Wikipedia talk:Good article nominations: Difference between revisions

Browse history interactively

← Previous edit Next edit →

Content deleted Content added

VisualWikitext

Inline

Revision as of 19:54, 6 July 2017

WT:GAN

This is the discussion page of the good article nominations (GAN). To ask a question or start a discussion about the good article nomination process, click the New section link above. Questions may also be asked at the GA Help desk. To check and see if your question may already be answered, click to show the frequently asked questions below or search the archives below.

view · edit

Frequently asked questions (see also: GAN instructions)

Good article nominations

Shortcut

WP:GAFAQ

This is the list of Frequently asked questions about nominating and reviewing Good articles. If you cannot find the answer to your question here, you might want to ask for assistance at the GA nominations discussion page.

Nomination process

The backlog is huge! Is there something we can do to restrict nominations?: There have been complaints about the perceived backlog in reviewing since the Good article status was created in 2006. Generally speaking, we don't want to restrict nominations along their path to GA. In the beginning, as many as 100 nominations were waiting for a reviewer to volunteer. By 2011, each day typically listed 330 nominated articles, of which 260 were waiting. By 2016, 580 were listed, 460 waiting; by 2021, 470 were listed, 300 were waiting. For comparison, today there are currently 614 nominations listed and 510 waiting for a reviewer.; While it may seem overwhelming, a large backlog isn't a bad thing. It shows that many nominators want to use GA as a tool to improve the encyclopedia. It also allows reviewers to choose from a wide selection of articles that interest them. From a nominator's perspective, the main concern is the expected wait time before receiving a review, not the number of articles on the nominations page.
Can't we force nominators to review articles?: Quid pro quo reviewing (editors must review an article before nominating, perhaps after a grace period) was regularly proposed and always rejected as likely leading to lower quality reviews and fewer nominations from excellent content creators who may not wish to review another person's work.
What order do nominations appear on the Good article nominations page?: One advantage of reviewing articles to help the Wikipedia community is the GAN appearance order: the more articles you have reviewed relative to articles you have nominated, the higher up in the queue your nomination will appear on the GAN page. Note that above this, order preference is given to nominators who are new to GA.
What if the nominator is a (perhaps dynamic) IP address?: Any editor with significant contributions to the article may nominate an article for GA status (while only registered users may review), so significant-contributor IP nominators are permitted. Non-significant contributors should follow the advice above for editors who have not significantly contributed to the article. Dynamically changing IP nominators may want to clarify to the reviewer that, despite their signature, they remain the same person.
Can I nominate an article I haven't significantly contributed to?: No. Nominations from editors who have not substantially contributed will be removed unless the nomination shows that the article's regular editors were consulted on its talk page alongside a note on the GAN template showing their commitment to the process, e.g., |note=Adopted following extensive improvement by another editor. In the past, many drive-by nominators did not know the article or its sources and did not respond to questions by the reviewer.

Review process

I want to review an article. Do I have to review the oldest unreviewed nomination first?: Thank you for deciding to review an article for GA. You may review any nominated article you are not involved in, regardless of the nomination's age or position in the queue.
As a reviewer, I want to help the nominator improve the article, but it already meets the criteria. What am I supposed to do now?: The purpose of a GA review is to determine whether the article meets the GA criteria. Article-improvement discussions are intended to prevent near-misses in nominated articles that almost meet the criteria. If the article already meets the criteria when you first review it, then explain exactly what you've checked in each of the main criteria (e.g., "I have verified there is no original research, copyright violations, plagiarism, unreliable sources, bias, edit wars, or untagged images.") and list it as a Good article. "Quick passes" are as legitimate as quick fails, although they are less common. Please do not make a list of nitpicky details or exceed the written requirements of the Good article criteria in an effort to make the review look rigorous. Instead, thank the nominator for presenting such a polished article, and encourage them to submit another.
Who can respond to the review?: The nominator is expected to respond to the reviewer's suggestions to improve the article, while everyone interested in the article is encouraged to participate in the review, not just the person who happened to nominate it. Nominators have no special privileges over other editors except that they can withdraw the nomination.
Should nominators respond to reviewers' concerns? And what should reviewers do if they don't?: The nominator has an interest in seeing the article become GA, so the nominator will likely want to respond to the reviewer to improve the article. In fact, all editors interested in the article are encouraged to respond to reviewers' concerns. However, nobody, including the nominator, is required to. If the reviewer identifies concerns directly related to the Good article criteria and no one addresses the concerns, then no one should be surprised if the reviewer declines to list the article. If the article does not meet the Good article criteria after the reviewer has waited a reasonable amount of time for the nominator to make improvements, the reviewer is sure to fail the nomination. Future article editors may benefit from review comments on how to improve the article.
The nominator disagrees with the reviewer. Can another reviewer take over?: If your GAN experience is not going well or if you are disagreeing with the reviewer's decisions, then you may allow the review to fail, take the reviewer's suggestions into account, then renominate the article immediately (to get a different reviewer). If the reviewer has not yet failed the nomination, you may try asking them to ask for a second opinion. Other than these, another reviewer does not normally take over an active review. You might want to read What the Good article criteria are not. To prevent avoidable disputes, reviewers should either not make comments unrelated to the Good article criteria, or they should carefully label those suggestions as optional (e.g., "I know that citation formatting is not required by the Good article criteria, and I will not consider this when making my decision, but if you might send this to Wikipedia:Featured articles later, then that process requires consistent citation formatting").
Does an article have to be on hold for exactly seven days?: No. Whether to place the nomination on hold at all, and the length of any such hold, is for the reviewer to decide. Depending on the quality of the article and the responsiveness of the nominator, a hold may not be necessary. If the reviewer decides that it is, they may choose longer or shorter periods of hold time. The reviewer may even modify the {{GA nominee}} template on the article talk page to include a "time" parameter, for example, "time=fourteen days", and the {{GANotice}} template used to convey messages to the GA nominator to include a "days" parameter, for example "days=fourteen". Keep in mind that protracted reviews show up as exceptions on the GA nominations report page.
What should I do if a review page becomes inactive?: This can happen for a number of reasons. Review pages should only be started by reviewers who are willing to take an active interest in the article and are committed to completing their review of the article in a timely manner. Sometimes another editor (such as the nominator) starts the review page by mistake. A reviewer can fix this by placing their signature after "Reviewer:" towards the top of the review page, but if no reviewer is forthcoming, it may be best to delete the review page: requests for such deletions may be posted at the discussion page. If a new reviewer is truly needed, follow the instructions page under "What to do during a review if it seems abandoned". Do not use this process to void a review you disagree with.
I failed the article, and the nominator just nominated it again without fixing the problems I identified!: That's okay. There is no time limit between nominations, and this is the recommended process if the nominator disagrees with your review. Let someone else review it this time. The new reviewer is sure to read your comments while independently deciding on their assessment. If your concerns were legitimate, then the new reviewer will doubtless agree with you and also decline to list the article as GA. If the article is passed and you do not believe it meets the Good article criteria, you can initiate a reassessment.
How can GA be reliable when a single reviewer decides?: The quality of a Good article is only as reliable as the most recent review and articles may deteriorate if unattended. The GA process deals with both of these issues by allowing repeat reviews by any registered user at any time. The process aims to encourage article improvement and build consensus on quality through multiple reviews—even though a single reviewer makes the decision whether to list the article according to the GA criteria. Any editor may contribute to any review discussion and reassessment is available when the "one reviewer decides" model breaks down.
What if I have concerns about the quality of a review or need to resolve a dispute over the GA process?: You can bring those concerns to the discussion page to get help from other editors. Remember, however, to notify all users about whom concerns have been raised or who are involved in any dispute that you have.

Archives

Archive 1	Archive 2	Archive 3
Archive 4	Archive 5	Archive 6
Archive 7	Archive 8	Archive 9
Archive 10	Archive 11	Archive 12
Archive 13	Archive 14	Archive 15
Archive 16	Archive 17	Archive 18
Archive 19	Archive 20	Archive 21
Archive 22	Archive 23	Archive 24
Archive 25	Archive 26	Archive 27
Archive 28	Archive 29	Archive 30
Archive 31	Archive 32

This page has archives. Sections older than 7 days may be automatically archived by .

Dated archives

This page has archives. Sections older than 7 days may be automatically archived by .

Talk:Proconsul/GA1

Cwmhiraeth posted to my talk page asking my opinion of this review, which was done by a new user, User:Suspicious eyes, on his or her second-ever Wikipedia edit. Not a single issue was found, which is odd, because the sourcing is rather sparse, including a "classic study" by Max Weber that is mentioned twice but never cited, and a Roman Empire section that has a single inline citation, at the end of the second of three paragraphs.

My plan is to revert the review and place the nomination back into the pool of articles waiting for a valid review. I'll wait to see what opinions there are on the matter, but not long. Note that the review has been disallowed at the WikiCup. BlueMoonset (talk) 05:03, 30 April 2017 (UTC)[reply]

I have already fixed the issues mentioned above, if anyone cares. Please read it, everyone. There are FAs that don't meet this standard. I reviewed an article a while back. The author blew his top and has somehow convinced people not to trust me. 10W40 (talk) 08:45, 30 April 2017 (UTC)[reply]

Maybe, but it still needs to be properly reviewed by someone. An editor that reviews an article as their 2nd edit and leaves no comments is not really good enough and more than a little suspicious. I would support a revert. AIRcorn (talk) 16:26, 30 April 2017 (UTC)[reply]

It might pass now, but it certainly shouldn't have passed at the time of the review. I support a revert as well. – Finnusertop (talk ⋅ contribs) 16:38, 30 April 2017 (UTC)[reply]

I have no thoughts about any issues having to do with this review from consideration of it outside the GA parameters. In my opinion the Review is incomplete/flawed, for one thing File:Scipio Africans Major.jpg is lacking the Commons-required United States public domain tag. Sorry but I support a revert. Shearonink (talk) 17:03, 30 April 2017 (UTC)[reply]

I suggest a reassessment as more appropriate. It is the procedure recommended for cases where there is doubt whether an article which has been listed as GA is up to standard. I thing reverting would be overreacting. No actual rule has been broken that I can see, so what grounds would be cited for reversion? • • • Peter (Southwood) ^(talk): 20:16, 30 April 2017 (UTC)[reply]

No, the review was clearly inadequate, by a brand new editor who obviously didn't understand what was required, and we typically revert such attempts. I note in passing that 10W40 did not, in fact, address the Weber issue, but it wasn't part of my decision; this needs a review by a competent reviewer. I have reverted the review and the listing. BlueMoonset (talk) 21:22, 30 April 2017 (UTC)[reply]

Where is this written? • • • Peter (Southwood) ^(talk): 05:35, 1 May 2017 (UTC)[reply]

Not everything needs to be written. At the end of the day we operate on consensus and consensus is clear. AIRcorn (talk) 06:24, 1 May 2017 (UTC)[reply]

Fair enough, but:

Are you referring to this instance, and the consensus assumed from the discussion above?
A consensus achieved at some other time? In which case, how do people who were not involved in that discussion know of its existence?
Something else? In which case please explain as I do knot know what you are referring to.

The point remains that the guidance provided for GA nominations and reviews is misleading and is the cause of dissatisfaction because of lack of clarity. • • • Peter (Southwood) ^(talk): 06:41, 1 May 2017 (UTC)[reply]

I meant the consensus above (which is 4:1 for a revert). The guidance for GAs is already massive and I don't think we need to document every eventuallity, especially with commonsense cases like this. AIRcorn (talk) 07:57, 1 May 2017 (UTC)[reply]

But is this really a common sense response, and should we look at the cause as well as the treatment?
Regarding the response: A review was done that does not actually appear to have broken any rules (I am judging this on my interpretation of the rules, and possibly the reviewer's. Others may differ, which is part of my point), by someone new, possibly acting in good faith and following the instructions to the best of their ability. Instead of following the reasonably predictable route as described in the GA nominations instructions of reassessment, the review is reverted. By a 4:1 consensus. Is this now to become the standard procedure in similar cases?
The GA instructions state: Articles can be nominated by anyone, though it is highly preferable that they have contributed significantly and are familiar with the subject, and reviewed by any registered user who has not contributed significantly to the article and is not the nominator. (my emphasis) It seems that in practice this is not really true. We should stop pretending that the instructions are not misleading, and explain what is really expected. • • • Peter (Southwood) ^(talk): 18:31, 1 May 2017 (UTC)[reply]

Reviews can be performed by anyone, but they can also be brought here and judged deficient. The important thing is not so much that this is a new reviewer but that it is a bad review. We should not let through articles that are not in Good Article status just to be nice to newbie reviewers. —David Eppstein (talk) 20:04, 1 May 2017 (UTC)[reply]

Right. If the reviewer had provided a more typical review with comments explaining how the article met some criteria, suggestions for improvements to help it meet others, etc., then the newness of the reviewing account would probably not be a big concern. --RL0919 (talk) 20:37, 1 May 2017 (UTC)[reply]

I would support adding the word "experienced" to characterize eligible reviewers in the instructions. – Finnusertop (talk ⋅ contribs) 22:15, 1 May 2017 (UTC)[reply]

Everyone has to start somewhere. If we only allowed reviewers who are experienced at GA reviewing, then progress here would grind to even more of a halt than it usually is. —David Eppstein (talk) 23:24, 1 May 2017 (UTC)[reply]

Finnusertop may have meant experienced editors rather than experienced reviewers. It is an important distinction. • • • Peter (Southwood) ^(talk): 23:38, 1 May 2017 (UTC)[reply]

Yes, that's what I meant. Thank you for clearing it up. – Finnusertop (talk ⋅ contribs) 18:09, 3 May 2017 (UTC)[reply]

(ec) That would be a step in the right direction, but how does one know when one has sufficient experience? As I see it, that is also something that only comes with experience. We need some way of informing people that if they don't know the English Wikipedia environment, policies and jargon pretty well they have virtually no chance of getting it right. This needs to be right up there in the lead section. GA review is decidedly not a good way to start editing, and there is nothing in that quoted instruction to suggest the complexity of what is actually expected. Even reading through the full instructions is not enough to prepare one for the job without the background of having done most of the required steps before. • • • Peter (Southwood) ^(talk): 23:32, 1 May 2017 (UTC)[reply]

I agree with the idea that there should be a minimum number of edits before doing a review. But I note that is not currently a rule. I tried just putting an article in the queue and waiting patiently. It took months and I ended up with this travesty. My conclusion was that next time I needed to ask somebody to do the review for me. It seemed like a good idea, and I'm sorry it led to this fiasco. Next time, I'll find someone with more experience. 10W40 (talk) 03:39, 2 May 2017 (UTC)[reply]

Yes, it is not the most specific way of putting it, but we already have a bit fuzzy requirements for the nominator ("contributed significantly and are familiar with the subject"). That is not a "rule" either, but a preference. – Finnusertop (talk ⋅ contribs) 18:09, 3 May 2017 (UTC)[reply]

Number of edits is easy to measure, but not necessarily meaningful in a way relevant to GAN skills. Getting an article through GAN as a nominator would be an apprenticeship/rite of passage that would probably be sufficient most of the time, but may be a bit more than strictly necessary. It is also easy to measure. It would be a recommendation rather than a requirement. I put it out as a straw man in the expectation that it will be shot down in flames. Perhaps for good reasons. It is a place to start. • • • Peter (Southwood) ^(talk): 06:09, 2 May 2017 (UTC)[reply]

I note that the reviewer for Talk:Encyclopedism/GA1 had 60,000 edits and 10 years of experience on Wikipedia when they made that review. I would certainly not have quickfailed the article at the time, but any experience requirement for GA reviewing that would have prevented that review would destroy the GA process, as the number of eligible reviewers would plummet. Caeciliusinhorto (talk) 19:02, 2 May 2017 (UTC)[reply]

Also a lot of experience with both GA and FA, from the evidence on their user page. Lack of experience cannot be the problem in that case. However, in most cases a quickfail is not a useful result, if there is a reasonable possibility that the article is fixable in a reasonable time span. It would be far more constructive to provide the same feedback on the talk page without calling it a review, and leave the nomination in the queue. • • • Peter (Southwood) ^(talk): 09:13, 7 May 2017 (UTC)[reply]

My idea is not to change the rule, but to make the written rule correspond to the real rule. No one challenges the review of Encyclopedism although the quality of the review itself is no better than one for Proconsul. 10W40 (talk) 23:34, 3 May 2017 (UTC)[reply]

Getting the written rules to correspond to the real rule would require a real rule to exist, which may not be the case. What appears to happen is a whole bunch of interpretations of the written rule, which expand it in various directions according to personal reasoning and preference. There is some consensus, but also a lot of stretching interpretation of the consensus. There are even people who try to follow the existing rule as closely as practicable, but I have no idea who and how many they might be. • • • Peter (Southwood) ^(talk): 09:49, 7 May 2017 (UTC)[reply]

@10W40: Of your previous GA nomination, you stated "I tried just putting an article in the queue and waiting patiently. It took months ...". This is inaccurate, it actually took four weeks before someone took that review on, about average for a GA review I would think. You also stated "My conclusion was that next time I needed to ask somebody to do the review for me." Not so, the nominator should not choose the reviewer; a reviewer needs to be independent and should voluntarily choose to review an article. The nominator should absolutely not ask a friend to review their article as you did; especially someone with no experience who has created an account two days earlier, and whose only editing action is to have added two commas to one article. Cwmhiraeth (talk) 17:35, 3 May 2017 (UTC)[reply]
- Cwmhiraeth, You say this, and it would probably get consensus as good practice, but it is not specified in the instructions: Articles can be (...) reviewed by any registered user who has not contributed significantly to the article and is not the nominator. If there is any other text limiting who can do a review, it is not easily seen by a potential reviewer or nominator. No mention of experience, no prohibition on asking someone, and there would not be much point in a prohibition as it would be unenforceable. There are even editors who specify on their user pages or on project pages that they would be available to review articles in specified fields, implying that they would be open to requests. If this contravenes any policy or rule, I am not aware of it. It may contravene some people's interpretation of best practice, but that is not admissible as a serious objection. • • • Peter (Southwood) ^(talk): 09:36, 7 May 2017 (UTC)[reply]

@Pbsouthwood: I am adopting a common sense approach. Someone who is unfamiliar with Wikipedia and the criteria by which a GA is judged, cannot possibly be in a good position to review a GA nomination. Are you seriously defending the choice of reviewer in this instance? Cwmhiraeth (talk) 09:53, 7 May 2017 (UTC)[reply]

Cwmhiraeth, I am assuming good faith, and pointing out that as far as I can tell, they have not contravened any written rule that I have been able to find. If that constitutes defence, then I suppose I am. Common sense is notoriously uncommon, varies depending on the observer, and is not always sensible. Besides that I agree completely that Someone who is unfamiliar with Wikipedia and the criteria by which a GA is judged, cannot possibly be in a good position to review a GA nomination Your analysis of the problem is clear and in my opinion, accurate, but I could not find it in the rules. • • • Peter (Southwood) ^(talk): 10:13, 7 May 2017 (UTC)[reply]

I did not attempt to do so. :) Cwmhiraeth (talk) 10:31, 7 May 2017 (UTC)[reply]

I'm the person who took up the second review of Proconsul. While I didn't look at the language & article structure as closely as many reviewers do, I did look at the content (which IMHO is more important), & pointed out a number of faults. However the nominator did not respond to my criticisms in a meaningful manner, & after giving him/her sufficient time to address them, I now find that person has been blocked for socking. If no one objects, I'll just fail this article. But I wish someone had pointed me to this thread, because I would have acted faster on this matter. Further, there has been another GA review I would have started on before now; I've been limiting myself to handling one GA review due to real-life commitments (a.k.a., family, full-time job, time spent commuting). And lastly, having stumbled upon this & being surprised, I feel like I've been made a fool of. -- llywrch (talk) 21:08, 1 June 2017 (UTC)[reply]

Archive URLs automatically

I highly recommend that all GA noms without archive links have the nominator run the article through iabot (Internet Archive bot), which will |archiveurl=s to all its reference templates with a single click (for example). czar 20:44, 5 May 2017 (UTC)[reply]

So you think it's a good idea to clutter up all references with visible "archived from the original on xxx" information even when all links are currently valid? Why? This is certainly a procedure which, if it were generally applied to all GA nominations, would discourage me from submitting articles to GA, because I wouldn't want to see them gunked up in this way. —David Eppstein (talk) 21:16, 5 May 2017 (UTC)[reply]

It depends on whether it is possible to archive a currently invalid link - I do not know how it works - It is easy to archive a currently valid link, how easy is it to archive the same site after the link is dead? How sure can one be that the link today has the same content as the link as accessed for the citation? Also, would it be possible to suppress display of the archived link provided the original link is working? That would cut down on unnecessary clutter. • • • Peter (Southwood) ^(talk): 09:57, 7 May 2017 (UTC)[reply]

Would it be better to do as suggested above and comment out the archived link (if this is possible), or request a modification to the cite templates to suppress display of the archived link while the original is still working? • • • Peter (Southwood) ^(talk): 17:32, 20 May 2017 (UTC)[reply]

Only suggested as a courtesy, not a requirement. Having been through this before, links eventually die, and I'd rather do the archival work now than regret it later. Also sometimes sites go down temporarily so I think we do better for our readers by offering backup inline with the original link. As for cluttering refs, I think the "access date" lines are a better target for loathing. WT:CS1 wants to keep them, last time I checked, so I personally remove them in my site stylescript. czar 19:04, 5 July 2017 (UTC)[reply]

Military history project: FA/GA discussion

MilHist project is hosting a discussion on GA / FA articles that have been tagged for a number of issues. The discussion can be found here:

MILHIST: GA / FA articles.

Interested editors are invited to participate. K.e.coffman (talk) 21:41, 10 June 2017 (UTC)[reply]

"Quick fails"

As written, the current GA criteria seem to be going out of their way to discourage "quick failing" articles. See, for example, footnote 2. I was rather surprised to see, however, this: "In all other cases, the nominator deserves a full review against the six criteria from the reviewer and is given a chance to address any issues raised by the reviewer before the article is failed." My concern is the demand that reviewers are obliged to "a chance to address any issues raised ... before the article is failed". This seems to be problematic on three counts.

It is not always appropriate. Articles are often not going to meet the GA criteria to a degree that reviewers are confident that the article is not currently ready for GA status, and in which they believe that placing the article on hold for x days is not going to help. Nonetheless, the articles in question may not meet the "immediate fail"/"quick fail" criteria.
It does not seem to describe current practice. Many experienced reviewers will, from time to time, close GA reviews after offering a full review and coming to the conclusion that the article is not ready to promote at this time. (I'm not naming names, but I know that I am far from the only person who does this.)
It does not seem to match up with the "instructions" page, which seem (rightly!) to be quite open to failing articles without placing them on hold. (Concerning failing an article, it says simply "If you determine that the article does not meet the good article criteria, you may fail it by doing the following". On putting articles on hold, it says "If you determine that the article could meet the good article criteria if a few issues are fixed and you wish to prescribe an amount of time for these issues to be corrected (generally seven days), you may put the article on hold by doing the following". Certainly none of the prescriptivism of the criteria page.)

I propose that we bring the criteria page into line with good sense, current practice and the instruction page, and change

In all other cases, the nominator deserves a full review against the six criteria from the reviewer and is given a chance to address any issues raised by the reviewer before the article is failed.

to something like:

In all other cases, the nominator deserves a full review against the six criteria from the reviewer. While it is common to place articles on hold if they do not yet meet the criteria, this is not necessary, and reviewers may fail articles outright after completing a full review if, in their judgement, the article does not meet the GA criteria and is unlikely to do so in the near future.

Comments welcome. Two final notes: First, this is not a solution in search of a problem. I have seen several arguments (some fairly bitter) about fails and quick fails that rest on the assumption (currently perpetuated by the criteria page) that failing articles without putting them on hold is in some way deeply problematic. We don't need to get into those arguments here; particular cases of articles being failed are not really the point. Second, this appears to be the de facto discussion page for all GAC issues, but I will leave a note on other relevant talk pages. Thanks, Josh Milburn (talk) 22:04, 3 July 2017 (UTC)[reply]

This change looks like a good idea to me. It has my support. —David Eppstein (talk) 22:10, 3 July 2017 (UTC)[reply]

Given the precedent of articles in general being failed without being quick failed, this proposal makes sense and thus, I support it. This wording is similar to the original wording before the meaning was changed by accident during a copyedit in the edit I referenced above. Y E ^Pacific ^Hurricane 00:30, 4 July 2017 (UTC)[reply]

Define "unlikely to do so in the near future". If an editor puts in 10 hours of work on an article in a week, anything can be fixed. It's unclear under what circumstances it makes sense to not give a courtesy seven days to a nominator. What's the benefit of failing instead of placing on hold? ~ Rob₁₃^Talk 01:05, 4 July 2017 (UTC)[reply]

I think all of this is answered in my original post. I gave three reasons for the change, and no benchmark was defined other than "in the reviewer's judgement". The nature of these review systems is that a judgement call has to be made- exactly the same thing is true in comparable professional and academic procedures. If the reviewer's judgement call was not the best it could have been, that's fine; there's nothing at all stopping a quick renomination, a reopening of the review, or a reassessment. We already have systems in place for this, and this is what goes on in practice. If you don't personally think that a fail without placing an article on hold could be appropriate outside of the narrowly defined "immediate fail" criteria, so be it; I support more reviewer autonomy, not less. Josh Milburn (talk) 01:17, 4 July 2017 (UTC)[reply]

I do not see the harm in waiting seven days. If they do not resolve it in seven days, fail it. Is there any particular reason you would not want to wait seven days? Kees08 (Talk) 01:27, 4 July 2017 (UTC)[reply]

Kees view is basically my own. The damage of failing a review is frustrating the nominator, who's perhaps waited 3+ months for the first review only to find he can't actually respond to it. If we're going to change things, there should be at least enough benefit to outweigh that damage. So what's the benefit? ~ Rob₁₃^Talk 02:05, 4 July 2017 (UTC)[reply]

Again, no one is forcing either of you, or anyone, to close reviews before you want to. But I do think that it is sometimes correct to close reviews without waiting the "expected" x number of days. Let me use myself as an example (noting that I have seen other experienced reviewers do both of these things). I've sometimes kept reviews open for months. Other times, however, I have felt that an article is not close to GA status, and I do not envision it being ready for GA status in the near future. I could run through the motions, and spend a few hours re-reviewing the article x days later, but this does not seem to be a good use of my time and does not seem to be a particularly respectful way to deal with the nominator, insofar as it seems to involve a degree of dishonesty. (I can envisage other good reasons to quickly close reviews, but perhaps shouldn't say them. Maybe other reviewers have other reasons for quickly closing reviews.) Incidentally, can I ask why you are suggesting that closing reviews leads to nominator frustration? That's not necessarily my experience. Also, I'm not really suggesting we "change things"; as I explained in my opening comment, I am suggesting that we should make WP:WIAGA consistent with current practice and WP:GAN/I. Josh Milburn (talk) 02:36, 4 July 2017 (UTC)[reply]

In case there is any genuine doubt, may I testify that closing a good article review as a fail without a hold leads to nominator frustration. 13.54.152.171 (talk) 23:00, 4 July 2017 (UTC)[reply]

Oh dear. Only if you've sworn on your holy book of choice, presumably. I have no doubt that some people have been frustrated by closes of this sort. I've no doubt that some people have also been frustrated by standard quick-fails, fails after holds, and perhaps even passes. And I've no doubt that people have been happy with all of these. I don't think that was really Rob's point. Josh Milburn (talk) 23:28, 4 July 2017 (UTC)[reply]

Oppose relaxing our commitment to holds before failing a good article review. A good article reviewer expresses respect for the effort of the nominator with a hold prior to a fail. The hold expresses our community's commitment to collaboration. A good article review is a collaboration between two editors: a nominator and a reviewer. The nominator expresses their view that an article meets the good article criteria with the nomination; a reviewer who disagrees is at best a one-on-one push, not a concensus. The reviewer already has the upper hand here, no need to make their club bigger. The reviewer's role should focus on a careful read and generating useful notes, not to second guess what the nominator or another editor or editors are or are not capable of accomplishing in a week. We have a wide range within our editors of skill levels and available time. After all, good article is our entry level quality rung, and we are trying to pass more articles, not fail more articles. Yes, the current policy as written strongly encourages a hold before a fail, limiting fails without holds to four specific situations, and is good policy since it expresses common practice; a hold before a fail is typical, customary, expected, and useful. 13.54.152.171 (talk) 05:48, 4 July 2017 (UTC) May I add I am not a WikiCup participant; previous discussion at WT:WikiCup#Concern: GAR fail with no hold on last day of round. Thank you. 13.54.152.171 (talk) 16:22, 4 July 2017 (UTC)[reply]

Thanks for your thoughts. To clarify: No one is supporting "relaxing our commitment to holds before failing a good article review" and this discussion has nothing to do with the WikiCup. Josh Milburn (talk) 16:27, 4 July 2017 (UTC)[reply]

I came across this after seeing the discussion in the context of the WikiCup, in which I participated once some years ago. I have looked at the review and what was noted, and I would say that it does not fit a reasonable reading of "the article does not meet the GA criteria and is unlikely to do so in the near future" unless predicting whether or when the nominator will return is reasonable. I would say it is not reasonable because placing a hold is the obvious way to test for a return. I think the IP editor has been treated very poorly here, especially having addressed the concerns in well under the seven day grace period usually offered, and I think that fails like this are highly counter-productive as they discourage editors and (in the event that there is a re-nomination) do not alter the backlog. On the specific case, since a fail apparently cannot be reversed, I think YE might offer to review the renomination as soon as it is made as a gesture of respect to the IP editor. Cwmhiraeth may also find the temperature of complaints would reduce if YE did not claim points for the review. This would (I think) address the specific question.

On the general question, Josh, I think the "unlikely to do so" clause as a reason for an immediate fail is wide open to gaming, to upsetting editors, and to fostering ill-will. Gaming in the WikiCup is not really the GA project's problem, but it could cause problems for the GA project. Suppose an editor irritated by the current case decided to fail several of YE's nominations by noting issues and then judging they were unlikely to be addressed quickly enough... since fails are irreversible, this would delay consideration of YE's articles for potentially several months. Alternatively, WikiCup editors could go looking for cases to review and immediately fail (as opposed to quick fail) to earn points more easily by not having to follow up on improvements. Either of these examples (and they are meant only as examples) would face WikiCup judges with problems but also produce disgruntled editors, about which the GA project should be concerned. I can think of articles that look complete but have something large missing where a review-and-immediate-fail would be justifiable under this caveat, but it strikes me as something to be used in only the most egregious cases because the cost of waiting a week is small compared with the potential issues with editor retention and enjoyment of Wikipedia participation.

I readily admit that my GA experience is small, but I would like to offer an example. When I nominated the rhodocene article for GA, it looked like this and was 745 words long (page size 17 336 bytes) supported by 15 references. The review was encouraging but offered some general comments that pointed to substantial areas for improvement. These are the changes that I made in response to the review, and when it was made a GA it was 1260 words long (page size 25 989 bytes) and supported by 24 references. Would it have been reasonable for the reviewer to say that improvements adding 49.9% to the page length, 69.1% to the word count, and 50% more references is article development that is "unlikely to occur" inside a week? It could certainly be argued that that amount of development is unlikely, yet it happened, and my first GA nomination was successful. Had I been failed, I would have been very discouraged and not looking to renominate (it was early in my wiki-career)... instead, I pushed on and got the article to FA standard, adding even more during the FA review than I had during the GA, bringing it to 4289 words (page size 75 620 bytes) with 60 supporting references. 6 years later, it's still an FA, I've contributed plenty more content, and the encyclopaedia and its readers are (I hope) better off. My points are that what development can be achieved in a week and what is unlikely are difficult to guess and that the consequences in discouragement / disillusionment for not allowing a hold are unknown and potentially significant. EdChem (talk) 12:15, 5 July 2017 (UTC)[reply]

Thanks for your thoughts. To repeat, this proposal has nothing at all to do with the WikiCup. It's about making a particular page in the GA system consistent with good practice, current practice, and other pages in the GA system. Josh Milburn (talk) 13:04, 5 July 2017 (UTC)[reply]

Incidentally, do you think there is a better way to word the alternative text so that it is not open to gaming? Josh Milburn (talk) 13:39, 5 July 2017 (UTC)[reply]

Maybe just return to the pre-early 2016 wording that was changed for no apparent reason that I can find? The wording was "For most reviews, the nominator is given a chance to address any issues raised by the reviewer before the article is failed." Y E ^Pacific ^Hurricane 15:16, 5 July 2017 (UTC)[reply]

I've no opposition to that. I certainly think it's preferable to the current wording. Josh Milburn (talk) 16:20, 5 July 2017 (UTC)[reply]

I think it's a judgement call. Without wishing to name names, I seem to remember TonyTheTiger had a habit of bringing half-formed articles to GA then doing all the legwork including substantial expansion during the review. Although I got annoyed with this, as I think people should bring articles to GAN in the best possible state they can up-front, it worked because I knew his track record of article writing. If a brand new editor turns up and offers a severely deficient review (see below thread), I'm less accommodating. Ultimately, I think it's a judgement call on how likely you think the article can meet the standards in the time specified. Ritchie333 ^(talk) ^(cont) 13:04, 5 July 2017 (UTC)[reply]

Support - I don't oppose failing w/o a hold period - I've done it once myself - but I do think such circumstances are rare. I do think that when done, the reviewer should provide a detailed explanation. I also think the reviewer should add the page to their watchlist and be prepared to review immediately if the changes are made in a timely manner. Argento Surfer (talk) 13:12, 5 July 2017 (UTC)[reply]

A detailed explanation is good advice, but I disagree that they should be required to add the page to their watchlist and review immediately if changes are made. Besides defeating the point of not holding it puts pressure on a volunteer workforce already lacking. We need to encourage reviewers, not add extra workload to the few who do. AIRcorn (talk) 06:32, 6 July 2017 (UTC)[reply]

Oppose as a net negative. The policy already states one can quick fail an article if it is far from meeting one of the six criteria. That is equivalent to being unlikely to meet GA criteria in the near future. Tacking on a sentence that says it's not necessary to hold when it usually is will just result in confusion for reviewers and frustration for nominators. The "positive" of quick-failing is to prevent the reviewer from needing to enumerate all the ways an article falls short if it's clear that it falls well short. There is no positive to quick-failing if you've already delivered a full review, which is why we use holds in that circumstance. There is simply no downside to saying "wait 7 days" and seeing what happens and a massive downside to quick-failing when the nominator actually does return promptly and make fixes. ~ Rob₁₃^Talk 14:14, 5 July 2017 (UTC)[reply]
Would you be opposed to changing it back to what it was early last year, as per Yellow Evan's suggestion? I'm happy with that as a compromise position. It is striking (troubling?) that this change, apparently made without discussion, has led to what many have taken as a key prescription of the GAC process. Josh Milburn (talk) 16:27, 5 July 2017 (UTC)[reply]
@J Milburn: First, we need to figure out what the positions actually are of everyone here. I think we're mostly talking past each other. I'm saying "Oppose, we should only quick fail when things are far from meeting the GA criteria". You're saying "Support, we should quick fail when things aren't likely to be fixed quickly". Ignoring the Oppose/Support, those positions sound like the same desired outcome of policy. If we're not actually fundamentally agreeing on what action is appropriate, then it's an issue of unclear wording and that's what we should focus on. Am I correct that we're basically taking the same position? ~ Rob₁₃^Talk 16:30, 5 July 2017 (UTC)[reply]
Yes, thanks, great question. Here's what I think we all agree on (all allowing the usual IAR caveat): 1) If an article meets the quick-fail criteria, it is permissible to fail it without placing it on hold and without even offering a full review. 2) If an article does not meet the quick fail criteria, it warrants a full review. 3) It is usually appropriate to place an article on hold after offering a full review to allow nominators/others to make fixes/respond before closing the review. 4) It is sometimes appropriate, after a full review, to pass an article without placing it on hold. If there is a substantive disagreement and we're not just quibbling about wording, then I think it's this: I hold (I think along with David Eppstein, Yellow Evan, Argento Surfer and Ritchie333) that it is occasionally appropriate, after a full review, to fail an article without placing it on hold (in symmetry with how it is sometimes appropriate to pass an article without placing it on hold). Some people (e.g., 13.54.152.171), if I am understanding them correctly, do not, and would rather see all articles that do not meet the quick-fail criteria and are not ready to be promoted at this time be placed on hold (IAR situations aside). There is, I think, a separate and less interesting disagreement about who is proposing that policy is changed. As I see it, we've always failed articles in this way, and it has always been OK according to at least some GAC process pages, though it has never been common (and I don't think anyone wants it to be common). Thus, anyone who thinks that we should force reviewers to put articles on hold is proposing a change. I think others see me as trying to "invent" some "new" way to fail articles. Josh Milburn (talk) 16:54, 5 July 2017 (UTC)[reply]

Symmetry? Really? Where did that come from? Would you support requiring holds on passes in order to, what, preserve "symmetry"? Obviously, symmetry between passing and failing is not required: if a reviewer wants to pass an article, that makes two editors who agree it is good. Yes, it is occasionally appropriate to fail a good article without a hold, in the case of a frivolous nomination, as specifically provided at WP:GACR#Immediate failures, not at the unfettered discretion of the reviewer, sorry, and a hold is required in all other cases; this is not the opinion of "some people" but rather our community's long-standing practice and also btw explicit P&G. 13.54.152.171 (talk) 18:58, 5 July 2017 (UTC)[reply]

With respect, you are twisting my words (I have not argued that symmetry requires anything), begging the question (I, and many others, dispute your claim that "a hold is required in all other cases") and making claims that are straightforwardly incorrect (it is not "our community's long-standing practice" that a hold is always required before a fail, and it is only explicit in WP:WIAGA [neither, incidentally, a policy nor a guideline] because someone put it there apparently without discussion, before which the page most certainly did not say that a hold was required). It's very hard for me to engage with you in a productive way while you are doing these things. Josh Milburn (talk) 19:41, 5 July 2017 (UTC)[reply]

I would disagree with "It is sometimes appropriate, after a full review, to fail an article without placing it on hold", yes. Frankly, I'd usually disagree with passing without a hold, since it tends to indicate a low-quality review. There are some exceptions – I've had at least one review quick-passed by an editor I know gives good reviews – but it's the exception, not the rule. I'm still waiting to hear the benefit of skipping the hold, and I'd need to see that before talking wording. There should be some sensible argument (even if I disagree with it) for why a change is a net benefit before we fully consider it. What's the argument? ~ Rob₁₃^Talk 04:15, 6 July 2017 (UTC)[reply]

Ok, then there is a substantive disagreement; some people believe that failing without holding can be appropriate, some people believe it can not. As for "what is the argument", I'm not really sure what you're looking for that has not already been raised. Here are a few thoughts, but mostly just repetition from above: 1) The acceptability of failing without holding was explicit in WP:WIAGA until it was changed without discussion or fanfare some months ago. 2) The acceptability of failing without holding is still suggested by WP:GAN/I. 3) Our guidelines should be descriptive, not prescriptive. Many experienced reviewers will fail without placing on hold. 4) The fact that many people are (occasionally) using this approach suggests (though I accept, does not prove) that there are at least a few cases in which it is appropriate. 5) Some articles, upon being reviewed, appear to be a very long way from meeting the GA criteria (or contain other issues) such that they are deeply unlikely (in the good-faith judgement of the reviewer) to be GA-ready in a reasonable amount of time. 6) Placing articles doomed to fail on hold may not be a particularly productive use of time, and may not be the most honest/respectful way to treat nominators. // There may be other reasons; instruction creep springs to mind, and clarifying the criteria page may help avoid future upset/tedious procedural arguments, it seems to be respectful of reviewer autonomy/judgement, there may be certain cases in which keeping a review sitting around and open would be in some way damaging (I'd rather not say more)... And so on (perhaps). Let me be clear: I do not envision my proposal as being about changing practices, and nor do I see it as being about forcing anyone to do anything. It's just about clarifying an instructional page. Josh Milburn (talk) 13:02, 6 July 2017 (UTC)[reply]

Under benefits, might you mention that fails without holds simplify the scoring of the WikiCup, which is organized in multiple rounds each with strict start and end dates, and a fail without hold avoids the problem of how to score a good article review that straddles rounds? 13.54.152.171 (talk) 18:14, 6 July 2017 (UTC)[reply]

I wouldn't personally include that, no. It's hardly a secret that I'm a big supporter of the WikiCup, but I don't really see that consideration as a particular advantage (or disadvantage), I don't think that WikiCup considerations have any place in a discussion like this, and it's not something that has affected my view of the matter. Josh Milburn (talk) 18:48, 6 July 2017 (UTC)[reply]

@J Milburn: Of the above, 1 and 2 are simply a description of what things used to be (appeal to tradition, naturalistic fallacy). 3 states that because some may break a rule or guideline, we should change it (Nirvana fallacy – all rules are broken sometimes, nothing is perfectly descriptive). 4 is a claim that the fact someone has done something means it must be positive (a rather odd twist on appeal to tradition, naturalistic fallacy – if I shoot someone in the street, it doesn't imply I was justified even though I would only shoot someone if I felt justified by self-defense, etc. You actually need a reason why it's appropriate). 5 is a statement of what is, not what should be done about it; you don't identify any benefit. And 6 is the only point that actually attempts to identify some benefit of not holding. But what time? The time lost is the 30 seconds it takes to write the hold comment and check back into the article a week from now. That outweighs the costs I detailed? Clearly no. ~ Rob₁₃^Talk 18:59, 6 July 2017 (UTC)[reply]

You're overlooking another cost of holding: The added frustration to the nominator of rushing to get through the week deadline for improvements when the reviewer wasn't going to pass the article anyway. A slow fail could be more frustrating than a quicker one, so once we reach a decision to fail we should do so expeditiously. —David Eppstein (talk) 19:10, 6 July 2017 (UTC)[reply]

David: Agreed! Rob: I reject your claims that I am engaging in all the forms of fallacious reasoning of which you have accused me; I think you were correct earlier when you said that we were talking across each other. I fear at least one of us has misunderstood the other. As such, I'm not sure anything productive will come from us continuing this back-and-forth, so I'm going to stop here. Josh Milburn (talk) 19:33, 6 July 2017 (UTC)[reply]

Support Failing without holding (which is different from quick-failing) has and should always be an option for a reviewer. The reviewers time is important too and if in their opinion the article is a long way from passing they should not be obliged to nurse a substandard article through the process. Leaving some advice and encouragement is often the kindest and best option for some nominations. AIRcorn (talk) 06:28, 6 July 2017 (UTC)[reply]
Comment - if the wording was changed without any discussion, it should be reverted back immediately. That might make the guidelines less ambiguous too. FunkMonk (talk) 13:08, 6 July 2017 (UTC)[reply]

Estimate of the frequencies of fails without holds

The prevalence of good article review failures without holds in current practice is estimated.

The good article reviews commenced in the first week of October 2016 were selected as a sample. This interval was chosen as outside the Wiki Cup.

Article	Hold?	Pass/Fail
British National Corpus	Y	Fail
California (Blink-182 album)	Y	Pass
Cliff Clinkscales	Y	Pass
Deus Ex: Human Revolution	Y	Pass
Dorchester, Dorset	N	Pass
Dual graph	N	Pass
Fortinet	Y	Pass
FUNCINPEC	N	Pass
Geography of Somerset	N	Pass
Indium	N	Pass
Madeline (video game series)	N	Fail
Metroid Dread	N	Pass
Northampton War Memorial	N	Pass
Ocepeia	N	Pass
Pop Warner	Y	Pass
SMS Erzherzog Albrecht	N	Pass
Sri Lanka
Worms Armageddon	Y	Pass

18 good article reviews were started between 1 October 2016 and 7 October 2016, inclusive. Of these, 15 passed, 2 failed, and 1 was not reviewed. One of the fails was after a hold; the other was not, but the fail outcome was arrived at by consensus of the nominator and reviewer after discussion.

7 of the completed reviews offered a hold and 10 did not. Some of the reviews (for example, Madeline (video game series) and Ocepeia) that were not formally held exhibited an "informal" hold, that is, the reviewer never assigned the article a hold status, but the nominator improved the article by addressed the review issues before the conclusion of the review. Of the 7 formal holds, 6 resulted in good articles.

13.54.152.171 (talk) 17:58, 6 July 2017 (UTC)[reply]

I'm not really sure what this proves, not least because it's a tiny sample. As far as I am aware, no one believes that failing without hold is a particularly common practice, and no one believes that it should be. What's your point? Josh Milburn (talk) 18:55, 6 July 2017 (UTC)[reply]

It is not the case that fails without holds are common practice in good article reviews. Fails are not common. Holds promote passes. A record of collaboration in promoting articles to good is clear in our community. A random sample of good article reviews is stronger evidence than "in my experience..." or "I myself have..." or the ever-popular "many editors...". 13.54.152.171 (talk) 19:54, 6 July 2017 (UTC)[reply]

Good article criteria are consistent with good article instructions

The good article criteria WP:GACR explain how to decide pass, fail, or hold, and the good article instructions WP:GAI explain the mechanics of implementing the decision. Neither the criteria nor the instructions are deficient. The instructions are clearly intended to be read once the reviewer has a decision in mind. Our good article instructions do not and need not re-iterate our good article criteria. There is no gross inconsistency that needs to be addressed here. 13.54.152.171 (talk) 18:08, 6 July 2017 (UTC)[reply]

Talk:Hungary/GA2

I am not happy with this drive-by review from an IP, plus according to the instructions only registered users can conduct reviews. What's the recommended procedure for reverting this? Ritchie333 ^(talk) ^(cont) 13:01, 5 July 2017 (UTC)[reply]

Ritchie333, we just revert it. IPs are not allowed to conduct GA reviews, so we hat the review page and undo the talk page edits, restoring the GA nominee template with the page number incremented. I'll take care of it. BlueMoonset (talk) 15:25, 5 July 2017 (UTC)[reply]

@@ Line 206: / Line 206: @@
 [[Special:Contributions/13.54.152.171|13.54.152.171]] ([[User talk:13.54.152.171|talk]]) 17:58, 6 July 2017 (UTC)
 :I'm not really sure what this proves, not least because it's a tiny sample. As far as I am aware, no one believes that failing without hold is a particularly common practice, and no one believes that it should be. What's your point? [[User:J Milburn|Josh Milburn]] ([[User talk:J Milburn|talk]]) 18:55, 6 July 2017 (UTC)
+::It is not the case that fails without holds are common practice in good article reviews. Fails are not common. Holds promote passes. A record of ''collaboration'' in promoting articles to good is clear in our community. A random sample of good article reviews is stronger evidence than "in my experience..." or "I myself have..." or the ever-popular "many editors...". [[Special:Contributions/13.54.152.171|13.54.152.171]] ([[User talk:13.54.152.171|talk]]) 19:54, 6 July 2017 (UTC)
 === Good article criteria are consistent with good article instructions ===