Jump to content

Wikipedia talk:Content assessment/Archive 5

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1←Archive 3Archive 4Archive 5Archive 6Archive 7→Archive 9

Assessment questions: Disambig and china C

  • What assesment for quality and importance should disambiguation pages have?
  • Is is okay for the China banner to lack ability to display the C-quality class?

--Kiyarrlls-talk 22:44, 1 January 2009 (UTC)

As a general rule, Disambiguation pages are not given 'standard' 1.0 quality ratings. Many projects maintain a 'Disambig-Class' to keep track of these: see Category:Disambig-Class articles. Generally their importance is set to "NA". Vis your other question, it's not really a question of whether it's "okay". If there are good reasons why WikiProject China doesn't want to use C-Class, then they are at perfect liberty to do whatever they like with their banner to support that. However, the most common reason for a project not "using" C-Class is because they judge the effort required to reassess all their B-Class and Start-Class articles is too great. The most common result, then, is that they set their WikiProject banner to allow C-Class and just ignore it thereafter. So If other editors come along and reassess an article as C-Class, that's fine; no effort spent, some benefit still gained. What is not 'okay', however, is for the banner to handle C-Class wierdly; remember that some editors will try to rate China articles as C-Class, and the banner needs to handle these sensibly. The only really 'okay' solution is for the banner to treat C-Class as Start-Class; marking them as 'unassessed' wastes that reviewer's time as they try to work out why C-Class isn't working for the China banner. Even worse is where the banner displays "this article has been rated as C-Class" but categorises into "Category:Unassessed Foo articles", which is actively misleading. So as long as the banner handles |class=C in a sensible fashion, you don't need to get an 'ok' from anyone else. However, there is still the question of why you don't want the banner to handle C-Class. Why is that? Happy‑melon 23:04, 1 January 2009 (UTC)
Happymelon! thanks a lot! Really helpful answer. I will now rate 'Phags-pa as NA.
It's not that I don't want the banner to handle C-class, I understand that so far China wikiproject banner does not handle C-class. I tried to rate an article as C-class once, was surprised when it didn't work, left the article as start class instead. In fact C does not exist in the China assessment table also. I attempted this just now just to verify and it says "C is an invalid class"
--Kiyarrlls-talk 01:27, 2 January 2009 (UTC)

Needed class?

In the assessment scale, there is a Needed class. I wonder how a page can be tagged as Needed. A talk page must exist in order that such an assessment can be tagged on to the page. For the talk page to exist, the main page must exist - otherwise it will be deleted. So, the page must already exist!

  • Am I missing something here?
  • Or is this a hangover from some past practice?
  • Or is there some other method by which a Needed article is included in a Needed-Class xxx category?

If my understanding is correct, about 102 always-empty categories can be eliminated. Also all the appropriate inappropriate verbiage in assessment related pages can go too, I guess. VasuVR (talk, contribs) 16:52, 7 January 2009 (UTC)

I would guess that it dates from a time when it was acceptable to have a talk page without an article. Most subcategories of Category:Needed-Class articles seem to be empty, though see Category:Needed-Class Anthroponymy articles for some legitimate usage. PC78 (talk) 17:57, 7 January 2009 (UTC)
Please don't delete it. A few projects use this system for articles they are creating, and they will scream if the category is deleted. There is a long history of using this class. Walkerma (talk) 18:58, 7 January 2009 (UTC)
Pages tagged with {{Needed-Class}} are an exception to CSD G8, under the "useful to the project" clause. Titoxd(?!? - cool stuff) 22:32, 7 January 2009 (UTC)
Also, often they will be used with a redirect page, where the article currently redirects to a related topic but deserves its own article eventually. I actually think it is underused, personally, but definitely useful and yes, people will throw a fit if it goes away.-RunningOnBrains 02:02, 8 January 2009 (UTC)

Thank you for the information and history. Sure, then we don't need to eliminate them (no screaming, throwing fit and all that). If it is underused, then it is going to be empty for 90+ % of cases. Many projects have their To Do sections, which cover such situations (that is requests for articles). VasuVR (talk, contribs) 08:05, 8 January 2009 (UTC)

set index

Perhaps Set index should be added to the Wikipedia:Version 1.0 Editorial Team/Assessment#Non-standard grades section? OlEnglish (talk) 02:34, 12 January 2009 (UTC)

I'd mark it as List-Class and be done with it. There aren't really enough of them IMO to justify a new class. Happy‑melon 10:21, 12 January 2009 (UTC)

Some WikiProjects & Rating

Just a datum to help the rest of you usefully discuss an efficient way to improve things: some WikiProjects are too small to really support a meaningful rating system beyond B-class. At this moment, yours truly is the whole active membership Wikiproject Ethiopia, & due to Real Life(tm) committments I can only participate part-time in editting & rating. Therefore, my priority at this moment is trying to improve stubs into start class (or better) articles -- one way I can contribute without the benefit of a community. (This also means not creating any new articles if I can help it.) That is the primary reason why there won't be any "C-class" articles in the WP:ETHIOPIA space: I would rather make articles "B-class" than to repeat article evaluations. So it's more than likely that there are more "A-class" & "GA-class" articles in Wikipedia than reported: there are simply not enough qualified people to either judge them or improve them to the appropriate level, so some topics will be overlooked. And I don't consider this necessarily a bad thing in itself, nor something that can be easily fixed; it's just something that needs to be remembered when proposing new solutions. -- llywrch (talk) 18:53, 9 February 2009 (UTC)

Thanks; I think that smaller projects which can't maintain a full A-Class review system will be considered extensively in any discussions... one option that I can think of is to have more active, but related, projects do the reviews. -Drilnoth (talk) 18:58, 9 February 2009 (UTC)
like, Wikiproject Africa, in this example? (I want to check that I have this straight.) ~user:orngjce223 how am I typing? 00:16, 11 February 2009 (UTC)
I think that's the idea, except I doubt they'd be able to handle it. They've got a lot of members who signed up, but a lot of them did so quite a while ago, and they're talk page only needs two archives. Looking at their Wikipedia:Africa-related regional notice board/Peer review page, the first response is from a milhist editor. - Peregrine Fisher (talk) (contribs) 00:37, 11 February 2009 (UTC)
The problem with looking for help from another project is even if it's on a related subject, you just miss out on the knowledge & confidence to make useful decisions. In my case, while someone in WikiProject Africa might be a good fit for suggestions with Ethiopia subjects, their expertise is about as relevant as someone from WikiProject:Poland helping with WikiProject:England articles. (And I'll admit everything I've written about Ethiopia I've learned as I write Wikipedia articles; the more I learn about this subject, the more I learn what I don't know.) One weakness I've noticed about the Wikipedia community is the lack of confident generalists: someone who can, say, look at a biography of an Ethiopian politician consider its subject matter as an Ethiopian politician rather than an Ethiopian politician. These kinds of people are far more useful to those of us dealing in areas with a lack of peers than related WikiProjects. I suspect this is why I've found, over years of participating off-&-on at Wikipedia:Peer review, that many commentators tend to focus more on trivial things like mangled sentences or copy editting, rather than considering what an article needs to cover, & how its subject should be covered; it takes probably as much skill to grade an article as A class or a Featured Article as it takes to write one. This is not a common skill (I would be clueless about reviewing articles on chemistry, for example), but the larger a WikiProject is the larger the chance is these people will be available to that WikiProject. -- llywrch (talk) 07:07, 12 February 2009 (UTC)

Grade deflation

Appartement du roi was mentioned here a few days ago as "a lot of good content but currently backed by few inlines". In fact (to a non-specialist), it seems to be as complete as one would wish for an encyclopedic article on the subject, and is fully referenced. WikiProject Architecture had rated it A-class, no doubt for those reasons. However, it is far from the prettiest article you will find on WP (it has no section headings, for example) and WikiProject France had rated it B-class, probably for that reason. I'd agree with them (WP:FRANCE), but it shouldn't take much to fix up the worst of the style problems, and I might actually go and do it myself!

The day after it was mentioned on this page, another editor came along and changed both assessments to C-class! Fair enough, we're each entitled to our opinions, but if such practice became generalized it would be very dangerous for the encyclopedia.

As far as I'm aware, we haven't formally changed our asessment criteria on WP:CHEMS since 2005. An article which was pretty much comprehensive (ie, A-class) in 2005 is probably still pretty much comprehensive in 2009, give and take a few recent developments and of course the effect of "unhelpful" editors. Other areas of Wikipedia certainly have changed their criteria since 2005: the MoS, for one, is much longer and more intricate than it was four years ago! This is a classic example of Parkinson's law: "Work expands to full the time available for its completion".

All the time we spend on ensuring compliance with ever more stringeant MoS compliance is time that we're not spending on encyclopedic content, either in article improvement or article assessment. If article assessment grades are tied to the same level of content but an ever more stringeant MoS compliance, we have grade deflation, rather than the grade inflation which people tend to worry about. This would be thoroughly pernicious, because it means that we would be giving ever less recognition to those editors who improve our encyclopedic content. Are we really saying that content is ever less important to our encyclopedia? Physchim62 (talk) 16:18, 15 February 2009 (UTC)

When I assessed that, I took one look at it, and as it was so convoluted with no section headers and a {{morefootnotes}} tag that I just went the the talk page and assessed it as C. Is a really-hard-to-read-article really helpful to the reader (and therefore deserving of even a B?). I didn't think so...but maybe I assessed it too fast. :) I'd assess it as B now. —Ed 17 (Talk / Contribs) 17:38, 15 February 2009 (UTC)
Hmm, the article had no section headings, but it was divided into easily identifiable sections. And was it really that hard to read? It only has twenty paragraphs of text, several of those paragraphs being quite short. It's not any longer in text than a medium-size newspaper article, which wouldn't usually have section headings either. I think that anyone who was looking for information could have found it (if it were in there, of course).
The point of my comment is not to criticise your assessment in particular, but to show the general perils of assessing primarily by style rather than by substance. I improved the article in about half an hour (and most of the improvement in the first five minutes) without knowing anything about the subject. The only 'specialist' knowledge I used was to correct a typo in French! I couldn't have done that had the encyclopedic content not been there in the first place. Who deserves more credit for the current state of the article – myself or the original contributors? Physchim62 (talk) 18:29, 15 February 2009 (UTC)

French Wikipedia grading JS

I was wondering how those talks were going about importing the assessor script that the French Wikipedia uses. Thanks, §hepTalk 02:27, 18 February 2009 (UTC)

If I recall correctly, I was told by User:Kelson that it wouldn't do what the old Outriggr script can do, and there wouldn't be anything that Pyrospirit's script doesn't already do. I plan to talk with Kelson on the phone next week, and ask him about the French 1.0 project, so I'll ask him then about what scripts they are using now - they may have something new. Walkerma (talk) 05:15, 18 February 2009 (UTC)
Oh. That's what I get for reading French a la Google. Thanks, §hepTalk 05:17, 18 February 2009 (UTC)

A-Class

Should A-Class officially be placed below GA-Class? It looks right now like A-Class articles can already be GAs, but they don't have to be, so it's kind of inconsistent. Perhaps a reordering to Stub, Start, C, B, A, GA, FA would be good at this point.

Also, maybe an official system for gaining A-Class assessments should be put into place (similar to the GA system, but a bit more open). Most WikiProjects right now either don't have an A-Class or their reviews to make articles A-Class are inactive... even WP:BIOG's A-Class review department is inactive. Thoughts? -Drilnoth (talk) 20:29, 3 February 2009 (UTC)

WP:MILHIST's A-Class reviews are super-active. And not only do we consider A-Class higher than GA, we use the FA criteria, not GA critiera so a downgrade such as this would be completely against how we have set-up our review department. See this: Wikipedia:WikiProject Military history/Assessment/A-class FAQ, does that look like it is less than GA, it is considerably more stringent. -MBK004 20:40, 3 February 2009 (UTC)
Concur with MBK. This is an instance in which the bar varies, and as such each of the A-class assessments need to be looked at in context of the project running them. Ours considers A-class vastly superior to GA-class, so much so that we actually encourage our members to bypass the GA-class assessment and move straight to A-class. Ranking A-class below GA-class would be slap in the face to milhist. TomStar81 (Talk) 21:27, 3 February 2009 (UTC)
(ec) Good point, although it would seem that that is more of an exception rather than a rule... Template:Grading scheme states that A-Class articles don't have to even be GA. I think that what MILHIST has is great, and I would be completely opposed to any downgrade of that, but for the majority of projects A-Class is used much less than any other assessments. However, that means that there is a big difference between the A-Class articles of one project and another project, and this should be standardized. I propose that A-Class be put directly below GA-Class, and maybe MILHIST could have a GA+-Class, because the articles that it currently has at A-Class are better than GAs (having undergone a more thorough review), but not quite FAs (if only in name). -Drilnoth (talk) 21:32, 3 February 2009 (UTC)
Or, that A-Class be standardized as better than GA, since that's what it's universally supposed to be, even if some people get it wrong.  :) Might just have to formalize a set of standards (like, this is how you get it better than a GA, to be used as a springboard for FA), allowing other interested projects to tack specifics onto that as needed. BOZ (talk) 21:44, 3 February 2009 (UTC)
Responding to BOZ: I agree that A should probably be between GA and FA, although Template:Grading scheme says that A-Class articles don't have to be GA, even though they're listed higher on the chart.
General note: An alternative might be to take the MILHIST A-Class assessment system and standardize it across all projects so that it is a solid "middle point" between GA and FA, open to all articles. -Drilnoth (talk) 21:46, 3 February 2009 (UTC)
(e/c) It's also true that GAs don't need to be A-Class. :) I concur with the other comments above; certainly at WP:FILM we treat A-Class as superior to GA-Class. Unless A-Class is actively being used by a majority of projects for sub-GA articles (and I don't believe that is the case), then I see no reason to change the long-standing status quo. PC78 (talk) 21:48, 3 February 2009 (UTC)
Maybe the 1.0 assessment team needs to open an RFC on this to formalize it like you did with C-class, otherwise this problem is going to persist. TomStar81 (Talk) 21:49, 3 February 2009 (UTC)
I support that. :) BOZ (talk) 21:51, 3 February 2009 (UTC)
(ec) Sounds good. I'll look into opening an RFC; my personal !vote will be to keep A-Class between GA and FA, but implementing a formal review system across all projects. -Drilnoth (talk) 21:51, 3 February 2009 (UTC)
Same here, as I have said above. BOZ (talk) 21:56, 3 February 2009 (UTC)
(e/c) Hold your horses for a sec - why a whole new review system? WP:FAC and WP:GAN are already totally backlogged becuase of a lack of reviewers. While your idea is good in theory, I don't think that another review system would work without a massive inflow of reviewers... —Ed 17 (Talk / Contribs) 21:57, 3 February 2009 (UTC)
If you open on rfc on this it would be a good idea to note that the 1.0 team did adopt milhist B-class standards around the time of the C-class referendum, so there is president for this sort of thing. TomStar81 (Talk) 21:59, 3 February 2009 (UTC)
To Ed: Thats the point. RFC doesn't formailze anything, it merely seeks input from contributers on a given issue. TomStar81 (Talk) 22:00, 3 February 2009 (UTC)
To Ed: Presumably, creating an A-Class review could remove some of the burden from the GAC and FAC backlogs. -Drilnoth (talk) 22:01, 3 February 2009 (UTC)
(@ Drilnoth) - I think that I will state this below too, but taking more reviewers away from FAC is not going to help the problem, IMHO...sure, right away there will be less, but as time goes by there will be the same # of articles at FAC as there is now, with less reviewers reviewing them. —Ed 17 (Talk / Contribs) 22:08, 3 February 2009 (UTC)

In some projects with no formal A-class review, then the mean and median of As is sometimes worse than GAs. But not in this WikiProject. Having said that with the evolution of standards, some FA/A relics from 2006 and before are worse than a lot of GAs, which are more fluid, so that archaic GAs are booted off more quickly. YellowMonkey (click here to vote for world cycling's #1 model!) 23:13, 3 February 2009 (UTC)

Can't A class be placed below GA, and the letter simply swapped. So all current GAs would become A class and vice versa? That would solve the confuing naming system of the moment, and could be done by bots.Yobmod (talk) 13:17, 5 February 2009 (UTC)
This overlooks the review structure already in place. All A-class articles in milhist can be GAs, but not all milhist GAs can automoatically b A-class. TomStar81 (Talk) 16:45, 5 February 2009 (UTC)

Can we maybe a get a link to this up on the watchlist page? The more input we get here the better prepared we will be to arrive at a discion on this matter, whatever that may eventually be. TomStar81 (Talk) 16:45, 5 February 2009 (UTC)

I don't think that's the case. Where was there a decree saying that A-Class articles are automatically GA. It doesn't hold for FA, if a FA is binned then it drops back to B. YellowMonkey (click here to vote for world cycling's #1 model!) 00:10, 6 February 2009 (UTC)
I think it is a pity that this happens: we should really track WikiProject ratings separately from GA and FA, so that a removed FA or delisted GA retains a WikiProject A-Class rating if it had one. The current logical structure is something like
Stub-Start- C - B - A
                 GA - FA
However, the relationship between A-Class and FA, and B-Class and GA is not a straightforward one, because the community review processes (GA, PR, FA) have a different focus to WikiProjects. Geometry guy 21:31, 6 February 2009 (UTC)

Removal of A-Class or its merger into GA

I have just tried to read this. Has there been a proposal made to remove A-Class or merge it into GA as a default somewhere? (Is this one of those perennial proposals then). Can someone point me to it if there has been. I just think that with PR, GA and FA all needing reviewers something has to give. If we remove the A-class as such, there might be more impetus on a project working those articles for FAC instead (as birds, dinos and fungi did/do) rather than keeping it in-house. Striking while the iron is hot and having everyone around, FA seems a small step beyond A-class, so maybe it is time to do away with A-class review. Casliber (talk · contribs) 00:47, 3 March 2009 (UTC)

We're discussing the role of A-Class right now over here. We're having a broader discussion about the need and role of A-Class in general, perhaps revamping the system of how A-Class operates (there isn't consensus support for abolishing it). Please join us! Walkerma (talk) 00:53, 3 March 2009 (UTC)
Ahaa, thanks. I knew it'd be somewhere...Casliber (talk · contribs) 10:45, 3 March 2009 (UTC)

Request for Comment regarding A-Class assessments

After the above discussion started, it has been determined that an RFC is needed. The question at hand is: How should A-Class assessments and reviews be handled? Template:Grading scheme states that A-Class articles do not need to be GA, but both the WP:MILHIST and WP:FILM WikiProjects have A-Class review systems placing them above GAs. Additionally, many WikiProjects don't use A-Class at all at this time.

Options for what should be done include: leaving A-Class as it is, up to each project separately; standardizing A-Class below GA-Class, and standardizing A-Class between GA and FA-Class. Also up for discussion is the possibility of making A-Class reviews project wide, rather than being covered by each WikiProject separately.

There is precedent for adopting MILHIST-style review systems because the currently-used B-Class review system (the six-point checklist) was implemented after MILHIST had already been using it.

Discussion

  • Put A-Class in between GA and FA and implement a Wikipedia-wide A-Class review system per my comments above. -Drilnoth (talk) 22:02, 3 February 2009 (UTC)
  • Concur This would satandardize the critera, but allow the project to adopt them when and if the projects decide to open A-class review department. TomStar81 (Talk) 22:05, 3 February 2009 (UTC)
  • Isn't this a bit premature coming off the back of a 90 minute discussion? A-Class is already between GA and FA; are we discussing keeping things as they already are? The current system seems perfectly fine. The {{grading scheme}} and Wikipedia:Version 1.0 Editorial Team/Assessment/A-Class criteria seems quite clear. PC78 (talk) 22:09, 3 February 2009 (UTC)
    • Although the table at {{Grading scheme}} shows A-Class as being in between the two, it says in the description that "Good article status is not a requirement for A-Class. " -Drilnoth (talk) 22:12, 3 February 2009 (UTC)
      • As I said above, the reverse is also true; A-Class is not a requirement for GA status. PC78 (talk) 22:14, 3 February 2009 (UTC)
        • No class is a "requirement" for any other class, although there is a logical order of progresstion that each article makes. Right now, A-Class doesn't have an exact position in that order, being kind of tied with GA and alternately being considered higher and lower than it. -Drilnoth (talk) 22:18, 3 February 2009 (UTC)
          • Of course it has an exact position is the current order, it's there in colour in {{grading scheme}}! You want to formalise something that already exists and has done for quite some time? I'm sorry, but I find this whole discussion quite pointless. PC78 (talk) 22:52, 3 February 2009 (UTC)
  • Support standardizing A between FA and GA, though that might already be the case. Opposed to a project-wide A-class review system, as less editors at WP:FAC and WP:GAN is the absolute last thing we need. In the beginning there will be less articles at FAC and GAN, and it will look like a good idea, but as articles start going through, FAC will fill up again - and there will be less reviewers to deal with them. Oppose forcing GA to be a requirement for A, unless we want to say that articles on certain topics (MILHIST, FILM, basically only topics that are covered by an active WikiProject's review system) can skip GAN for A as long as they go through the review system of the relevant project? (sounds creepish to me, but...throwing it out there.) —Ed 17 (Talk / Contribs) 22:21, 3 February 2009 (UTC)
    • Comment I'm also opposed to making GA-Class a requirement for A-Class if there is a Wikipedia-wide review system, although if there isn't such a system then I think that GA should either be made a requirement or A-Class should be standardized as lower than GA. -Drilnoth (talk) 22:28, 3 February 2009 (UTC)
      • That's why I put that last sentence in. Though it sounds creep-ish, I think that it would not be fair to the milhist/film editors to force them to WP:GAN and wait a month for a review...especially when milhist's FAC-style reviews (a) are shorter (about two weeks, give or take; also, you normally get initial comments about a day or two after nominating) and (b) are, IMO, far better than GAN's one person-style...not in the least because of the additional feedback you get from having people (plural) review it. —Ed 17 (Talk / Contribs) 22:39, 3 February 2009 (UTC)
  • A class higher than GA Good articles aren't evne that good, A class articles might as well be featured.--Pattont/c 22:30, 3 February 2009 (UTC)
  • Support formalizing A-class being placed between FA and GA, with a review system indicating that the article must at minimum meet the GA requirements, as well as any requirements dictated by type of article, and/or applicable WikiProject; reviews should be handled by those familiar with the topic, such as WikiProject members, and not someone who has done significant work on the article. An article should not be required to have a GA review to make A, but should meet all the GA requirements. BOZ (talk) 22:34, 3 February 2009 (UTC)
  • Ugh, where did the !votes come from? This is a request for comment, not a request for hit-and-run one-line opinions. I think many people misunderstand the intention of the grading scheme, and the distinction between a Good Article and GA-Class. GA-Class is an expression of quality, while the Good Article Assessment process is a way of identifying that quality. An article can be the quality of a GA without being a GA, simply by virtue of not having been through the formal GAC process. As such, there is absolutely no inconsistency in saying that A-Class articles need not be Good Articles; they need to be of the quality of a GA, indeed higher in general, but there is no need to go through the extra step of a formal GAC. The situation is entirely analogous to Featured Articles: many if not most of them have never been assessed by GAC, should we then say that there is inconsistency in that situation? Of course not, and it's the same with A-Class. An article that could not pass the GAC criteria should not be an A-Class article, but any article that would meet the A-Class criteria should breeze through GAC, although there is no requirement that they do so. As such, the linearity of the scale is quite clear in my mind, as B→GA→A→FA. Happy‑melon 23:23, 3 February 2009 (UTC)
  • I think that there's a fundamental misunderstanding about how A-class assessments work for MilHist and Film. At least in these two projects, the review process is fundamentally designed to prioritize content issues over style, MOS, and other site-wide guidelines, which tend to be better addressed by PR, GAN, and FAC. The idea is that the WikiProject review is more likely to be used by editors with expertise in the subject matter being reviewed (as well as particular project-level guidelines), and thus the majority of the focus is on the text itself. Reviewers at site-wide processes cannot be assumed to have specialist knowledge of any given material, and thus may support in good faith articles which have major content issues that may not be apparent to a layperson. (Case in point, I just found several in the Star Trek VI article currently at FAC, which did not go through our review process.) The idea is to help editors vet their articles for content before FAC, so as to help ensure that the article will pass smoothly and with a minimum of objections. If editors choose to pass over that to go directly to FAC, it's their (potential) loss - no different than with PR or GAN. Girolamo Savonarola (talk) 23:56, 3 February 2009 (UTC)
  • Comments (since this is a RfC!):
    • I have to concur strongly with Girolamo and H-M here. A is not better than GA, nor is GA better than A. Which is better, a complete article with a few style problems (A), or an article that has all the style issues correct which may be missing a few important pieces of content (GA)? This system originated at WP:CHEMS; could a typical GA reviewer or FAC review really judge whether or not an article like 2-Pyridone was complete or not? It's true that a complete article usually has most of the style issues addressed, and a GA is often fairly complete, but it's very possible for an A-Class article to fail GA. So I'd oppose requiring GA as a prerequisite for A, and I'd also oppose A as a prerequisite for GA or FA, though there are clearly benefits where such a sequence is followed. I wish there were a nice way to display the table without it appearing that there is a GA to A to FA sequence! As Roger Davies put it very nicely last year:
      1. B-Class - assessment by a single editor representing the project
      2. GA - assessment by a single editor representing the community
      3. A-Class - assesssment by a team of editors representing the project
      4. FA - assessment by a team of editors representing the community
    • What I sense behind all this is that we'd like to get A-Class to work more effectively - something we can all agree on. Some argue that we should require a strict peer review like the one that works so well at WP:MILHIST and a few large, big projects; others say that such a system couldn't survive at many small projects. How would they deal with reviewing/delisting existing As? I think what we need to do is to come up with a peer review system for A-Class that is scalable to small, sporadically active projects (even the A-Class review at WP:CHEMS has got pretty inactive, but we are quite small). That might seem impossible, but WP has a history of doing the impossible, and it would nail this interminable debate once and for all. So - does anyone have any clever ideas on how to move us towards a workable A-Class peer review system for small WikiProjects? Walkerma (talk) 03:52, 4 February 2009 (UTC)
      • I hadn't really though of it that way. I'll try thinking about possibilities for A-Class reviews for small projects. -Drilnoth (talk) 03:56, 4 February 2009 (UTC)
      • Well, the main problem with small projects having A-Class reviews is that they, for the most part, don't have the manpower to perform them properly. So a potential solution is to have several projects "pool" together to perform A-Class reviews of articles that come under their combined scope. An example would be having WikiProject U.S. Roads assessing articles that fall under the several state route projects (already done), or WikiProject Tropical cyclones assessing articles that fall under WikiProject Meteorology (not done). This would take advantage of what are essentially overlapping editor pools. So essentially, we wouldn't have one project grading another, but content-centric review centers that have authority voluntarily delegated by its "member" (for lack of another word) WikiProjects. So we would have Wikipedia:Content review/Roads and Wikipedia:Content review/Weather for the two examples above, and these pages would not concern themselves with style issues at all. That's one solution that might be able to pull it off. Titoxd(?!? - cool stuff) 04:22, 4 February 2009 (UTC)
      • I've always been under the impression that larger projects were the only projects responsible for the implementation and continuation of A-class, and while it may be true that smaller projects may lack the ability to effectively run an A-class department most of those projects are as noted above part of a larger project that could open an A-class department. There is one other option: Owing to the scope of certain projects any number of other projects could petition to go hand in hand for the creation of a task force in which case the smaller project would gain access to a larger project that has the capacity to run an A-class department. For example, a project on medicine could opt to go either to milhist (MASH units, comabt medics and such) or to films (medical movies, instructional videos, and such) with a task force proposal and then if the latter project is open to sharing reach an agreement wherein articles from the former project go through the latter project's A-class review process. The former project's editors can comment, the latter projects editors can close. A similar agreement of sorts is already in force at milhist for partner peer reviews, and no complaints have been received. TomStar81 (Talk) 05:48, 4 February 2009 (UTC)
  • I'm not a fan of A-class. It's too uneven, and shouldn't be used for 1.0 assessment. Most projects don't use it anyways. Now isn't going to be the time that this happens, but we need a number scale. No one would go for an F-class instead of a stub-class, but becuase we have C and B classes, it seem obvious that we should have an A-class. It's a left-over because we didn't think ahead. Stub = 1, Start =2,..., GA = 5, FA = 6. If some project wants high standards, comply with the style guides and go for FA. - Peregrine Fisher (talk) (contribs) 04:32, 4 February 2009 (UTC)

Complete, total, and utter support for the buddy system. I thought there were only a few projects which do a project Peer Review A-class review, but as I look through Category:WikiProject peer reviews, this search I see a butt-load. Granted, many of these aren't very active, or haven't been active at all for a while, but it shows the interest is there. The question is how to promote such a system...I was thinking it might be best to set up a subpage, here or at Peer Review, for WikiProjects seeking a Peer Review A-class review partner. We could then mass-message all the WikiProjects, seeing if they'd be interested in partnering up. They wouldn't even necessarily have to be close in scope (although that would be desirable), since a fresh set of eyes from a non-specialist is always great. Or we could set up some sort of system where we construct sort of a WikiProject Tree, where we trace all the listed "Parent WikiProjects" upwards and worked towards setting up peer reviews there. Anyway, I'd be in favor of almost any system that gets more projects into Peer Reviewing doing A-class review for their own articles.-RunningOnBrains 15:51, 4 February 2009 (UTC)

Fair enough...I thought they really were one and the same (I typically think of Peer Review as the step between GA and FA), but I suppose not. Doing a search still reveals a lot of projects doing/attempting a legitimate A-class review than I thought were doing it, some more active than others.-RunningOnBrains 17:12, 4 February 2009 (UTC)
On a related note, would anyone be opposed to me creating and populating a Category:WikiProject A-Class Review?-RunningOnBrains 17:14, 4 February 2009 (UTC)
Go ahead; that sounds like a useful cat. -Drilnoth (talk) 17:20, 4 February 2009 (UTC)


Most of the articles i work on are in projects that do not have A class reviewing systems in place (and they are not small projects), and it really is a bit annoying to find that writing a GA has to be followed up by months of work to get to FA, if it's possible at all, with little or no input from outside reviewers (peer reviews are too variable). Hence i would find a site wide A-class review very helpful. Most of the FAC reviews i read seem to be held up by very simple things, and last a long time while editors make obvious fixes, so an official pre-FA class review could take a lot of the work load from there.
I don't mind if it is only recommended before FA but not compulsory, but something to help editors out who are not interested in military history or spiders would be great, preferably something more structured than peer review, and that takes note of things GA does not ask for as well as having generally higher standards. Eg Formatting refs.
Note that if A class focussed on content, and left GAR to review style, then most articles that passed GA and A would still not be at FA level. MoS and prose requirments of GA are nowhere near FAYobmod (talk) 13:09, 5 February 2009 (UTC)
a 2D scale

I've always thought of it as a two-dimensional space, with "quality" (maybe "style" would be a better term) and "content" the two axes; like the image to the right. Articles usually start in the bottom-left and (hopefully) end up in the top-right. But the assessment grade boundaries aren't straight: some areas prioritise improvements in quality, while others focus on content improvement. So for instance an article does not need to increase significantly in "style" to move from GA-Class to A-Class, but it does need to increase significantly in "content". The reverse is true for B to GA. That's just a rough sketch: some things are still significantly misplaced (C-Class article with no content at all?) and I don't really think the square box is a good model for the limits, but you get the general idea. The important thing is that there's no loss of linearity: you can't have an article that continues to improve get to A-Class without meeting the GA-Class requirements. Happy‑melon 17:09, 5 February 2009 (UTC)

I have been trying to find out which projects have active A-Class assessments. How can I find out which projects do so.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTM) 23:24, 9 February 2009 (UTC)
I think that most of the active and semi-active ones are listed in my user sandbox (permalink). -Drilnoth (talk) 23:46, 9 February 2009 (UTC)

I suggest that projects be able to award A-class as they see fit, but that it should take at least two project editors' agreement that an article has reached A-class to so designate it. -- Ssilvers (talk) 06:21, 11 February 2009 (UTC)

Proposal

As a possible method to implement A-Class as a "content, rather than style," while being on the same "level" as GA-Class in regards to articles quality, I have a proposal.

All WikiProjects large enough to handle it have an A-Class review department, as do all of the projects which are "top" in the hierarchy (that is, all "parent" projects). Whenever an A-Class review is requested for an article that doesn't have a project associated with it that does A-Class reviews, the parent projects look at it instead... e.g., WP:RPG articles would probably go to WP:BTG. This way, all articles will be able to recieve A-Class reviews.

The exact method by which A-Class reviews would work is to be determined... certainly a group effort, rather than one person, and probably modelled after the MILHIST system.

Regardless, GA and A-Classes are fully separated: GA evaluates style only, with a bare minimum requirement for content. A-Class is the opposite, focusing almost entirely on an article's content. This way, the two assessments are separate but compatible and equal with one another.

Once an article has passed both GA and A-Class reviews, I see there as being two options, since neither A-Class or GA-Class assessments would work for such an article. The first option would be to create something like an "Excellent Article"-Class, which isn't quite featured but has passed both of the lesser reviews. The other option would be to actually do away with the entire FA review system and say that, once any article has passed both GA and A-Class reviews, it becomes Featured. This would probably lead to an increase in total FAs, but it would also take some of the strain off of that part of the review system so that more effort could go into GA and A-Class reviews. This latter would be my personal preference, although I understand that it would be a rather drastic change.

If there is any support for this type of idea, I can write up an official proposal. -Drilnoth (talk) 14:35, 4 February 2009 (UTC)

It isn't quite accurate to say that A-class (at Milhist at least) focuses almost entirely on content. Ours has five criteria, super-versions of the B-Class criteria, and the article has to pass each of them. While there is considerable focus on content, Milhist A-class is seen as the last step before FAC and is a way of certifying that the article is (nearly) FAC ready. For instance, articles are frequently failed because they need copy-editing. Please see the Milhist A-class criteria and support Milhist A-class FAQ. --ROGER DAVIES talk 06:06, 5 February 2009 (UTC)
That sounds like the kind of thing that doesn't need to go into a sequential ranking of articles. You could use a category to say it's ready for FAC. - Peregrine Fisher (talk) (contribs) 06:08, 5 February 2009 (UTC)
It does really need to go in, because Milhist A-class is practically FAC and GA isn't. Having it set up like this is probably the single biggest reason why Milhist has so many FAs (about 12 a month) and why the Milhist pass-first-time rate at FAC is so high. --ROGER DAVIES talk 06:16, 5 February 2009 (UTC)
But why, if it's only used by a small segment of wikipedia? Any article could have its class changed to A, but apparently it's only on a GA/FA level with a few projects. Doesn't sound like a good sitewide class. - Peregrine Fisher (talk)(contribs) 06:24, 5 February 2009 (UTC)
The simple answer is that Milhist A-class is demonstrably a proven method of getting articles to FA and therefore its wider adoption can only improve standards all round. Also, FA participation only involves a small segment but that doesn't stop us pushing it as an ideal :) --ROGER DAVIES talk 06:31, 5 February 2009 (UTC)
Well, there are two ways to go, as I see it. Either the whole site treats A class like milhist, or milhist should devolop something that isn't messing up the classes. I doubt we can change the whole wiki, although changing milhist isn't going to be easy either. I know they like it, and aren't concerned with how other A class articles are. - Peregrine Fisher (talk) (contribs) 06:36, 5 February 2009 (UTC)
Well WP:AUS has 120 FAs but might not be able to sustain a A-class review, and most other WikiProjects have less than 20 FAs, and those projects usually only have 2-3 guys who bother with full/polished articles so that would be useless. I can't see what the fuss is with all these A-GA-B-C class is when a lot of relic FAs are so out of date that they don't even pass modern GA. YellowMonkey (click here to vote for world cycling's #1 model!) 00:21, 6 February 2009 (UTC)
The suggestion that we should dumb down Milhist A-class for the dubious benefit of tidiness won't go down particularly well with the members :) What happened to inspirational leadership by the way? --ROGER DAVIES talk 06:43, 5 February 2009 (UTC)
I don't think you should dumb it down. I just don't think it should be part of sitewide assessment. - Peregrine Fisher (talk) (contribs) 06:47, 5 February 2009 (UTC)
Well, there we differ. My own view is the gulf between GA and FA is far too great for all practical purposes and, unless GA is completely redesigned, will always remain thus. --ROGER DAVIES talk 06:54, 5 February 2009 (UTC)
It's cool. We're not going to fix assessment any time soon. - Peregrine Fisher (talk) (contribs) 07:03, 5 February 2009 (UTC)
!? But that's exactly what you're talking about doing :) Fiddling with hierarchies and so forth. --ROGER DAVIES talk 07:07, 5 February 2009 (UTC)
We're talking mupltiple RfCs and watchlist notices. Look at what C class took, and this is probably more controversial. - Peregrine Fisher (talk) (contribs) 07:10, 5 February 2009 (UTC)
I think the basic idea here sounds great. I think it needs a couple of things - first, I think it should be raised over at WT:COUNCIL, since they tend to oversee coordination between WikiProjects. We'd probably need to sort out how the hierarchies of projects will work. One stumbling block is that often you will have an inactive parent (such as the US project) with very active projects under it (Texas, Dallas, Chicago, etc). Maybe we could allow "sister" projects to help each other (e.g., Illinois might do a review for the Idaho project, which I think is fairly inactive) Second, I think it needs at least one person to take charge of it for the long haul, to make it happen then to maintain it. Without that, I think things will just tend to fall apart after the initial interest wanes. Is there anyone who has the time to take on this commitment? Uf so, I think we may be able to get something going.
Regarding the more drastic changes suggested, I think that would be a bit optimistic. It also assumes that FA = (GA + A), but I think the folks at FAC would disagree strongly, and FA is supposed to be much stricter on the style. Even if you raised GA standards, the politics between FA and GA would make it very hard to get people to ever regard them as being at the same level even for style aspects alone. I think if we were designing the system from scratch I might have supported this idea, but trying to change something as well established as FA would be very difficult, especially as it does seem to work (albeit slowly).
Thanks too, RoB, for getting that category together - I'm amazed that it didn't already exist! Walkerma (talk) 05:30, 5 February 2009 (UTC)
As usual when I read threads here (and I have read them), I find myself agreeing with Walkerma. I am generally in favour of proposals like this which aim to clarify the distinction between WikiProject assessments and community review processes. Girolamo put it well in the lead thread of this RfC: "At least in these two [Wiki]projects, the review process is fundamentally designed to prioritize content issues over style, MOS, and other site-wide guidelines, which tend to be better addressed by PR, GAN, and FAC.". However, I agree with Roger Davies that WikiProjects can take style issues very seriously in preparation for FAC.
I would also like to mention policy issues here. Most of the work done at GA is not about style (the process only requires an extremely weak compliance with MoS), but policy: "is it neutral, verifiable, and free of copyright violations". These aspects are often of lesser interest to WikiProjects, because editors there know what is true and accurate, and are not interested in copyright law.
But just as some WikiProjects do pay attention to these issues, so too GA has a "broadness" criterion covering content, and this is often a major focus of GA discussions. GA is not equipped to do this as well as many WikiProjects, but in some areas, WikiProjects are absent or dormant.
So a simple recipe won't work. In particular "GA+A=FA" has no chance of success, not only because of the politics, but because GA is no longer interested in the intricate requirements of the MoS, and is much more focused on policy compliance.
However, I support the idea of decoupling WikiProject assessments from community review processes, as it would resolve so many anomalies. We all have the same goal in mind: to improve the encyclopedia. The current set-up is making that not as easy as it could be. Geometry guy 21:58, 6 February 2009 (UTC)
Then, what about:
  • GA focusing on policy
  • A focusing on content
  • PR focusing on style and compliance with MoS
  • FA focusing on everything, somewhat stricter than GA, A or PR
Then we still have FA > GA+A+PR, meaning we could strongly advise people to go to GA and A and PR before going to FA, but with no direct guarantee of success. Then the assessment would work more like scout badges, each of them focusing on one area, and then you have the final-level-boss: FA. Please remember that Wikipedia:Wikipedia is an MMORPG ! SyG (talk) 12:17, 15 February 2009 (UTC)
I think it's a Bad Idea to try and separate out "content" and "style": both are subjective issues and both are important to most people's idea of "quality". I took a look at a short chemistry article last night that would be a pretty good B-class on content: but given that it took two separate PhD chemists (one of them a full professor in a related field) to agree what the article was actually saying, I think the style definitely needs improving!
I would distinguish the levels more on who does the reviewing
  • GA: a relatively quick run through by an uninvolved editor, as at present
  • A-class: a decision by the WikiProject that the article is "basically done". WikiProjects will have different criteria as to what is "basically done", just as editors have different ideas of article quality.
  • PR: a more formal review, open to the whole community, as at present
  • FA: scrapped and replaced by a system that works :P
If smaller WikiProjects don't think their assessment procedures for A-Class are solid enough, they can always send an article or two to WP:PR to get some outside input: I don't think we need to be adding yet more bureaucracy to WP in the quest for some Chimera of "absolute quality". Please remember that Wikipedia is a self-perpetuating Process an encyclopedia. Physchim62 (talk) 14:04, 15 February 2009 (UTC)

Statistics

  1. According to Wikipedia:Version 1.0 Editorial Team/Statistics there are only 615 A-class articles.
  2. 139 of these (22%) are Military history articles (see Wikipedia:WikiProject Military history#A class articles)
  3. 113 of these (18%) are lists in Category:A-Class Spiders articles
  4. 93 of these (15%) are lists in Category:A-Class U.S. Congress articles
  5. Hence, more than half of all A-class articles are accounted for by 3 projects.
  6. We can deduce that there are 276 (45%) articles in other A-class categories.
  7. There are 1472 A-class article categories.
  8. Hence, it is reasonable to assume that approximately 1200 A-class article categories are empty.
  9. Most of the A-class review processes in Category:WikiProject A-Class Review are inactive.

So, while I admit that in one or two projects A-class works well, for the vast majority of projects, it doesn't. It's unnecessary, or extraneous, or irrelevant. I think there are two options:

  1. try merging the successful A-class review schemes into FAC somehow;
  2. remove A-class from the assessment grading scheme of projects where it doesn't work, and only retain it for those projects with an A-class review system. DrKiernan (talk) 12:45, 5 February 2009 (UTC)
There is a third option - try to get the system to work better! Walkerma (talk) 22:42, 5 February 2009 (UTC)
Note, incidentally, that:
  1. WikiProject Spiders places A-Class below GA-Class (Wikipedia:SPIA#Article progress grading scheme), and doesn't appear to have any formal review process
  2. WikiProject U.S. Congress doesn't appear to have any formal review process
I'll try to put together some more exhaustive statistics shortly. Kirill [pf] 13:11, 5 February 2009 (UTC)
DrKiernan's analysis may be a bit flawed since, may I point out, articles are not limited in the number of WikiProjects to which they may belong. I think the conclusion drawn in point number 8, especially, may be off by a bit. Category:A-Class Ships articles contains 22 articles, most of which also fall under the Military history umbrella. — Bellhalla (talk) 13:21, 5 February 2009 (UTC)
Well, I've just looked at the first ten categories listed at Category:A-Class articles and nine of them are empty. The tenth has one article in it. So, I don't that assumption 8 is that far off. DrKiernan (talk) 13:28, 5 February 2009 (UTC)
Okay, I've uploaded more complete statistics at Wikipedia:Version 1.0 Editorial Team/Assessment/Statistics as of February 2009.
The dropoff in A-Class articles isn't that dramatic—there's a fair number of projects with a dozen or so—and I'm not sure how much of it is simply due to poor article quality rather than underuse of a particular class per se; but it's food for thought, in any case. Kirill [pf] 14:05, 5 February 2009 (UTC)
Article counts over time
This may also be relevant: despite the number of articles assessed (and the number assessed at the higher levels) rising continually since the scheme was first introduced, the number of A-Class articles plateaued in mid 2008 and has been declining since. Happy‑melon 14:13, 5 February 2009 (UTC)
I'm not sure how relevant that is since most Milhist articles soon go on to become FA. In other words, we don't have a large number of AC articles because we have a high turnover of them. That may well be true for other A-class articles. --ROGER DAVIES talk 17:55, 5 February 2009 (UTC)
But unless the average time that an article remains A-Class is also changing, a decline in absolute A-Class numbers (running against roughly constant rates of increase of all other classes of articles) is still indicative of a decline in A-Class participation generally. The percentage of A-Class articles as a total of all articles has fallen from 0.13% two years ago to 0.03% now. By contrast the percentage of FA-Class articles has declined from around 0.23% to 0.13%, and the percentage of GA-Class articles has actually increased from 0.23% to 0.29%. Whichever way you look at it, A-Class is bucking the trend. Happy‑melon 19:34, 5 February 2009 (UTC)
A's should be demoted to B's, GA should be renamed A, and milhist should track these things some other way. - Peregrine Fisher (talk) (contribs) 20:31, 5 February 2009 (UTC)

(Unindent) I still don't see the problem here. If a system only works well in a few cases, but doesn't work well in most cases, there are two choices:

  1. Scrap the system - but this hurts the cases where it works well.
  2. Fix the system - find out about the cases where it works well, and try to grow things from there.

It seems to me that we have not really tried #2, and it seems logical to try getting the system to work before we jettison it entirely, especially when it clearly fills a gap in the assessment scheme. Also, if GA were renamed A, it would really confuse everyone even more than they are already, mixing two different types of assessment in one scale! And if we went ahead and did #1, scrapped A-Class altogether and renamed everything, I predict that within a year or two this page would have newbies proposing that we create a new class. We wouldn't fix anything, we'd just move things around a bit and annoy MILHIST and FILMS and a few others. I'm pretty certain that some projects would retain A-Class, even if the 1.0 scheme abandoned it, so we'd end up splitting up the system.

Let's also remember that MILHIST and FILMS are not the only projects that use A-Class properly. A-Class is supposed to be a small class, so 0.03% is not unreasonable when set next to 0.13% that are FAs. I'd have said the ideal would be around 0.1%, because we still need to get a peer review system working on other projects.

When we got this system going, I always expected A-Class to be patchy in the early days - i.e., while projects were starting up and learning about assessment. It would've been premature to talk about A-Class review when a project only had 50 articles tagged out of 5000. So the statistics above don't surprise me at all. But many projects have now been assessing articles for a year or two now, and they have both the experience and the "inventory of articles" to institute a formal A-Class review. In other words, the startup phase is over, and the majority of articles have been tagged and assessed; this is where projects can move to the next stage of refining their systems, and doing things like A-Class reviews. Let's facilitate that! All we need is to create an environment that supports projects with mentoring, support from sister projects, scripts/templates/bots to make life easier, etc.

I think the negative comments above are (in effect) abandoning the system before it's really been tried. How can a project know that A-Class is useless for them, when they've never really known how to use it? And we've never made any effort to support and grow the A-Class review system.

If we abandon A-Class, the only way for an article to progress to FA is via GA. In other words, our standard system for generating our "ideal" articles involves no review by subject-experts, unless they happen to hang around the FAC pages. Look at an article like Aldol reaction; how much of that article do you understand without clicking on hundreds of links? How can you tell if it's nonsense, and if it's really comprehensive? Do we really want to send the message that we don't want the input of WikiProjects in preparing articles for FAC?

In summary, we need A-Class anyway, so let's make a decent effort to get A-Class review to work well throughout all the projects like it does in MILHIST and FILMS. This discussion should focus on how best to achieve that - unless someone has a really good alternative (which I haven't seen yet!) Only if that effort fails to propagate peer-review should we talk about merging A-Class into other classes. Walkerma (talk) 22:42, 5 February 2009 (UTC)

I concur with Walkerma; if wanted, I could draft up a full version of my proposal above which may help fix these issues. -Drilnoth (talk) 22:57, 5 February 2009 (UTC)
I think that A class serves a purpose in getting an article up to featured status. One I've been suggesting as useful is stuck waiting to get references. This is not an easy task. So my opinion is that it is better to work on clearing problems that would keep an article from reaching FA status. So delays in getting articles upto A are not a problem, but a valid checkpoint. I suspect that for many articles the issue is lack of references. If that's the case, then maybe a list of these so that editors could work on this and gt more article up to A status could be a good thing. Vegaswikian (talk) 23:03, 5 February 2009 (UTC)
I agree with Walkerma's assessment of the situation, and I think we should try putting together a proposal with this in mind. I like the idea brought up earlier in this discussion of having the more active WikiProjects help with A-Class review for less active child or parent projects. Members of a WikiProject will generally be familiar enough with the topic to accurately review articles under the scope of related WikiProjects.
As I see it, the main problem right now is one of framework; that is, for most WikiProjects, what is lacking for an A-Class review system is not editors, but the system itself. If there is a good, fairly consistent system for A-Class review at every active WikiProject, there will be enough editors to make use of it—both through the main editors of individual articles (which virtually every article meeting A-Class standards will have a fair number of), and through other editors of the WikiProject (or related WikiProjects) who wish to help identify articles near FA-level. Pyrospirit (talk · contribs) 23:39, 5 February 2009 (UTC)
I'm in concurrence with the arguments of Walkerma. The entire idea of in-project review is so that subject "experts" pick it apart for every factual error and fallacy. GA rarely does this, and is based almost entirely on "is it seemingly comprehensive, does it follow style guidelines, etc". The MilHist Systems that have already been put into place wikipedia-wide (the b-class assessment system that is somewhat derived from ours) have proven to be successful. The solution here is to improve A-Class, not abandon the idea entirely. Cam (Chat) 23:48, 5 February 2009 (UTC)
I would have to disagree about A-class review and subject experts. Most of the articles in A-class or FAC review aren't known to anyone but the author and I doubt many of the reviewers even have a particularly strong amateur knowledge of the topics to be honest. To be frank if an article didn't even exist until 2007, eg, most of my MILHIST FA/A articles, probably nobody cares about it. I hadn't even heard of 90% them until I started writing about it. YellowMonkey (click here to vote for world cycling's #1 model!) 00:17, 6 February 2009 (UTC)
(e/c) But will WikiProject A appreciate having to do reviews on behalf of less active WikiProject B? Perhaps I'm being too cynical, but I wouldn't have thought so (if that's what you're suggesting). I'm inclined to disagree with your assessment; I would say that the main problem is a lack of participation. That's ultimately why smaller projects will not be able to sustain their own A-Class review systems. Heck, even WP:BIOGRAPHY failed to keep their A-Class reviews going, and one would assume that to be one of the most active projects. PC78 (talk) 23:57, 5 February 2009 (UTC)
Follow-up thought: even if X project didn't care if they had to review project Y's articles, would there be enough reviewers to ensure quality reviews on the articles of project X? —Ed 17 (Talk / Contribs) 00:39, 6 February 2009 (UTC)
(at PC78) I think that WP:BIOG's A-Class assessment system failed because the project is too diverse... no two high-quality articles are likely to interest the same person. On the other hand, projects like MILHIST and the various country or TV show projects, have a more focused subject area, even if it is just as large a project. Gary Gygax and Albert Einstein wouldn't be worked on by the same people, by Gygax would be of interest to everyone in an RPG project and Einstein would to a science project.
(at ED17) That really depends on what size the project was and what specific members it has. -Drilnoth (talk) 02:10, 6 February 2009 (UTC)
G.A.S (talk · contribs)'s view on this

With the introduction of C-class, we (WP:ANIME) set up a proper assessment page, set up assessment guidelines, came up with a proper list of examples per class and reassessed all of our articles. We also had a long debate regarding the whole A-class issue (The latest being Wikipedia talk:WikiProject Anime and manga/Archive 33#A-Class, for real), and our conclusions was as follow:

  • That A class does not seem to serve a purpose A-class above/beyond GA/FA which are more widely recognized.* (Even the Selectionbot rates it on par with GA class.)
  • Since it's led by individual projects, it doesn't get as much notoriety or regulations to make it mean more than a GA when editors already have to go through tons of loops to get a GA passed nowadays.
  • It is uncommon for editors to edit articles only up to A class, as they either stop at GA, or push through to FA class. In the former case A-class is not worth the trouble, and in the latter case, a mere hurdle on the way.
  • We simply do not have the manpower to do it: "The only use of A-class is when we have a rigorous review process for it that preps articles for going to WP:FAC (the WP:MILHIST A-class review is perhaps the model of this). Past that, I don't see any use for it, as we're generally not going to have any sort of formal review process with enough participation (or to be honest really, enough quality review for the purposes of FAC). If people want feedback for FAC, they can open a peer review. I would simply not recognize it for the purposes of the project." — sephiroth bcr

Walkerma makes some valid comments above regarding the use of a project review to validate articles — I would expect those potential issues to be fixed by the time an article gets to GA class, though. At WP:ANIME, we have set the bar for B-class rather high in order to achieve this.** (Even if a B class assessment is not requested, any and all B class upgrades are reviewed).

*: I have yet to see an editor that lists the articles they have gotten up to A-class, as is widely done with GA and FA class.
**: Walkerma, have a look at the B class examples. Would you say that they are at (what you define as) the B class level?

In short: we have deprecated A class for the purposes of WP:ANIME until further discussion, recommending a GA review, a Peer review, and a B class review instead. If a workable system comes up, we would be happy to reconsider it again. I hope the highlights some of the problems that needs to be addressed for the smaller Wikiprojects to properly use A class.

G.A.Stalk 05:35, 6 February 2009 (UTC)

Maybe we should solicit comments from some of the the wikiprojects. - Peregrine Fisher (talk) (contribs) 05:40, 6 February 2009 (UTC)
GAS, quite a few MILHIST editors list the articles they have raised to A-class. Especially since that milestone qualifies them for a MILHIST-specific award: WP:MILHIST#ACM. -MBK004 05:53, 6 February 2009 (UTC)
Re GAS's points: 1) Some projects are using the class effectively for specific review, as per above. I agree with Walkerma - just because this is not working equally well across the entire site doesn't mean that we revert to the lowest-common denominator - especially since projects seem to implement it as their structures mature. 2) I wasn't aware that recognition had anything to do with article quality or classification. In any case, editors should not regard the work involved in responding to reviews as "tons of loops to get passed" - this almost sounds like the ends is more important than the means, when the review is the important part. Getting classed up is the consequence, not the point. Recognition, if it happens, is the gravy. As for whether or not GA or A means as much, that's subjective and depends on what the editor is more concerned with in their prose: style or content. 3) It is only meant to help as a step along towards FA-class. Everyone wants them to continue to FA, and the idea of ACR is to midwife the article towards a smooth and easy FAC. Editors who see this as a bothersome hurdle are of course welcome to ignore it - just like they can ignore PR or GAN, but may in some cases also be doing so at their peril either immediately at FAC or down the line when issues that could have been caught at A-class instead have to be raised at FAR. It's simply a tool, no more or less. 4) It's hard to know that without at least steadily attempting to gather editors towards the task for a consistent and considerable length of time. Not offering the option at all certainly will not improve the articles, while referring them to PR may not get the attention of anime editors either. An article that only seeks external reviewers from PR, GAN, or FAC may have errors which ares glaring to those familiar with the subject, but otherwise opaque to a random editor. Were I to review an anime article, I highly doubt that there is much that I could review about the article other than the style, writing quality, and basic conformance with our general policies and guidelines. Girolamo Savonarola (talk) 06:11, 6 February 2009 (UTC)
The projects who use A-class well sound admirable. I have yet to hear why it's a good thing to mix in with the other classes, though. Why not call it WA (MHA?), for wikiproject assessed, or something. I think that people like it since it sounds like one of the rungs on the Start, Stub,..,FA ladder, but it isn't. Point blank, we don't have the manpower to implement it project wide, unless we want to roll it into peer review, which currently struggles mightily. Getting other projects to use it well, is probably as easy as making all projects as good as milhist. - Peregrine Fisher (talk) (contribs) 06:24, 6 February 2009 (UTC)
Who's forcing it on projects? It works well in some projects - that's fine. Why the crab mentality? Girolamo Savonarola (talk) 06:35, 6 February 2009 (UTC)
(e/c) My point is that the above matters needs to be addressed. We do have an effective request system at the main talk page. If you need anime editors peer review an article, just ask:) The main reason for it being deprecated was unfortunately (1) a lack of manpower (see the discussion -- we have tried for quite a while not to deprecate it. The A class review system has been dormant for more than a year by the time we decided to deprecate it, nontheless the fact that the new assessment system attempted to pick it up), (2) a lack of requests (Only 17 requests were ever made, and only 2(?) went on to FA and FL.), and (3) the fact that we were unable to set up guidelines for when we can consider an article to be A class. (Style? Comprehensiveness? Both? How should it be different from FA? GA? // By hurdle I mean an easy one, not bothersome)
As I said though, we have incorporated the review function into our B class review, so it is not a matter of articles not being reviewed before GA or FA at all.
G.A.Stalk 06:32, 6 February 2009 (UTC)
Why don't we put this on WP:CENT? - Peregrine Fisher (talk) (contribs) 06:33, 6 February 2009 (UTC)
Isn't it on the template already...? —Ed 17 (Talk / Contribs) 06:35, 6 February 2009 (UTC)
Made me chuckle. It is. Kinda shoes how good CENT is at drawing attention to things. - Peregrine Fisher (talk) (contribs) 06:39, 6 February 2009 (UTC)

What's the point???

Well, given the persistent debates over A/GA/B/C class I'm not sure why there is such a big fuss, given that within any "class" of any WikiProject there is a big variation in quality!

  • A lot of modern FAs are properly done. A lot of FAs from the old days, or deteriorated ones, are much worse than many GAs or even a clean B-class article which is on the rise towards GA and higher honours.....
  • Ditto for MILHIST A-class articles, some of the older ones are poor
  • B-class and C-class articles. Regardless of what the criteria says, with these grassroots levels of article rating, there is massive variance in rating articles and almost anything goes. One of the main founding proponents of C-class, after the criteria was implemented, went on to shift all two-paragraph start-class articles from his WikiProject to "C-class", basically doing nothing except to foster grade-inflation and devalue it. (He always votes support/keep on every FAC/GAN/GAR from his WikiProject even when half the article is unreferenced and even the odd single-sentence article is rated "Start")

I would say the most glaring trouble is that the undisputed top rank of Wikipedia, the FAs, have an increasing variance in quality due to the tail in the distribution due to old FAs that are now B-class articles, and that too, disorganised-type B-class articles. YellowMonkey (click here to vote for world cycling's #1 model!) 00:30, 6 February 2009 (UTC)

Perhaps a Wikipedia wide drive is on order then to locate and remove the faulty articles? I agree with YellowMonkey that class varies from project to project (amusingly, both milhist and films share the A-class rank, as noted above, but neither project has yet embraced C-class. Granted, this later point may change in time). It could be said then that projects employ systems that work best for them, based on what they feel the contributors to the project want. Ultimately, the decision to embrace or forgo an assessment should be left to those who work within a project, not those who govern the color coded chart. Overhauling the system, changing its settings, or moving to discard part or all of A-class will undoubtedly have a ripple effect. TomStar81 (Talk) 08:02, 6 February 2009 (UTC)
I think that's right, except I don't think a wikipedia wide drive is feasible. We could probably get non-milhist articles demoted, but I doubt we could up-assess the ones that would meet whatever A-class means for them. - Peregrine Fisher (talk) (contribs) 08:12, 6 February 2009 (UTC)
Well some WikiProjects just assume that because MILHIST is the biggest, whatever that they do must be good for MILHIST and must also be good for them (not necessarily true), MILHIST could be strong in spite of some policies. But I have seen one WikiProject that has not had an FA for 18 months, has only 2000 articles, but has three taskforces and tried to run a competition...Another with 70 odd FAs and about 50,000 articles has about 40 taskforces, which seems over-the-top given that the said WikiProject has tanked and only makes a single-digit number of FAs per year....YellowMonkey (click here to vote for world cycling's #1 model!) 00:14, 7 February 2009 (UTC)

There's a big list of Wikipedia:Unreviewed featured articles. It's just that hardly anyone nominates the articles, and let's face it, a lot of people do the bare minimum to drag across the FAC barrier and won't fix up a nominal FA that is only of B-grade quality unless the star gets put under the hammer. YellowMonkey (click here to vote for world cycling's #1 model!) 00:14, 7 February 2009 (UTC)

C-class

YellowMonkey complains that C-class articles can be no more than Start-class.[1] Some wikiprojects do not recognise C-class.[2] GA-class interrupts the symmetry of the A-B-C hierarchy.

How about instead of a system which assigns a letter, we switch to one that builds on the Start-Stub classification. A standard article should comply with the B-class criteria. So, how about renaming B-class to "Standard-class" and merging C-class back into Start-class? This provides a new structure: Featured - WikiProject Approved (call it A if you will) - Good - Standard - Start - Stub. DrKiernan (talk) 10:03, 6 February 2009 (UTC)

I like this idea. Maybe using a scale of Stub, Start, Average, Good, and Featured, and then A-Class articles can be renamed to be "WikiProject Reviewed," with this last rating separate from the standard scale... so that there can be a Start-Class WikiProject Reviewd article of a Featured-Class WikiProject Reviewd article. This would have the benefit of allowing projects to choose exactly how they want to handle their WikiProject Reviews (currently A-Class reviews), without changing the standard ranking system. MILHIST would then be able to keep their current system as-is, except for a name change, and other projects would be able to choose their own ways of handling reviews, if they want them at all. -Drilnoth (talk) 14:45, 6 February 2009 (UTC)
I also like it, but I would prefer a numerical rating. FA = 1, GA = 2, B = 3, etc. - Peregrine Fisher (talk) (contribs) 21:41, 6 February 2009 (UTC)
The current system is approximately
Stub-Start-C - B - A
                \   \
                 GA -FA
so the anomaly between the first two names and the CBA system doesn't have a lot to do with GA and FA. The simplest solution would be to decouple community review processes from WikiProject ratings (which can vary from project to project anyway) so that we can keep track of them separately. I have commented in a couple of places in the previous threads. Geometry guy 22:06, 6 February 2009 (UTC)
I have always like that suggestion the best. Just put a separate variable in the project templates for the community reviews. That simplifies it all very nicely, and that still allows the projects to set their own ratings on the article. It would also allow A class articles to continue to display their GA status. Charles Edward (Talk) 23:54, 6 February 2009 (UTC)
It strikes me that reacting to an imperfect system by wanting to change the components that are working is at best trying to throw the baby out with the bathwater. Happy‑melon 00:23, 7 February 2009 (UTC)

I didn't say/mean that C-class articles can't be more than start. I just pointed out that the lower grades of ranking are never applied consistently and that the mean proponent of C-class apparently views assessment only as an inflationary statistical means of feeling better. YellowMonkey (click here to vote for world cycling's #1 model!) 00:10, 7 February 2009 (UTC)

Take a look at WP:GA, which has a nice counter for how many articles are FA, FL, or GA. I believe that, as GAs are graded by the community, not its relevant Wikiproject, and that they are considered to have a higher distinction than A-class, they should be separate and above A-class. An article cannot be both A from its project and GA from WP:GAN. If I were revising the rating system, it would be FA, GA, A, B, C, Stub, with A distinct from GA, and C-class essentially being the same as start, though that's another story. Projects can assign an article up to A-class to indicate that it is close to GA, and it receives visible distinction at GA. Many very good, acceptable articles are B-class but are so much better than other Bs. With A-class below the review-required GA, Wikiprojects can more easily assign its best articles A that have not been independently reviewed. Reywas92Talk 21:08, 7 February 2009 (UTC)

Um...A-class is considered to be higher in most projects, and they can be A and GA at the same time... —Ed 17 (Talk / Contribs) 21:14, 7 February 2009 (UTC)
I know that, but isn't the purpose of this discussion to rethink the A-class? That's my opinion about how to classify them. My problem is that so few articles are A-class because projects think they are higher than GA, which requires external reviewing, but A doesn't. Reywas92Talk 21:52, 9 February 2009 (UTC)
OK, a few thoughts. G.A.S. specifically asked me to look over a few things so let me reply to those first. I looked over the Anime A-Class discussion, and I think it is very reasonable. What I note is that some of your B-Class articles, such as Naruto, look close to A-Class to me, though of course I know nothing about Anime so I couldn't judge that! I see that it recently failed at GA (mainly because of grammar and a few of the sources - a very thorough GA review, well done PF!) but I'd guess that in terms of content the article contains all the main points that should be there. It wouldn't take much work to get it to A-Class, and that would provide another route to FAC rather than another GAN. But if the WikiProject has a system that works well, I'd say keep on working that way. FYI, the decision to make SelectionBot allocate the same score to A and GA was made by me.
I agree with the comment about variability - I think that's inevitable, given the range of projects. It's not just FA, it's at every level (including Stub). However, I think the biggest variability is not between reviewers, but based on the time of the assessment. When we started assessing at WP:CHEMS in early 2005, proper inline citations were a novel idea rarely seen, even in FAs. Something that was B-Class back then would probably fail C-Class today (likewise, most FAs of 2005 would fail GA today). So the biggest way to reduce variability is to re-do some older assessments - and I know we have some really old ones at WP:CHEMS that are probably way off. I think I'd like to get a bot to generate (for WikiProjects) lists of articles with assessments over a year old - that should be easy to do.
One thing I've always liked, though, is the fact that we have a system where organisation is devolved to the people who use it. That approach works well on the English WP, where people will always do what the heck they want anyway! (The term "herding cats" comes to mind....!). There is no decree saying, "If you don't adhere to X,Y,Z standards your WikiProject will be excommunicated." That freedom comes at a price - there is some variability - but in practice we were able to put together quite a nice selection for Version 0.7. Having a flexible system with devolved decision-making allows some projects to have A-Class review, where others are happy to have articles go B to GA to FAC. It allows some projects to tag List-Class, and even disambig. pages etc, whereas others may choose not to tag even lists. Some projects even have B+ class and other interesting things. WikiProjects can organize their articles nicely, yet I believe we can still use their metadata to select articles in a reasonable way for 1.0.
I like the suggestions for decoupling community reviews from WikiProject reviews. I think I'd like to propose an IRC meeting to brainstorm some ideas - perhaps next weekend on #wikipedia-1.0? If people like that idea, I think there should be two main things to brainstorm - I'd prefer one, but these two are very much interrelated. Once we've got some ideas, we can see if there is the will and the resources to carry them out. So the agenda would look something like:
  • What might a good mechanism for getting WikiProjects engaged in A-Class peer review, possibly with support from other projects? I believe that Drilnoth is putting together a formal proposal for that, and we should discuss that, plus any new ideas.
  • How might we clarify the difference between community reviews from WikiProject reviews - in the main assessment page, or in the talk page templates themselves, or what? Would we allow an article to be both A and FA, or A and GA, or B and GA, for example? (WP:CHEMS already does this, BTW).
  • Once we have discussed the above ideas (not before!) and decided on possible courses of action, we will be in a position to ask: Should we try out these courses of action? Or should we instead abandon A-Class reviews? Should we leave community & WikiProject reviews merged together as they are now?
That way (I believe) we can turn this discussion into a plan of action. Does this sound OK? Walkerma (talk) 04:24, 8 February 2009 (UTC)
Sounds good to me; would you be able to get the IRC set up? -Drilnoth (talk) 14:23, 8 February 2009 (UTC)
To address some of the old FA and GA (and also a few A class) that won't meet today's standard, I suggest we should start a project-wide sweeps on all three of those. We can set an arbitrary date and any articles before that day should be reviewed briefly to make sure that it meets the current standard (if not, at least very close to the standard). GA has sweeps and the results are promising. It didn't affect the rate of promoting GA. So how about that? OhanaUnitedTalk page 05:11, 10 February 2009 (UTC)
I think the idea of sweeps should be on the agenda; however, I think that WikiProjects should focus on B and A-Class rather than community-based assessments (GA & FA). Having projects reviewing Bs to find possible As, and reviewing As to see if any should really be Bs, I think that is definitely one approach to consider. Thanks, Walkerma (talk) 06:15, 10 February 2009 (UTC)
I'm hoping to reinvigorate the GA sweeps shortly. -Drilnoth (talk) 14:23, 10 February 2009 (UTC)
Hi, discussion has spread into several sections. One issue which i don't think has yet been discussed fully is A-class value across Wikipedia. In some projects there are no controls on A-class, anyone can assign even to C standard articles, e.g. Appartement du roi has a lot of good content but currently backed by few inlines. i've done my own sweeps to change down to better assessments. This is a problem because it devalues the A-class identity community-wide and takes away from high standard A-class e.g. milhist. Most people understandably don't know that A-class milhist or film is different from some other projects, because it still stays A class in bright turquoise, it doesn't say "A-class milhist being prepped for FAC". Needs to be consistenty and a floor on the standard of A-class. Interesting quantitative analysis and discussion above, thanks Tom B (talk) 17:05, 10 February 2009 (UTC)
I like the idea of making WikiProjects focus on A/B rather than GA/FA. Indeed, would it be practical to make A a top individual article goal, with GA and FA being goals targeted by a WikiProject as a whole? For example, most projects will advise things to work towards a more generic Wikipedia-wide assessment. A-Class is much more project and topic oriented. Maybe that doesn't make much sense. Basically, we have project members instead aim towards getting their articles to A-class, with the frequent reviews being more pertinent to the subject the project and the article itself is based around. When the article is felt by the WikiProject to have achieved a level of quality adequate for the actual topic covered by the article, the project then aims for the more overall-quality measure of GA and FA. This system of working towards A-class before GA and FA would not be a requirement by any means, but more of a precedent. Currently, it is the reverse (to an extent). GA and FA are set generic Wikipedia-wide standards for articles, while A-class sets a standard more specific to an individual project. It makes more sense for a project to at first aim for the latter, have standards focusing on particular aspects of the articles under their scope before gunning for that article to be excellent in all areas. I do hope any of the above makes sense, as I don't think I've explained it very well at all. A-class always came across to me as more of a "Top WikiProject article" while the other two seem to be more of a "Best of Wikipedia." --.:Alex:. 22:05, 10 February 2009 (UTC)
The problem is that some WikiProjects (and taskforces) simply do not have enough members, familiar with reviewing, not feeling comfortable, or resources to merit its own A-class review board. WP:MILHIST has more members than most other projects but that does not mean the planning or brainstorm session should neglect smaller projects. OhanaUnitedTalk page 14:29, 11 February 2009 (UTC)

Death to WikiBureaucracy

Many very committed editors have already pointed out the problem of the smaller WikiProjects. I think we should be clear that all the WikiProjects are small when it comes down to having enough people with enough time to do the things which everyone agrees are desirable. WP:CHEMS is one of the larger and more active projects, but we have fewer than twenty active editors, many of whom have other WikiInterests or WikiResponsabilities as well. A "solution" to the "A-class problem" which involves our best editors spending more time checking articles which are already pretty good – and so less time improving those that are very much less than optimal, or even writing those which still don't exist – that is no solution at all.

When A-class was invented, it was meant to mean "no need for any more work on this one". That isn't the same as saying that the article can't be improved, simply that it has covered all the points of the subject in a reasonable manner; the concept was entirely linked with encyclopedic content, although obvious style faults would fall under the heading "needs more work". Of course there is still FA-status but, given the number of simple grammatical errors which pass WP:FAC, I have to question whether the star is worth anything. Both WP:FAC and WP:GAC are snowed under with candidatures, and couldn't cope with any extra workload on those editors whose choose to spend their time there.

Any "A-class review" system would have the same problems as the current WP:FAC: too much work and not enough editors. Why can we not assume good faith on the part of the WikiProjects, and raise queries through those projects' talk pages? Physchim62 (talk) 01:01, 12 February 2009 (UTC)

The problem is not that we think that small WikiProjects can't handle the load and need help; the issue is that we've been asked several times for help already. How we can help WikiProjects that wish to have stronger assessment processes is probably one of the most critical functions of WP:1.0, so we need to figure out a way of helping small projects without burdening functioning larger projects. Titoxd(?!? - cool stuff) 08:37, 12 February 2009 (UTC)
I would turn that argument on its head and say that the assessment procedures are wrong if the projects can't cope with them. Our first attempts at assessment on WP:CHEMS took one minute per article, which is why we were sure that the system was viable. We also assured ourselves that the three editors involved were giving equivalent grades to the articles, dispite having slightly different assment criteria. In time (see Wikipedia:WikiProject Chemicals/archive01, from 2005), we adapted the criteria to what we were actually doing, not the reverse. It doesn't take long to see that an article is missing important material, or is simply badly written: on the other hand, it takes a great deal of time to pretend to show that an article has no faults. I deliberately say "pretend to show", because I don't think that any assessment system could actually demonstrate this, the criterion is fatally subjective. Experience at WP:FAC shows this very clearly. Physchim62 (talk) 17:14, 12 February 2009 (UTC)
Haven't we already suggested a solution: merge small projects into larger ones, by turning them into "task-forces" of the bigger projects? DrKiernan (talk) 08:48, 12 February 2009 (UTC)
That works for some projects, but in some cases it isn't feasible because a child project is too large. -Drilnoth (talk) 13:30, 12 February 2009 (UTC)

What about the 50,000 B-class articles? Would it not be a productive use volunteer time to review the B-class articles to see which of them can be brought up to A-class with a little work? Obviously, only the projects could do such a review. Physchim62 (talk) 20:54, 15 February 2009 (UTC)

I'd argue that for most general topics, those without any technical content, could be reviewed by anyone. While a few mistakes might be made, I'd bet that most of these upgrades would be spot on. Vegaswikian (talk) 22:24, 15 February 2009 (UTC)
I think you're probably right, but how would you organize such a "B-class review" of over 50,000 articles without breaking it down into, say, WikiProject size chunks? Physchim62 (talk) 22:59, 15 February 2009 (UTC)
Can anyone tell me on which basis an article would be A ? As it seems still unclear at least for me :( --KrebMarkt 22:51, 15 February 2009 (UTC)
"The article is well organized and essentially complete." from Wikipedia:Version 1.0 Editorial Team/Assessment. Physchim62 (talk) 22:59, 15 February 2009 (UTC)
You manage to solve the GA VS A issue then ? --KrebMarkt 23:03, 15 February 2009 (UTC)
I don't have a problem with the GA vs. A issue! ;) Not unless someone tries to make it obligatory, in which case I'd cry "pointless bureaucracy" very loudly. Style is important – "well organized and essentially complete" includes having references to tell people where the information came from, images (where possible and appropriate) in the right places, etc. – but surely content is more important in an encyclopedia? We're not here to design a beautiful website or the perfect term paper, last time I checked. Physchim62 (talk) 23:20, 15 February 2009 (UTC)
:) Yea, very messy GA & A issue. If you do start Operation (A) just don't forget to contact the various projects assessment department when they exists, courtesy require so. --KrebMarkt 07:14, 16 February 2009 (UTC)

I would support completely scrapping the A-class process myself. GA and FA are close enough that going through another hoop is just a waste of time and doesn't really do anything to improve the article. Wizardman 15:41, 20 February 2009 (UTC)

At this point I agree with Wizardman. The number of A-class articles is low and the review processes are mostly inactive. In the past processes such as these are labelled historic or deleted, so we should treat A-class the same. I think it should be removed from the version 1.0 grading scheme. If the military history and other projects wish to continue using their own grading scheme, and continue to have an A-class then that's fine too. Wikipedia:WikiProject Mathematics had/have a scheme of their own with an additional (B+) grade, so there's no reason why military history and others shouldn't either, if they wish to do so. DrKiernan (talk) 15:56, 20 February 2009 (UTC)

That depends on what you define as "important" and "improving the article". If an article jumps from B to GA to FA, that means that nobody has asked themselves a formal question about the article's content at the highest level. Maybe problem will be picked up at WP:GAN or WP:FAC, but there's no real guarantee, especially as neither of those processes are really structured for assessing the factual accuracy and completeness of an article. Do we really want to say to the world that we don't care about the content of an article after a fairly low level (B-class)? That after that level we are only worried about its style? That we can't be bothered with a quality control system on content because it's too difficult and irritates a few of our more independently-minded editors? I don't think we should go down this path because MySpace already does it better :P Physchim62 (talk) 16:06, 20 February 2009 (UTC)

I completely disagree that GA and FA are about style over content. I always review content. DrKiernan (talk) 16:39, 20 February 2009 (UTC)
Scrapping A class is beyond the scope of just the English wikipedia. I would prefer the current statut-quo if the article improvement per time cost for each A peer review is too low. Some project like WP:Anime have set the B level rather high so B class articles aren't all that poor content, quality & style wise. --KrebMarkt 16:33, 20 February 2009 (UTC)
Or WP:MILHIST. :) —Ed 17 (Talk / Contribs) 18:46, 20 February 2009 (UTC)
Lucky you. We have a astounding percentage 0.67% of B rated article in the Anime/Manga project while MILHIST is around 3.90% and that percentage includes B rated lists. B class rating is very step in Anime/Manga. --KrebMarkt 19:15, 20 February 2009 (UTC)

I agree with DrKiernan here. Few articles use A-class rating; it should be maintained for those articles that have it, and for the WikiProject(s) that still have an active A-class review system, but it should be deprecated for all other purposes. With the new WP:Article alerts system available, and being subscribed by most wikiprojects I interact with, I expect much fewer content embarrassments at GAC/FAC; before article alerts was available, one could easily sneak through articles at FAC/GAC with superficial reviews from editors not familiar with the content (I can give you examples of 2008 FA bombs if you insist). Xasodfuih (talk) 13:54, 26 February 2009 (UTC)

IRC meeting to discuss proposals for A-Class

It looks as if some key people can't make an IRC meeting this Sunday, and I don't want to drag people away on Valentine's Day, so I think we will have to hold the IRC meeting instead on Saturday 21st or Sunday 22nd, probably around 1900h UTC. I you want to discuss this on IRC, please let me know if a particular date or time is good or bad. Walkerma (talk) 21:28, 12 February 2009 (UTC)

And what of those of us who refuse to use IRC and still want to discuss this issue? -MBK004 21:34, 12 February 2009 (UTC)
Would it maybe be easier to just have a time to all be on Wikipedia at once to have a "semi-live" conversation? -Drilnoth (talk) 21:39, 12 February 2009 (UTC)
E/c's to the max, though... —Ed 17 (Talk / Contribs) 21:53, 12 February 2009 (UTC)
Heh... good point. -Drilnoth (talk) 22:02, 12 February 2009 (UTC)
For the purpose of transparency, could the contents of the IRC meeting be made public so that people who aren't there can see what was discussed? Pyrospirit (talk · contribs) 22:45, 12 February 2009 (UTC)
Yes, we can make it public - I'll announce that at the start and here on-wiki. Anyone who refuses to use IRC can still post their ideas on wiki; if you come up with a detailed proposal, I'll present it for you, if you request that. It's a brainstorming meeting to clarify the options, not a place to decide policy, so the conclusions will be brought here for comment. Walkerma (talk) 03:16, 13 February 2009 (UTC)
Actually, we can't post public logs without the permission of every participating user. –Juliancolton Tropical Cyclone 03:21, 13 February 2009 (UTC)
If I'm not mistaken, freenode policy allows public logging on the request of the channel owner if all participants are informed of it, regardless of whether they explicitly consent to it. Kirill [pf] 03:31, 13 February 2009 (UTC)
Well, if the users are informed of it, and they accept it, chances are they give their permission. :) –Juliancolton Tropical Cyclone 03:34, 13 February 2009 (UTC)
So to make it clear: If you show up to the meeting, you should expect a logger to be present, and for that log to be made public on-wiki. Titoxd(?!? - cool stuff) 16:21, 17 February 2009 (UTC)

(←) If I could convince someone who particpates in the discussion to write up a brief overview of the discussion (10-15 sentences) I would like to include it in the next DRAMA report for the Signpost due Feb. 23. I have a feeling that discussions will be a lot to read if I end up not being able to monitor everything live. Let me know if you're interested. Thanks, §hepTalk 02:32, 18 February 2009 (UTC)

I will probably want to write a summary anyway, and I will try to make it useful for the Signpost. However if others want to write something, that's fine with me too. Walkerma (talk) 02:49, 18 February 2009 (UTC)
Sorry to push; but, details anyone? §hepTalk 00:28, 23 February 2009 (UTC)
Yes, I would like to know too. :) I screwed up at UTC and got on at the wrong time, so....I have no idea what happened. —Ed 17 (Talk / Contribs) 00:32, 23 February 2009 (UTC)
Sorry about the delay! The basic summary is this: There is consensus that GA and FA assessments should be separated from the rest on WikiProject banners (an incremental step towards the goal of having separate scales for project-wide and site-wide assessments), so that, for example, Dungeons & Dragons would be a Featured A-Class article. This isn't completely related to A-Class reviews, but was brought up and there was agreement. Secondly, there is going to be a work group dedicated to better organization between WikiProjects, with its first goal being to help coordinate A-Class reviews and assessments for each interested project. -Drilnoth (talk) 02:01, 23 February 2009 (UTC)
Very cool, thank you! §hepTalk 03:02, 23 February 2009 (UTC)
If you have any questions or comments about what I just mentioned, please feel free to ask and I'll answer to the best of my ability. -Drilnoth (talk) 03:06, 23 February 2009 (UTC)

Summary of the IRC discussion

(Sorry I was a bit busy till now, though it is still 22 Feb in my time zone! Please correct anything I missed/misunderstood.) The complete meeting log is here].

Much of the discussion focused on how to encourage wider use of A-Class, given that it is hard for a small (or even medium) WikiProject to sustain a rigorous A-Class review. There were several solutions suggested, any one of which might be helpful for a particular project:

  1. Consider making small projects into task forces of larger parent projects
  2. Small projects may choose simply to adopt a rigorous B-Class system, and not use A.
  3. A project may work with closely related projects, perhaps in a consortium, to conduct reviews close to its subject area.
  4. A small but active project may work with a "light touch" review, using only 2-3 reviewers.

The consensus was that there was no single solution for all projects. Centralizing the work to a community-wide review was seen as duplicating GA/FA, and ducking the content issue. All participants seemed to value the use of A-Class for allowing a "content first, style issues later" approach to article-writing. One suggestion was well received, to have GA/FA separated from WikiProject assessments in project templates. It was felt that A-Class has to mean something if a project is going to work towards more A-Class articles; clearly a project review would foster that, and be driven by personal/project pride, as well as the desire to get an article's content right before FAC. It was proposed that a small team – perhaps coordinated by Walkerma and Drilnoth – work to promote A-Class reviews where appropriate. Coordinators of active WikiProjects will be contacted and asked to share their ideas and viewpoints, on A-Class and other issues, at http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Council/Coordinators%27_working_group . Walkerma (talk) 04:37, 23 February 2009 (UTC)


Details of the IRC meeting

FA proposal

As you may already be aware, I am finalising a draft proposal at Wikipedia:Featured articles/Proposal which would move the primary responsibility for choosing Main Page articles to the WikiProjects. The proposal still retains an important element of "Community oversight", but not in the current form of WP:FAC (which would be, quite simply, abolished). The aim is to ensure – better than at present – that the articles which represent Wikipedia on the Main Page have good and reasonably complete encyclopedic content on their chosen subjects and, in whatever case, represent our "very best work" in that particular subject area.

Comments are welcome at the talk page. I raise the matter here and now because it might have implications for any system of Community A-class review. Physchim62 (talk) 00:57, 20 February 2009 (UTC)

assessment of article for which there is only limited information

Ok, here is the issue. Saint Croix Macaw is a short article, for which most information is already included in the article. Sometimes, there is just not much info for a species. So, when I go through the criteria of the various classes, this is my assessment:

  • Stub: A very basic description of the topic.
    Nope, much more than that.
  • Start: The article has a usable amount of good content, but it is weak in many areas, usually in referencing. Quality of the prose may be distinctly unencyclopedic, and MoS compliance non-existent; but the article should satisfy fundamental content policies such as notability and BLP, and provide enough sources to establish verifiability.
    It has all the info there is, so it cannot be weak in many area's. Everything is fully referenced, Prose is not distinctly unencyclopedic and MOS is followed as far as possible. So, this one is out.
  • C: Not start anymore, not B yet.
  • B: Six criteria:
    1. The article is suitably referenced, with inline citations where necessary. It has reliable sources, and any important or controversial material which is likely to be challenged is cited. The use of citation templates such as {{cite web}}: Empty citation (help) is not required, but the use of <ref></ref> tags is encouraged.
      Meets criteria.
    2. The article reasonably covers the topic, and does not contain obvious omissions or inaccuracies. It contains a large proportion of the material necessary for an A-Class article, although some sections may need expansion, and some less important topics may be missing.
      Meets criteria.
    3. The article has a defined structure. Content should be organized into groups of related material, including a lead section and all the sections that can reasonably be included in an article of its kind.
      Meets criteria, the amount of info available makes it difficult to make subheaders, and not needed.
    4. The article is reasonably well-written. The prose contains no major grammatical errors and flows sensibly, but it certainly need not be "brilliant". The Manual of Style need not be followed rigorously.
      Meets criteria.
    5. The article contains supporting materials where appropriate. Illustrations are encouraged, though not required. Diagrams and an infobox etc. should be included where they are relevant and useful to the content.
      Meets criteria, there are no free images available.
    6. The article presents its content in an appropriately accessible way. It is written with as broad an audience in mind as possible. Although Wikipedia is more than just a general encyclopedia, the article should not assume unnecessary technical background and technical terms should be explained or avoided where possible.
      Meets criteria.

So, my questions is, what do I miss that requires the article to be stub or start class only and not a B? -- Kim van der Linde at venus 00:35, 9 March 2009 (UTC)

Proposed change of colour for Portal-Class

Not sure where the best place for such a discussion is, but I've started one here as that's where Template talk:Portal-Class redirects to. Comments welcome. PC78 (talk) 16:33, 10 April 2009 (UTC)

Proposed colour changes for Merge, Category & Template-Class

Following on from the above, I have made a proposal to adjust the colours used for Merge, Category and Template-Class. Discussion can be found here. PC78 (talk) 17:40, 16 April 2009 (UTC)

Is it Possible ...

It is relatively easy to view all articles for a Project by either class or importance. But is it possible to view a category of articles by both: e.g., Christianity articles that are class:Start, Importance: High? Athanasius • Quicumque vult 20:14, 30 April 2009 (UTC)

Yeah, it is. WP:CHINA already does that, as you can see from Category:Top-importance China-related articles, for instance. It just means some serious alterations to a project's banner to accomodate it. John Carter (talk) 20:18, 30 April 2009 (UTC)
Big task to be sure - I know that banners have certainly grown into things that I don't want to mess with, for fear of breaking the darn thing. Some point down the road, however, that should probably become a standard. It just makes it much easier to identify and priorititize where work is most needed. Athanasius • Quicumque vult 20:33, 30 April 2009 (UTC)
The new WP1.0 bot will be integrated with the toolserver and will be able to provide intersections like this dynamically, for all projects, with no need for on-wiki changes. Eventually :D Happy‑melon 22:16, 30 April 2009 (UTC)
That will make things much easier. Eventually. Thanks for the info. Athanasius • Quicumque vult 00:01, 1 May 2009 (UTC)
I believe that CBM will be getting this new bot running this spring or summer - also including lots of other nifty features! Walkerma (talk) 02:29, 2 May 2009 (UTC)

Template:Good Article

{{Good Article}} has been nominated for deletion. See WP:TFD.

76.66.202.139 (talk) 10:02, 5 May 2009 (UTC)

A-class??

I'm not quite clear on what "A-class" means. 1) There don't seem to be many A-class articles. 2) I've seen articles that say, "_____ is a former Featured Article", and they're always GA-class. 3) The few A-class articles I've seen appear to be on the same level, quality-wise, as Good Articles. 74.33.174.133 (talk) 20:29, 8 May 2009 (UTC)

We had a big debate on clarifying the issue earlier in the year. Once our group has time to work on it - after the release of Version 0.7 - we will work on that. The short answer is - it depends how the assessment was reached, by the GA reviewers or by an internal WikiProject review system. The former tends more to address style issues, where the latter can focus more on specialist knowledge of content. Walkerma (talk) 01:39, 14 May 2009 (UTC)
I believe that a majority of A-class articles come from the Military History WikiProject; more information can be found at our project A-class criteria and our project A-class review. The normal Wikipedia A-class guidelines (non-MILHIST) can be found here. Cheers! —Ed 17 (Talk / Contribs) 04:46, 14 May 2009 (UTC)

Proposed addition to assessment table

I propose a new column for articles that are tagged for needing immediate attention. Crash Underride 18:27, 12 May 2009 (UTC)

Can you give us some idea how many projects currently use that classification? Cheers, Walkerma (talk) 01:36, 14 May 2009 (UTC)
Wouldn't that be better handled by content tags, rather than talk page assessments? It's much more likely to be noticed that way, and it then serves the dual purpose of warning readers of the issue until it's resolved. Pyrospirit (talk · contribs) 04:47, 14 May 2009 (UTC)
My idea was to add it to the table because I include the assessment table in every issue of the projects newsletter. Each time the bot is run and re-organizes the numbers for the columns, etc. it's not only automatically updates on the project page, but the newsletter as well. Crash Underride
Sorry I was so slow to reply, I got deep into final exams and so on, and didn't check back until now. Can you solve the problem by using the subst command? I had a similar problem with transcluding a template for our collaboration, and Subst solved that nicely.
As for articles that need immediate attention, if there are specific cleanup tags on an article page, we could consider adding a column to the "list by quality" pages that indicates this (similar to the way we currently indicate articles included in Version 0.5). Would that be a useful feature? Cheers, Walkerma (talk) 07:15, 20 June 2009 (UTC)
Well, have a look at the bottom of the up coming newsletter. I place a link at the bottom by the table. As for the the subst command, I'm afraid that I'd jack somethin' up. I'm paranoid like that. My idea was to add a new column that was for the ones that needed immediate attention, and have them grouped like they are by priority and class. Crash Underride 07:26, 20 June 2009 (UTC)
OK, that clarifies things a lot - thanks. Let's see if any of the tech people respond here - if not, I'll post some questions. Cheers, Walkerma (talk) 07:40, 20 June 2009 (UTC)
I hate duplicated effort, so I'll go ahead and ask: How is this different from the work that User:WolterBot does? (c.f. Wikipedia:WikiProject Tropical cyclones/Cleanup listing) Titoxd(?!? - cool stuff) 03:19, 21 June 2009 (UTC)
It's different because I'm talkin' about havin' a new column placed on the table. Not a listing made. I've already created the project category. Crash Underride 16:53, 21 June 2009 (UTC)
Hmm... if there were a way of listing article talk pages in that category, we could certainly treat is as a "review category", similar to what we do with Category:Wikipedia featured articles and Category:Wikipedia good articles. Titoxd(?!? - cool stuff) 06:35, 25 June 2009 (UTC)
Could that be done in the second generation bot, do you think? Walkerma (talk) 07:16, 25 June 2009 (UTC)
Depends on how it is done. If we have a large category that encompasses all articles with cleanup tags, then it's a matter of adding one line of code. If there's a category for each WikiProject, it becomes considerably more painful. However, both are possible. Titoxd(?!? - cool stuff) 07:25, 25 June 2009 (UTC)
I agree with Titoxd, however I know jack about the codes for the bots, lol. Crash Underride 19:36, 25 June 2009 (UTC)

Stupid question?

Does an article need to be adopted by a project to be able to be assessed? I can't find the templates to do it if not. Nimbus (Cumulus nimbus floats by) 21:43, 27 August 2009 (UTC)

I'm pretty sure they don't exist; the main premise of this system is that assessment is done by WikiProjects, not outside of them. Kirill [talk] [pf] 02:21, 28 August 2009 (UTC)
Thanks, I thought that might be the case. In theory there could be some quite good articles around. I ventured into a few radio-controlled model related articles, improved one at least to 'start' class but it remains unassessed, the aviation project does not want to adopt them from past discussions. Nimbus (Cumulus nimbus floats by) 09:45, 28 August 2009 (UTC)
If you have an interest in radio-controlled models, you can set up a WikiProject for it - see WP:WikiProject#Creating and maintaining a project. Have a look through article histories for regular, current users and see if they would like to join. You can always ask for advice at WT:WikiProject Council. --RexxS (talk) 13:14, 28 August 2009 (UTC)
Thanks, I am quite heavily involved with the aircraft project (especially aero engines) at the moment, I'll have a think about it. As an RC aircraft modeller it would be nice to group these articles together with the ultimate aim of improving them. Cheers Nimbus (Cumulus nimbus floats by) 16:13, 28 August 2009 (UTC)
FYI: On occasion (mostly a long time ago!) the WP:1.0 project has done assessments of untagged articles upon request. The problem with this is that we often don't know the subject well, but if you need this, let us know. Walkerma (talk) 22:18, 28 August 2009 (UTC)
Thanks, I guess most of the bigger (better?!) articles are covered by projects now. Nimbus (Cumulus nimbus floats by) 18:30, 29 August 2009 (UTC)

Objection about how multiple project ratings are combined for an article

I had an interesting experience with the John Keats article just now. It looks seriously deficient to me, but has been rated "B-class," i.e., by the definitions here, "mostly complete and without major issues." The rating is, in my opinion, simply wrong. Digging into this a bit, it looks like WikiProject Biography has rated it a B on the basis of a plausibly complete "Life" section, but WikiProject Poetry has rightfully rated it a start-class article, apparently based on the obvious lack of any substantive discussion of Keats's poetry.

I experimented by reducing the Biography rating to "start" (I immediately restored it to "B" after this test) and saw the article's overall rating drop to "start." This verifies my impression that whatever combines ratings from multiple projects is taking the maximum rating from any project. I contend that the minimum rating ought to be used, at least if the WikiProjects involved are equally significant to the subject of the article. In the case of John Keats, WikiProject Poetry ought to be given at least an equal say, if not the final say: if the poetry discussion is inadequate, the entire Keats article is inadequate. I wonder if whoever has developed the article-rating mechanism can reconsider this aspect. -- sharpner (talk) 01:26, 18 October 2009 (UTC)

  • I have edited this article to note that the best WikiProject rating is taken for any given article. That's the behavior I've observed based on an admittedly minuscule number (1) of trials. If the bot actually behaves differently, I look forward to being corrected, but I hope this edit attracts attention to the issue. -- sharpner (talk) 01:45, 18 October 2009 (UTC)
I have noticed that the WP:BIO B class ratings are frequently off the mark. So I question any logic that uses the highest rating as the default. Vegaswikian (talk) 01:55, 18 October 2009 (UTC)
Bear in mind that the B-Class assessment of that article was over two years old. Assessment standards have changed over time, and no doubt the content of that article has too (it's not unreferenced, BTW). Anyone is free to change an assessment they don't agree with, or request a reassessment by the project. PC78 (talk) 07:14, 18 October 2009 (UTC)
Use the maximium not the minimum, otherwise projects rating will not change. This article in question had a problem with the Biography rating, it's not a general problem of combining multiple project ratings. SunCreator (talk) 13:11, 19 January 2010 (UTC)
Respectfully, no. Perhaps there was a problem with the Biography rating, but that's irrelevant. It doesn't matter what the Biography project rates this article if the Poetry project thinks it's so deficient. It could have been a legitimate "B" from the Biography angle and still wouldn't deserve a "B" overall, since coverage of Keats's poetry (at the time I wrote the original objection) was almost non-existent. From my viewpoint, this is purely a problem of how Wikipedia chooses to summarize the ratings from multiple projects in a single overall rating. -- sharpner (talk) 19:57, 29 January 2010 (UTC)

Sharpner: What do you mean by "overall rating"? Where did you see it displayed? I am trying to decide if this is an issue with my software, or with some other javascript. — Carl (CBM · talk) 13:49, 19 January 2010 (UTC)

Use the beta Wikipedia interface. You'll see the article's overall rating right under the article title at the top of the page. -- sharpner (talk) 19:57, 29 January 2010 (UTC)

A-Class (revisited)

It seems that A-Class is extremely underused. As far as I know, only around a thousand articles (please correct if I'm mistaken) make use of it. I feel this is regrettable, as there is a large but often unrecognized gap between WP:GA and WP:FA. Only one WikiProject really makes widespread use of A-Class, WP:MILHIST. Most other projects simply ignore this. I was wondering if it's possible to make an effort towards "revitalizing" A-Class and making it more readily usable. Any thoughts? –Juliancolton | Talk 02:17, 31 December 2009 (UTC)

I agree, but I had no real time to devote to it during 2009. I was thinking of working on this a lot during January & February (I'll get busy again after that). Shall we make it a joint effort? If so, perhaps we can bring in some other interested parties, and plan what to do. Then we just need to get on and be busy beavers for a few weeks - I think we can at least get a few projects woken up. I will promise to start doing A-Class reviews in chemistry, starting tomorrow! Cheers, Walkerma (talk) 02:41, 31 December 2009 (UTC)
I will watch from afar what you are going to do in order to "revitalize" A-Class meanwhile i will be hanging around Strategic Planning. So far there is no potential recommendation that may directly impact A-Class. However having someone from English wikipedia Assessment in Task Force Wikipedia Quality discussion would be plus. --KrebMarkt 07:35, 31 December 2009 (UTC)
Thanks for making that connection. I was originally going to be on that task force (I even got an email to that effect), but then I got asked instead to work on the Offline Releases task force, and I didn't think it appropriate to be active on both. Maybe I'll gatecrash a meeting - unless someone else from here wants to represent our work? Walkerma (talk) 15:38, 31 December 2009 (UTC)
You can only be member of one Task force team but everyone can participate to any Task force discussions. Randomran from Community Health Task force also give his opinion in Wikipedia Quality discussions and i usually hang around Community Health & Wikipedia Quality without being part of any Task force. --KrebMarkt 17:31, 31 December 2009 (UTC)

(unindent) Thanks KrebMarkt for raising this. I took a good look at the quality task force discussions. I see that you did discuss article assessment on around 29th, but it has been dormant since then. I do think believe that if you want to improve quality, you need to have assessment to measure that, though such a system should be kept simple and universal (which is what I think we have on en:WP). If an assessment thread pops up again, I'll be sure to add my 2c.

Meanwhile, as for reviving A-Class, I think the way forward for me personally is to start doing this much more at WP:CHEMS, including writing content to raise articles to A-Class, and then try to spread the interest (with help from people like your good self, if you can). I think that several projects already do something like what was intended for A-Class but they call it "strong B-Class"; if we can find the right tools to organize A-Class many of these strong B-Class might become A-Class again. Ideas are welcome, but personally I need to put in some hours of hard work first! Walkerma (talk) 06:15, 4 January 2010 (UTC)

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Content_assessment/Archive_5&oldid=1152587023"