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.
Support for this proposal. There isn't much recognition out there for GA reviewers, even though it can (especially with thorough source checks, see below) be a very time-consuming task. And who doesn't like collecting shiny things? Not sure how detailed the awards could be, but maybe there could be something for people who review old and/or very long nominations. Unexpectedlydian♯4talk‽
What about an incremental awards system like the one at DYK? It seems to be hampered by a lack of visibility, but I'd be lying if I said those awards weren't a strong motivation for me to get involved in DYK. Thebiguglyalien (talk) 17:42, 2 January 2023 (UTC)[reply]
I rate this suggestion a sure, why not. I don't expect this to have a particularly large positive effect, but on the other hand I think the likelihood of significant negative outcomes is negligible. That is to say, this might entice some editors to put in the time and effort to review nominations a bit, but I don't think we're going to see any notable increase in low-quality reviews from editors who just want to collect awards if we implement this. TompaDompa (talk) 22:43, 2 January 2023 (UTC)[reply]
Trainsandotherthings, spot checking is already required : see Wp:reviewing good articles. Also, I do not know how a reviewer could certify that the article is free of original research without checking some of the sources. I would, however, be in favor of making the requirement more prominent. (t · c) buidhe 21:52, 30 December 2022 (UTC)[reply]
Separating this proposal out into its own heading as it is related to proposal 2 but may benefit from another thread. To repeat what @Mike Christie has written above, a rule could be "check a minimum of 5% of the sources, and no less than five in total". To give my opinion, I was thinking more "around 10% of the sources, and no less than five in total". Unexpectedlydian♯4talk‽ 20:45, 3 January 2023 (UTC)[reply]
I'd just like to preempt any such concerns by explaining that any implementation of QPQ would probably require a grandfather clause. I.e., a prolific nominator with 100 nominations and only 1 review would not suddenly need to review 99 articles to start nominating again. Rather, they'd need to review 1 article in order to nominate their 101st article. There are questions as to how we would monitor that. The easiest would be to simply reset nomination and review numbers, though that might prove unpopular. Krisgabwoosh (talk) 22:24, 30 December 2022 (UTC)[reply]
I have gotten some great GAN reviews in the past; these two stand out for their thoroughness, but I'm not sure they are great examples for exactly this reason. Has anyone got a really well done, but less lengthy review that could serve as a model? (t · c) buidhe 08:51, 30 December 2022 (UTC)[reply]
Open a page to collect noms on a rolling basis for a perennial "Exceptional Review of the Month" recognition. Let's not get lost in the details of what qualifies, as the point would be mainly to recognize exceptional reviews affirmed by the community. I.e., reviews of difficult subject matter or broad/Vital topics, reviews that are exceptionally welcoming to a new reviewer, reviews from an expert in the field who wouldn't normally review. Basically we'd encourage nomination of these and any reviews over some voter threshold would get barnstars and their review shown in a place of honor for some limited time. This would effectively address the heart of Props 4 and 1 while providing a format for peer recognition. I'd like for editors to "collect" exceptional review recognition the way that other editors collect GA and FA icons. czar 09:23, 14 January 2023 (UTC)[reply]
Hi "name", I see you are relatively new to reviewing Good Articles. If I can help in anyway feel free to ping me
There are some editors who have several open GANs who are choosing not to review, perpetuating the backlog. By limiting the number of nominations, editors will hopefully spend more time reviewing. This in turn will improve their review/GAN stats and encourage more editors to review their own noms. It would also encourage editors to nominate their best work first, as less comments in a GAN will get their article promoted faster, freeing up another spot for their nomination and reducing the amount of time a reviewer needs to spend giving comments on the article. I think five open GANs would be a good number, but other suggestions are welcome. Z1720 (talk) 02:03, 2 January 2023 (UTC)[reply]
The most prolific/high quality contributors tend to receive the quickest reviews anyways
Addendum: Following implementation, consensus was determined that 20 would better be implemented. This is in regards to the interpretation of this thread and the general layout of the nominations page.
There's been a fair amount of debate about a limit, with more support for more generous limits. I think a cap of 20 is more than reasonable, and to get consensus I'm proposing it as a separate item. Trainsandotherthings (talk) 04:35, 15 January 2023 (UTC)[reply]
Limiting nominations per nominator targets mainly industrious contributors and may result in fewer high quality nominations without more reviews.
Anyone who is nominating this much is after it for reward culture
Some nominations remain open for several months and no one wants to review them. In the past, when I reviewed very old nominations the nominator had sometimes left Wikipedia and the time I spent on the review is wasted. This proposal would cause nominations that are very old to expire and be automatically quick-failed. This can hopefully be set up by a bot to lessen editor time. The article can be nominated again. Z1720 (talk) 02:03, 2 January 2023 (UTC)[reply]
assume that the article is static and they'll need to pass or fail as-is
I'm going to split this out into its own proposal so it doesn't get buried under #7. Assuming it's technically possible via ChristieBot, a flag of some kind should be added to nominations whose editors have not edited in 30 days. The flag would be removed if they start editing again. ♠PMC♠ (talk) 23:22, 6 January 2023 (UTC)[reply]
Edit the template in my sandbox: https://en.wikipedia.org/wiki/User:Rjjiii/sandbox (update: moved to its own page: https://en.wikipedia.org/wiki/User:Rjjiii/GA ) This is a rough draft to give a concrete idea of what I mean about a template with instructions. The actual instructions should come from someone more involved with GA, and I've tried to just copy and paste from the GA instructions and essay as much as possible. I don't think experienced editors will be greatly inconvenienced by deleting something like this before a review. For new editors I think having a kind of form to fill out, will be more clear than the current blank slate. The language of these instructions should mirror whatever is on WP:GAI, which should also give prescriptive, actionable steps. This does not change any policies; it provides a default path for reviewers.Rjjiii (talk) 05:29, 2 January 2023 (UTC)[reply]
Here is an example of how I'm imagining a static template. The idea is that by default, the review box would present a reviewer with a form that would have clear actionable steps:
==== No original research ==== It contains [[Wikipedia:No original research|no original research]]: <!-- Check at least 1 of every 15 sources to see if the content in the sources backs up the sentence, clause, or paragraph where it's attached in the article. Make note of any sources that do not verify the article's content. -->
Sorting the GAN page so that the format is unchanged but nominators who review frequently come higher on the list will reward reviewers (since some reviewers will work from whatever is top of the list) without limiting the number of nominations or requiring quid pro quo. Mike Christie (talk - contribs - library) 11:29, 2 January 2023 (UTC)[reply]
{{laal}}
There is an social expectation, enforced by some editors but absent from the GA criteria, that reviewers provide an extended, written peer review to accompany the standard assessment against the GA criteria. We should codify in the GA criteria that there is no such requirement. This will return GAN to being a simple checklist assessment against the criteria with written text only as needed to justify the assessment. czar 04:45, 3 January 2023 (UTC)[reply]
Detailed peer review at GA should explicitly be optional, determined by the nominator's wishes. A peer review/FAC prep/whatever parameter should be added in the {{GAN}} template so the nominator can flag whether or not they're looking for that style of review. This parameter could also be another useful sort parameter for the sorting overhaul being discussed at proposal 9. ♠PMC♠ (talk) 06:18, 7 January 2023 (UTC)[reply]
Drive by nominations are currently allowed, but discouraged, by the rules: "Articles may be nominated by anyone, though it is highly preferable that they have contributed significantly to the article and are familiar with the subject." Change this to something like, "Articles may be nominated by anyone who has significantly contributed to the article. Drive-by nominations are not allowed." I think this actually fits practice better: many drive by nominations are just reverted by those of us who maintain the page; other drive by nominations, however, can use up reviewer time. Changing the policy to explicitly forbid them will discourage such nominations or at least make it a bit easier to quickly revert them. (t · c) buidhe 09:50, 4 January 2023 (UTC)[reply]
Nominators who are not significant contributors to the article must consult regular editors of the article on the article talk page prior to a nomination
"Drive-by" nominations, which are permitted
Open GAR pages would be much easier to find and might get more attention if the GAR listings were on the same page as the GANs. The topic parameter of the GA template would allow the GARs to be listed in the same section as the GANs for that topic. Mike Christie (talk - contribs - library) 13:10, 7 January 2023 (UTC)[reply]
Part of what makes GAR a cumbersome, months-long endeavor is that it attempts to copy the GA model of addressing problems as they are pointed out – but often times, the original GA nominator isn't around to be the point person and fix all of the issues that come up. To that end, GAR should just stick to reviewing, and not to attempting to fix major flaws in the article. GARs should be limited to two weeks; the result can either be to keep the article, to delist the article, or send it back to GA for a full review if someone (who isn't the GAR nominator) volunteers to field it through the GA process. One nice benefit about this is that if someone volunteers to take it through, they can get a GA credit for their troubles. It also consolidates the two backlogs where appropriate. theleekycauldron (talk • contribs) (she/her) 08:57, 18 January 2023 (UTC)[reply]
Streamline the GAR process by merging individual and community good article reassessment, with a process loosely modelled on the following concept:
All GARs are centrally listed, with anyone free to weigh in. A GAR can be closed as delist after a week by the editor who opened it, unless someone has objected to the delist. At that point participants are expected to discuss the article's issues, make any necessary changes to the article, and come to a consensus about whether the issues are resolved or not. If a clear consensus develops at that point, the opener can close the GAR on their own recognizance. If participants are not able to resolve the discussion, an uninvolved editor or the GAR coords (if that proposal gets consensus) should step in and determine next steps.
With apologies for booting you down slightly Femke - I figured I would clarify what exactly is being proposed, so editors can support or oppose based on a clear concept. ♠PMC♠ (talk) 03:24, 10 January 2023 (UTC)[reply]
Have a bot leave an invitation to review GAN on first-time nominators talk pages if Proposal 9 passes. The invitation would briefly explain Proposal 9 and the benefits of doing a GAN review.
If proposal 9 passes, change the backlog box, Wikipedia:Good article nominations/backlog/items, to list the five articles that sort to the top using the sort order in proposal 9. The containing page, Wikipedia:Good article nominations/backlog, would also change to say "highest priority" rather than "oldest".
The nominations page is by far the most watched and active. Everything discussed at the Criteria and reassessment page concerns the process so it makes sense to put it keep in all in one centralised place. Aircorn (talk) 07:15, 14 January 2023 (UTC)[reply]
Put a notice banner for active GA nominations atop the article (in mainspace) as encouragement for editors reading the article to review it. czar 08:27, 14 January 2023 (UTC)[reply]
Require the nominator to run through a self-review of the GA criteria as a checklist on the talk page to show how the article meets the criteria. This shifts the burden of the first passthrough comments from the reviewer to the nominator. It also shows where the nominator needs coaching if they didn't understand the criteria. For instance, if we want the reviewer to spotcheck citations for controversial statements, the nominator should be able to say they checked that in advance. czar 08:50, 14 January 2023 (UTC)[reply]
When an article is nominated, have a bot provide a low-level pre-review with advice/suggestions for improvement, so the nominator is prompted to clean up the basics before the reviewer arrives. (This moves some of the reviewer's burden to an auto-reviewer bot.) I.e., anything afoul of the quickfail criteria, notice of any open maintenance tags, Earwig %, recent edit warring, reversion of drive-by nom (if approved above), paragraphs lacking citations. czar 08:50, 14 January 2023 (UTC)[reply]
{{good article tools}}
Hold an RFC on options to make GA status more prominent in mainspace. Showing and explaining GA status in a more prominent way would both help with reader literacy of status (and prompt curiosity about the criteria) while rewarding editors for their efforts. I.e., one RFC proposal could be to turn the article title the color green with a tooltip that explains how the article has been reviewed.
In a former career, I observed a high school teacher that happened to be teaching research skills on using Wikipedia. She said that the lock icon in the corner meant that it had been reviewed. Readers have no idea what our esoteric icons mean, so a little explanation of what exactly is verified and what isn't could go a long way towards mutual incentives. czar 09:33, 14 January 2023 (UTC)[reply]
you'd be surprised how little the teachers know about Wikipedia at times
{{GANentry}}
I mentioned this above but wanted to break it out to a separate proposal for consideration. We now know that most editors find new reviews through WP:GAN, so it behooves us to make the page legible. Right now, each row looks like this:
When it could instead be shorter, with something along the lines of:
Can play with the specifics but this would give more relevant info to a potential reviewer (the short description) and reduce the clutter of a bunch of shortlinks most editors won't use. I'd also be in favor of moving the review/GA counter to the second line and encourage more nominators to use the {{GA nominee}} |note= field to make WP:GAN entries more appetizing. czar 01:36, 16 January 2023 (UTC)[reply]
{{GA nominee}}
|note=
Split the topical sections of WP:GAN off into transcluded subpages, so that people can watchlist them individually. Do the same with User:SDZeroBot/GAN sorting.
Change the categories used to sort nominations on WP:GAN to match those used by WikiProjects, Featured Articles, ORES topic predictions (as used by User:SDZeroBot/GAN sorting), or something else widely-used elsewhere, making it easier for potential reviewers to find articles that interest them.
Allow nominations to be placed in multiple categories on WP:GAN, WP:DELSORT-style.
Make a good article version of the Wikipedia:WikiProject Military history/Academy
Following Prop 9 passing, I wonder if it might be fairer if those that have a very high number of reviews shouldn't get some form of minimum weight. With a ratio of 4.46 (259 reviews to 58 GAs), I am basically guaranteed to be at or near the top of any sorting; however, some, like User:Chiswick Chap (311 reviews to 488 GAs) and User:Sturmvogel 66 (914 reviews to 821 GAs) are (to my mind) unfairly penalized by this. Certainly, their ratios are not as good, but I would defy anyone to say they haven't contributed very significantly to the project; indeed per the stats, they are some of the most prolific reviewers in our history. I am strongly in favor of our new sorting system, but think it could be made fairer by setting minimum weights to nominators with a certain number of reviews, perhaps 0.5 for every 50 reviews, or something similar (perhaps a simple decimal value for every single review would be easier), such that Chiswick and Sturmvogel would get ratios of 6 and 18, in recognition of their extensive work on both sides of GA, rather than their present 0.63 and 1.13. And of course, any ratio higher than the minimum assigned by the number of reviews would be retained. Iazyges Consermonor Opus meum 19:15, 17 January 2023 (UTC)[reply]
One of the strengths of FA reviewing is that people can play to their own strengths; some people review prose, some people review sourcing, some people review image licensing. If reviewing were itemized by criterion, instead of having one person doing all the work for every nomination, that might help streamline the process. Of course, we don't want people stepping on each other's toes – GA reviewing shouldn't have community members duplicating each others' efforts or anteing the bar up too high – so one reviewer per criterion. I mean, one of the big things we're finding in this discussion is that people aren't doing reviews because it's a really big time commitment for a volunteer project – wouldn't we attract more new reviewers if we allowed people to dip their feet in the water with one criterion or another first, before doing a full solo review? theleekycauldron (talk • contribs) (she/her) 08:37, 18 January 2023 (UTC)[reply]
Hopefully the changes we are proposing won't result in us needing backlog drives in the same way we have needed them in the past. But, I think there could still be value in allocating certain months where the GA community focus solely on reviewing. There could be associated barnstars or awards for particularly impressive reviews. It could also be a good way of getting new people to participate in reviewing if we phrase it in a friendly "come along and review" way. Unexpectedlydian♯4talk‽ 08:45, 18 January 2023 (UTC)[reply]
Most special stuff on wikipedia gets organized by every month of the year. FA has it, DYK has it, FL has it, FP has it. Yet, Good articles don't have a separation by month of the year. So I propose adding categories that separate GAs based on when they were promoted. For example, GAs promoted in December 2022, January 2018, May 2012. It'd be a nice way to categorize GAs and analyze GAs promoted long ago. Onegreatjoke (talk) 02:10, 19 January 2023 (UTC)[reply]
Rather than notifying each listed WikiProject for a GAR nom, only notify those projects for which the article is high or top importance.
Per above: leave the notification of Wikiprojects up to the discretion of the nominator. Change the text from "Notify major contributing editors, relevant WikiProjects for the article, the nominator, and the reviewer." to "Notify major contributing editors, the nominator, and the reviewer. Consider notifying closely related WikiProjects"
Since the above sections say we want WikiProjects to be notified of new GARs, this sounds like a job for a bot. Notify the WikiProjects for which the article is tagged and make the bot respect a "do not notify" opt-out template.
As the name implies, this drive does have to close inevitably. I propose we continue discussion here at the talk page but close off Wikipedia:Good Article proposal drive 2023 from further proposals. 🏵️Etrius ( Us) 23:59, 21 January 2023 (UTC)[reply]
Have a bot post a monthly digest of open reviews associated with each WikiProject to increase review visibility. The bot should comply with a "do not notify" opt-out template.
Wikipedia:Delisted good articles
This is more of a personal opinion of mine but I propose the idea of making a page listing delisted GAs in a similar way to other pages like WP:FFA and WP:FFL. This is because I feel that the way showing delisted GAs through Category:Delisted good articles is pretty cluttered to me. Sure it works somewhat but I feel like it could be done better in the ways that I've listed. Would like to hear some opinions. Onegreatjoke (talk) 19:36, 25 January 2023 (UTC)[reply]
A list of GA-related backlogs can be found at https://bambots.brucemyers.com/cwb/bycat/Good_articles.html. Cleaning up backlog as a contest helps editors to stay engaged and it is also a good way to draw attention to improving poor good articles as a whole. CactiStaccingCrane 16:11, 2 February 2023 (UTC)[reply]
Hi. There currently appears to be a bug with the bot such that it is not accounting for any reviews (or GAs) that a user has made under a previous account name. This is leading to the page erroneously displaying that I have reviewed precisely 0 GAs, when that is not true. The issue lies with the bot that updates User:GA bot/Stats recording reviews in a table based on the username that was in-use at the time of the review rather than the current username of the user who made the review.
Until the bot that updates the table at User:GA bot/Stats is fixed to accurately reflect statistics of GA reviews for users who have changed usernames, the sorting of the GA table should revert to being a reverse-chronological order (i.e. oldest-at-top)
The bot has a table of name changes, which I add to manually; adding an oldname/newname pair fixes it for a given editor. Red-tailed hawk is right that the data would be fairly easy to get -- I could simply read through all the users on the review stats page and check if the user page is a redirect, and if it is insert it as an oldname/newname pair. There are also cases where the old user page does not redirect to the new user page; I can't find those automatically but I can add them on request. I've added a handful already, either on request or where I know the old list had merged the names -- e.g. Malleus Fatuorum/Eric Corbett. I can produce the list of redirects fairly easily if we want to see it, but I'd want to see consensus that we should automatically add those. I think it would probably be OK -- the user page wouldn't be a redirect if the user weren't willing to have the accounts associated. Mike Christie (talk - contribs - library) 11:47, 5 February 2023 (UTC)[reply]
Here is a list of all the old name / new name pairs that would be created if I were to do this. Mike Christie (talk - contribs - library) 12:47, 5 February 2023 (UTC)[reply]
Although I am not as active as I would like to be with this project anymore, I have been involved in many of the previous discussions on improving GA. I think we need to be careful that any changes we make do not deviate too much from what makes GA unique from other peer review processes. To my mind that is its lightweight approach and flexibility in reviewing. I have always seen this as an easy way to improve articles from both a reviewer and nominator perspective. It has its flaws, but I think we often do better to accept and embrace them than make this something it isn't. FA, DYK and peer review (is it still around?) fill some of those gaps, although they all have their own flaws.
I am going to be contrary to a few others here by saying the backlog doesn't matter. Or more that it is something we should just accept. As long as I can remember a backlog has existed and everytime we reduce it it comes back bigger than ever. The main issue is new nominators who do not realise it could take a while for their article to get reviewed, but that is solved somewhat with the new GAN format.
In fact I think the new format is the biggest improvement to GA in years and hope it reduces many of the issues or perceived issues. I would like to see what that does before making major changes to the format. This is still a worthwhile exercise though.Aircorn (talk) 21:57, 1 January 2023 (UTC)[reply]
Just starting this discussion as I'm sure it will come up eventually. Do we have a sense of why people don't review GA noms? In my mind, there are three categories:
Focussing on the final two categories, I assume that one reason might be the slightly thankless nature of being a reviewer (a GA is a recognisable milestone and the nominator gets a nice green plus they can add to their belt, whereas a reviewer does not get the same kind of recognition). Another reason is being put off by the intensive nature of checking sources, especially on longer articles (although this is a separate discussion). Personally, I don't review articles in a subject area which I know little about (like science), and I am sometimes put off by very long articles in subject areas that I normally would be interested in. What other reasons do people have? Unexpectedlydian♯4talk‽ 10:56, 31 December 2022 (UTC)[reply]
and you wish to prescribe an amount of time for these issues to be corrected (generally seven days)
|status=
|status=onhold
will use GANotice to let the nominator know that the article is on hold.
can perform most of these steps automatically
A few reasons I have come across for people not reviewing is because they don't think the have the skills to cover all the criteria. While I think they would be fine it obviously is an issue for them. Maybe we could somehow have co reviewing as an option? No idea how it would work, maybe just through second opinion, but make it more explicit. Aircorn (talk) 17:28, 2 January 2023 (UTC)[reply]
Which surface/pages do most reviewers use to find new articles for review? WP:GAN? WikiProject article alerts? Something else? czar 06:53, 7 January 2023 (UTC)[reply]
Is ChristieBot's review count, as listed at User:ChristieBot/GA reviewing stats, limited in some way? I keep a tally of my GA reviews and it hit 50 today; I've just gone through my Talk contributions and re-counted and you can subtract 1 or 2 off that were second opinions but just by page creations of "/GA1" pages I should be much closer to 50 than my currently listed 33. If the bot only goes back to a certain date or the limitation is known then that's fine—both 33 and 50 should get the point across that I'm not a nominator who never reviews—but I wondered what the reason is. — Bilorv (talk) 15:35, 7 January 2023 (UTC)[reply]
a log file kept by the review tool
One thing occurs to me as a possible problem with any change that nudges nominators in the direction of reviews, and that's competence in reviewing. Doug Coldwell did not review much, and I think his prose skills did not enable him to be a good judge of what is acceptable as a GA, so I would not have wanted to see him doing more reviews. This is more common with non-English-speakers, though. I can think of a couple of nominators whose work I've reviewed who had a good grasp of most of the criteria, but whose English was not strong and who were not good at avoiding close paraphrasing. In some cases I think I recall helping to push the article over the line by doing some copyedits myself. I don't think this should prevent us encouraging more reviews, but perhaps it's something to bear in mind when choosing what to review -- if I don't think nominator X is a good reviewer, I shouldn't hold it against them if they haven't reviewed much. Mike Christie (talk - contribs - library) 15:11, 9 January 2023 (UTC)[reply]
Is there a timeframe for when proposals will be assessed and/or implemented? ~~ AirshipJungleman29 (talk) 01:54, 11 January 2023 (UTC)[reply]
{{cent}}
I won't open it up as a proposal for now given the aim of this proposal drive is partially to eliminate the need for backlog drives. However, it may be the case that a backlog drive is a good way to get the community together and focus on reviewing for a month or so, if we find that there is still some sort of a backlog in the future. Maybe it could be reframed as a GA review month-long competition or something, if the number of noms doesn't necessarily hit the "backlog" criteria. Unexpectedlydian♯4talk‽ 22:38, 14 January 2023 (UTC)[reply]
I've implemented proposal 7a (flag inactive nominators). If anyone would like a different way to flag them, please say so. I set the "inactive" timer to 21 days; that's easy to change if we want to. I also flagged inactive reviewers; we didn't discuss it but it was easy to do and seems harmless. Inactive reviewers are not flagged if the status is in "second opinion", since in that case there's already a request for another reviewer. I can remove the "inactive reviewer" notes if we don't want them. Mike Christie (talk - contribs - library) 23:22, 14 January 2023 (UTC)[reply]
Moved discussion to feedback section. Mike Christie (talk - contribs - library) 11:58, 17 January 2023 (UTC)[reply]
Any feedback, or coordination of implementing passed proposals, can be discussed at Wikipedia:Good Article proposal drive 2023 Feedback. Once the Proposal Drive is closed, I will create a separate tab for the feedback page. Unless there is any opposition, I will close the drive itself towards the end of January and hope to wrap up pass/fail discussion by early February. Feedback will go till the end of February. 🏵️Etrius ( Us) 04:17, 16 January 2023 (UTC)[reply]