonight, not sure which one yet). I'd say reviving any one of those to avoid FAR is just as important as putting through a new FA through FAC. It's a less glamorous task, though, and we can't seem to drum up as much interest in repairing/maintaining as in nominating. Hog Farm Talk 13:36, 26 October 2021 (UTC)
No, the problems facing each process are very different. Conflating them and then claiming the FAC process is broken is false. The Rambling Man (Keep wearing the mask...) 14:58, 26 October 2021 (UTC)
Hog Farm is on the right track. Outside of the small FA community, other Wikipedia editors will see an FA is an FA is an FA, and when viewing an article that is an FA, almost half of the time, they will be viewing an article that no longer meets the criteria. This damages the reputation of all FAs and can lead to the declining interest in working towards an FA we have seen over the last decade. (How many here are old enough to remember that writing an FA used to be a pretty standard part of seeking adminship-- today, those who care about FAs are more likely to be found on this page and nowhere else.)
Also, looking at historical trends, we are nowhere near cutting the swaths through deficient FAs that we have been in the past. Simply, the pool of FAs show that a huge percentage aren't, and that is what active editors will see. The overall FA process declines in prestige when FAs are not maintained. And it is very discouraging to reviewers to watch nominations sit at FAR for months, unimproved, while the main nominator for those articles is pushing through new FAs. Should we be promoting new FAs for nominators who aren't maintaining their old ones? Separating FAC from FAR from TFA is folly, because all of them are a piece of the picture in terms of editor motivation, examples of Wikipedia's best work, and many other factors.SandyGeorgia (Talk) 23:47, 28 October 2021 (UTC)
I'm not claiming that they should be conflated. I'm saying that while we're coming up with improvements to FAC, we also need to keep in mind that "throughput throughput throughput" without any plan to keep these up to code is just going to result in everything being delisted in 10 years, which is obviously what we don't want. Hog Farm Talk 15:02, 26 October 2021 (UTC)
FAR is completely different. Often you have to scrape around Wikiprojects or retired editors to see if someone will pick up an article with no tangible benefit to them at all. I'm sure this will get laughed out of town, but we do GAN drives every so often and all that's available are barnstars but it motivates some people. FAR has literally nothing going for it if it's not the principal editor doing all the heavy lifting just to maintain a bronze star. FAC is completely the opposite. The Rambling Man (Keep wearing the mask...) 16:03, 26 October 2021 (UTC)
That is not at all accurate, and I don't believe (could be wrong?) you frequent FAR enough to understand that. James Joyce is a very good current counterexample, and there are more. As to GA sweeps, you may remember that I proposed we deal with the backlog of very old deficient FAs by doing something akin to GA sweeps, and that was rejected on this page. SandyGeorgia (Talk) 23:49, 28 October 2021 (UTC)

Hog Farm, thanks for doing this; very interesting data. One thing that strikes me is that you can see specific editor trajectories affecting some of the smaller categories. For example, without checking, I'd bet that "Geology and geophysics" is almost all the work of Mav, Ceranthor, and Jo-jo Eumerus, and that the growth pattern you see there corresponds very closely to the high-activity periods of those editors. If Jo-jo were not currently active that would be a stagnant category at the moment. Mike Christie (talk - contribs - library) 11:51, 26 October 2021 (UTC)

Great work with this! I only really pay attention to the part of the table I can effect, which is the sport part where I work; which is going up year on year. The one thing this table does show is that some categories get more nominations (well, successful nominations) than others. Best Wishes, Lee Vilenski (talkcontribs) 12:56, 26 October 2021 (UTC)
Although many of my FAs are under Geography, seems like? JoJo Eumerus mobile (main talk) 20:36, 26 October 2021 (UTC)
The decrease in "geography and places" is likely because that's where the number of cities delisted at FAR over the last year or so were probably categorized, I think? Hog Farm Talk 20:39, 26 October 2021 (UTC)

For those who like me thought "I haven't seen many business ones going through, I wonder why that one went up", it's because numismatics is included in that section. Hog Farm Talk 13:37, 26 October 2021 (UTC)

Wonderful stuff Hog Farm, some fascinating data. How difficult would it be to produce a "Pct chg Dec 2011 to 2021" column, to give us a set of ten year trend figures, rather than the rather arbitrary start date used now. At worst, it would be interesting if/how much that moved the figures. Thanks, Gog the Mild (talk) 15:26, 26 October 2021 (UTC)
Won't be too hard, maybe 30 minutes work? Will look at that after work. Hog Farm Talk 15:45, 26 October 2021 (UTC)
@Gog the Mild: - Added, with no color. My migraine is saying I need to get off the computer. Hog Farm Talk 23:55, 26 October 2021 (UTC)
@Hog Farm: Can you explain how the 10 yr trend is calculated, & probably add it to the Notes below the table. It isn't obvious to me, for example, why the figure for the total is 76%, when the number of FAs has more than tripled. Thanks. Johnbod (talk) 00:38, 1 November 2021 (UTC)
@Johnbod: - Added as note C. I think what's going on here is that the number you're looking at for the tripling is the column on the left, which is from 2008. The 2011 number which is used for the 10-year trend, is the 4th from the left, with a total FA count of 3,417. I calculated it by (6022 - 3417)/3417, which is 76% after rounding. Hog Farm Talk 00:45, 1 November 2021 (UTC)
Yes, that was it - thanks! Like other measures from these figures, this shows the strong areas are getting still stronger, and the weak ones weaker. Johnbod (talk) 04:37, 1 November 2021 (UTC)
So work harder. Get more people involved the weaker areas. But don't for a moment nay-say the efforts of those working in the areas that are somehow considered trivial. There's a terrifying element of snobbishness which is beginning to overwhelm FAC to the point where it will become a walled garden. I always viewed this place as somewhere I could take some themes in which I had a serious interest and improve them to the betterment of the project. What I've experienced here over the last year or two is a welcoming advocacy of that sort of approach, only tempered with a really hostile, passive-aggressive, dismissive approach to the topics which I (and many others like me) have contributed. The problem of lack of diversity or improvement in under-represented areas is not an excuse to berate those of us who contribute about topics we're interested in. The sooner some users here get that understood, the better, and the sooner we have fewer passive aggressive posts about our contributes, the better too. So, stop it. If you're worried about lack of medical articles or vital articles, DO SOMETHING ABOUT IT, but not to the detrement of the areas of FAC which are flourishing. Talk about cutting off your nose to spite your face. The Rambling Man (Keep wearing the mask...) 18:13, 2 November 2021 (UTC)

Since the effects of FAR have been brought up a few times, I've gone through WP:FARCHIVE and tallied up FAs delisted by category at WP:FFA from January 2021 to October 2021:

  • Literature and theater: 4
  • Geography and places: 17
  • Warfare: 10
  • Education: 13
  • Physics and astronomy: 3
  • Sports and recreation: 9
  • Royalty and nobility: 5
  • Biology: 6
  • Business, economics, and finance: 4
  • Music: 9
  • Culture and society: 2
  • Religion, mysticism and mythology: 4
  • Politics and government: 6
  • Health and medicine: 3
  • History: 11
  • Media: 9
  • Computing: 3
  • Art, architecture and archaeology: 5
  • Heraldry, honors, and vexillology: 7
  • Meteorology: 4
  • Transport: 3
  • Law: 3
  • Geology and geophysics: 2
  • Video gaming: 1
  • Language and linguistics: 1
  • Engineering and technology: 1
  • Mathematics: 1

All but three categories (chemistry, food and drink, psychology and philosophy) have seen at least one delisting. Comparing to the FA statistics table, some of these represent areas being decimated by FARS (heraldry/honors/vexillology, education, geography and places), while this suggests that several of the high improvement categories are seeing a lot of "churn" (sports and recreation, warfare, and music all have comparatively high delistings yet significant positive growth. Some of the ones that seemed less active (such as politics and government) represent areas where there have been recent FA nominations, but they have been offset by FAR losses. Don't know if this is useful, but thought I'd throw these numbers out there. Hog Farm Talk 06:56, 3 November 2021 (UTC)

FA stats in the vital category

Total FAs and GAs
Number of WP:VIT3 good and featured articles

I thought this would be another statistic of interest for the reader here in terms of the type of article we get at FA compared to the days of yonder. We don't seem to get many "Vital" articles anymore (or at least, VIT3 level). I made the graphs for a presentation about the core contest, but seemed at least a bit relevant here too. GA did see an increase in vital articles, but relatively modest compared to the years before I could get good data. Femke (talk) 17:01, 26 October 2021 (UTC)

Aren't the years before you could get good data inherently an unreliable thing to compare to because, well, the data wasn't as good? Perhaps perception was different? Best, Tyrone Madera (talk) 17:19, 26 October 2021 (UTC)
I don't quite understand your question. We know that we got about 90 VIT3 FAs at some point before 2010, and we're slightly below that now. Logically, we must have fewer successful FAC / more FARs since 2010 in that category than before each year.
If I remember correctly, I couldn't get earlier data because VIT3 wasn't formed or stable enough. Femke (talk) 17:37, 26 October 2021 (UTC)
What strikes me about these two graphs is the % decline in "vital" GAs. The total number of GA's have quadrupled from about 8,000 to 32,000 (just going off the graph), but the vital ones have only increased by about 20! The higher quality of our articles on narrower or smaller topics is not at all restricted to FA & GA; as I was complaining ten years ago: "So the more general and abstract the article subject, the worse our articles tend to be. This is the reverse of the pattern generally found in published reference works. It also means that some of our weakest work within a subject area gets the highest number of viewers. This is not good." This remains true, although possibly there has been some improvement. Johnbod (talk) 20:01, 26 October 2021 (UTC)
But who is really involved in determining what is a "vital" article? The Rambling Man (Keep wearing the mask...) 20:06, 26 October 2021 (UTC)
Dodging the issue - whichever definition or selection you use it will be true. Architecture is pretty poor, Renaissance architecture worse (it failed FAC in 2007, when that was hard to do), but many articles on individual buildings and architects are pretty good (though of course more are not). Johnbod (talk) 20:12, 26 October 2021 (UTC)
Not dodging anything. We have a quantitative discourse over numbers of so-called vital articles. I understand how GAs and FAs are made, but not at all how "vital articles" are selected, other than my experience which appears to find these articles elected by perhaps three or four hard-core "VA" projecteers. Of course, the relevance of this is precisely zero in toto because unless we have volunteers who are prepared to work on these articles and then go through the glacial slog of FA only to be thwarted because someone doesn't understand plain English, it's little wonder several of our "fundamental articles" are not seized upon with vigour and pursued to excellence. The Rambling Man (Keep wearing the mask...) 20:16, 26 October 2021 (UTC)
Well, missing the point, then. I'm no fan of the "vital" hierarchy at all, but conceptually one has to recognise there is a hierarchy within any subject area (if that, very perversely imo, is not accepted the article views will get you to much the same place). I don't follow your last bit, but the main reason for the weakness of big topic articles is surely that they are much, much harder to do, let along get through a GAC or FAC. Johnbod (talk) 20:29, 26 October 2021 (UTC)
Well maybe you're missing the point? It's not about how hard something is, more about finding the right people for the job. We can't do that with fundamental topics, and that's because we don't have enough volunteers qualified to work articles up to the standard required and go through the turgid, glacial, bulky process. The Rambling Man (Keep wearing the mask...) 20:38, 26 October 2021 (UTC)
Actually, "hard" is an important point - the more popular a subject the more sources are there to consult and use and eventually they are too many. Then you also get arguments about length, why certain sources or viewpoint weren't included and so on. A topic that draws more attention also draws more opportunities for arguments, and more "improvements" that aren't. JoJo Eumerus mobile (main talk) 20:48, 26 October 2021 (UTC)
It's definitely harder, but I would disagree with the statement that we have a lack of qualified editors. I trust most FAC regulars can do this if they would like. I see a couple of different barriers:
  • culture; we're not encouraging people to do this.
  • our gamification system: for instance, the Wikicup doesn't provide enough bonus points for larger topics to make them attractive. The core contest was quiet for a few years.
  • The atmosphere at FAC for newer editors that are unfamiliar with the unwritten rules may not be pleasant.
  • The focus on details that do not matter that much on the grand scale of things. It's more difficult to get source formatting and commas perfect on these large topics, where you'll have a cohort of editors with intermediate experience contributing. Femke (talk) 20:59, 26 October 2021 (UTC)
  • It's also harder to get ahold of enough sources to be sourcing-comprehensive for the bigger topics. I could get First Battle of Newtonia with a handful of web sources and two or three print books; I'm working on the much-more-significant Battle of Raymond and have consulted a lot of print sources and still am at least one (Grabau's Confusion Compounded) short of FAC-able. Hog Farm Talk 21:14, 26 October 2021 (UTC):::::::*
    Like Featured articles, Vital articles are determined by consensus. Anyone can nominate any article as a vital article, but the number of Vital articles at each level is limited, so usually adding one means removing another. This means that you normally need to provide a case for article X warranting inclusion more than article Y. See Wikipedia:Vital articles/Frequently Asked Questions. By their nature, Vital articles are hard to write. The wealth of sources is one part of it. It's also the case that many Vital articles, especially the higher level ones, cover very broad topic areas indeed, with literally thousands of subarticles. (But not impossible; it's been done.) The Rambling Man is quite right in that there is no incentive whatsoever to work on them at FAC (although they are worth extra points at WikiCup) despite recurring complaints that FAC is the domain of obscure articles. 21:31, 26 October 2021 (UTC) — Preceding unsigned comment added by Hawkeye7 (talkcontribs)
Hawkeye7 although they are worth extra points at WikiCup are they? I don't believe that's true. I think WikiCup ignores that micro-project too. The Rambling Man (Keep wearing the mask...) 23:45, 26 October 2021 (UTC)
You're quite right; I confused this with the way it scores additional points for every five Wikipedias (including the English Wikipedia) in which an article appears, which tends to favour the vital articles, but neither explicitly nor directly. Hawkeye7 (discuss) 00:16, 27 October 2021 (UTC)
Although I am not a fan of how the "vital" article process is executed, there has to be some way to measure the quality they pretend to be after, and that process is all we have for now. And although I could take some minor differences, I think Femke has it mostly right. A reminder that she restored Earth at WP:FAR in time for Earth Day, and has worked on a number of very difficult, very significant articles, and has several WP:MILLION awards.
There are specific parts of the culture that not only discourage, but outright encourage, production not of articles that get millions of views a year, rather niche articles that get less than 15 pageviews a day. That these are valued equally by those who seek points in contests on the internet baffles me; what comes to mind is that niche topics lend themselves to churning out three or four a month, while an FA in a broad/"vital" topic can represent years of work, and will need constant maintenance and updating to reflect new research. Somehow, the kind of work Femke does needs to be represented and valued by these processes, and yet, every time we attempt that discussion here, it turns acrimonious and is personalized. This debate has existed for as long as FAC has. Where we used to debate "short" articles, or the proliferation of "hurricane/video game" articles, then "cookie cutter" articles, we now have other niche categories dominating FAC.
Whatever the designation was that has always implied something akin to the opposite of "vital", this issue has always been with us. We should be able to talk about it without discussion being personalized. We should find ways to encourage, or at least not discourage, the topics that are dense, hotly debated, technical or have worldwide significance and that get a million views a year, over the one-time recreational event that gets 10 views a day. There should be room for both; right now, we're missing something important, regardless if we call if "vital". And the argument that we can't help what editors want to write doesn't pan out when we do things at FAC, FAR and TFA that actively discourages those editors. Medical experiences with FAC, FAR, TFA and ERRORS over the last two years come to mind. Medical articles can't buy a review, while niche topics get many reviewers (some who demur when asked to review denser topics), and this is utterly discouraging to new medical FA writers, who then see their work damaged via ERRORS by editors who don't know the topics. There are too many ways that the overall process is too difficult for broad and core topics. And we should be able to talk about those issues. SandyGeorgia (Talk) 00:10, 29 October 2021 (UTC)
  • I have just returned home. As I'm tired, please pardon the occasional tongue-in-cheek lapses. This post is serious. Let's start with the "considerably above average" in HogFarm's table. There is only one member: "Business, economics, and finance." It shows a growth of 644%. But the US Census Bureau says, "The goal is to count everyone once, only once, and in the right place." So, we have to discard commemorative coins. Why? Because college-level textbooks on Business, Economics, or Finance don't usually discuss commemorative coins.
  • Also as we are now in the realm of generalized demography, and demographic trends are most commonly examined in intervals of decades (decennial), let's start in 2011—October to be precise, judging from the maple leaves clogging the backyard drain. In other words, I shall be comparing the FA count in the period:October 26, 2011 to October 25, 2021. The basic articles (i.e. non-numismatic and not about specific companies) numbered 14 in 2011; they numbered 11 yesterday, i.e. they went from constituting .41% of FAs in 2011 (when there were 3,408) to .18% yesterday (when there were 6,023). The is a proportional decrease of 56%.
  • The next category is "above average." The table shows a growth of 407% in Biology. But if you count the articles in Biology proper, not about individual organisms in microbiology, zoology, or botany, there were 23 in 2011, there are 28 now, a decrease from .674% of the total to .464%, i.e. by 31%.
  • Let's consider, "considerably below average." The perennial favorite in that category has always been mathematics, the only subject people take pride in not knowing. In 2011, there were 17 or 0.50% of the total; yesterday there were 17 or 0.28% of the total. The is a decrease of 44%.
  • Finally, in "below average," allow me to pick a subcategory at random. How about physics and astronomy biographies? Well, I obviously cannot count astronauts. Why? Because none in that group will be able to derive Kepler's third law from Newton's law of Gravitation (taught in first-year college physics). So, discounting the obvious interlopers, there has been a modest growth there —for which someone or other needs to be congratulated. Their proportions have gone from 0.235% (8) in 2011 to 0.298% (17+1) in 2021, a 27% increase. Why the "+1?" Because I've added Neil Armstrong: he did teach physics in his post-lunar life.
  • Summing up: the growth in FAs in WP is uneven. Their poverty in many traditional areas seems to have been ameliorated mainly by changing the areas. For example, there is nothing in common between the 100 best English novels according to Robert McCrum and the best-written articles on novels on WP. (Well, except for To Kill a Mockingbird). In my view, these tables are of very limited value, because there is a kind of gerrymandering going on of subject areas. Fowler&fowler«Talk» 21:08, 26 October 2021 (UTC)
  • Finally, I should add that the reason "vital articles" are hard to write is that people both nominators and reviewers conflate (a) acceptable tertiary sources such as undergraduate textbooks or first-year graduate ones and (b) specialty secondary sources such as journal articles or research monographs. The more vital (I prefer "high level") is an article, the less it needs (b). A high-level article can and should be written with acceptable tertiary sources because only those can summarize a broad field. Three or four such textbooks max should suffice. Fowler&fowler«Talk»
  • So, in summation, FAs are written by people interested in the subject matter. FACs are probably reflected in our demographic. FAs continue to be promoted quicker than they are demoted. Some FAs are considered "more vital" than others. The FAC process and the "characters" who mandate attention to minutiae or review one another's work are unwelcoming. The attitude of some users here is a total turn-off for new volunteers to get involved at FAC. The snobbish "values" attributed to one specific candidate over another is divisive and destructive to the process. I think we've got it covered here.  % changes in categories is somewhat irrelevant when such systematic issues dominate the problems at this project. The Rambling Man (Keep wearing the mask...) 22:33, 26 October 2021 (UTC)
  • You are both spouting and shunting garbage. I am saying quite the opposite. Vital FAs are not being written because both nominators and (sometimes) reviewers are being too pretentious. The nominators don't need journal articles or specialty monographs. By using them, they make the task harder both for themselves as they don't have the knowledge to integrate the material (it can't be done in a Wikicup) and for the reviewers because lacking that knowledge or experience the nominators present an incoherent description which any reviewer who is truly looking to understand can readily spot. The nominators are their own worst enemies. They need to get off the high horse of pretense, of subscribing to WP JSTOR access, this fancy journal access, or that (and the reviewers do too for asking for such sources). The vital articles, for the most part, don't need that nonsense. They, however, do need nursing, reading much more that just a sentence or two here or there, mulling things over that can take months. Fowler&fowler«Talk» 23:14, 26 October 2021 (UTC)
  • Ok, well personal attack (again) and poor formatting aside, you miss the point once again. This isn't about being "pretentious". This is about a lack of interest in the topics that some have declared as "vital" and the usual vitriol and glacial progress made at FAC once a nomination is in transit. There are literally users here whose interest is in belittling nominations. Literally. And that has created a vile and toxic atmosphere. We have users who comment on other reviewer's reviews. That's toxic too. The snobbery here is overt and destructive, and reinforced time and again. Little wonder no-one from the real world wants to be part of this "community" other than the nuclear-hardened amongst us. The Rambling Man (Keep wearing the mask...) 23:25, 26 October 2021 (UTC)
    When the issue of a non-milhist coordinator was raised, my first thought was you, being an editor mainly associated with sports articles. Hawkeye7 (discuss) 00:16, 27 October 2021 (UTC)
    Hah, very kind, but the "seniors" would never allow that ever. And yes sports, but my reviews are not constrained to sport, and I thoroughly enjoy wading through a MILHIST-approved FAC to provide insight from a different perspective, but that's as a reviewer....! The Rambling Man (Keep wearing the mask...) 20:10, 27 October 2021 (UTC)
Thanks Hog Farm. In a former life I used to audit auditors. Really. Gog the Mild (talk) 18:44, 27 October 2021 (UTC)

I am still working to catch up, now turning to these numbers (most appreciative that Hog Farm updated the charts).

The summary is that the overall trend that has long concerned me has not changed. Femke’s graph of File:TotalFA and GA.pngs shows that other areas of reviewed content have grown considerably, and her File:VitalFA and GA.pngs graph show that FAs in a certain (somewhat arbitrary, but I think we get the point) category of "vital" have not kept pace.

More to my point, if you look back at the older charts at Wikipedia talk:Featured article statistics#FA category tallies, you can see that the declining and increasing categories have changed. (Hog Farm, this is one reason I'd like to see you add the new numbers to that page ... so they will all be in one place for future reference.) If you start at the 2010 update and scroll through subsequent updates, sorting by percentage change, you will see some constants (like Warfare and Biology), but some areas that were once growing healthily are now either declining or not growing at a healthy pace.

Health and medicine saw a respectable (considering how hard it is to write and maintain an FA) increase in FAs between 2008 and 2011 of 39%, which was not in the bottom group of concerning categories ("considerably below average" growth). Today it is #8 from the bottom. I focus on my own area of editing because it is what I understand best, and I know what the factors are there, but this is an example of how others can look at these trends over time-- by going back to look at historical data. What is increasingly growing over time versus what is not. There are areas in which we are losing contributions. What originally launched this whole discussion was my premise that, if we choose all/most of our Coords from an area that is not typical and is not declining, we don't know what we may be missing about the other kinds of experiences people are having at FAC, FAR, TFA and ERRORS. The MILHIST experience, by virtue of the size of the project, is completely different from many other areas, in good ways. They have much less concern about reviewers showing up because there are so many of them, they have many shared resources, they have many who can mentor others, they have many who can answer questions on a broad range of sub-topics ... well, we could all go on and on. This is not necessarily a typical experience; many of us struggle to find anyone who will review. This is but one example-- there are others-- of why I am concerned that in the future we try to broaden the Coord base to represent more/other areas.

So, when looking at the numbers, MILHIST/Warfare can be separated out as a perhaps stand-alone situation. Biology stands out as an areas where a large number of different editors have worked on diverse topics over time, and an area that has never lost ground. Maybe those people can explain what the "plus" factors are that keep that as a growing group. Is it just mushrooms? And birds? Or ease of finding sources? I wouldn't know, but they do. But they are an example, along with Warfare (MILHIST) of a category that has continually flourished.

Another category of increasing groups, while representing a good thing, are subject to the kind of change we saw over time in, for example, Awards and vexillology, because those increases are largely fueled by individual editors. We have seen at FAR how we do lose those "broad swaths" of articles when one of those editors moves on and no one maintains those articles; in some cases, we have lost almost every article written by a prolific FA writer.

But in between those areas, we find the categories (like my example of Medicine, once flourishing, now not) that we can keep an eye and try to understand in terms of trends affecting the FA process.

Because the months of August and September showed a high number of promotions, I drilled down to see what was happening there. August had 37 promotions, and September had 36.

  • Of our 30 categories at WP:FA, 40% (12) of the 30 categories had no promotions, while
  • only 5 categories accounted for over two-thirds of the promotions—
    • Sport with 27% (20), followed by
    • Music (9),
    • Warfare (8),
    • AAA (art, architecture, archeology, 7) and
    • Biology (6).

I think we mostly know what is going on in those areas, but what is going on in the missing and very low growth areas? Here are the rest of the numbers from those two months (maybe someone is interested in filling in the missing areas-- like Health and medicine?)

    • 4 Lit & theatre
    • 3 Video games
    • 2 History; Geography; Education; Cuture and soc; Business, economics, finance;
    • 1 Geology; Law; Physics; Media; Meteorology; Eng. & tech

Besides the missing areas, I digress to the TFA situation wrt these categories. TFA slots were once so hotly sought after that the queue of people asking for a slot became unmanageable, and a convoluted point system for how to allocate the limited slots was put in place. That point system actively encouraged FAC to produce articles in "Vital", Core, or underrepresented (by the number of FAs in the category) topics. This was an example of something that was explicitly done to encourage growth in areas where we need more FAs. We have nothing like that today, and TFA slots are barely in demand and not at all the subject of hot competition as they once were. Why aren't we talking about these issues as ways to encourage FA production?

Whether one thinks that having one-quarter of the production in those two months be of niche articles about individual one-time recreational events, that are generally viewed about 15 times a day, relative to Femke's Earth that is viewed around 10,000 times daily, is a function of where you stand on the niche/vital debate that has raged forever, and which I hope we won't repeat, but I hope we can see that there are areas we could work on. My own opinion is that this is not an optimal situation, does not represent data that will help FA regain the status it once had in the community, and does not represent what FA could be striving for if we would just talk about these issues. Medical articles practically by common sense definition are "vital", but we can't get people to review those FACs. So we have Health and medicine FAs in decline-- what is happening in other areas and how can we change that? I do not see a healthy FA process in the data.

Sorry, out of steam, will continue tomorrow; my back doesn't allow me to sit at a computer as long as I used to, will fix typos and any grammar issues tomorrow. SandyGeorgia (Talk) 01:11, 29 October 2021 (UTC)

Regarding biology articles, I think you're right; it's a combination of very easy sourcing as well as a diverse topic area. Also worth noting the number of taxa that probably exist reaches up to the 10 million mark... Dracophyllum 09:30, 29 October 2021 (UTC)
one-quarter of the production in those two months be of niche articles about individual one-time recreational events, that are generally viewed about 15 times a day this just isn't true and is unnecessarily derogatory towards people who make good faith edits on such "recreational" articles. Many of which achieve well in excess of "15 times a day" viewership. It's probably worth dropping this ongoing sniping at FAs about topics which certain users consider "beneath" the encyclopaedia. If people want to create a "Vital Featured Articles" task force, that's great, go ahead and do that, but don't continue to "remedy" this perceived issue in this harmful fashion, there's nothing to be gained from veiled criticism of the items being promoted, better to focus on the items that aren't and leave us minions to our trivia. The Rambling Man (Keep wearing the mask...) 09:04, 1 November 2021 (UTC)
As a biology editor, I wouldn't say sources are necessarily easy to find, many of my animal FACs (a bunch of them on the "vital" list) would have been impossible without the generous people at WP:RX, so that should go for most other subjects. On more than one occasion have I received PDFs consisting of articles that editors there had scanned just for me. I think such subjects may be easier to get into for non-professional editors, like myself, which could explain why that area is stable. FunkMonk (talk) 10:11, 1 November 2021 (UTC)
Sure, journal articles can be paywalled but very rarely will you need a physical book to get it to FA... Dracophyllum 18:18, 2 November 2021 (UTC)
Well, as with everything else, that depends. I've needed physical books for all the articles about extinct birds I've written, for example. I own about 5-6 books about this subject that I use repeatedly, plus some pdf books. Would be almost impossible to be comprehensive on such obscure topics only with journal articles, as much of the relevant literature is books rather than articles. FunkMonk (talk) 19:16, 2 November 2021 (UTC)

FAC reviewing statistics for October 2021

Here are the FAC reviewing statistics for October 2021. The tables below include all reviews for FACS that were either archived or promoted last month, so the reviews included are spread over the last two or three months. A review posted last month is not included if the FAC was still open at the end of the month. The facstats tool has been updated with this data. Mike Christie (talk - contribs - library) 01:07, 4 November 2021 (UTC)

Reviewers for October 2021
# reviews Type of review
Reviewer Content Source Image Accessibility
Nikkimaria 4 18 24
The Rambling Man 16 1
Buidhe 2 13
Hawkeye7 7 3 3
Hog Farm 5 7 1
Z1720 11 1
Jo-Jo Eumerus 9 1
Kavyansh.Singh 4 3 3
Casliber 10
Chidgk1 9
Tim riley 8
Aoba47 6 2
Lee Vilenski 8
ChrisTheDude 7
Amakuru 5 2
HJ Mitchell 7
Wehwalt 5 2
Gog the Mild 3 4
Graham Beards 6
Grapple X 5 1
Femkemilene 5
Iazyges 3 1
JennyOz 4
Nick-D 3 1
Eddie891 4
Ian Rose 3 1
Jens Lallensack 3
David Fuchs 2 1
Pendright 3
Dudley Miles 3
Peacemaker67 3
Edwininlondon 3
Hurricanehink 3
Aza24 1 2
DMT Biscuit 2
Wingwatchers 2
Dracophyllum 1 1
Zawed 2
Wtfiv 2
FunkMonk 2
Double sharp 2
Kosack 2
Realmaxxver 2
Mr rnddude 2
Zmbro 2
Moisejp 2
Heartfox 2
Eewilson 1 1
Sportsfan77777 2
Epicgenius 2
Volcanoguy 2
Harrias 1 1
Homeostasis07 1
Nrco0e 1
Girth Summit 1
ProtoDrake 1
Clayoquot 1
Oldelpaso 1
Materialscientist 1
Parsecboy 1
Richard Nevell 1
GreenMeansGo 1
GhostRiver 1
Tintor2 1
DanCherek 1
CodingCyclone 1
Ichthyovenator 1
61.205.249.123 1
CactiStaccingCrane 1
Ww2censor 1
Ceoil 1
Famous Hobo 1
Noswall59 1
Display name 99 1
Trainsandotherthings 1
Urve 1
Damien Linnane 1
LightandDark2000 1
Jimfbleak 1
Chariosacha 1
SounderBruce 1
Sturmvogel 66 1
Laser brain 1
Tkbrett 1
Sanfranciscogiants17 1
Mike Christie 1
Usernameunique 1
Igordebraga 1
Hurricane Noah 1
Vami IV 1
Destroyeraa 1
Therapyisgood 1
Kusma 1
Pamzeis 1
BennyOnTheLoose 1
Kwamikagami 1
Gerda Arendt 1
J Milburn 1
Balon Greyjoy 1
Sdkb 1
MareBG 1
Indopug 1
TheDoctorWho 1
Extraordinary Writ 1
TheSandDoctor 1
Lingzhi.Random 1
Totals 248 64 53 0
Supports and opposes for October 2021
# declarations Declaration
Editor Support Oppose converted to support Struck oppose Struck support Oppose None Total
Nikkimaria 1 1 44 46
The Rambling Man 11 1 1 4 17
Buidhe 1 14 15
Hog Farm 4 1 8 13
Hawkeye7 7 6 13
Z1720 11 1 12
Jo-Jo Eumerus 10 10
Kavyansh.Singh 3 1 6 10
Casliber 10 10
Chidgk1 2 1 6 9
Tim riley 7 1 8
Lee Vilenski 8 8
Aoba47 5 1 2 8
Gog the Mild 1 6 7
Wehwalt 5 2 7
ChrisTheDude 7 7
HJ Mitchell 6 1 7
Amakuru 5 2 7
Grapple X 5 1 6
Graham Beards 3 3 6
Femkemilene 4 1 5
JennyOz 4 4
Nick-D 2 1 1 4
Iazyges 3 1 4
Ian Rose 3 1 4
Eddie891 4 4
Dudley Miles 3 3
Pendright 3 3
Edwininlondon 3 3
David Fuchs 1 2 3
Peacemaker67 3 3
Jens Lallensack 3 3
Aza24 3 3
Hurricanehink 3 3
DMT Biscuit 2 2
Double sharp 1 1 2
Eewilson 1 1 2
FunkMonk 2 2
Heartfox 2 2
Moisejp 2 2
Harrias 2 2
Volcanoguy 2 2
Zmbro 1 1 2
Wtfiv 1 1 2
Zawed 2 2
Epicgenius 2 2
Mr rnddude 2 2
Dracophyllum 1 1 2
Sportsfan77777 1 1 2
Realmaxxver 2 2
Kosack 2 2
Wingwatchers 2 2
Ceoil 1 1
ProtoDrake 1 1
Igordebraga 1 1
Hurricane Noah 1 1
Vami IV 1 1
Ww2censor 1 1
Homeostasis07 1 1
Nrco0e 1 1
Girth Summit 1 1
Usernameunique 1 1
Mike Christie 1 1
61.205.249.123 1 1
CactiStaccingCrane 1 1
TheSandDoctor 1 1
Lingzhi.Random 1 1
Extraordinary Writ 1 1
Tkbrett 1 1
Sanfranciscogiants17 1 1
Ichthyovenator 1 1
CodingCyclone 1 1
TheDoctorWho 1 1
Laser brain 1 1
Tintor2 1 1
DanCherek 1 1
MareBG 1 1
Indopug 1 1
Sturmvogel 66 1 1
GhostRiver 1 1
GreenMeansGo 1 1
SounderBruce 1 1
Sdkb 1 1
Jimfbleak 1 1
Chariosacha 1 1
LightandDark2000 1 1
Richard Nevell 1 1
Parsecboy 1 1
J Milburn 1 1
Balon Greyjoy 1 1
Trainsandotherthings 1 1
Urve 1 1
Damien Linnane 1 1
Materialscientist 1 1
Kwamikagami 1 1
Gerda Arendt 1 1
Display name 99 1 1
Noswall59 1 1
Oldelpaso 1 1
BennyOnTheLoose 1 1
Pamzeis 1 1
Famous Hobo 1 1
Clayoquot 1 1
Therapyisgood 1 1
Kusma 1 1
Destroyeraa 1 1
Totals 192 1 2 0 14 156 365

Time to stop transcluding FAR

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.


Although I've opposed this in the past, loading the FAC page has become increasingly difficult for me. Today I was trying to promote a FAC and it took several tries to get the page to load so that I could remove its FAC. We've got traffic at FAR up sufficiently that it doesn't need the extra visibility. (t · c) buidhe 02:44, 8 November 2021 (UTC)

I just tried loading the FAC page on my iPad via a hotspot from my phone (4G, full strength signal) and didn’t experience any issues. Sandbh (talk) 05:26, 8 November 2021 (UTC)
As I've said anytime the question's come up in the recent past, I favour removing FAR from the FAC page, not because I have trouble loading the page but because I believe FAR has its own momentum again and doesn't need to be conjoined with FAC. From a practical point of view I'd also prefer to be able to go to the end of the page and find myself looking at the oldest FAC, not a FARC page... Cheers, Ian Rose (talk) 06:36, 8 November 2021 (UTC)
As I have said above somewhere, I have had problems with the page for quite some time and the nominations viewer stopped working for me months ago. We should stop transcluding FAR; it was only ever meant to be a temporary arrangement when FAR was in the doldrums. Graham Beards (talk) 11:48, 8 November 2021 (UTC)
I think that Ian has said it well, and I entirely agree with both of their points. Thanks to Buidhe for raising it. Gog the Mild (talk) 11:51, 8 November 2021 (UTC)
Agree with Ian Cas Liber (talk · contribs) 13:06, 8 November 2021 (UTC)
  • Sure --Guerillero Parlez Moi 15:31, 8 November 2021 (UTC)
  • Go frit. ——Serial 15:34, 8 November 2021 (UTC)
  • Agree there's no further need to have FARs on the FAC page. There was a problem, that problem is plainly solved.--Wehwalt (talk) 17:02, 8 November 2021 (UTC)
  • Be aware of the consequences (for when criticism arises in the future). James Joyce is being restored to FA status at FAR entirely through the efforts of a new FA writer, who saw the FAR because it was linked from this page. Does FAR have enough of its “own momentum”? I doubt it; it has increasingly more editors who are committed to maintaining FA quality across the board (rather than just pushing through new FAs), but the risk of decreasing prominence is that we lose more core/vital FAs without new blood such as has occurred at James Joyce, precisely because the page was linked from here, which drew in new editors. Long story short: if you agree with ending the transclusion, then please watchlist WP:FAR and WP:URFA/2020, and get involved in those places with helping restore older FAs. SandyGeorgia (Talk) 15:44, 8 November 2021 (UTC)
  • If we do this, which I don't oppose, we should maybe have some more prominent banner on FAC linking to FAR, as well as concisely explaining what it does. Johnbod (talk) 15:59, 8 November 2021 (UTC)
    Great idea … and not just in the sidebars. While I am not suggesting the same format, notice how WP:GA gives equal placement to their equivalents, with both nominations and reassessments listed in the tabs at the top. This page could more prominently include both nominations and reviews. SandyGeorgia (Talk) 16:04, 8 November 2021 (UTC)
    Agree with Sandy that this is a good idea. I'd like to see FAR still transcluded, but it's becoming difficult for technical reasons. I'm sure that we can think of a solution to still clearly advertise FAR here, without necessarily doing it the same way we have been. Hog Farm Talk 16:09, 8 November 2021 (UTC)
    When I asked earlier why on the Wikipedia:Featured_articles page the link WP:FAR on the right cannot be listed above WP:FAC, there was a deathly silence. Fowler&fowler«Talk» 18:09, 8 November 2021 (UTC)
    F&f, I got very sick from my COVID booster, and haven’t yet caught up with this page, but in general … the tone was not conducive to healthy discussion anyway. Hopefully that has changed. The WP:GA setup gives some ideas of how we might improve, but the design issues are beyond me … perhaps the GA setup would work for us … I dunno? SandyGeorgia (Talk) 18:49, 8 November 2021 (UTC)
    Sorry to hear, SandyG. Hope you are feeling better. WP:GA is better than WP:FA, but I'm even suggesting tabs from L to R: Main | Criteria | Reassessment | Instructions | New Nominations | Instructions | etc. for a redesigned FA page. Fowler&fowler«Talk» 19:06, 8 November 2021 (UTC)
  • No objection but I agree with Johnbod. FAR has only entered my consciousness really because it's been transcluded on FAC recently. Also, if we're looking at load times, can we perhaps de-clutter the header (which is enormous!) and enforce the rule about no graphics/templates on FACs? I've seen a lot of people using {{xt}} and similar lately, for example, which is only a small template but lots of them will take a toll. HJ Mitchell | Penny for your thoughts? 16:11, 8 November 2021 (UTC)
    Just a reminder: enforcing the “no graphics/templates” is one of those (many) chores I have been harping about that any editor can do— it should not only fall to the Coords to check and enforce the instructions, and those templates are THE main problem in this issue (but I have said this before … ) Further, if regular (non-Coord) editors aren’t able to make any headway on this issue, then the Coords can do the sort of thing I used to do—appoint and empower an editor delegate to go through and remove them when the nominator or reviewer doesn’t, hasn’t, or won’t. This is work the entire community can and should enforce. SandyGeorgia (Talk) 16:24, 8 November 2021 (UTC)
    I'm guilty of some template use as it was my understanding that text-formatting templates like {{tq}} were fine, but I can quit using it if it helps. I will say that lately on several different computers (and with terrible home internet) I've had no trouble loading FAC or getting nominations viewer to work, but obviously these things aren't universal. ᵹʀᴀᴘᴘʟᴇ 16:37, 8 November 2021 (UTC)
    I've always kinda wondered about "and {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions" in the instructions. If it's so off-topic that it needs hatted, shouldn't it just be moved to the talk page of the individual FAC, with a link pointing to the new location where it was removed? And if it's not so off-topic that it'd be useful, then why hat it? Hog Farm Talk 16:45, 8 November 2021 (UTC)
    Yes. You can also empower a FAC delegate to move off-topic discussions on a FAC to the talk page of that FAC, providing a link back on the FAC. Or do it yourself … SandyGeorgia (Talk) 17:03, 8 November 2021 (UTC)
    @Grapple X and Hog Farm: I'm far from an expert, but as I understand it the server loads Wikipedia:Featured article candidates first, then every page transcluded onto it (ie the header, several hatnotes, FAR, and 40-50 FAC subpages at any one time), then everything transcluded on those pages (eg {{done}} and {{tq}}/{{xt}}) so you can see how even those small templates add to the load when used repeatedly. The page won't load fully until every last one has been loaded. You can always subst them if you want to keep the formatting without transcluding a template. HJ Mitchell | Penny for your thoughts? 18:21, 8 November 2021 (UTC)
    That makes sense. Worth rewording the banner to encourage this or to encourage regular wiki mark-up instead, perhaps? ᵹʀᴀᴘᴘʟᴇ 18:27, 8 November 2021 (UTC)
    Nobody's going to read that banner. It's way too big and is a massive information overload. About 80% of it needs moving to a subpage and replacing with a "see here for more information" link. HJ Mitchell | Penny for your thoughts? 18:37, 8 November 2021 (UTC)
    And we just need to start ENFORCING the no templates thing. Here we are, un-transcluding FAR (which helps maintain the overall FA pool) only because of template proliferation at FAC alone (because I do enforce it at FAR [2]), which will eventually affect FAC in precisely the same way EVEN AFTER we hamstring FAR by untranscluding (the original template limit problem occurred *before* FAR was transcluded). Sorry for caps et al … iPad typing stinks. SandyGeorgia (Talk) 18:52, 8 November 2021 (UTC)
And I reminded folk here [https://en.wikipedia.org/w/index.php?title=Wikipedia:Featured_article_candidates/Nonmetal/archive3&diff=prev&oldid=1053928136} Graham Beards (talk) 14:31, 9 November 2021 (UTC)
  • Let's put this to the test: If I go in and subst all these templates (and remove collapsed ones to the talk page), would people be fine with it? Jo-Jo Eumerus (talk) 19:06, 8 November 2021 (UTC)
    Moving collapsed discussions without an assurance that they contain nothing relevant to the promotability of the article would not be helpful to me as a coordinator. Gog the Mild (talk) 19:19, 8 November 2021 (UTC)
    Jo-Jo, per Gog ... this is why my suggestion in this area is that the Coords would or could appoint someone they know to understand the relevanced of various comments ... for example, I believe I had empowered Laser (before he was a Coord) to do similar for me, after having observed ample examples of his work ... SandyGeorgia (Talk) 20:09, 8 November 2021 (UTC)
    Jo-Jo, please don't; those templates can leave an unholy mess when substed. It's fine if you don't have to edit afterwards but that's unlikely to be true here. And I'm one of those who are not keen on material being moved to the talk page; I'm fine with a coordinator doing it when something is offtopic, but otherwise I'd prefer to leave material in place. I don't really like the collapse template in FACs either but it does have its uses. Mike Christie (talk - contribs - library) 19:27, 8 November 2021 (UTC)
    Coords aren't doing it at any rate, and pages are too long-- this is part of the problem. So FAR (which impacts overall FA quality) pays the price ? Grumble; this is not a good thing. The problems at FAC need to be addressed. SandyGeorgia (Talk) 20:13, 8 November 2021 (UTC)
Substing something like the collapse templates will produce a mess of markup afterwards but replacing them with strikethrough formatting to denote that things are addressed would save a lot of size; if people are okay with that I can go through and do that. I'm about to go remove any colour templates I've used in current reviews anyway. As an aside, I would be happier to keep FAR transcluded, although it's not somewhere I visit (shame on me) it does serve a purpose staying here. ᵹʀᴀᴘᴘʟᴇ 19:36, 8 November 2021 (UTC)
I would prefer as little strike through as possible. I struggle to read the text beneath it. Gog the Mild (talk) 20:14, 8 November 2021 (UTC)
With this in mind, and per Mike Christie's doubts that the collapse template is a huge drain on the inclusion limit--would you and the other co-ords prefer resolved comments be moved to the nom's talk page? They would still be viewable at a glance but would then not be involved in the transcluded section, keeping size down. I can start doing that in my reviews going forward if it's helpful, even as a trial run. ᵹʀᴀᴘᴘʟᴇ 20:25, 8 November 2021 (UTC)
I'd say it depends on the type of comments, per my personal opinion. If there's 5kb worth of minor prose comments that are mainly personal preference, then I'd strongly support kicking that onto the FAC's talk page. But if it's more major comments on say sourcing or scope/comprehensiveness, then I'd like to see the major elements stay up on the main FAC page. Stuff like "He contracted dysentery" vs "He became ill with dysentery" are the sort of comments we can kick to the talk page, IMO. Hog Farm Talk 20:38, 8 November 2021 (UTC)
Why aren’t they put on talk to begin with ? SandyGeorgia (Talk) 20:55, 8 November 2021 (UTC)
Removing colors helps. Putting lengthy reviews on talk to begin with would help even more. And please do start visiting FAR ... in some ways, the work there is much easier than at FAC (that is, when no editor will engage to improve, the commentary needed is minimal, and when editors do engage to improve, they just need guidance .. and there is no time pressure). SandyGeorgia (Talk) 20:11, 8 November 2021 (UTC)
I'm not sure that the colour and collapse templates will help to any significant degree; they're very small relative to the limit, even when multiplied by extra levels of transclusion. Mike Christie (talk - contribs - library) 20:14, 8 November 2021 (UTC)
So, considering different responses on this thread, reboot.
First, Buidhe as a Coord, why are you proposing to untransclude FAR without having clear consensus on how to solve the underlying problem, which is the size and duration of FACs? FAR is not the problem, and there has not been recent consensus on how to solve the FAC problem. As a Coord, what is your proposed solution? (I ask as several of the newer Coords reject the ways I managed the size of the FAC page …. What is the new proposal for managing the page ?
Gog the Mild would similarly like to hear your views. If you don’t like strikethroughs and don’t like comments moved to talk, what is your preference?
I continue to suggest that the most significant way the FAC limits problem could be solved would be to stop putting lengthy, line-by-line reviews on these pages to begin with. Lengthy reviews can be conducted on talk, and summarized back to the FAC. It works at FAR; it has worked here in the past. I don’t think it helpful to be damaging FAR when the FAC problem causing the limits problems has not been addressed. SandyGeorgia (Talk) 20:39, 8 November 2021 (UTC)

Is there a way to figure out which FACs are taking up the most transclusion space in general? My gut instinct is that the color-templates within the collapse box at Wikipedia:Featured article candidates/Harry S. Truman 1948 presidential campaign/archive1 are likely taking up a disproportional load. Although I do personally lean towards the side of thinking that the collapse boxen aren't the best things to use on a transclusion-heavy page, especially when they contain templates within. Hog Farm Talk 20:44, 8 November 2021 (UTC)

I'm on mobile at the gym right now but when I'm home it should be a simple matter of checking page sizes for each nomination which I'm happy to do. The biggest ones may be template heavy, or they may simply be the most prose heavy. ᵹʀᴀᴘᴘʟᴇ 20:51, 8 November 2021 (UTC)
It’s not only size … it’s the number of templates that affect the limit. SandyGeorgia (Talk) 20:53, 8 November 2021 (UTC)
Poor DrKay has answered this question so many times … and based on that, you are probably right about Truman. So, the point has always been, once you untransclude FAR (which is the canary in the coalmine which indicateds when FAC is in trouble), the FAC reviews will just grow longer and longer, ala Truman, until the same problem re-occurs. SandyGeorgia (Talk) 20:51, 8 November 2021 (UTC)
Yeah, removing FAR is not going to be as big of a boon as is thought. It's just a one-time quick fix. We need to find a solution that actually solves the root of the problem as to why FAC transclusion length keeps multiplying, or we'll be back here in a year. Hog Farm Talk 20:56, 8 November 2021 (UTC)
I have given my view above Sandy. What was unclear about it? Gog the Mild (talk) 20:56, 8 November 2021 (UTC)
Gog, I have re-read the page to see if I missed a comment, and can find no indication that I did. I read above that you are against moving collapsed discussions to talk, and you don't like strikethroughs because you have a hard time reading through them. (I am not sure why reading through struck discussions of resolved prose nitpicks is helpful, but I guess you mean you like to read even all of those, even when resolved.) Your responses tells us what you don't like. My new question is what you prefer to see or what alternatives you propose in terms of addressing the length and duration issues that lead to page loadtime and template limit problems. Many alternatives have been proposed over the years. What is your preference? Removing FAR will not make this problem go away. SandyGeorgia (Talk) 22:14, 8 November 2021 (UTC)
The title of the thread is "Time to stop transcluding FAR". My comment above, which seemed and seems clear to me, was to agree with Ian Rose's opinion that "I favour removing FAR from the FAC page, not because I have trouble loading the page but because I believe FAR has its own momentum again and doesn't need to be conjoined with FAC. From a practical point of view I'd also prefer to be able to go to the end of the page and find myself looking at the oldest FAC, not a FARC page." Gog the Mild (talk) 22:57, 8 November 2021 (UTC)
Yes, this has been stated and repeated. My question is a different one. How would you solve the underlying problem that leads to this proposal? “!Voting” without discussion rarely leads productive places, and I see this discussion started right away with the !voting. Would you support an envelope for all the FA pages, similar to what is at WP:GA? Or going to a GA-like page, where all FACs are not transcluded to one page that has to load all at once? Or doing something different to contain sprawling FACs? The issue of transcluded FARs is inseparable from the problem of FACs stalling the page, and should FA become (perceived as) a one-way process (once a FAC, always a FAC), that is detrimental to its reputation, just as seeing more and more FAs delisted if participation at FAR declines is. SandyGeorgia (Talk) 23:44, 8 November 2021 (UTC)
Removing FAR would be an improvement at least for now. In the long run I'd like to see other ways of reducing page length and complexity. For example, putting prose comments on talk pages instead of the FAC, which would also make it easier to evaluate consensus for promotion and quicker opposes and archival for FACs that aren't quite there yet. (t · c) buidhe 21:11, 8 November 2021 (UTC)
Re: moving "prose comments" to talk pages, does that mean reviewers will need to split their reviews across the nomination page and the nomination talk page? The Rambling Man (Keep wearing the mask...) 21:15, 8 November 2021 (UTC)
TRM, the way I'm imagining it, the review stays in one place while is active, and then once all comments are addressed and a support is made, the more minor comments (such as pointing out choppy phrasing or typos) would then move, in order to save space. Hog Farm Talk 21:24, 8 November 2021 (UTC)
Ah, that doesn't appear to be what buidhe is saying below. I would have thought that FAC is keen to retain reviewers, so making their lives more difficult or belittling their efforts as "personal preference" or whatever doesn't seem like the best way to achieve that retention. I would say that once a reviewer has indicated a support, their comments could be moved to the talk page, however I am aware of a number of reviewers who also read through other comments before making their own. Once again, having comments, even resolved ones, over two pages, just makes reviewers' lives more difficult. I suggest that should be be the last thing this process needs right now. The Rambling Man (Keep wearing the mask...) 21:32, 8 November 2021 (UTC)
I'm not trying to belittle anyone's reviewing efforts. It's just that it's a basic fact that there are "major" and "minor" comments at any FAC (although there's not a clear bright line between the two). And some comments are frankly preference, and we all make them. Cases where there's nothing purely wrong with the current wording, but we believe that an alternate wording would be better. There's nothing wrong with that, just that those aren't major-level comments. As to reading through other comments, I think we would both agree that comments potentially relevant to further FA reviewing would be good to keep on the page. But I don't reviewers or coords are going to need to see comments such as "This sentence is missing a full stop" or "this word is misspelled" or "you start too many straight sentences with 'the'" once they are resolved. Although I think nipping the templates issue in the bud is going to be the better result. Yes, we can fix the problem now by removing the transclusion to FAR, but that doesn't solve the issue long-term. We need a real long-term solution, or we'll be here again next year. Hog Farm Talk 21:50, 8 November 2021 (UTC)
These types of comments might not be considered high level, yet yet. Back in the day I remember learning quite a bit from Brian Boulton's comments that were often a list of prose issues. An example that comes to mind is when to hyphenate or not - I sandboxed the rules somewhere because I can never remember them exactly. This is important for others who might read the review and learn from it, but wouldn't see it on the talk page. There's nothing wrong with trying to use as few words as possible, striking (when did striking go out of fashion?), and once the review is done moving to talk with a visible link, but asking reviewers to split between pages might be problematic imo. Victoria (tk) 22:05, 8 November 2021 (UTC)
Indeed. Prose comments (I thought) are equally, if not more important to our readers than MOS compliance etc. I used to do a fair bit of fixing issues for nominators, but realised that sometimes those nominators need to pick up some of the skills for themselves. Much like you've said Victoria, Brian's comments have almost certainly improved a number of Wikipedians writing style. Perhaps flip it, add all the nausea of image/source/MOS reviews to the talk page, but leave the actual prose review on the nom page. That's more logical. But in no way is splitting a review logical, unless it is completely concluded. But even then, you probably just move the whole review, not try to cherry-pick the "prose preference" elements. The Rambling Man (Keep wearing the mask...) 22:18, 8 November 2021 (UTC)
Not saying that you are trying to belittle anyone's efforts, but there's a theme that runs through this talkpage and its archives of snobbishness about certain review styles. But in any case, as I said, the kinds of reviews I leave are a mixture of all kinds of things, and I couldn't split the review without it being a major pain in the arse. I guess all I'm suggesting is that anything which makes any reviewers' lives more difficult should not happen. And yes, this is merely trying to fix the symptoms of a technical constraint, I understand. I would have thought applying the "talk page for comments" would be ideal for FAR and FARC, leaving just the names of the nominations listed here. The Rambling Man (Keep wearing the mask...) 21:58, November 8, 2021
I am saying/have always said something a bit different from what a few others are saying here. We must see major concerns about things like copyvio, reliability of sources, comprehensiveness, POV, improperly licensed images, and so on ... on the FAC page. Those kinds of things should be looked at quickly as they often indicate non-starters; they are the kinds of things that can lead to quick archivals. But once a reviewer is satisfied that the article has sound bones, and are only doing prose nitpicks, can the reviewer not say something like, "I am satisfied on x, y and z, but have some prose concerns that I will review on talk"? Then the Coords would be expecting them to come back and say yea or nay, once done. We have some reviewers who will literally add a hundred lines of prose nitpicks to a FAC, with the associated templates and color coding, and those are always the ones stalling the page overall, affecting load time, and triggering template limits. (Begging the question that, when there is that much prose repair needed, why were not just a few samples given, followed by an "Oppose, come back after a copy edit"? But that is a different topic ... ) SandyGeorgia (Talk) 22:22, 8 November 2021 (UTC)
Well I would consider reviewing the prose for quality writing and professional standard not to be nitpicking, I think it's one of the WIAFA criteria. And prose review includes more than just so-called nitpicking, it includes comprehension issues, like understanding terminology, jargon etc. And comprehensiveness concerns, like "and what happened then?" It's completely unreasonable to expect (good faith) reviewers to make two separate passes at an article and generate a "nitpick" list and a "non-nitpick" list. This will just drive reviewers away and result in even slower progress at this process. The Rambling Man (Keep wearing the mask...) 22:30, 8 November 2021 (UTC)
I support Grapple X's idea of moving resolved comments to the talk page. FACs become such sprawling, unwieldy messes that when I'm reviewing, I'm never sure how to organize my comments in order to maintain clarity, even though my reviews aren't as lengthy as some. I often resort to section headings, even though you're not supposed to do that. If the talk pages were sectioned, and resolved comments were replaced by support votes followed by a note, that would allow coordinators to focus on any outstanding comments and check the relevant section if necessary. Hog Farm wants to leave the main points on the FAC page, but if reviewers left a note about what they checked, that could help. Say something like:
  • "Support. Reviewed prose and comprehensiveness; reviewed and spot-checked sourcing; see Wikipedia talk:Featured article candidates/X/archive1#Comments by A. Parrot".
A. Parrot (talk) 21:18, 8 November 2021 (UTC)
As a coord, it would help me to put straightforward matters like spot-checks, proofreading, or resolved image copyright questions on the talk page. There should still be a comment on the main review page as A. Parrot notes, so split reviews are not an issue. (t · c) buidhe 21:26, 8 November 2021 (UTC)
But when I review an article, I'm doing prose checks, MOS checks, reference checks, image checks. Am I supposed to write these split reviews concurrently then? The Rambling Man (Keep wearing the mask...) 21:29, 8 November 2021 (UTC)
Just to clarify, I was only suggesting moving reviewer comments to the talk page once resolved, not while in progress which could get clunky. I don't mean to criticize reviewer efforts, but if there are extensive improvements necessary, consider opposing. (t · c) buidhe 22:16, 8 November 2021 (UTC)
But I often pick up FACs from the urgent page. Usually they have already gained a couple of supports. It seems churlish and wasteful to just oppose those instead of giving a thorough review, simply because there's a template transclusion issue. From my point of view, and I know some people prefer me not to review here (indeed even review my reviews), I would not sanction splitting reviews at any time, but I would probably be almost okay with moving the content of a review, wholesale, after a support has been offered. But, as before, just addressing the symptoms of the problem really. The Rambling Man (Keep wearing the mask...) 22:23, 8 November 2021 (UTC)
Buidhe, I am unconvinced that a global or general "Move to talk page" is needed; it is more typical for a few sprawling prose reviews to be the cause of the page stalling. First, convince reviewers if their reviews need to be that long, they should just give samples and oppose. Second, the limited times when you might need to move prose reviews (and I believe these are quite limited) to talk is when they indeed are huge, sprawling, and mostly resolved, such that they offer little of benefit for Coords to review, and discourage other reviewers from engaging at all. These few and far between are atypical, and yet we are proposing to lose helpful FAR feedback because of these few FAC reviews, which could be handled differently ! SandyGeorgia (Talk) 22:28, 8 November 2021 (UTC)
I don't do "samples". I review articles line-by-line, in chronological order. I often do it in sections allowing for a nominator to pick up the themes. You can't "pick out" prose review elements from my reviews because they are amongst all the other comments relating to other FA criteria. Reviewers will be discouraged if they are not allowed to make comments in the way which best suits them, not some curious method to appease a coding failure. Don't encourage "reviewers" to modify their behaviour, encourage "nominators" to be more active in picking up on what is expected of them. The Rambling Man (Keep wearing the mask...) 22:34, 8 November 2021 (UTC)
I didn't intend to suggest it as a new norm, but personally just something I would do going forward in order to bear in mind the impact of my own reviews on the overall load. For example Gog the Mild & Buidhe, the reviews I have added at this and this nomination are entirely resolved, my suggestion would be to move those subsections to the nomination's talk page and leave behind a message like "Support: For resolved comments see talk page". I wouldn't intend for it to be the default but for anyone who took it up it would reduce some size and template calls (even things like {{U}} pings will added up surely). As I see it, right now I have no trouble having FAR transcluding, and if I can further reduce my own footprint on the page, it'll only lessen any load. ᵹʀᴀᴘᴘʟᴇ 22:50, 8 November 2021 (UTC)
I think this would be just about tolerable, as long as copy/pasting the comments by a co-ord still met any requirements of copying within Wikipedia. It would make other reviewers' lives more difficult to have to cross-refer to another page to see if something they were bringing up had already been discussed though. The Rambling Man (Keep wearing the mask...) 22:56, 8 November 2021 (UTC)
Grapple X, you don’t need anyone’s acquiescence to do that; it is the established norm. The FAC instructions are:

To withdraw the objection, strike it out (with <s> ... </s>) rather than removing it. Alternatively, reviewers may transfer lengthy, resolved commentary to the FAC archive talk page, leaving a link in a note on the FAC archive.

That is a valid option for the few FAC pages that are causing these stalls. Removing FAR will not solve the problem, and will not help the overall quality of the FA pool, upon which FA reputations depend. As TRM mentions, though, be sure to follow WP:CWW. SandyGeorgia (Talk) 23:06, 8 November 2021 (UTC)
Those instructions should be modified to put the emphasis on co-ordinators to do that. That is, after all, the kind of thing we'd expect someone co-ordinating to make a judgement call on. We should remember how lucky we are to have enthusiastic, even nitpicky, reviewers at this process, placing more burden on them than is already expected is unnecessary. The Rambling Man (Keep wearing the mask...) 23:10, 8 November 2021 (UTC)
(edit conflict), (edit conflict) Grapple X, reviewers present their reviews in many different formats and find an interesting number of ways of not following the approach suggested at the top of the FAC page. I am grateful for each and every one of these and rarely ask a reviewer to change their approach. If they are more comfortable with a particular style, fine: I want them to feel comfortable, they will then be more likely to review more often. If you are asking my personal preference, then as a reviewer I have a preference for all comments to be on one page, uncollapsed, for reasons similar to TRM's; as a coordinater I have a strong preference for all comments to be on one page - I worry that I may overlook comments which have been moved, or originally written, elsewhere. But don't let this effect whatever you are comfortable with doing. Gog the Mild (talk) 23:12, 8 November 2021 (UTC)
(edit conflict) I know it's currently allowable but every tranche of co-ords will have their preferences when it comes to these things and if it slows down their processing of nominations it's counter-productive—a leaner nomination is nice, but a closed one is leaner still. So I'd be reticent to start if it was going to something the current co-ords would find cumbersome. I'm with you in not wanting to remove FAR so threading the needle in terms of options that streamline the page but don't impede is key; I don't mind doing extra legwork myself but I don't want to give someone else any. ᵹʀᴀᴘᴘʟᴇ 23:14, 8 November 2021 (UTC)

There are multiple topics being discussed in this thread, partly because some commenters see them as interlinked. No doubt they are, but it does make it harder to get to a decision on the original point (or any point). The topics I see are

  • Should FAR be dropped from the FAC page, or remain, or be linked in some way, such as by a system of page header tabs like the ones used by WP:GA?
  • What should be done about the template transclusion limits? Would it help to ban all templates, including the smallest, or would that not help?
  • Should commentary be moved to talk on FAC pages in some circumstances? At the coordinators' discretion? Only if the commentary is perceived to be offtopic, or just to reduce page size?

I'd like to suggest that discussion of the second and third points be continued in new sections or subsections, and we use this section to try to get to a consensus on what to do with FAR.

I !voted above to remove FAR, and I very rarely get involved with FARs myself, but I have to say I'm sympathetic to the point of view that FAR is healthier now partly because it has been transcluded here. So I'll change my comment to say that I think it has to be removed at least temporarily because of the transclusion problem, but I'd be open to returning it if we can solve that. I also like F&f's suggestion of using something like the GA tab layout; it's clean, loads quickly, and it means that no matter what page you're on you get an overview of the whole process. Mike Christie (talk - contribs - library) 23:20, 8 November 2021 (UTC)

The issues are intertwined, Mike; the unfortunate thing is that people started !voting on this proposal before discussion. If it comes to this, I must, reluctantly, lodge an OPPOSE on the matter of untranscluding FAR, because no one yet has offered a solution to the underlying problem at FAC, which will continue whether FAR is untranscluded or not. In fact, the problem of sprawling FACs is probably even a much bigger problem, as they are so off-putting to subsequent reviewers. SandyGeorgia (Talk) 23:49, 8 November 2021 (UTC)
  • Support removal of FAR, the processes do not have to be conflated in this way at this time. As for "sprawling FACs", I think this is a misnomer: what is happening is highly detailed, scrutinising reviews. I have seen plenty of evidence to support the fact that these reviews do not dissuade subsequent reviewers from participating; they may give pause to subsequent reviews while the detailed review is being undertaken so as not to duplicate effort. I do not believe that trying to mandate how a reviewer makes their review is helpful in any sense to the process, and forcing some odd split in reviews for "nitpicks" vs "non-nitpicks" just won't work. Perhaps all nominations should go into a "pending" area so that co-ordinators can assess their suitability for the process and make a determination as to whether the candidates are suitable for a full and comprehensive FAC process. Otherwise, there is no alternative other than moving completed reviews to the talk page, and that, as I have noted, is detrimental to some reviewers/co-ords who prefer the complete picture in one place. Once again, to reiterate, making reviewers lives harder is absolutely not the answer: we should be grateful for the time and effort these volunteers put in and receive nothing other than negativity toward their reviewing style. The Rambling Man (Keep wearing the mask...) 08:18, 9 November 2021 (UTC)
  • Support. I agree with TRM. It's saddens me that this has turned into a vote when it's a technical problem. We should be helping reviewers and not trivialising their input. Nitpicks are just as important as factual inaccuracies in my view. They give our FAs their final shine, and banishing reviews to the article's Talk Page is not the way to go. And, I get the impression that FAR trumps FAC in the minds of some. The problem is the FAC page is riddled with transclusions and the software can't cope for all users. Graham Beards (talk) 09:17, 9 November 2021 (UTC)

I'm still concerned that this is not a long-term technical solution. We can only do this once. It'll only buy us a little bit of space, we'll fill that space back up, and be back here all over again in the future. I'm afraid we're going to have to make actual changes to what we do, or else this'll happen over and over again, and we won't have a quick fix next time. It's the templates that seem to be what's driving it up, especially templates within templates. Hog Farm Talk 14:28, 9 November 2021 (UTC)

I put together the outlines of a long-term solution last year, but then did not pursue it further once Ealdgyth went on break, and it appeared that rigorous sourcing reviews weren’t going to be done any longer at FAC anyway. That proposal hinged around getting the “good bones” (good sourcing) resolved before moving to the next steps. But without someone doing what Ealdgyth used to do, FAC pages seem to be mostly about prose, with less emphasis on core policies, so I don’t see how that proposal could be implemented without a stronger emphasis on core things like sourcing, NPOV, etc over prose. No one is advocating against prose nit-picking; prose review has a place in every FAC, but reviewing prose that might be based on poor sourcing can be a timesink.
I do agree this problem will just be right back in a matter of time (after leading to more FAs delisted because of less participation at FAR), because the FAC template limits problem first occurred years ago, is not due to the FAR transclusion, and the problems have never been resolved by addressing the issues here at FAC.
The notion expressed above of a sense that FAR is more important than FAC is odd: an equal footing for promoting and re-assessing is what is being advocated. SandyGeorgia (Talk) 14:40, 9 November 2021 (UTC)
While I don't know what the solution is, I agree with Hog Farm and Sandy that removing FAR won't solve the deeper problem. And I think of it as more than a technical issue. The FAC page is enormous and hard to read, even to me, a fast reader who has been participating in FACs for a decade. How dauting must it be for anyone thinking about nominating, or even reviewing, for the first time? There has to be some way of streamlining this page and making the process more accessible. I supported moving comments from support voters because it seemed like the approach that would cause the least disruption to individual editors' reviewing styles. A. Parrot (talk) 15:10, 9 November 2021 (UTC)
  • Support. As a reviewer I have found it very useful to have all the commentary in one place; it helps focus attention on outstanding issues, and also makes it more likely that disputed or problematic parts of the review will receive greater independent attention. I would also support a transition to a GAN-type model where the FAC pages themselves are not loaded onto a single parent page, but only summarized in some way. But this is a good first step. Also, per TRM and Gog. Vanamonde (Talk) 15:20, 9 November 2021 (UTC)
  • Extrapolating from Graham Beards below, it seems quite likely the proximate cause of the "FAR is broken" problem is the extra length of FAR, which is the result of increasing activity at FAR. If that level of activity is going to continue, it seems likely to me that there really isn't a viable way to keep FAC and FAR transcluding to the same page, regardless of template usage and the transfer of prose reviews to the talk. I'm surprised this hasn't come up before. Vanamonde (Talk) 01:16, 10 November 2021 (UTC)
    Vanamonde93 That is backwards; when the FAR transclusions drop off the FAC page, it is always because the FAC page is exceeding limits on its own. FAR is the canary in the coalmine which lets you know FAC has a problem. If you didn’t have FAR at the bottom of the FAC page, as the FAC problem grows, you will just lose the last FACs on the page— the same thing we are seeing in the FAC archives. When you look at some of the very long FACs filled with templates, you can see why FAC has a problem, while FAR doesn’t (these templates don’t take over FAR pages). And the subject has come up … over and over and over … people talk past each other, or don’t read, or don’t understand the template issue, or are simply rude, with responses like TLDR rather than engaging a complicated matter and responding to direct questions when asked. SandyGeorgia (Talk) 04:12, 10 November 2021 (UTC)
    I appreciate that there's more to the issue than FAR, truly I do. I'm all for more discipline with templates, and/or a broader restructuring of the nomination pages. But with respect, the notion that FAR is irrelevant to this problem is simply incorrect. There is a non-zero number of templates on this page as a result of FAR, and therefore a non-zero period of time when the page breaks solely due to the presence of FAR on it. Removing FAR would solve the problem for those periods, and as such is an improvement. Relative to the average Wikipedia discussion, this one is being conducted with remarkable decorum, I'd say: If the coords disagree with you (and they seem to), surely it's worth considering they have good reasons for doing so? Vanamonde (Talk) 05:41, 10 November 2021 (UTC)
    There is a non-zero number of templates on this page as a result of FAR ... correct ... and the template limits are exceeded via FAC (which is first) before the FAR page is ever reached. The page stalls mid-FAC, and the enormous majority of the templates are on the FAC page. I have no indication that all the Coords agree that we have found a solution, or that we have even discussed possible solutions to the actual problem. SandyGeorgia (Talk) 08:27, 10 November 2021 (UTC)

Support removing FAR. Victoria (tk) 21:24, 11 November 2021 (UTC)

|}

Underlying problem

Wow; scroll down the FAC page and look at the sea of green and collapse templates, and that the source of this problem is not FAR is clear. Look alone at the amount of templates in just this one FAR !!!

So … whatever solution is proposed, please be aware of history. The template problem first surfaced not at FAC but in the FAC archives. The stalling of FAC is not the only issue here; it is that, once FACs are promoted or archived, the archives are truncated when the templates limit is reached. Removing the FAR transclusion (where template limits are not the problem) will not solve the eventual problem in FAC archives, when these nominations are closed. This hasn’t been a problem lately because FAC throughput is so low, but if FAC throughput continues to pick up, the problem will resurface in FAC archives. SandyGeorgia (Talk) 15:49, 9 November 2021 (UTC)

    • Sandy, it already has. Hog Farm Talk 15:51, 9 November 2021 (UTC) (INTERJECTION: now fixed, see Wikipedia talk:Featured article candidates/Featured log/October 2021. SandyGeorgia (Talk) 16:53, 12 November 2021 (UTC)
      Yes it has. The final FACs in archives are truncated and not showing.
      Mike Christie how will the inability of FAC archives to display all pages affect your data gathering? October overall had 46 promoted, but 43 are displayed; what gives with none of the Coords addressing this? (The note was added by the editor who helps maintain WP:FAS and noticed the problem.)
      To the rest of the Coords; by allowing this charade to continue, you are not leading FAC out of a long-standing problem with comprehensive solutions, and when I ask what your comprehensive solutions are, you are not answering the direct question. The proposed solution (to remove FAR because of a FAC problem) will not solve the problem, which is now also in the FAC archives. SandyGeorgia (Talk) 16:03, 9 November 2021 (UTC)
      And now someone has to manually go in to the FAC archives, find the offending FAC, and remove the excessive templates— exactly the problem we are discussing here. SandyGeorgia (Talk) 16:04, 9 November 2021 (UTC)
      A few of those will be text formatting in my reviews, if no one minds editing after the fact I'll go remove those from my contributions to see if it helps. ᵹʀᴀᴘᴘʟᴇ 16:16, 9 November 2021 (UTC)
      Grapple X, it's worth a try. (No one can tell you not to fix your own edits in a way that complies with the FAC instructions, so go for it.) SandyGeorgia (Talk) 16:50, 9 November 2021 (UTC)
      Changes made, still only showing 43 of 46 there so the impact hasn't been much. That's not to say it's not something which would help if more widely followed though. ᵹʀᴀᴘᴘʟᴇ 18:12, 9 November 2021 (UTC)
      Sandy, there's no effect on my data gathering -- it added a few seconds to my effort to capture the data, but it wasn't really an impediment. I agree removing FAR doesn't fix the problem, but nor does leaving it where it is. And I think we're going to need consensus on what "the problem" is before we could make other kinds of changes such as banning all templates in all cases, or moving text to talk pages, and I don't see that consensus emerging quickly. So I still think we need to drop FAR for now, and then we can take a step back and see what a longer-term solution might look like. Mike Christie (talk - contribs - library) 17:25, 9 November 2021 (UTC)
      Thanks, Mike. On the other hand, if we were to run reviewer stats the way I used to, the ability to read the entire page was key. My method was not to indicate who had the most reviews, but who had the reviews that provided the most information that was key to promoting or archiving the FAC (quantity over quality)-- that is, the most useful reviews. I read through each FAC at the end of the month and assigned scores: +3 (info essential without which the review would not have been promoted/archived), +1 (helpful but not essential), 0 (neutral, passing commentary and the like), -1 (a review that went the opposite direction of subsequent reviewers or failed to identify key shortcomings, just a driveby, etc), and -3 (similar to -1 only worse). This method meant I had to read the entire archive, top to bottom. And I think it resulted in us awarding the truly valuable reviewers, while it provided a quantitative method of identifying the less helpful reviewers (although they were never named, they could read the writing on the wall and hopefully adjust). I have no indication that our current Coords are interesting in replicating this kind of effort, but it provides an example of when one may need to read the entire FAC archive, top to bottom. It is not hard to imagine other scenarios where one might want to search the entire archive for given keywords; right now, we can't do that.
      I agree with you that we should identify the problems and potential solutions; that should have been done first, ala discussions such as those you mention here. If the Coords allow this premature discussion to close without addressing the real problems (and let's be clear, the Coords' voices speak louder than others, and this proposal was advanced by a Coord), they do us all a disservice, as we have been talking about those issues for years, and once this thread closes, we are unlikely to see solutions advanced.
      The cart went before the horse here. And there is another troubling aspect to how this proceeded. This premature proposal was put forward by Buidhe, who is now speaking not just as an editor, but as a Coord. When a Coord has a clear conflict, they should proceed very carefully. But ... As it has been revealed in the last few hours:
      1. Buidhe did not give FAR the courtesy of a preliminary discussion (see here).
      2. One of the most problematic FACs on the page, causing page load and template limits issues, turns out to be one of Buidhe's (Wikipedia:Featured article candidates/Armenian genocide/archive2)
      3. The FAR page is running slower than need be because Buidhe (and a few others) failed to follow FAR instructions: see here.
      I don't intend to imply any malice, or intent, or neglect, or deviousness, but Buidhe as a Coord should not be advancing something in which they are so personally invested and involved-- it's not a good look in terms of conflict of interest. The appearance is that this is what we want FAC to look like. Is it? We don't know; we haven't had that discussion. Neither is it a good look for Hog Farm to be the one to finally notice that there is a problem in FAC archives, as I have been saying, this far in to the discussion, which no other Coord even mentioned (yea HF !!!)
      My suggestion is that the Coords should put this premature discussion on hold, and instead heed your call for a discussion of the underlying issues first. They should encourage us to hold all of those discussions, and explore all options ... like a hat similar to the one at GA, a two-step process similar to my proposal, any kind of redesign that will address these issues, elimination of all templates at FAC ... whatever it takes. Remember that when I first encountered the template limits issue in the FAC archives, we did not have that blooming pingie-thingie, with everyone using templates to ping everyone else, so the problem is much bigger than it was before. A discussion focused on all of those issues is much more likely to get us to a place that is beneficial for all: FAC, FAR, reviewers, nominators, and the overall process. Making the overall process a "once an FA, always an FA" by minimizing re-assessment of old FAs via WP:FAR puts us even below the WP:GA process, which gives equal billing to reassessment. What have we come to ? SandyGeorgia (Talk) 18:30, 9 November 2021 (UTC)
      I'm at work, so really can't reply at length, but I just want to register my opinion that I think Buidhe did nothing wrong at all. There's been plenty of recent discussion of the FAR issue and I saw her post as an attempt to determine if those discussions had led to a consensus -- plus as I said above it's clear to all of us that the kind of broad discussion you're asking for will not lead to a quick resolution, and FAR is broken right now. I think Buidhe's post was appropriate for a coord. Mike Christie (talk - contribs - library) 18:39, 9 November 2021 (UTC)
      Mike Christie did you mean to type “FAR is broken right now”, or did you mistype FAR for FAC? FAR is functioning better than ever, while FAC is stalled. If FAR is broken, could you explain how? Is a “quick” solution better than a functional long-term solution? If we need a “quick” solution to the temporary FAC brokenness right now, we already know what that is and I explained it in my post above. SandyGeorgia (Talk) 20:07, 9 November 2021 (UTC)
      Yes, I meant FAR -- it is broken in that it is not transcluding here. I agree the process itself is functioning well. I wouldn't say FAC is stalled, though; from my perspective it seems very healthy -- better than in years. Mike Christie (talk - contribs - library) 21:43, 9 November 2021 (UTC)
      Ah … I completely misunderstood your comment; the transcluded FARs have not fallen off the page for me for quite some time, so I thought you literally meant FAR was broken, rather than the transclusions :) :) SandyGeorgia (Talk) 03:52, 10 November 2021 (UTC)
  • I have not read Sandy's post immediately above, it falls well past my TLDR threshold. However, I would like to repeat my comment from my first post "Thanks to Buidhe for raising it". Sandy, can I strongly recommend that you let the thread "Time to stop transcluding FAR", ie this one, play out to a conclusion. And that if you wish to discuss other matters on this page, including ones which you feel are intimately linked, you do so in a separate thread. Gog the Mild (talk) 19:01, 9 November 2021 (UTC)
    I appreciate the suggestion, but I have already explained why that is a sub-optimal approach to a very big problem that FAC has. Perhaps after you carve out time to read my posts, you will see that. SandyGeorgia (Talk) 19:57, 9 November 2021 (UTC)

Glancing at the top only of the October archive, the likely problems leading to page load and template limits are easy to spot. (I haven’t even moved beyond the first dozen).

But the real problems are probably ones like this:

Fixing those last two would probably resolve the archive problem, and discontinuing same on FAC page (per instructions already in place) would probably solve the load time issues. I stopped reviewing archive after those two. SandyGeorgia (Talk) 20:30, 9 November 2021 (UTC)

I took the step of removing most of the templates (all bar the ping ones, for fear of suddenly re-pinging all involved) from the Sound and Vision page; it has had no impact on loading the truncated nominations from the archive but it does seem that now the I'm Goin' Down nom is appearing as a link rather than a transclusion at the bottom, which it hadn't earlier for me. I can self-revert if that's a step too far but there's no material change to the archived nom, just some formatting no longer present. ᵹʀᴀᴘᴘʟᴇ 20:51, 9 November 2021 (UTC)
Thx, Grapple X. But if the new FAC trend looks like those two (and I only reviewed the top of the list), and if there is lots of that, it may take more to reduce the load. SandyGeorgia (Talk) 21:04, 9 November 2021 (UTC)
Part of it is my fault I did some pretty extensive quoting from sources at the James Longstreet FAC with the color template, when I definitely knew better. I can go through and remove the {{xt}} instances from my review there after work. Hog Farm Talk 21:11, 9 November 2021 (UTC)
Not sure ‘bout that … what we found in the past when DrKay looked at the data is that the templates within the collapses are the big problem. SandyGeorgia (Talk) 21:15, 9 November 2021 (UTC)
Sandy, I'm going to try to find time to look into this in more detail, but I'm pretty sure you're only technically correct. Yes, templates within a collapse template that includes everything in the collapse were a problem, but any collapse template in a FAC (or FAR) should be using {{collapse top}} and {{collapse bottom}}. Those are very cheap and unlikely to cause a problem. Mike Christie (talk - contribs - library) 21:47, 9 November 2021 (UTC)
I removed my template usage from Longstreet, and now 44 are showing on the archive, when only 43 were showing earlier. Hog Farm Talk 01:19, 10 November 2021 (UTC)
Verrrry interesting :) So, with a few more editors removing their templates, we might recover October’s archive. And, this means we need to find a new way of doing things here. This was always a problem in the past because of the high throughput at FAC. As FAC declined in recent years, we rarely had as many FACs in archives as October had. If throughput picks up regularly, this will be a constant problem. It needs to be comprehensively solved— not rushed, not blamed on FAR, not done piecemeal. Templates and archives are not getting along at FAC. SandyGeorgia (Talk) 04:06, 10 November 2021 (UTC)

For years there were only a few articles listed in FAR; now we have around 28! I still think this is stopping the page loading properly for some of us. (This is an issue for me both at home and at the hospital where I work, i.e. different servers). Graham Beards (talk) 21:22, 9 November 2021 (UTC)

I believe you :) Where we differ is that removing FAR will not solve the problem, because it is FAC that it is overloaded with templates and long pages. We can remove FAR, to the overall detriment of the process, and the same problem will still surface down the road because of the FAC pages. Bst, SandyGeorgia (Talk) 22:14, 9 November 2021 (UTC)
And I believe you. Clearly, we need to solve this problem. It might need a radical change in how we list FACS and FARS that is not so dependent on transclusions, which btw are often a total mystery to new editors as they were to me in 2007. Transclusion makes nominating FACS ridiculously complex as we all know: ) --Graham Beards (talk) 22:32, 9 November 2021 (UTC)
We are in violent agreement :) I want to see FAC solve FAC problems, and make it a priority to do so. These constant and inaccurate discussions blaming problems on FAR are … disengenous … and as soon as the threads close, things go right back to business as usual. And every time a discussion of FAC is attempted, it becomes personalized and goes nowhere … Best, SandyGeorgia (Talk) 22:46, 9 November 2021 (UTC)
  • Comment I am suggesting that the problem of emphasis might be on the FA page. Perhaps tabs along the lines of User:Fowler&fowler/Proposed tabs for the Wikipedia FA page (as suggested by Sandy and Mike Christie, though not necessarily in this form) might reorient the priorities, and therefore redistribute the effort. Fowler&fowler«Talk» 00:04, 10 November 2021 (UTC)
    Sure, why not. I also support simplifying the current header. (t · c) buidhe 00:13, 10 November 2021 (UTC)
    I really like the tabs idea. Could keep FAR at a higher visibility without tacking it onto the end of FAC. Hog Farm Talk 00:28, 10 November 2021 (UTC)
    (edit conflict) While it's a good step to clean up the header, I think Sandy is right that it's not going to address the issue with the size of review pages still being the main factor. Enforcing template usage is a step too but I'm wondering; is it possible to include an edit notice in the FAC preload (similar to the edit notice you would see upon editing a BLP page for example) which again draws attention to not using templates and offers mark-up alternatives to common templates? If it's something that could be inserted by default I would work on the text for it; I know it's not guaranteed to be followed but it would catch first-time nominators/reviewers' attention and likely also be seen by experienced reviewers now noticing a big bright box for the first time. ᵹʀᴀᴘᴘʟᴇ 00:33, 10 November 2021 (UTC)
    Not sure more instructions will help, Grapple, as we have now people breaching and openly disregarding the instructions anyway. I have found that collaborative editors only need to be asked once, and they don’t typically repeat offend. If someone just went through every FAC chunked up with templates, explained the problem, and asked editors to remove their templates—exactly as we do at FAR—that usually does the trick. Only takes one request.
    The much bigger problem is how to get back to the old adage “FAC is not peer review”, and get the lengthy peer reviews shut down and sent over there, to come back here better prepared. FAC is to determine if an article meets criteria; it is not the place to pull every article through until it does. We don’t have enough resources for that. SandyGeorgia (Talk) 01:24, 10 November 2021 (UTC)
    What you've missed, Sandy, is that the FAC instructions explicitly allow the use of colour, collapse, and quote templates: The only templates that are acceptable are xt, !xt, and tq; templates such as green that apply colours to text and are used to highlight examples; and collapse top and collapse bottom, used to hide offtopic discussions (internal curly brackets removed). This is why editors use them. There is no disregard for written instructions; but if these templates are the cause of the problem, then the instructions need to be updated. Mr rnddude (talk) 03:04, 10 November 2021 (UTC)
    You are correct, Mr rnddude, and my apologies for the oversight and to anyone I may have offended. (I would strike part of the comment above, but this specific issue template aside, we still do have reviewers and nominators ignoring other instructions, so I’ll not muck up the post above, while acknowledging the change I did miss.) Searching the history:
    The ten-year convention FAC had on no templates (between when we discovered the problem and when this change was enacted in November 2018), was based on
    this discussion. I was pinged to that discussion and apparently weighed in although as I indicated there, this was while I was still recovering from the subarachnoid hemorrage from the tree falling on me, and from the post I made, it is apparent I wasn’t fully digesting the issue or responding properly. Considering the state of my brain at the time, I am not surprised I forgot this, and I apologize.
    So, the change to a ten-year convention was made three years ago because people thought it wouldn’t cause a problem. Now we know it has, and wasn’t a good decision. We have many options. We could go back to the simpler instruction of no templates. Or redesign the whole page. Or many other possibilities … but removing FAR is not the something that will solve this problem. SandyGeorgia (Talk) 03:48, 10 November 2021 (UTC)
    I would not be opposed to moving to a header such as F&f suggests, but that still won’t solve the problem we are having now. FACs cannot show in archives because of the template problem. GA uses a header like that, but they don’t keep archives. I am having a hard time envisioning how the FA process would function without our archival system. The way we operate now is that, by definition, an article gains the star when a Coord moves it to the promoted archive, and then FACbot codifies that on talk. Someone would have to walk me through how we would do our basic bookkeeping, along with things like Mike stats, if we ditched the FAC archives So, we still need to address the problem with how FACs are currently conducted, even if we move to header that puts all parts of the process in tabs at the top. SandyGeorgia (Talk) 03:57, 10 November 2021 (UTC)
  • Support removal per nom. There are some interesting and sensible ideas being suggested for improving FAC, in particular enforcing the no-template rule and perhaps splitting the page into tabs. But those are separate suggestions, which can be discussed separately, while this is a concrete proposal from a coord to fix an issue that's occurring right now. Of course, I wouldn't want to risk FAR returning to the moribund state it was in a few years ago, so we can monitor the situation, but from Buidhe's comments and others here, as well as my own observation, it looks like for now there is enough enthusiasm and activity there that it doesn't need the extra publicity of being on the FAC page at this time.  — Amakuru (talk) 08:54, 10 November 2021 (UTC)
  • Comment there's an overwhelming majority in favour of removing FAR for the time being, so that should be enacted immediately. I do not believe it is irreversible and should the situation improve through other methods, it can be replaced. Other such suggestions should be better formed and proposed here instead of this rambling mass of commentary. The Rambling Man (Keep wearing the mask...) 15:47, 11 November 2021 (UTC)

New proposals

There are several new proposals in the new thread on this page, #Additional solutions to page limitations, with options that could change how we view this proposal (eg, transclude each FAR individually rather than via one transclusion, which ups the template count). I don’t believe we can view this proposal outside of the context of the other proposed changes to address the overall problem. SandyGeorgia (Talk) 17:23, 11 November 2021 (UTC)

Other proposals have been mentioned throughout this debate, and several people (myself included) have made it clear that they support removing FAR even while those other points are still being discussed. I think the coord who closes this would be justified in regarding this section as a consensus to remove. Mike Christie (talk - contribs - library) 17:34, 11 November 2021 (UTC)
Then they will surely say that :). Do they still support or oppose this proposal, or do they want discussion of the other options brought forward to be explored first? Those weighing in on this thread, besides yourself, are @FAC coordinators: @FAR coordinators: and @Graham Beards, Guerillero, Serial Number 54129, Wehwalt, Johnbod, Fowler&fowler, HJ Mitchell, Grapple X, Jo-Jo Eumerus, The Rambling Man, Victoriaearle, A. Parrot, Vanamonde93, Z1720, and Amakuru: (with apologies if I missed anyone, please add).
My preference, considering the new options advanced on this page at #Additional solutions to page limitations is that we hold all discussions open, as we pursue comprehensive rather than piecemeal solutions. The best case is we discuss a bit longer and come to a stronger and broader consensus. The worst case is FAR is removed and we lose more historic and important featured work, in which case, we will know why. SandyGeorgia (Talk) 18:09, 11 November 2021 (UTC)
PS, considering the positions some Coords have taken, I suggest we seek uninvolved closure. SandyGeorgia (Talk) 18:13, 11 November 2021 (UTC)
As one of the coords, I can say that I feel to WP:INVOLVED here to do a closure of any of the recent discussions here. Hog Farm Talk 18:22, 11 November 2021 (UTC)
Unless anyone disagrees, or has other suggestions, we might ping Wugapodes, who closes many discussions posted at AN (not always to my liking), and has closed here before. SandyGeorgia (Talk) 18:36, 11 November 2021 (UTC)
I think there's little point in having coordinators for this process if they can't close these discussions; if we don't trust them to do that they shouldn't be coordinators. (Though, Hog Farm, I think individual coords recusing is fine if they want to.) Sandy, I'm afraid I'm not going to repost my !vote to remove FAR below unless a coord asks for a fresh response from those who have commented above. I don't think it's necessary. As I've said above I'm not against FAR being part of this page, but agreeing on how we address the include-size problem is not going to happen in days. Mike Christie (talk - contribs - library) 18:48, 11 November 2021 (UTC)
Indeed, the consensus is not exactly controversial or in any doubt. The Rambling Man (Keep wearing the mask...) 18:59, 11 November 2021 (UTC)
Mike, you may remember that we once had a director, who avoided taking strong positions so he could close them. We now have four Coords, no director, and some of these Coords have taken positions—one brought the proposal forward, and one acknowledged not having read posts. The one most eligible in terms of not having taken a strong position one way or the other (Hog Farm) has had the good sense to recuse.
At any rate, if the consensus is as strong as you and TRM believe, you would have no reason to object to having an independent closer make that call. We are all on the same team here, right? Wanting the best and most comprehensive solutions for articles, nominators and reviewers, and Coords, and the betterment of the overall FA process? SandyGeorgia (Talk) 19:04, 11 November 2021 (UTC)
You may also recall that within the last month, the very community on this page discussed what they perceived to be the role of the Coords—a limited view that I disagree with, but nonetheless, this is what many on this page argued when Gog the Mild described exactly what he thought his job to be in the post:

I have always worked off the job description at the top of the FAC page. "The FAC coordinators ... determine the timing of the process for each nomination. ... the coordinators determine whether there is consensus. A nomination will be removed from the list and archived if, in the judgment of the coordinators: ... None of the nominators may nominate or co-nominate any article for two weeks unless given leave to do so by a coordinator; if such an article is nominated without asking for leave, a coordinator will decide whether to remove it. A coordinator may exempt from this restriction an archived nomination that attracted no (or minimal) feedback." I imagine that this is what Buidhe and Hog Farm think they have volunteered for. Gog the Mild (talk) 16:11, 25 October 2021 (UTC)

If that is the only “job description” and all they feel responsible for in terms of the overall FA process (something I argued against), then closing talk page discussions isn’t part of their “job description”. Closing nominations is. Some on this page have alleged that Raul654 had dictatorial tendencies. I beg to differ. What is the role of the Coords? Gog sees it very narrowly. We can’t have it both ways.
Any objections to pinging in Wugapodes? SandyGeorgia (Talk) 19:13, 11 November 2021 (UTC)
I nominated Wugapodes for adminship; I trust his abilities as a closer implicitly. To me, consensus seems abundantly clear, to the point where I'd suggest there's no purpose to a closure. But to lay at rest any dispute, and since there's no reason we shouldn't have a closure; Wug, would you be willing to do the honours? Vanamonde (Talk) 19:18, 11 November 2021 (UTC)
Thank you so much, Vanamonde … I haven’t always agreed with Wug’s closures, but believe them to be fair and thorough … but didn’t dare do the pinging myself! SandyGeorgia (Talk) 19:20, 11 November 2021 (UTC)
I've only skimmed the first section and am a little pressed for time, but I'm not sure a formal close is required. Consensus seems clear enough and there's not too much nuance. To me it seems that FAR should no longer be transcluded, but that a prominent link should replace it and better long-term solutions discussed. I can take a closer look tomorrow, but a formal close would probably just be saying that with a lot more words. Wug·a·po·des 19:55, 11 November 2021 (UTC)
What a difference a new set of eyes can make. I would be satisfied if we came up with some sort of temporary “prominent link” (and since few people scan to the bottom of the page, to be noticeable, eg prominent, it would be at the top of the FAC listings) instead of the FAR transclusion, while more comprehensive and longer-term solutions are being discussed. Emphasis on the word prominent; FAR is now lost in the sidebars and the excessive instructions at the top of FAC. Thanks, Wugapodes. And I have to hope that the rest of the FAC issues that led us to this problem will not be pushed aside and forgotten, and that we will pursue the other options. SandyGeorgia (Talk) 20:23, 11 November 2021 (UTC)
Thanks, Wugapodes; I think a close here, though you may not think it necessary, would help this group move to the next part of this discussion. Mike Christie (talk - contribs - library) 02:32, 12 November 2021 (UTC)
  • Remove FAR for now, the proposals below will take months to come to any kind of conclusion. The Rambling Man (Keep wearing the mask...) 18:11, 11 November 2021 (UTC)
  • Remove FAR until a solution is implemented. If the status quo is retained, the issue will be quickly forgotten and editors such as me will be left with a non-functioning FAC page and they will move to other parts of the project or just scale down their involvement. The inclusion of FAR was for a one month trial (suggested by SG) in 2014.[3] It was never meant to be a forever solution; it was just to get FAR out of the doldrums. The arguments that transclusion of FAR is only a part of problem haven't been persuasive IMHO – they have just convinced me that removal is a good place to start. And, I don't really see why our coords can't close this discussion; it's just housekeeping in my view. Graham Beards (talk) 19:36, 11 November 2021 (UTC)
  • Transclude FARs individually, as discussed in the Template usage section below. It saves a lot of cap space, but still keeps FAR up. Seems like a reasonable compromise to me. Hog Farm Talk 19:45, 11 November 2021 (UTC)
  • I prefer a comprehensive approach to the broader problem(s) that have led to constantly hitting template limits, which would involve many elements discussed in Grapple X’s new section. Go to a two-stage process (which if used as described will minimize the template and length problem); implement several of Grapple X’s proposals below (in the sections #Template usage and #Review length/nomination talk pages); deal with excessive sigs (from Kusma’s #Discussion section); and if all of that still results in template limit and load time problems, then move to #GA-type header per F&f to get all parts of the FA process on equal footing. I fear that individually transcluding FARS will just push the can down the road (the “can” being the problems in FAC length and conventions that have taken hold such that the limits will continue to be a problem, and are already a problem in archives). Without addressing the underlying problems that will continue at FAC unless some of these measures are implemented, removing FAR from the page will accomplish nothing except the loss of some Featured articles that might have been saved, and community perception that FAC is a one-way street. SandyGeorgia (Talk) 20:06, 11 November 2021 (UTC)

Towards closure

Wugapodes suggests above “that a prominent link should replace it (the FAR tranclusion) and better long-term solutions discussed”. I find this a workable solution, with emphasis on “prominent”. FAR is lost among sidebars and content at the top, and putting the “prominent” link at the bottom of the page (where FAR is now) will mean few people see it. The goal is to engage as many editors as possible in re-assessing our (still very large) backlog of unreviewed older FAs. Almost three-fourths of our current FAs are very old and need re-assessment, and at times, need to be delisted: see WP:URFA/2020#Monthly stats. FA is only as strong as its weakest links, and the value of all bronze stars is lessened by FAs that no longer meet criteria.

If you scroll down WP:FAC now, what is “prominent” and easily noticed are the green collasped bars containing “How to nominate an article” and “Commenting, supporting and proposing”. What if we added a new green collapse section with a header something like … “Re-assessing articles that already have Featured status”, and put a small blurb inside that green? If that blurb simply made use of Category:Wikipedia featured article review candidates, that would produce a list of the articles at FAR. That is how we can avoid losing the new editors like the one who tackled James Joyce at FAR because they saw it at FAC. They saw James Joyce listed here, and came over to work on it … I suspect we can accomplish same by somehow making use of the category.

Not a design type; others may have better ideas. Somehow, using the category might be the solution here. SandyGeorgia (Talk) 21:16, 11 November 2021 (UTC)

You can see this proposal now in my sandbox (click on the third green collapse box—everything else is the same). This puts the list out there, to help us capture folks who might see an article of interest, without any transclusions at all. SandyGeorgia (Talk) 00:24, 12 November 2021 (UTC)

Additional solutions to page limitations

Given that a few comments above have suggested making a separate discussion for solutions to the page limitations other than removing FAR, I'm opening one up, hopefully to get more focus on individual issues. Template use in reviews has been raised as a contributing factor, as has the length of review text left on the review page. I'm splitting this into two sections to discuss both of these issues; along with my own proposed ideas. ᵹʀᴀᴘᴘʟᴇ 16:06, 11 November 2021 (UTC)

Template usage

Immediately struck by how we still have the following text in the instruction header: The only templates that are acceptable are {{xt}}, {{!xt}}, and {{tq}}; templates such as {{green}} that apply colours to text and are used to highlight examples; and {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions. I would propose removing this entirely. Optionally we can replace the text with a proposed alternative to template use. My suggestion there would be to replace this:

Please do not use graphics or templates on FAC nomination pages. Graphics such as  Done and  Not done slow down the page load time, and complex templates can lead to errors in the FAC archives. The only templates that are acceptable are {{xt}}, {{!xt}}, and {{tq}}; templates such as {{green}} that apply colours to text and are used to highlight examples; and {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions.

With something along these lines:

Please do not use graphics or templates on FAC nomination pages. Graphics such as  Done and  Not done slow down the page load time, and complex templates can lead to errors in the FAC archives. If formatting your comments for clarity, use plain wiki markup; bold, italic, or underlined text may be used for emphasis or to denote quotation. User notification templates such as {{ping}} or {{U}} should be replaced with a direct wikilink, such as [[User:Example|Example]]. Offtopic discussions should not be collapsed within templates, but may be moved to the nomination's talk page.

Suggestions welcome on this. ᵹʀᴀᴘᴘʟᴇ 16:06, 11 November 2021 (UTC)

I very strongly agree with the collapse template suggestion. Leaving aside the question of moving on-topic review comments, purely off-topic or purely disruptive material should be moved to the talk page, such as happened with Wikipedia_talk:Featured_article_candidates/archive86#Wikipedia:Featured_article_candidates/Koh_Tao_murders/archive1 Hog Farm Talk 16:14, 11 November 2021 (UTC)
I like your suggested replacement (there is a possibility that the pinging alone is adding to the overall problem). But I would raise one consideration: before we went to the no templates/no markup stuff more than ten years ago, some FACs were dreadfully hard to read through because of bold, italic and underlined. Perhaps we could limit those to one choice? SandyGeorgia (Talk) 16:28, 11 November 2021 (UTC)
Sure. I have no skin in that since I think just quoting in quote marks works for me but if the current co-ords find one kind of markup easier to read (ultimately they're the ones expected to read it all) then we could narrow it down. The aim was just to show that there are simple alternatives. ᵹʀᴀᴘᴘʟᴇ 16:30, 11 November 2021 (UTC)

Here are some details that may help the discussion -- maybe we should capture this as a cheat sheet for when the issue next comes up.

  • The include limit is about two million bytes.
  • Every character in a FAC counts for one byte: "Now is the time for all good men to come to the aid of the party" counts 66 bytes.
  • A collapse template (top plus bottom) counts about 1000 bytes. It doesn't matter how much text is between the templates.
  • A {{tq}} template costs four bytes for each character in it: Now is the time for all good men to come to the aid of the party counts 264 bytes.
  • A colour template such as {{green}} costs two bytes for each character in it: Now is the time for all good men to come to the aid of the party counts 132 bytes.
  • Transcluding FAR using {{WP:FAR}} is extremely expensive; it doubles the cost of each character in every FAR, and it redoubles any transclusions inside FARs. When I tested this a day or so ago, the individual FARS added up to about half a million bytes, and when transcluded using {{WP:FAR}} they cost about twice that. Transcluding FAR using individual FAR templates would be a big savings.

I'd say that collapsing is completely harmless; it's never done more than two or three times per FAC, and most FACs don't do it, so the cost is negligible. I'd see no problem with banning tq and xt in favour of colour templates. It looks like we're about to (at least temporarily) remove FAR from this page; if/when we re-add it it would help tremendously if we could do so with individual FAR templates, not the parent FAR page template. Mike Christie (talk - contribs - library) 16:41, 11 November 2021 (UTC)

Original template limit explanations from FAC archive 50
The issue is not the "expensive parser count" (which is rarely a problem as very few templates use expensive parser functions like #ifexists, and in most of the cases I know the template is substituted, which eliminates the problem). The problem is the "post-expand include size" (which I will abbreviate as PeIS). I took the PeIS of the current FAC page ;) and obtained:
NewPP limit report
Preprocessor node count: 1828/1000000
Post-expand include size: 645624/2048000 bytes
Template argument size: 1513/2048000 bytes
Expensive parser function count: 2/500
Each individual FAC page is transcluded like a template onto WP:FAC, which makes the PeIS of WP:FAC quite large. Fortunately the maximum PeIS (2048000 bytes/characters) is large enough that the total content of FACs would have to more than triple to break the limit. You got it easy here compared to WP:PR, which operates close to the limits regularly (current PeIS: 1432316/2048000).
The problem with templates is that they contribute disproportionately to the PeIS, due to a bug in the mediawiki software in which a transclusion within a transclusion counts "double". For example each raw use of {{collapse top}} and {{collapse bottom}} on an individual FAC page adds about 326 bytes to the PeIS of the individual FAC page, but almost twice that to WP:FAC. If editors add a section header then the length of that header is added to once to the PeIS of the individual FAC, and twice to the PeIS of WP:FAC.
To evaluate the size of the problem, suppose there are 40 reviews, with an average of 10 collapsed comments per review, and about 74 characters (=bytes) of section header in each. That adds about (40 * 10) * (2 * (326+74)) = 320000 bytes to the PeIS of WP:FAC.
This doesn't seem to me to be large enough to cause a problem on its own, and it could be that other templates were more responsible for past problems. For example it is a complete disaster to use {{collapse}} or {{hidden}} to collapse comments, because these templates read in the entire content of the review, thus doubling contribution of the review to the PeIS of WP:FAC: editors should only use the "top/bottom" versions of these templates for extended content.
As a final suggestion, the contribution of collapse templates to the PeIS could easily be halved by using template substitution. Both {{subst:collapse top}} and {{subst:collapse bottom}} are templates which can be substituted. Then the template code (basically wikicode for a collapsible table) would contribute only once to the PeIS of WP:FAC. Geometry guy 04:08, 27 March 2011 (UTC)
Thanks for the explanation; that's very useful. I went back to look at the old collapsible sections to see what template was being used; I found that here, for example, it was {{hidden}}, which is like {{collapse}} in that it puts the entire contents of the collapsed section in the template, which leads to the higher PeIS.
The FAC page can get as large as fifty or more candidates, though the delegates get more aggressive about archiving at that point. I think a more conservative estimate would be 55 candidates, with 10 collapsed comments each. That would be 645,624 * 55/40 for the base page size, which is 887,733; then (55 * 10) * (2 * (326+74)) = 440,000 for the additional PeIS due to the templates. The total size would then be 1,327,733. That's still less than two thirds of the limit, and I think that's a fairly conservative calculation.
It occurs to me that if we decide to do this it would be hard to tell by glancing through a FAC whether a nominator had used an "approved" (i.e. low-PeIS) template. Perhaps a way around that would be to have slightly modified versions of these templates that have some easy way to identify them -- "OK for FAC" in small letters in the collapsed header, for example. Then readers of the FACs could easily see if an "expensive" template were being used. Mike Christie (talk - contribs - library) 13:21, 27 March 2011 (UTC)
This sounds like a good idea. Note that we can't change the existing collapse templates: the slight modification ("Is the transclusion of this template at FAC?; if so print a message") is possible, but costly in terms of template size, and not all of Wikipedia revolves around FAC. Slightly modified and approved collapse templates can, however, be made immediately identifiable, for example by using the FAC blue instead of the standard green. Geometry guy 01:20, 28 March 2011 (UTC)

Gimme had explained it differently, but yes, basically the problem was that transclusions within transclusions count double, and we have editors persisting in using all kinds of templates at FAC. Also, it's not only the size of the FAC page that matters-- it's the size of archives. When I first discovered the problem it was because some of our featured articles were disappearing from the featured archive. The long and short of it is that there are just about no templates that are helpful at FAC (I do add the red color on FFAs), so I don't know why we want to use them. Very long commentary on FACs is still very long even if it's hidden, still discouraging to subsequent reviewers, still adds to the page load time, and still indicative that the article might not have been ready for FAC. SandyGeorgia (Talk) 13:35, 27 March 2011 (UTC)

I hadn't realized you felt that capping would be unhelpful even if it were permitted -- I think the delegates' opinions need to carry extra weight in an issue like this. I'd also like to hear from Andy and Karanacs (and Raul, if he wants to weigh in). The FAC reviewers' opinion also counts, of course, and from the discussion above there are some (Gyrobo, Lecen, Jinnai) who would like to see capping, and some (Johnbod, Truthkeeper, RHM22) who would not. So there's not a clear consensus in favour of allowing capping, but nor is there one in favour of forbidding it; I think the current rules derive from the template limits issue and not from a consensus on whether capping is beneficial or harmful.
I took a look at the archives; we haven't had an featured log in years with more than 55 articles. The largest I was able to find (without looking at every single one) was January 2008, which had 82; if every single FAC in that list had had ten collapsed comments it would have exceeded the template limits. I checked the PeIS and in fact it would have been OK with three collapsed comments per FAC, but no more (though there were many fewer long reviews back then). However, that may not be relevant since we don't plan to go back to old logs; the question is how many articles are we likely to see in future logs.
I suggest that we deal with this as two separate questions -- is it technically feasible to use collapse templates in the FAC page without causing us to hit the template limits problem? And if so, do we want to? I think the answer to the first question is very likely yes. I don't have a strong opinion on the second question. Mike Christie (talk - contribs - library) 14:16, 27 March 2011 (UTC)
So my question is, when commentary belongs so excessively long that capping is required, why aren't reviewers just Opposing? The Oppose is underused at FAC, which contributes to the backlog. And then there's the issue of having to step back through diffs to see who capped what and whether it was done appropriately ... but then, I should know by now we can't force editors to read instructions, no matter what we decide. SandyGeorgia (Talk) 14:43, 27 March 2011 (UTC)
Well getting to long because of problems is one thing; getting to long because of a discussion on a disagreement between the reviewer and the nominator is another.Jinnai 23:35, 27 March 2011 (UTC)

Mike Christie left a message for me. GeomGuy seems to have explained the technical issue with inclusion limits. WP:FAC is made of review pages included as templates, so everything on a review page counts toward the include limit of WP:FAC. As I recall, some people were using a collapse template that put the collapsed text as one of the parameters, such as {{hidden}} - such text was counted twice, once for the collapse template and once for the review inclusion. Collapsing lengthy comments could quickly cause WP:FAC to hit the inclusion limit. That may have changed, but I don't think so. Collapse templates that don't use parameters, or only use a parameter for a brief caption, should be much less costly as suggested by the above analysis. Gimmetoo (talk) 15:11, 27 March 2011 (UTC)

My view is that we should not let technical issues affect process decisions, and that a blanket ban on templates is akin to a ban on extended review comments. However, we should, per Gimmetrow's comments, not be wasteful.
I support Sandy's general point that FAC is primarily intended as a summative review, rather than a formative one. However, justifications for supports and opposes are vital, not only for decision-makers: a good review can bring some articles up to standard very quickly, and feedback helps to improve the encyclopedia.
Detailed reviews add more to the PeIS than most templates, despite the bug (if a template is small, then twice small is still small).
The thing to avoid is wastefulness: expensive templates can be removed or replaced by cheaper ones; self-indulgent reviews or arguments over reviews can be moved to the talk page. Such changes improve the quality of the review as well as saving space. Geometry guy 00:56, 28 March 2011 (UTC)
I've tried using subsections at Wikipedia:Featured article candidates/La Stazione/archive1, does that make it more readable? It certainly makes it easier for me to see how many editors have participated, and respond to each reviewer on a per-section basis. --Gyrobo (talk) 01:10, 31 March 2011 (UTC)
It makes it easier to tell how many reviewers/commenters you have and probably easier to edit, but as far as making it easier to tell what, if any unresolved issues whether those who support/oppose have good reasons it doesn't do much. There are still long sections there where its unclear what's been resolved and what hasn't. That's why I still prefer collapsed comments. It makes it easier not only for the closer, but also other reviewers. If I see something somehad took issue with, I might respond to it and agree or disagree, but if its buried under a bunch of clutter it makes it less likely I'll see it.Jinnai 02:37, 31 March 2011 (UTC)
Collapsed comments do not make it clearer (to delegates) what has been resolved. Neither collapsed comments nor section headings shorten the review or help load time, although they do improve readability. If reviewers aren't going to oppose, entering commentary on talk is what improves readability and load time. SandyGeorgia (Talk) 03:19, 31 March 2011 (UTC)
Putting all that here so we don’t have to go look for it next time. Mike, I’m not yet convinced that your cheat sheet is accurate. I wish we still had Gimmetrow, Geometry guy and RexxS, but we don’t, and people tend to glaze over in these discussions. But from my memory of the very first explanation (which I still have to search for in archives, I have the impression your numbers may be not be accounting for everything. Also, FACs are also transcluded, so help me understand that you are accounting for the doubling of a transclusion within a transclusion, and that collapsing is completely harmless. SandyGeorgia (Talk) 01:09, 12 November 2021 (UTC)
Sandy, I just rechecked and there are some minor mistakes in what I posted, though I don't think they affect the conclusion. Here's how I got those numbers -- if you follow along these edits you should be able to check; I've corrected the details in the version below. Go to WP:FAC, click on edit source, and click on "show preview". Scroll to the bottom of the page that results -- not the bottom of the edit box, but the bottom of the entire page -- and expand the "Parser profiling data" section. You'll see "Post-expand include size" (PEIS) which is currently 1,801,039. Now go to the edit window, and try the following tests, clicking on "Show preview" after each one and checking the PEIS after each edit:
  1. Add "fred" on a new line in the edit window. Effect is no change on the PEIS -- this was not an include.
  2. Add {{green|fred}} in the edit window. Effect is to add 38 characters to the PEIS. In other words, {{green}} costs 34 bytes to do nothing, and 4 bytes to add "fred". (The 34 bytes is what I missed, above.)
  3. Now change that to {{green|fredx}}. The effect is to add 1 more character to the PEIS. This shows that adding more text in green only adds as many bytes as you add to the template.
  4. Now change it to {{tq|fred}}, and then to {{tq|fredx}}. I got 1,801,170 and 1,801,172, showing that text using tq adds a baseline of 123 to do nothing, and then 2 bytes for each additional characters. (Again I missed that baseline value of 123.) So {{tq|<...100 characters...>}} would cost 100 *2 = 200 characters + 123 for the baseline = 323 characters.
  5. All that applies to directly transcluding into WP:FAC. If you transclude from inside another template, all the costs are doubled. So if you make the same edits to an existing FAC, with an edit summary saying you're testing and will revert, and then do show preview at WP:FAC, you should see the same numbers, but this time doubled. That is, you should see that the baseline cost for {{green}} is 68 bytes, and putting "fred" in green in a FAC causes the PEIS for WP:FAC to grow by 68 + 2*4 = 76 bytes. Putting "fredx" in green would cost 68 + 2*5 = 78 bytes. Similarly tq is doubled: putting tq in a FAC costs 246 characters before you include any text, and putting "fred" in tq in a FAC costs 16 more bytes in the WP:FAC PEIS. Putting "fredx" in tq would cost a total of 266 characters.
  6. Finally to see the cost of collapse, add a collapse top and bottom template to a FAC and show preview WP:FAC again; this time you should see an increase in the PEIS of about 1,000 regardless of where you place the collapse template.
Let me know if any of this doesn't work for you. Mike Christie (talk - contribs - library) 02:10, 12 November 2021 (UTC)
Too tired to do all that now, but at least we have a record now of the methodology. As I was reading through archives looking for the post from Gimmetrow that I never found, I was impressed that we have been having this problem and discussion since 2007. We need to fix what’s broken in here, and not just temporarily. Thanks for the lengthy explanation, and I hope it will help everyone, but I fear we keep going ‘round in these circles because people just can’t or don’t want to wrap their heads around the technicalities. I wish we could save this whole section somewhere, in case we are still having this discussion another 14 years from now <sigh>. All the best, my friend, out for the day, bubble bath beckons, SandyGeorgia (Talk) 02:18, 12 November 2021 (UTC)

Review length/nomination talk pages

I believe reviewers should be encouraged, but not mandated presently, to move extended discussion to a nomination's talk page once it has been resolved. This may mean a lengthy review is moved across with a courtesy link, or it may mean simply forking out extended asides which are not directly relevant to the outcome. As it would not be a rule I don't believe we need a strict guideline on what is moved but at this point just breaking the ice regarding the use of talk pages would be a good step; once we see a few users trying it in practice it will be easier to see what effect it has on readability and on the ease of assessment for co-ordinators but presently I see next to no use of it, so there is plenty of room to start. Suggestions on how much this should be used are welcome, particularly input from current co-ords on what would be ideal for their jobs here. ᵹʀᴀᴘᴘʟᴇ 16:06, 11 November 2021 (UTC)

I would not go so far as “mandate”; this is a case-by-case issue, and the problem occurs on very few FACs. At times, there are lengthy discussions that are critical (sourcing, POV). Recently, it has been the line-by-line prose reviews with quoting templates that are causing problems, and I question whether seeing all of the line-by-line resolution of ill-prepared prose concerns is helpful to subsequent readers of the FAC. But even those are not always ‘’so’’ lengthy as to warrant ‘’mandating’’ removal to talk. SandyGeorgia (Talk) 16:31, 11 November 2021 (UTC)
That's why I think it should just be an encouragement ("consider doing this where appropriate"); others have stated they would prefer not to do so, so I wanted to be clear this was not a suggestion to make it necessary. ᵹʀᴀᴘᴘʟᴇ 16:33, 11 November 2021 (UTC)
Sorry, it appears I misread that you were suggesting a mandate. So we are on the same page. SandyGeorgia (Talk) 17:04, 11 November 2021 (UTC)

Discussion

I'm wondering whether we can find out more what actually causes the issues so we can try to find solutions with the greatest possible impact. If we used no templates at all, we could still run into transclusion limit issues at some point because the FAC subpages themselves are transcluded onto WP:FAC and its archives. Do we just need to get them smaller? One option without reducing content or content-related formatting would be to minimise signing our posts with the full ~~~~ signature (some people have complex signatures and sign their replies to each and every small prose suggestion). If we can't get page lengths under control, we could have an editor or bot go through WP:FAC and replaces transclusion of every subpage larger than 30k by a summary plus link. But we should try to understand the impact of any action we take on load times and on readability for reviewers and coordinators. —Kusma (talk) 16:17, 11 November 2021 (UTC)

The transclusion limit is high enough that with just straight prose in the FACs, it wouldn't be too bad. What's going on, to the best of my knowledge, it that templates within transclusions have an exponential effect. If we can largely avoid transclusions within the FACs themselves, the exponential effect won't occur and it won't take up near as much space towards the transclusion limit. Hog Farm Talk 16:22, 11 November 2021 (UTC)
Complex signatures may be a factor, but for curiosity's sake I had a look at what kind of text is being transcluded with every instance of {{tq}}, which is in common use. Simply adding {{tq|example text}} essentially represents the following at each instance: <templatestyles src="Talk quote inline/styles.css" />{{#ifeq:{{NAMESPACE}}|{{ns:0}}|{{FormattingError|Template:Tq is only for quoting in talk and project pages. Do not use it in actual articles.}}|<!-- --><q {{#if: | title="{{{title}}}"}} class="inline-quote-talk {{#if: | inline-quote-talk-italic}} {{#if: |inline-quote-talk-marks}}">example text</q><!-- -->}} and this is multiplied by this appearing perhaps a dozen or more times per review, if not more. There's a lot to be saved here even incrementally. ᵹʀᴀᴘᴘʟᴇ 16:27, 11 November 2021 (UTC)
Yes, I've looked at that. I wonder whether we shouldn't just drop the namespace check and replace it by bots or edit filters... but that's a discussion for Template talk:tq or the Pump. —Kusma (talk) 16:32, 11 November 2021 (UTC)
Although I don’t know where to find it now (or even if it is still true), there was once a page indicating that editors should pay heed when other editors asked them to adjust their convoluted signatures. We should start enforcing that on FACs. I recall having to read through some very dreadful and disruptive signatures, that left me with a headache after one FAC (and I recall another Coord, maybe Ealdgyth, saying same), although I had to read 50 FACs every day. In particular, we had discussions in the past of FAC editors who added lengthy self-promotional text to their sigs. Coords used to deal with things like this on their own (I pinged the editors and asked them to adjust their sigs); apparently they no longer are willing to, so the community should take this on. SandyGeorgia (Talk) 16:42, 11 November 2021 (UTC)
Kusma, what actually causes the problem is that “FAC is not peer review”, but is currently being treated as peer review, even after we created and populated Template:FAC peer review sidebar. FAC is where we determine if articles meet the criteria. It has become instead where, by gosh, we will pull ‘em through no matter how long it takes, leading to lengthy FACs sapping limited reviewer resources. The solution is to more actively use the “Oppose” button, and suggest withdrawal and re-submission after PR. FACS are simply too long because they no longer conform with the purpose of this page, and that length (with the pingie-thingie) now means we exceed limits, and are lucky if we get three supports after weeks to months of review, because the FACs are so unreadable. This is not optimal functioning of FAC. SandyGeorgia (Talk) 16:36, 11 November 2021 (UTC)
PR isn't quite active enough to be used in this manner. Sometimes you get lucky or you know who to ask, but usually it just stagnates. Jo-Jo Eumerus (talk) 16:41, 11 November 2021 (UTC)
Jo-Jo, that is a self-fulfilling prophecy problem. If we shut down the use of FAC as peer review, our specially designated PR page would come back to life again, as it did at the end of last year and beginning of this year, when quite a few of us weighed in on Every Single Article that presented via that template. I gave up in here about the middle of the year, as FAC seems so entrenched, but quite a few of us were weighing in on the PRs, and they were working. But why would editors go to PR if they can get a peer review and guaranteed promotion after months of sapping resources at FAC because some of our most qualified reviewers are determined to pull ‘em through here, rather than sending the ill-prepared to PR? You and I made great progress in prepping your articles for FAC !! SandyGeorgia (Talk) 16:47, 11 November 2021 (UTC)

Move to a completely new reviewing model

See User:SandyGeorgia/sandbox4

Please don’t start !voting, as I am not ready to formally propose this. We discussed it last year, but I dropped it after we lost Ealdgyth, a top-notch source reviewer. (I think there may have been some changes discussed here on FAC talk that I did not incorporate yet.)

The basic idea is to move to a two-step process, like the one that now works quite well at WP:FAR. We wouldn’t advance to the lengthy prose reviews until critical issues like sourcing have first been vetted. This would help avoid expending our limited reviewer time on FACs where underlying problems are discovered only after considerable prose review has been done. I will continue work on what is at user:SandyGeorgia/sandbox4 only if there is some interest in exploring this possibility further. It works at FAR, which is a deliberative process that allows time for all to be heard, if they are willing. It allows us to implement the possibility of removing some lengthy text from the first stage to talk, once the nomination advances to the second stage (but that could be at Coord discretion, limited to very lengthy FACs). SandyGeorgia (Talk) 16:56, 11 November 2021 (UTC)

  • I agree. No amount of prose/MOS polishing will help if the underlying substance isn't up to snuff. This process would encourage us to thoroughly vet the sourcing before moving on to the other FAC criteria. (t · c) buidhe 17:08, 11 November 2021 (UTC)
  • Agree as well. Sourcing and prose are both important, but sourcing is the greater one. Hog Farm Talk 17:26, 11 November 2021 (UTC)
    Interjecting here that most editors who have worked at WP:FAR (eg Buidhe and Hog Farm) know the model works :).
    But for those less familiar with FAR, it may be helpful to scan down the full page, to see that reviewers declarations to the Coords are things like Move to FARC (the second stage, where articles may be deleted), Close without FARC (when reviewers deem an article has been restored to FA standards in the first phase, and the second phase is not needed), Speedy delist without FARC (in the unfortunate but recently increasing cases where FAs have been merged out of existence via AFD, such that the need for a second stage is gone, as the FA is gone), and so on. Even if an articles moves to FARC (the second stage, where the FA can be delisted), it can still be saved if someone engages belatedly to improve the article (see Wikipedia:Featured article review/James Joyce/archive2). The process allows considerable flexibility, as one goal is to save stars, and reviewers tell the Coords when and if the nomination should proceed to the second stage (which in FAC’s case, would be the stage where we !vote on promotion).
    For the purposes of FAC, the two-stage discussion would accomplish several goals, only one of which is allowing several options for dealing with length and template issues. SandyGeorgia (Talk) 17:48, 11 November 2021 (UTC)
    • Just a side note here... I do not (and will not for years) have the bandwith for what I did long ago. I strongly suggest that someones be lined up who will commit to this before folks try it. And it should be someone who does thorough source reviews, not just checking for formatting. If the people doing it aren't making waves, they aren't being thorough enough. And that'll cause burnout... it did for me (multiple times). Ealdgyth (talk) 17:40, 11 November 2021 (UTC)
      • I don't think it works if it's just one person doing it. That would clearly cause reviewer burnout. Ideally we are able to find several people to do source reviewing sometimes. (t · c) buidhe 17:42, 11 November 2021 (UTC)
      • This is why I did not advance the proposal last year; it is and should be a grave concern to all at FAC if the work Ealdgyth once did is no longer being done. One hope is that this proposal would force stringent sourcing reviews back to FAC; that may not be a realistic hope. But note that we have recently seen quite a few FARs of Featured articles that did not have Ealdgyth source reviews, and have since been found not to meet notability, and have been merged out of existence. What does that tell us? SandyGeorgia (Talk) 17:50, 11 November 2021 (UTC)
        Similar to Hog Farm, I have a fear of doing much in the way of source reviewing. Maybe if there was some sort of mentorship program? Eddie891 Talk Work 19:21, 11 November 2021 (UTC)
  • My initial thought on this is that, while it's certainly the case that there's no point polishing prose based on bad sourcing, the converse is also true. It would be a bit silly to do a thorough source analysis on an article that was so poorly written that we'd quick fail it on prose. It would be worth at least doing a sanity check that the article is in the FA ballpark first...  — Amakuru (talk) 18:01, 11 November 2021 (UTC)
    Amakuru the two-stage process allows you to OPPOSE at any time. You can still oppose on prose in the first phase. All of the “sanity check” you mention can be done in the first phase as well. You just can’t SUPPORT until the sanity pass is done and the nomination has moved to the second phase. :) I recognize it may be hard for people who haven’t worked at FAR to grasp that intuitively. Think of it like FAR, where you can lodge any concern you want in the first phase, but you can’t vote to DELIST until the second phase. Here, conversely, we hold off on SUPPORTS, hoping we will encourage the sourcing and other work to be done first, while allowing indepth prose analysis later. But one can always offer a few sample, poor prose, OPPOSE, in the first phase as well. SandyGeorgia (Talk) 18:18, 11 November 2021 (UTC)
  • In addition to Amakuru's concern, I'm not sure we can draw as clear a line between source reviews and other reviews as this implies. There's many aspects to sourcing; reliability, comprehensiveness, formatting, and neutrality, are all not separable from source use. And if we're covering all of those in the first round, there's not much left for the second. Furthermore, many reviewers (myself included, but Mike Christie comes to mind among prolific reviewers) won't do source reviews, but cover several aspects of sourcing, in addition to prose, as part of their review, while reading clear through the article. Trying to separate these aspects seems to me to create more problems than it solves. Addendum post edit-conflict: if we're also allowing for early opposition but not support, that's creating a remarkably unwelcoming environment, particularly for new nominators. Vanamonde (Talk) 18:20, 11 November 2021 (UTC)
    I agree with this. Many reviewers don't make a "nitpick on prose"-only review, they do that intertwined with many other aspects of review. Arbitrarily splitting these out and then deeming one aspect of the FA criteria to be somehow more important than the others is not appropriate and will devalue/dissuade reviewers even more than they are dissuaded now for various reasons. The Rambling Man (Keep wearing the mask...) 18:30, 11 November 2021 (UTC)
    Vanamonde, reviewers who review as you mention can still conduct those reviews. They can’t support in the first phase; those premature supports force the Coords to carry nominations longer than necessary, when issues are found after three or four supports are lodged. In those few cases where reviewers who prefer that model are doing very lengthy prose reviews, they can be encouraged to put that initially on talk, pending their eventual support in the second phase. Of course, if they uncover sourcing, reliability, comprehensiveness, NPOV etc concerns during their read-through, they would lodge those in the first phase. As it works at FAR, the two-phase process allows reviewers the best of both worlds.
    Re early opposition and unwelcoming environment, we may disagree, but I have always said the Oppose button is the fastest route to promotion. I suspect we put off just as many nominators when they have to sit here for weeks or months, only to have their nomination ultimately archived. They, and the article, benefit by getting that news in two weeks, rather than two months. SandyGeorgia (Talk) 18:31, 11 November 2021 (UTC)
    I disagree strongly with the notion that opposition here speeds up promotion, as I disagree with your contention that lengthy reviews here are a problem. We have a single body of editors who are able and willing to review articles at this level. We do not magically find more reviewers by going to PR. The quantum of reviewer effort available is more or less constant, and a lengthy review supplied during an unsuccessful FAC is no more wasted than the same review that would need to occur at PR. Partitioning that effort between FAC and PR is a book-keeping issue, not a fundamental fix for our overall lack of reviewer participation, which I agree is an issue. Indeed, given a limited reviewer body, a single venue is likely more efficient than two. Vanamonde (Talk) 18:44, 11 November 2021 (UTC)
    Then we disagree :) But every reviewer is different, and many of us remember when the standard was that “FAC is not peer reivew”. I, for example, am happy to spend hours, weeks, months on articles pre-FAC to help prepare them for FAC, but have little interest in rewarding nominators who bring ill-prepared articles to FAC by pulling them through via copyediting, often at the expense of sourcing. One of the biggest reasons I prefer working that way is precisely what led us to this issue: at other venues, I don’t have to worry about templates,  Done,  Not done, striking, and the like … I can indicate however I want. Here, we have a template and length problem because we are no longer simply opining yea or nay to promotion, rather writing entire FAs at FAC. And I’ve engaged hundreds of Opposes where I helped the nominator come back quickly for a strong promotion on a subsequent FAC, so I guess we have different experiences. SandyGeorgia (Talk) 19:39, 11 November 2021 (UTC)
    You say you'd much rather do long reviews at a different venue, and that's your prerogative; but I've yet to see anyone else express such a preference; and the only reason you've given for that preference is a software issue, nothing deeper. I'd argue the extra scrutiny at FAC is more than enough payoff for some templating rules: I've some problematic reviews at PR and GAN that would never fly at FAC. Vanamonde (Talk) 20:09, 11 November 2021 (UTC)
    I don’t think it’s accurate to say the only reason I’ve given is a software issue, but I’m not going to subject you to a replay of everything I’ve typed on this page in the last month :) SandyGeorgia (Talk) 20:35, 11 November 2021 (UTC)

A note for those who weren't around at the time: this was an RfC on a somewhat similar idea; there was a workshop here where we tried it on five articles. I think it's worth reading through the opposes on the RfC to see what people might object to in this idea. Mike Christie (talk - contribs - library) 18:23, 11 November 2021 (UTC)

Mike, that is mentioned in my sandbox draft, and the differences between that proposal and this one are covered. Not the same thing at all. SandyGeorgia (Talk) 18:32, 11 November 2021 (UTC)
  • Comment this proposal feels like throwing the baby out with the bathwater. The current reviewing process works just fine, we could use more reviewers, sure, but there's zero evidence that the current approach is producing sub-optimal FAs nor dissuading reviewers. The issue is simply the technical constraint, and that should not drive radical and potentially damaging changes to the review process. The Rambling Man (Keep wearing the mask...) 18:57, 11 November 2021 (UTC)
    I am not clear on which part of the baby or bath water is being thrown out ? All of the same pieces are present, just in a different order. Pls clarify. SandyGeorgia (Talk) 19:37, 11 November 2021 (UTC)
    Why do we need a new reviewing "model"? Is it because the current set of FAs being promoted are unsuitable, or is the process failing the community? I don't think so. Is this down to a transclusion issue, i.e. software constraint? I think so. We don't need to reorganise the entire FAC approach, we need to fix the software. The Rambling Man (Keep wearing the mask...) 19:40, 11 November 2021 (UTC)
    The question was, “what are we throwing out”. The proposal only changes the order of what we are already doing. If you are hanging up on my section heading (“new model”), well … semantics. We do need to address sourcing and many other issues if they aren’t being rigorously reviewed, or nomminations have gotten so long you can’t decipher what has been reviewed. SandyGeorgia (Talk) 19:46, 11 November 2021 (UTC)
    We are throwing out a currently user-friendly, successful reviewing process. I'm not "hanging up" on anything, just baffled that there seems to be a move to completely reconfigure the review process because of a software issue. The Rambling Man (Keep wearing the mask...) 19:49, 11 November 2021 (UTC)
    There is much more than a software issue involved here, as I laid out in the sandbox preamble. One side effect of this proposal is that it will help solve that software issue. SandyGeorgia (Talk) 20:31, 11 November 2021 (UTC)
    Well, I disagree. This is an odd time, with more FAs being promoted last month than in a decade or so, to fundamentally change the entire approach with a "side effect" to address a software issue. Abandoning a successful process is wholesale wrong. The Rambling Man (Keep wearing the mask...) 20:34, 11 November 2021 (UTC)
    What are we “abandoning” when we simply change the order of doing exactly what we are currently doing? When seeking solutions, hyperbole isn’t helpful. SandyGeorgia (Talk) 20:37, 11 November 2021 (UTC)
    The proposal 'abandons' the current highly successful and user friendly approach and seeks to adopt a much more contrived approach which will dissuade engagement. There's no hyperbole. The Rambling Man (Keep wearing the mask...) 20:41, 11 November 2021 (UTC)
    Our definitions of successful are highly different; so be it. (They hyperbole was partly mine: it’s not a completely new model, just a re-ordering of the current model … my fault on that one.) SandyGeorgia (Talk) 00:50, 12 November 2021 (UTC)
    As I am making myself dizzy re-reading old archives, still trying to find the first explanation of the template limits problem, I came across this post (which I applaud :) :).
    Yes, it's a problem (and I'm guilty of adding the comprehensive list of issues) - I usually "hide" comments when complete so that's a start. The alternative is to encourage comments on article talk pages rather than at the FAC. The Rambling Man (talk) 11:13, 26 January 2008 (UTC).
    And
    I just wonder if these comments are turning FAC into Peer Review Mark II. Often I'll either miss a PR or there won't be one and I end up with 30 comments on a FAC when really it should just be Support, Oppose or Comment (and I'm not sure if Comment is really supposed to encompass all the guff I come out with)... The Rambling Man (talk) 12:15, 26 January 2008 (UTC)
    What changed ?? Anyway, once again, I am unable to find the original (and best) explanation of the template limits issue, so it must have happened on a different page. I was struck by seeing in archives how many years we have been having this same problem and same discussion … which is why I am digging in to we must fix it this time. SandyGeorgia (Talk) 01:29, 12 November 2021 (UTC)
    13 years ago. Ancient history. And since then, having experienced much more robustness in the process itself, and the need to have a coherent approach to commentary, and in the background of a process which already turns a vast majority of the community off, I feel it's better to enable the reviewers to do their job the way they feel most comfortable doing it. That means not arbitrarily deciding that one FA criterion is more or less important than the others, it means not overwhelming reviewers with instructions, it means not splitting reviews into apparent "nitpicks" vs "non-nitpicks" (as demonstrated, this is practically impossible in many cases). I mentioned before, we could charge the co-ords with assessing viability of FACs, i.e. they go into a pending area where our most trusted users can give them a once-over before allowing them to proceed to the full review. Or we could encourage our co-ords to see when FACs are going too long because of issues and close them quicker (which of course will destroy contributor confidence). What we shouldn't do, and sorry for the repetition, is to make our reviewers' lives any more difficult. And that would even include creating a terrible bottleneck for "source reviews", this appears to be a highly specialised skill which requires access to a vast array of resources and that means only a handful of editors could contribute there. Meanwhile, the dozens of "nitpick" reviewers would just sit on their hands while the backlog grows? Nope, that's not right at all. The Rambling Man (Keep wearing the mask...) 08:07, 12 November 2021 (UTC)
    It strikes me that the community rather comprehensively defeated a similar two-step process in 2018. I don't think we should seek to implement one that did not address the opposes in that RFC. I've read the relevant portions of Sandy's draft page and I'm not sure it would address it in practice. But I'm keeping an open mind.--Wehwalt (talk) 20:12, 11 November 2021 (UTC)
    We disagree that it was similar; the objections to it were mostly that it was a separate process. This is not. SandyGeorgia (Talk) 20:30, 11 November 2021 (UTC)
    There appear to be a fair number of concerns that requiring an initial source review would bottleneck and make more complex the process, see for example SarahSV's "In addition, to set up a source-review process that must be completed first (whether on a separate page or, as proposed below, on the FAC page) will create a bottleneck, because we don't have enough editors willing and able to do comprehensive source reviews. Even during the pilot for this proposal, one article was left unreviewed and one source review appeared to have been abandoned. Reviewers will be under pressure to hurry up so that the content review can begin, which will mean source reviews will be cursory. "diff. There are others that are similar.--Wehwalt (talk) 20:43, 11 November 2021 (UTC)
    But they are left waiting for source reviews now as well (see Buidhe comment above). The wait now is at the end of the FAC process, after much effort has been expended, and some find their FACs archived after all that. The idea is to encourage more source reviewers. I trust our reviewers not to “hurry up so that the content review can begin” (or be called out by others when they do that), and by the way, nothing in this proposal stops the content review from running concurrently. It just allows the Coords to archive sooner, if serious sourcing issues are found. SandyGeorgia (Talk) 20:52, 11 November 2021 (UTC)
  • ANYWAY, as I said, not ready to propose this formally yet, but here we are, taking more time on this than on Grapple X’s proposals above. (Also. The hyperbole in completely new model was mine; it’s not “completely” new, just a re-ordering of what we already do.)
    Could everyone who is spending time on this section please spend time also in the three sections above and one below this, because those are getting lost. If those proposals help advance FAC and FAR, this can wait for another day, when and if source reviewing becomes a priority. SandyGeorgia (Talk) 21:22, 11 November 2021 (UTC)

GA-type header

I am copying (from above on this page) F&f’s proposed tabs to get all parts of the process on equal footing by using a header similar to WP:GA. SandyGeorgia (Talk) 18:54, 11 November 2021 (UTC)

  • Comment I am suggesting that the problem of emphasis might be on the FA page. Perhaps tabs along the lines of User:Fowler&fowler/Proposed tabs for the Wikipedia FA page (as suggested by Sandy and Mike Christie, though not necessarily in this form) might reorient the priorities, and therefore redistribute the effort. Fowler&fowler«Talk» 00:04, 10 November 2021 (UTC)
  • I would also support this line of formatting. Hog Farm Talk 20:20, 11 November 2021 (UTC)
  • I can't see a single drawback to this suggestion. The Rambling Man (Keep wearing the mask...) 20:27, 11 November 2021 (UTC)
  • We have all kinds of helpful reviewing information, guides, scripts, templates, etc now in Template:FA sidebar; could that kind of information have its own tab? SandyGeorgia (Talk) 20:33, 11 November 2021 (UTC)
  • Sounds reasonable --Guerillero Parlez Moi 20:42, 11 November 2021 (UTC)
  • I can see considerable benefit and no substantial downside given current software limitations. Vanamonde (Talk) 20:47, 11 November 2021 (UTC)
  • I can spend some time this weekend making a mock up if there's interest. Any opinions on the color of the bike shed? Based on the blue banner and gold star, I was thinking of     Wug·a·po·des 00:39, 12 November 2021 (UTC)
    • I've always found the GA one a little garish, so going more subtle with color is preferred here (the bronze star certainly offers nicer tints to use.) And making it more obvious they are tabs/which one is active would probably be good too. Der Wohltemperierte Fuchs talk 00:42, 12 November 2021 (UTC)
      • I definitely agree. You can see the first draft in my sandbox. The major color is the existing background color in the header  . The dark accent is   based on the FA star and used for content "below" the blue surface (i.e., inactive tabs, collapsed boxes). The light accent is   which is from the {{mbox}} background and is used for content "above" the blue surface (i.e., sidebars and uncollapsed boxes). The tab headers were taken from the headers in {{FApages}} but they van be easily changed if we want more or different tabs. Wug·a·po·des 01:00, 13 November 2021 (UTC)
  • I think this approach would work well, but this would institutionalize the separation of FAR from FAC. The FAR folks have made it clear they think including FAR with FAC has been helpful in raising awareness, so is that really OK with them? Another thought is that I recall at GA having a single talk page has helped unify discussions there. Each tab has its own page, but each talk page just goes to WT:GAN. Would that be a good idea here? Have a single featured article talk page that includes discussions of FARs, FACs, and FAs? It might raise visibility of FAR-related discussions, which might compensate for the separation of FAR from FAC. I haven't thought about this long enough to know if I like the idea, but the comparison to GA makes it an obvious suggestion. Mike Christie (talk - contribs - library) 02:16, 12 November 2021 (UTC)
    Mike, I have thought about those questions (ever since I first proposed the GA-type header).
    Yes, I see the drawback of the separation you mention, but it is a significant improvement over the clutter we now have in the FA sidebar, where FAR is completely lost, while TFA occupies most of the sidebar. It is very hard to even see that FAR exists. And we’d get a tab for all the helpful advice, scripts, etc. Yes, we get the separation you mention, but we have to find a compromise here, and at least that separation is on equal footing— they are all across the top, and a new user will be more inclined to notice that a page for re-assessing older FAs even exists.
    On the talk page issue, I think one page would not be detrimental, and even might help re-unify the FA processes and community as they were before the split that resulted from the (disastrous) firing of the director. Before that split, we were one. The processes worked together, towards shared goals. We basically discussed pretty much everything here, with occasional links from here to side discussions on the other pages. After Raul654 was fired, the community discussions got split and we lost the forest for the trees. We work hard to pass nominations at FAC, to find them end up FAR because they didn’t even meet basic notability (and yet FAC people aren’t even aware that is happening), or they end up attacked at ERRORS at TFA over issues that were, or should have been, covered in a rigorous FAC review … in short, a unified talk page could help restore some of that community and get everyone back on the same page and bring the overall process to a stronger, more cohesive place, as it was in the days when that journal article (which I can no longer locate) showed Wikipedia with FAC as the center of the universe—even more so than Jimbo’s page. It is illuminating to see the environment that exists on the FAR talk page compared to the one here; it would be helpful if FAC talk were exposed to the rare FAR discussions, and if we all got back together as an FA community.
    It may not be a perfect solution, but I think it is a perfect compromise, and I see some upside potential in re-building community. SandyGeorgia (Talk) 02:57, 12 November 2021 (UTC)

A new way of transcluding

It struck me earlier that perhaps there is also a future-proof option which would avoid this happening again even if permission to do X, Y or Z changes in future or if review lengths continue to increase—why do we need to transclude the entirety of each FAC/FARC onto the page at once? I'm sure most of us regulars use the nominations viewer add-on and it's suggested for everyone at the very top of this page. What if changes were made to the pre-load content for new FACs which by default wrapped any comments in <noinclude> tags, and left behind a bare bones template where the page is transcluded, showing nominator, number of supports and opposes, and whether an image or source review is needed? It should be possible to change the pre-load content to essentially a template which displays these figures, with each reviewer updating the review count when they make a declaration and/or pass on images/sourcing; viewing the individual page on a click-through would show the full review itself but not transclude that portion of it onto WP:FAC. I could get to work on writing up a mock-up of how this would work if it's something that the co-ords would be happy to work with (it would mean having to individually read each FAC page separately, but as honest question, do you do that already enough that it wouldn't be much more new work?). ᵹʀᴀᴘᴘʟᴇ 00:08, 13 November 2021 (UTC)

Interim fix

As we work through the various proposals above, in the spirit of Wugapodes recommendation to remove the FAR transclusion but replace it with a prominent link while we discuss long-term solutions, I suggest a collapsed box be added to the instructions for now, as in this mockup in my sandbox. Click on the third green collapsed box-- the rest is unchanged. This would allow an interim solution while we are working out a better design.

P.S. It produces a straight list of articles at FAR using {{#categorytree:Wikipedia featured article review candidates|mode=pages|hideprefix|namespaces="1"|hideroot=on}} —courtesy of a helpful IP at WP:VPT. SandyGeorgia (Talk) 02:07, 12 November 2021 (UTC)

OK, FAR is gone and right now my iPad is having trouble loading the page. (Stalling after only 16 FACs load.) Case rested. This exercise in debate helped neither FAC nor FAR. SandyGeorgia (Talk) 15:34, 12 November 2021 (UTC)

It must be hardware-specific then. I loaded the whole FAC page in less than one second on my iPhone. The Rambling Man (Keep wearing the mask...) 17:05, 12 November 2021 (UTC)
I doubt it; when I say “on my iPad” (verus my “real” computer), it has to do with which internet connection I am using. (In my case, related to the limited amount of time I am able to sit at a real desk on a real computer since the tree fell on me.) And not everyone, everywhere, has access to ultra-high speed internet. What I am saying is that we are still limiting access to FAC. SandyGeorgia (Talk) 17:09, 12 November 2021 (UTC)
Ok, I did the whole page on a 4G network in a second. I do have a measly 40 MB/s at home, but this was in the school playground. The Rambling Man (Keep wearing the mask...) 17:18, 12 November 2021 (UTC)
I would imagine there are vastly worse internet connections around the world than a UK 4G phone in the playground, to be honest. I've done entire working days through remote desktop on a laptop tethered to my phone before, so that thing's quite zippy these days. But then again, the sort of internet connection that can't handle the FAC page is probably not going to handle it however many savings we make through banning templates and other tweaks. For those people we really need a separate page which links to individual FACs without including the entirety of all of them on one page.  — Amakuru (talk) 17:24, 12 November 2021 (UTC)
I upgraded to (I don't remember what) last year for these very problems, but there are times when I can no longer sit at a desk, and I hope to not have to purchase something else just to cover those times I am forced to the sofa. More significantly, not everyone has or can afford fast coverage as we have, which in my case, is adequate for all of my other internet usage, but not for FAC. MEANING: we need to continue to press forward on solving the broader problem(s) with how the FA pages are structured. Over and out, heading for a car trip, from where I will have even more limited coverage as I pass through remote areas with variable access. SandyGeorgia (Talk) 17:28, 12 November 2021 (UTC)
Have a safe journey, Sandy.  — Amakuru (talk) 17:29, 12 November 2021 (UTC)
Back during the worst COVID stage of 2020, I was living with my parents. We were unable to get internet connections that averaged more than 10 MB/s. I couldn't open the FAC page and had to get to FAC pages by navigating to the article's talk pages and getting to the FACs from there. Gotta live rural USA infrastructure. Hog Farm Talk 17:32, 12 November 2021 (UTC)
Thanks, Amakuru (as weather and driving conditions are not good). HF, I did not have these problems where I used to live; where I am now, they are a constant fact of life, with or without the damn tree falling on me. SandyGeorgia (Talk) 17:33, 12 November 2021 (UTC)
Amakuru indeed, I was "working from hotel" near Reading earlier this week, a rather splendid 196 Mb/s on 5G somewhat outperformed the "premium WIFI" offered by the hotel! It would be interesting to know to whom we're now pitching Wikipedia, e.g. screen resolutions, page sizes, etc. I don't know, but I've been told that more than half of our readers now use the mobile view, which as I said, loaded the contents of FAC practically instantly, of course while no doubt background loading the content. But it wasn't a terrible experience by any means. I guess if the project is going to constrain to 3G or below mobile networks and "not really broad" broadband then FAC will actually probably never be accessible to some. The Rambling Man (Keep wearing the mask...) 17:45, 12 November 2021 (UTC)
  • Amakuru's entirely correct: if anything, he's probably understating things. Most of my FACs were submitted on a 2MB/s connection. If the FAC page is not loading for people with internet in their homes in the US/UK, no amount of tinkering with template rules is going to fix it; we need to move to a one-review-per-page system like GAN. No amount of moving long reviews to PR is going to fix it either; if we move all the excess length there, PR isn't going to load, and the access problem is not fixed. Vanamonde (Talk) 17:51, 12 November 2021 (UTC)
    That's a very good point. To whom are we catering for load speeds? If we had a set of parameters against which to test to results of any change to process, that would be a start, especially as it seems to be not just a transclusion issue but a hardware issue. The Rambling Man (Keep wearing the mask...) 17:56, 12 November 2021 (UTC)
    (edit conflict) Yes, although I suppose the drawbacks of this approach are (a) for coordinators looking through the list for what needs attention and what doesn't, and (b) similarly for reviewers looking for somethign that is in need of a further pair of eyes. We could always do both, have an overview page which gives the list of open candidates and perhaps some rough indications of where they're at, while retaining the page which shows all the FACs in one place for those that like that arrangement and have a suitable computer and connection...  — Amakuru (talk) 17:58, 12 November 2021 (UTC)
    Sounds like a 'bot could be used to give a summary of "number of reviews/supports/opposes/image review/source review" etc at a top-level. And even maybe "time since last edit" or similar? The Rambling Man (Keep wearing the mask...) 18:00, 12 November 2021 (UTC)
    Having the names of the nominators in the summary would also be good.--Wehwalt (talk) 18:20, 12 November 2021 (UTC)