Jump to content

Wikipedia talk:WikiProject Council/Assessment working group

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

First task: Setting up the group

[edit]

Contacts

[edit]

What do you guys think about importing this list into our list here? They've already listed themselves as contacts. §hepTalk 01:16, 2 March 2009 (UTC)[reply]

  • No need to import it, but there's no harm in linking it with an accurate description, eg "Wikipedia Council list of contacts for WikiProjects". I would guess that the list is somewhat out of date. Physchim62 (talk) 01:32, 2 March 2009 (UTC)[reply]
(edit conflict) A good thought, but unfortunately it looks as if most of the list dates from 2007, when WP:COUNCIL was more active. A few of them will probably still be appropriate, but quite a few will not. Maybe a personal invitation to comment would work? I don't think I'll have time tonight - I have 40 exams still to grade - but if you have time, go ahead. But don't just randomly post - check each user page first to see the person's status. I know some formerly active Council people like Badbilltucker are (sadly) no longer around. Walkerma (talk) 01:33, 2 March 2009 (UTC)[reply]
There's a toolserver tool somewhere that can check their status in mass. I'll see if any are active, and if they are still associated with their project and have not been surpassed by another as coord. If these match I'll compare my list against who has listed themselves here and leave a quick note about the project with whoever is left. Would active within the month be suitable, or should I make it shorter (ie a week)? §hepTalk 01:38, 2 March 2009 (UTC)[reply]
I don't get the impression that we're short of participants or points of view as it is. Maybe we could lay off the spamming for the moment, at the expense of taking a second round of consensus building at a later stage for the more far reaching proposals. Physchim62 (talk) 01:58, 2 March 2009 (UTC)[reply]

Here's the tool for those who'd like to know.Headbomb {ταλκκοντριβς – WP Physics} 07:50, 5 March 2009 (UTC)[reply]

Why do you need delegates?

[edit]

Feel free to reorganize this into another header if you wish.

OK, where on Earth did this part about "delegates" come from. I don't trust that a single person should be able to speak for an entire Wikiproject, especially for the larger ones. If you want to inform Wikiprojects of impending discussions, well, isn't that what Newsletter Bots are for? Though I have de facto-led a Wikiproject for the past while, I don't presume to speak for it. To me, this whole thing of delegates just seems like creating positions for the sake of creating them. Scrap that entire section, and put any Wikiprojects that want to be notified on a list. We should strive to promote egalitarianism, not build false power structures. NuclearWarfare (Talk) 02:54, 2 March 2009 (UTC)[reply]

That way the projects need only spammed to let them know of the project. Then a person/group of people can relay information and possible changes that are discussed here to WikiProject members that don't want to participate here but still need to participate at the "local" level. It creates less spamming and more communication between people in the long run. §hepTalk 02:58, 2 March 2009 (UTC)[reply]
(ec) Heh... you should have seen the page before. Anyway, the reason for the delegates is so that here are specific users to contact about new discussions, rather than spamming WikiProject talk pages every time a new discussion starts. Everyone's free to join in the discussions; we just want a list of users who are likely to be interested so that they can be notified quickly. -Drilnoth (talk) 02:59, 2 March 2009 (UTC)[reply]
This is illustrating an issue we're having. Everytime something even slightly -cratic happens, I feel parts of the Wikipedia community begin to panic for little reason. For most WikiProjects, representatives are more than likely are the only active people. I do agree with what NuclearWarfare is saying, we should have just had a mass invitation, and we should be careful about -cracy, but I don't think this is too much of a problem. It isn't WP:Esperanza. --pashtun ismailiyya 03:03, 2 March 2009 (UTC)[reply]
I'm also not sure why delegates/representatives are necessary. Can't interested folks also just watchlist these pages? Why the need for notification? And really, how often are large discussions that require lots of input really going to occur? I see the A-class discussion, but that's about it right now... --Rkitko (talk) 03:06, 2 March 2009 (UTC)[reply]
You can only feasibly hold so many large discussions on one pgae at once. I see there being mutliple large discussions over an extended period of time here. §hepTalk 03:10, 2 March 2009 (UTC)[reply]
Shep: If people want to participate, are you suggesting that they will be too scared to do so here but not too scared to do so at their wikiproject? And that even if they do participate at their project, is the delegate supposed to judge consensus and report back to here? Sorry, but that just smacks of easy process abuse and process creep to me. Also, WP:Archive and WP:Subpage do exist. Just utilize those on this page with a new discussion.
Pashtun: I'm not opposed to new processes by any chance. But if I don't see a need for something, I'm not going to support it. Simple as that.
Drilnoth: So, what will these specific users do to alert their Wikiproject. Won't they, you know, spam their Wikiproject's talk page? NuclearWarfare (Talk) 03:12, 2 March 2009 (UTC)[reply]
Bots spam, but it's not usually considered spam when an interested party alerts other potential interested parties on the talk pages of projects they're interested in. --KP Botany (talk) 03:20, 2 March 2009 (UTC)[reply]

(edit conflict) Why not just have an opt in as someone suggests above, then interested editors and/or projects can opt in to notification? I think the Wiki way is that you just watchlist what you're interested in. --KP Botany (talk) 03:15, 2 March 2009 (UTC)[reply]

(at NuclearWarfare): Hmm... you have a point. Maybe it would make sense to spam projects whenever there's a new discussion unless they opt out on this page? -Drilnoth (talk) 03:24, 2 March 2009 (UTC)[reply]

Or maybe you should take all of that system back to WP:COUNCIL where it came from and leave this page to a fairly fruitful discussion about assessment issues, which might well have different participants. Physchim62 (talk) 03:48, 2 March 2009 (UTC)[reply]

So what happens with projects who don't send a delagate? The LGBT project coordinator seems to find it undemocratic, and most members would agree, which leaves us with no contact? Can we get the spam to our project talk page?YobMod 14:25, 9 March 2009 (UTC)[reply]
Second that one as the majority of project have no representant. --KrebMarkt 14:27, 9 March 2009 (UTC)[reply]

Pagename

[edit]

I propose "Wikipedia:WikiProject coordination group". Headbomb {ταλκκοντριβς – WP Physics} 23:40, 5 March 2009 (UTC)[reply]

Agreed. –Drilnoth (TC) 23:42, 5 March 2009 (UTC)[reply]
Why not make this a subpage of Wikipedia:Version 1.0 Editorial Team/Assessment or move it to Wikipedia talk:Version 1.0 Editorial Team/Assessment/A-Class criteria, and restrict discussion to the single issue of A-class. The current page name and purpose appears to duplicate WikiProject Council. DrKiernan (talk) 14:33, 9 March 2009 (UTC)[reply]
Why should it be restricted to A-Class discussions? The WikiProject Council hasn't been doing a whole lot lately, so this could potentially create a better system for coordination between WikiProjects. That having been said, I wouldn't mind seeing this discussion being limited to A-Class if the Council started doing some more work to coordinate between existing WikiProjects. –Drilnoth (TC) 15:27, 9 March 2009 (UTC)[reply]
Like I said below, I'm quite willing to take a broad interpretation of "assessment", especially if that interpretation includes article improvement at all levels, but "coordination" for its own sake belongs at WP:COUNCIL, not at some ad hoc sub-group. The challenge, for those that wish to take it on, is to revitalise the council, not to parasite a group which has actually made progress in the restricted field that it has set itself, and might yet make more. Physchim62 (talk) 23:35, 10 March 2009 (UTC)[reply]

Counter-proposal

[edit]

Physchim62 (talk) 14:39, 9 March 2009 (UTC)[reply]

That's fine by me. DrKiernan (talk) 15:23, 9 March 2009 (UTC)[reply]
(ec) See my comment above. –Drilnoth (TC) 15:27, 9 March 2009 (UTC)[reply]

OK, I going to move the page back to it's old name, but I shaln't make the redirect just yet (that would complicate any further changes). I'm going on "no consensus for the original move" and "some consensus that 'assessment' means more than A-class" and that "assessment" can be interpreted widely. There is also a discussion on scope at the bottom of this page. Physchim62 (talk) 16:41, 9 March 2009 (UTC)[reply]


Question about format of this page

[edit]
This page is long enough already. :/ —Ed 17 (Talk / Contribs) 22:47, 3 March 2009 (UTC)[reply]

I think these discussions are moving a little too fast. The summaries seem to be assigning "consensus" to some topics that have barely been discussed, by a small number of people, with significant objections. 24 hours is not enough time to have a good discussion and come to consensus on any issue. A lot of people don't log in every day, or are not logged on all day. I worry that this method is going to rush the group into deciding that there is a "consensus" for something when there just hasn't been enough time-or participation-to figure that out. Karanacs (talk) 15:49, 3 March 2009 (UTC)[reply]

  • I'm not sure I've seen anything asserted in a summary that I've disagreed with, nor seen anything that has been deemed to be a consensus, and I'm kind of keen on what Walkerama is doing. I think I'm fairly opposed to some proposals to be able to make this a somewhat unbiased view, even though I am a contributor to the discussions. Hiding T 17:12, 3 March 2009 (UTC)[reply]
  • I've only seen two claims of consensus reached on this page: first, that there did not seem to be a consensus for eliminating A-class throughout Wikipedia; second, that there seems to be a consensus here to keep A-class. I think the first point is on solid ground, and given that absence of consensus for change leads to a continuation of the status quo, A-class would be kept. All the other claims of consensus seem to refer to previous decisions made elsewhere (individual WikiProjects, for example). I think the pace might be a little quick (I'd probably prefer a new question every 2 days), but I see the point, and I don't really have a problem with it. cmadler (talk) 18:12, 3 March 2009 (UTC)[reply]
  • This was an attempt to structure the discussion, and to make sure we covered the foundations of the issue before we moved to the applications. However, people are quite free to add comments to the earlier parts of the discussion; I'm reading those too, and I may add addenda to the summaries if something new and major develops. My experience of WP discussions, especially on issues where people hold strong but differing opinions, like this 2006 one, is that things can happen very fast, and this was my attempt to slow things down! I even had one person asking me to post the next question a bit sooner! I think that a 24 hour cycle means that every time zone has a fair chance to comment. My other experience of WP discussions of this sort is that all sorts of side discussions get started, on what colour the new X-Class template should be, or whether portals can be A-Class etc, and so the core discussion gets lost; we need to deal with the foundational points before we go to such details. I could easily have asked all three questions at once - indeed, we covered more topics in our two hour IRC discussion than we've done in the last three days - so I thought we were going quite slowly. If you know my style, I try to build a consensus over the long term (sometimes it goes against my personal views!), and I will oppose any attempt to make a major change without a clear mandate from the community. The little summaries are designed to do that - we don't want people's thoughtful opinions lost by the scroll bar. If there's any aspect you want me to answer or explain, please ask away. Cheers, Walkerma (talk) 19:40, 3 March 2009 (UTC)[reply]
  • As I said, I don't find it ideal, but I don't really have a problem with the pace. I think you (Walkerma) are doing an excellent job of moderating this discussion. cmadler (talk) 20:13, 3 March 2009 (UTC)[reply]
Agreed; this keeps us focused. —Ed 17 (Talk / Contribs) 22:23, 3 March 2009 (UTC)[reply]
Agreed as well. As a moderator, Walkerma has the confidence of many editors whose opinions are otherwise very diverse. I note that his moderation took this page from bickering about its own scope to constructive discussion literally overnight (in this time zone at least;). Physchim62 (talk) 22:39, 3 March 2009 (UTC)[reply]
Between you and Wilkerma, i think that we will get something useful out that discussion THAT is way better than how it was back at WP:ASSESS. --KrebMarkt 10:20, 4 March 2009 (UTC)[reply]

Coordination is hot

[edit]

bureaucracy less so... and why is this page move locked? I'm wary of the 'delegate' solution rather than letting anyone interested in spam sign on as a point of contact. +sj + 20:13, 5 March 2009 (UTC)[reply]

I will boldly create a section for non-delegate participants to sign up. Right now isn't the good moment to backtrack to the name issue. We lost already enough time bickering with it. First we create Something useful and palatable for any editors then we discuss the name/brand/tag we put on it. Funny that the current leading solution for A Class review is very decentralized. --KrebMarkt 20:35, 5 March 2009 (UTC)[reply]
It's intended that anyone can add their names to the list to represent a project with which they are involved. The page is move-locked because of previous edit warring over the name, although the lock should expire automatically in 2 days. –Drilnoth (TC) 20:41, 5 March 2009 (UTC)[reply]
I represent only myself here and there are also others contributors in the same case. I'm crazy bold to have created an additional section for i and them to sign up but i find that gesture Mandatory to show that discussion(s) and project(s) here are not just for Coordinators or so. --KrebMarkt 20:46, 5 March 2009 (UTC)[reply]
Excellent change. –Drilnoth (TC) 20:51, 5 March 2009 (UTC)[reply]
Move war and discussion of who can participate (answer: anyone) are on /Archive1. cmadler (talk) 21:05, 5 March 2009 (UTC)[reply]
Thanks! I was going to do that but I thought I was a bit too involved in some of those discussions not to attract controversy. Physchim62 (talk) 21:29, 5 March 2009 (UTC)[reply]
It wasn't me; I just pointed it out. cmadler (talk) 21:30, 5 March 2009 (UTC)[reply]

Scope

[edit]

I was wondering if Wikipedia talk:Template messages/Cleanup#More info? should be something discussed here, since each WikiProject could be a "topic" for template parameters. This would be taking it out of the "assessment" area, so a rename would be needed, but I think that that's where this project should be moving anyway. –Drilnoth (TC) 20:42, 5 March 2009 (UTC)[reply]

Perhaps, that is class not about the A-class system at all. It would be best to place it after item 3 on the agenda (if you feel it's worth talking here).Headbomb {ταλκκοντριβς – WP Physics} 23:44, 5 March 2009 (UTC)[reply]
Yes, it isn't related to A-Class. I just thought I'd bring it up now so that we could plan ahead. –Drilnoth (TC) 23:46, 5 March 2009 (UTC)[reply]
I don't even think that would be taking us too far from assessment, merely looking at a different point on the scale. It seems like a worthy idea, but I would need to be convinced that it's both workable and useful. There are other similar fixes which would need to be discussed at the same time, to try to ensure that we aren't somehow reinventing the wheel. Physchim62 (talk) 12:55, 6 March 2009 (UTC)[reply]
It's related to how articles are tagged for cleanup... how would that not be taking us too far from assessment? –Drilnoth (TC) 13:20, 6 March 2009 (UTC)[reply]

Scope revisited

[edit]

OK; there seems to be quite a schism here about whether this group should be focused on A-Class reviews or on general coordination between WikiProjects. There have been good points on both sides, so let's try to just voice our opinions here. –Drilnoth (TC) 15:30, 9 March 2009 (UTC)[reply]

  • General coordination: I believe that that was the original goal of this project. Since the Council itself hasn't been very active (and seems to generally help with creating guidelines for new projects, rather than organizing collaborations and other inter-project things), I think that this page can serve to help make the WikiProject system smoother and more efficient. –Drilnoth (TC) 15:30, 9 March 2009 (UTC)[reply]
  • Agreed - The name implies this. Not to say we can't talk about these other issues. But if this is an assessment discussion only then it should bear that name. :: Kevinalewis : (Talk Page)/(Desk) 16:01, 9 March 2009 (UTC)[reply]
  • One step at a time: I think that general coordination would be wonderful, but I don't think we have enough active people here to do that at the moment. There are over 1000 projects, as far as I know! I think that coordinating assessment is already a lot of work, and that is only a small part of the proposed scope. I will be absolutely thrilled if this page is still active two years from now! The WP:Council had many experienced and active Wikipedians - many more than we have here - and even that has fallen quiet, because of the commitment needed to organise such a large and diverse group. So let's do what we can handle for now; if we accrue a group of active and committed Wikipedians over the next year or two, then we may be able to grow the scope naturally. If we didn't accumulate a lot more long term help, I think that a grand coordination project would be dead within three months. Hopefully, though, we may be a catalyst to reawaken the Council. Walkerma (talk) 16:06, 9 March 2009 (UTC)[reply]
  • Assessment and improvement of articles. I think that many of Drilnoth's proposals could be fitted in within a wide interpretation of that header, but I also think that this group came together mainly to discuss assessment (not just A-class and whether we should "abolish" it or not). As a professional educator, I am convinced that there is no improvement without some sort of assessment, and that goes at all levels. Physchim62 (talk) 16:25, 9 March 2009 (UTC)[reply]
  • I agree with Walkerma and Physchim62; one step at a time and put the improvement of articles first. I'm not overly happy that the page was moved, but I'd rather not get sucked into meta discussion any more than I have to. Hiding T 16:39, 9 March 2009 (UTC)[reply]
  • General coodination → that way. Titoxd(?!? - cool stuff) 21:27, 9 March 2009 (UTC)[reply]
  • Agree that we should first consolidate our current scope before enlarging. right now the number of unique contributors to this discussion page is around 45-55 less than the number of representants who signed up. --KrebMarkt 07:59, 10 March 2009 (UTC)[reply]
  • General coordination, assessment as it first task, as outlined in the agenda.Headbomb {ταλκκοντριβς – WP Physics} 15:19, 11 March 2009 (UTC)[reply]

Some statistics

[edit]

I pulled these from the September 2008 statistics of the Version 0.7 Selection Bot, so they date from before the recent discussions (and the recent unilateral cull of A-class articles). I took the first 101 articles on the 0.7 importance score (101 because there was a tie for 100th place). I compared them between "quality" classes and the statistics of how "useful" the articles were. The "usefulness" criteria are: how many other articles link to the article in question; how many other language verions of wikipedia have an equivalent article; how many people visit the page over a period of "several weeks" (unspecified, but the same for all articles). I shall let the data speak for themselves. Physchim62 (talk) 21:31, 9 March 2009 (UTC)[reply]

Class No. articles Median page links Mean interwiki links Median hit count
FA 13 60,300 145 353,940
A 7 36,171 149 294,330
GA 8 14,864 130 262,335
B 72 16,857 128 157,590
Start 1 5,476 118 193,980

Proposals to reorganize the assessment scale

[edit]

Alternative discussion point: Separate assessment scales

[edit]

As something which could help standardize assessments across Wikipedia, I have another proposal: separate WikiProject assessments and Wikipedia assessments. WikiProject assessments would be D, C, B, A, and the criteria for each could be determined by each WikiProject separately, as they wouldn't be used for the v1.0 selection. The Wikipedia-wide selection would go Stub, Start, Good, Featured, which would measure the article's quality in comparison to all other Wikipedia articles.

My reasoning for this is that, right now, there kind of are two separate scales... Good and Featured don't seem to be in the same assessment scale as the others. This also causes much confusion about A-Class... is it above or below GA? If this idea was adopted, each project could choose how to use their A-Class assessments separately from other projects, but the Wikipedia-wide assessments would be universal the same way that Good and Featured are now.

As a less drastic change, we could just remove Good and Featured from the WikiProject assessment scale... allowing, for example, a B-Class Good article... but keeping the remaining assessments unchanged. Thoughts? -Drilnoth (talk) 15:04, 1 March 2009 (UTC)[reply]

As I mentioned above, such a proposal would have to be implemented very carefully to deal with the 5000+ articles which are currently GA- or FA-status. And I cannot see why you suggest that the project assessments should not be used for WP1.0, that's just nonsensical: if we have assessment information out there, we should use it. There is no "perfect data" or "objective quality", whatever some people would like to kid themselves. Indeed I think the hardest point of all is convincing people that we actually do have two separate scales at the moment, and that both are useful in their different ways. Physchim62 (talk) 17:03, 1 March 2009 (UTC)[reply]
I don't get this proposal at all. WikiProject are supposed to assess articles on behalf of WP1.0, not independantly. And I don't believe that we do have two seperate scales. PC78 (talk) 17:12, 1 March 2009 (UTC)[reply]
Article assessment is supposed to be useful for the WikiProjects themselves. The fact that WP1.0 also uses the data has certainly encouraged a lot more projects to get on board, but we are supposed to be creating an online encyclopedia as well as an offline one. Physchim62 (talk) 17:20, 1 March 2009 (UTC)[reply]
My hope with this idea is that the Wikipedia-wide assessment (used by 1.0) would not be nearly as subject to personal opinion as the current system... Stubs have stub templates, Good articles have passed a GAR, Featured articles have passed a FAR, and Start-Class articles are everything else. -Drilnoth (talk) 17:26, 1 March 2009 (UTC)[reply]
I think your falling into the fallacy of objective quality. Article quality is of necessity is a subjective concept, but that doesn't mean that judgements based (at least in part) on personal opinion are worthless: think of the billions of whatever currency which are spent each year on opinion polls and other forms of market research! Any system of assessment which pretends not to be based on personal opinion is in fact just based on the personal opinions of those who wrote the criteria and do the assessments: such people don't scale with the growth of Wikipedia, the WikiProjects can do. Physchim62 (talk) 17:42, 1 March 2009 (UTC)[reply]
I still don't see the benefit of entirely separating project assessment from WP1.0 assessments; Wikipedia is supposed to be a single project working towards a single goal. I don't believe that our assessments are so fatally flawed that they require such a drastic change. Let's try and stick to the issue at hand, eh? PC78 (talk) 17:31, 1 March 2009 (UTC)[reply]
One advantage of admitting that the systems are separate and independent is that they can provide a rough check on one another: an A-class article which fails GAN or an FA rated at B-class suggest real problems with the article which ought to be solved. Physchim62 (talk) 17:42, 1 March 2009 (UTC)[reply]
The systems are separte, sure; if we're talking about divorcing GA and FA/FL from the assessment scale then that's something that merits discussion, but let's take one thing at a time. But they don't represent a seperate "scale", and trying to draw a ring around FAC, GAC and stub sorting is a bit of a nonsense, frankly. PC78 (talk) 17:49, 1 March 2009 (UTC)[reply]

(outdent) They're separate but not orthogonal, that is, we would expect a high degree of correlation between the two scales but perhaps not perfect correlation in all cases. WP:MEASURE/A is an example that attempts to separate out the two scales while remaining within the general WP1.0 framework. Physchim62 (talk) 18:01, 1 March 2009 (UTC) (ec)(undent)Okay then, how about just separating GA and FA assessments from the main scale? Then individual projects can assess articles between Stub and A (giving A some more attention, hopefully), and the community as a whole can separately review articles to be Good or Featured. -Drilnoth (talk) 18:02, 1 March 2009 (UTC)[reply]

(response to Physchim): That's kind of what I was thinking for separating the scales, except that the parameters would actually be separate. So instead of just saying that all GA-Class articles are also B-Class, they would be B-Class with an additional "Good" parameter. That way, also, if the project thought that the article was better than GA but not ready for an FA review, they could say that it was A-Class without removing it from the GA categories. -Drilnoth (talk) 18:05, 1 March 2009 (UTC)[reply]
That's not at all difficult to code into the project banner, should a project want it. I think I'll do the experiment at WP:MEASURE/A: it can always be reversed if need be. See Talk:Kilogram for an example. Physchim62 (talk) 18:54, 1 March 2009 (UTC)[reply]
It seems like Smallville is A and GA class. Maybe it's already possible. - Peregrine Fisher (talk) (contribs) 19:00, 1 March 2009 (UTC)[reply]
Smallville is A-class instead of GA-class: Kilogram is B-class and GA-class, but Category:GA-Class Measurement articles has been taken out of Category:Measurement articles by quality to avoid double counting by the WP1.0 bot. Physchim62 (talk) 19:15, 1 March 2009 (UTC)[reply]
If GA and FA have been removed from the assessment scale at WP:MEASURE, do you really need to make note of these at all in the banner? It's already noted elsewhere on the talk page of Talk:Kilogram that the article has passed a GA review. PC78 (talk) 19:25, 1 March 2009 (UTC)[reply]
We would, to ensure that they are properly categorized (if the banners didn't have some GA or FA parameter, all of the articles would be put together without separation by topic/project). However, it would be possible to allow the cats to be added without any visual display, so that GA and FA are only mentioned once on a page, but still categorized via banners. -Drilnoth (talk) 19:51, 1 March 2009 (UTC)[reply]
It's easy to remove the visual note from the banner: in fact, the WP:MEASURE banner at Talk:Kilogram already has one such hidden parameter, to track vital articles. I would hide the note if it became more common not to use GA-class as a WikiProject assessment schemes, but I think there are only a handful of projects which work that way at present (WP:CHEMS is another example). Physchim62 (talk) 20:20, 1 March 2009 (UTC)[reply]
  • I don;t actually agree with the assertion that "there kind of are two separate scales". I think there are actually as many scales as there are reviewers, and the current system allows that kind of granularity that a more structured and detailed process would not allow for. I think there is only a problem here if you choose to see a problem here. No-one has to view the WikiProject assessments and Wikipedia assessments as separate, and a number of projects and editors are not. I mean, thinking things through, if you separate the processes out, what's going to stop some projects just using GA and FA as the review processes to determine their B and A class articles? I don't really understand the thinking that has gone into creating a problem and a solution to it. Hiding T 20:49, 1 March 2009 (UTC)[reply]
Indeed, I'm of the same view. Orderinchaos 00:47, 2 March 2009 (UTC)[reply]

Discussion on A-Class and FA-Class

[edit]

It occurs to me that if you follow the thinking to a logical conclusion, it would appear that it is possible some people think that an FA-Class article may not actually be as good as an A-Class one. I'm a bit puzzled by that. I thought it might be worth exploring that viewpoint, because I wanted to explore the possibility that a solid proposal might be for WikiProjects to unify A-Class assessment at Peer Review, and therefore better entrench the peer review process, A-Class and FA-Class in the system, as it were. Hiding T 21:01, 1 March 2009 (UTC)[reply]

A-class review at WP:PR is exactly the system used at WP:MEASURE, with the the added proviso that we've asked around related WikiProjects to try to find somebody – anybody! – who might know something about the Apothecaries' system. If we can't find anyone by the end of the review, we shall just have to go with our instincts and say that the article is "essentially complete" (bar a couple of minor points which might not even be addressed in the the published literature).
As for some FAs being just possibly not as good as some A-class articles, it is the opposite which would be truly amazing, given the different natures of the review process. This is especially true for the older FAs, although older A-class tend to be lower in quality as well. At WP:CHEM/WP:CHEMS we've just started a drive to review all our A-class articles and FAs against current A-class standards: we're probably going to have a lot of article improvement to do if we want to maintain some of the ratings. Physchim62 (talk) 22:27, 1 March 2009 (UTC)[reply]
I'm not sure I follow you. You seem to be suggesting that some A-CLass articles are just possibly not as good as FAs. Can you clarify that for me. Hiding T 22:35, 1 March 2009 (UTC)[reply]
No, our main worry (at WP:CHEMS) is that some of our older FAs are not up to our current A-class standards. The same probably applies to some of our older A-class articles. It would be best if we looked for the problems ourselves and tried to fix them, that gives us a little more control over the criteria and the timescale. As for FAs and A-class in general, it is nonsensical to say that FAs are systematically better than A-class, or vice versa for that matter: everything depends on how you define "better". Physchim62 (talk) 22:43, 1 March 2009 (UTC)[reply]
Technically and historically, FA-Class was designed to be above A-Class and was designed to be a step in quality above. That's the definition of better I'd be happy using, the one built into the scale from the start. Would you say your older FAs are up to FA standard? Hiding T 22:52, 1 March 2009 (UTC)[reply]

Historically, back in 2005 when the project assessment system was first developed, there were only handful of projects using it, and it was sort of assumed that projects would not put articles up to FA if they didn't think they were A-class. In any case, WikiProjects themselves were still in their infancy. It makes sense under that arrangement to have (FA+A)-class higher than a simple A-class: more people have looked at the article, and from different perspectives.

Fast forward to 2009, and only just over 10% of articles go to FAC with a WikiProject A-class stamp of approval, and the majority of those are from WP:MILHIST. That means that nearly 90% of the candidates for featured article status (slightly less if you look at promoted FAs, as articles which come up from the projects have a higher rate of success than others) haven't got a seal of approval from the corresponding WikiProject that they're "essentially complete". Now, of course, you can define FAs to be Wikipedia's "very best work" should you so wish, but it's not an obviously higher standard in the way it was intended to be when the WikiProject assessment scheme was set up. Everything depends on how you define article quality, and the A-class was designed to have a different quality criterion from FA, and a criterion which was (and still is) difficult to ensure during the FAC process. Physchim62 (talk) 23:46, 1 March 2009 (UTC)[reply]

I disagree that A-Class was designed to have a different quality criterion from FA, that's not my memory of how it was set up at all. I'll take your word regarding the fact that nearly 90% of the candidates for featured article status haven't been rated as A-Class by a WikiProject, because I haven't got the time to do the research, but I do not accept your conclusions or your proposed solution. Another way to gain a similar result to what you desire is for WikiProjects to have greater input into FA reviews. If every Project on the article talk page was notified when an article went in for FA, that would engage the community. But I'm starting to get a better grip of what happens now, see if you agree:
  • Article goes from C-Class to B-Class, with different projects using different review criteria.
  • Article goes from B-Class to GA-Class. Different WikiProjects handle this differently. I haven't worked out how differently yet, but it shows up importantly in the next step. I suspect some projects may skip GA-Class much as some projects skip A-Class.
  • Article goes from GA-Class to either A-Class or FA-Class. Different WikiProjects handle this differently. Some projects don;t bother with A-Class due to smallness of project or due to other reasons I'll outline below.
What's happening with some projects and some editors is this. Say I've got a C-Class article, and I work it up to a GA nomination. Now, if I'm a subject area specialist, I'm probably happy that it is B-Class, but I can't review it myself. I could put it in for a B-Class assessment, but I'm shooting for GA-Class, which I see as higher, so I go straight for that to save time. If it passes, well, I don;t need the B-Class review as much, do I? The same for FA-Class; if I've worked up a GA to the best it can be, I'm going to shoot for FA, not bother going A first, because I'm probably the best person to review it as A-Class anyway. If the articles fail their respective reviews, then I can see where A-Class might come in, but to my way of thinking, if an article failed FA then it isn't actually A-Class. Now if what we are looking to do is get more WikiProject involvement and assessment on our articles, especially our FA articles, which let's be honest, are perceived top of the tree, then I think the way to get that is by using Peer Review as a step before FA, and making that the A-Class review process. We could even change A-Class to become RA-Class, for Recommended Article, or PA, for Peer Article, or PRA, Peer Reviewed article. That way we can experts into the review process, we can get articles expert endorsed, and if we can get consensus that an article has to go through Peer Review before it gets to FA, we can ensure FA really is top of the tree. That's my proposed way forwards which would strike the middle path of saying that neither method currently used has to be deprecated. Thoughts? Hiding T 11:28, 2 March 2009 (UTC)[reply]
I agree with nearly all of your comments, except one of the last ones - saying anything "has" to happen assumes that there is something wrong with what happens presently that needs fixing. I do not see any need for a mandatory stage before FA - if the article meets the required standard, it will pass, if it doesn't, it will not. I'm all in favour of review (I often ask for it myself - the fact that very few ever respond is another matter) but introducing bureaucracy to fix a non-existent problem is something I think you'd struggle to find consensus for. Orderinchaos 11:44, 2 March 2009 (UTC)[reply]
Yeah, that's a good point. I think I was trying to ensure WikiProject participation was guaranteed during the process to make sure an FA was the best it could be. But here's the rub: If you take a C-Class article and rewrite it to what you think is FA standard, and then run at FA-Class, what is the article if it fails? Maybe the solution would be to have one grand review process, where articles were assessed for every thing higher than a C-Class. That would be the ideal, I grant it would never work, but it would be nice if we could just go to one place, and after discussion it was decided it needs this for FA, but at present it's A-Class, or that it's GA or B-class, and that the pointers needed to scale each class were given. Hiding T 12:41, 2 March 2009 (UTC)[reply]
We'll just have to disagree on "how it was all set up" ;) The main "problem" is that GAN and FAC review a minuscule number of articles compared to the WikiProjects and yet they are still perpetually backlogged. Some subsidiary problems include the concentration on a single editor taking the same article through the whole process (rather than promoting collaborative editing) and the fact that GAN and FAC have no mechanisms at all for gearing article improvement towards the articles that people are actually reading or would want to read if they were good enough. If you want to believe that FA and GA are better just because they are "community-based", I can provide you with contrary examples! Personally, I'm quite well known for my "featured crud" posts, but that is not the topic we're discussing here. I'm quite happy to accept that FAC and A-class are judging quality by different criteria and so will sometimes come to different results. What is ridiculous is to say that FAC is always better, just because it has been defined to be better: that is a completely circular argument. Physchim62 (talk) 15:03, 2 March 2009 (UTC)[reply]

A proposal

[edit]
  • Following the discussion above, it would seem usefull to consider seperating GA/FA from Stub/Start/C/B/A. This would finally resolved the confusion about what are the distinctions between A/B/GA. The banners can easily be modified to handle this, especially the metabanner, adding a |wikiclass=G/F parameter, leaving class=Stub/Start/C/B/A for projects. This way we'd know that an article is both GA and A class (or B or C class). FA would then automatically imply A-class for the project (unless for some reason they disagree, but those reason would probably have been mentionned the FA, or lead to an FAR).
Wikipedia
(|wikiclass=)
Project
(|class=)
Type
(|type=)
Featured A* Article (default), List, Topic, Portal
Good** A*** Article (default), List, Topic, Portal
B Article (default), List, Topic, Portal
C Article (default), List, Topic, Portal
Start Article (default), List, Topic, Portal
Stub Article (default), List, Topic, Portal
NA NA Template, Category, Image, Disambig,
Redirect, Project, Future, Merge, Needed ...
*A-class isn't a prerequisite for Featured, I simply mean that Featured implies an A-Class at the project level.
**Good is still given after a GAN and is independant of project class
***What A-Class means in this case is left to the project
  • Support: A much-needed change... I'd prefer to have Stub and Start moved onto the same scale as Good and Featured, as I outlined above, but this is still a great step forward in article assessment. -Drilnoth (talk) 23:54, 1 March 2009 (UTC)[reply]
    • Comment: No offense, but I dislike the idea of moving Stub- and Start-Class out of Project control. It's not that I don't see the logic of it, but I foresee a massive backlog of those classes never being assigned if the projects don't do it. hornoir (talk) 00:30, 2 March 2009 (UTC)[reply]
  • Support: I think that this makes perfect sense. It splits out the project-reviewed content from the community-reviewed content and makes it, IMHO, more efficient. I especially like the fact that lists can now be "classed" like articles. KV5Squawk boxFight on! 00:09, 2 March 2009 (UTC)[reply]
  • Qualified support. Ideally the project class should show "unclassified" if it hasn't actually been classified by a project: we shouldn't be pretending to have done quality control checks which we haven't actually done. On the other hand neither should I let my personal idea of "perfect" get in the way of something that is Good. I prefer this proposal to Drilnoth's original proposal for similarly technical reasons: for the 40+% of articles between Stub and GA, there is no community assessment of quality, so it seems better to retain the information gained from WikiProject assessments rather than dilute it in a broad "WikiStart-class". Physchim62 (talk) 00:25, 2 March 2009 (UTC)[reply]
    • Further quaification to support. While I welcome a concrete proposal, maybe it is a little premature to suggest doing this before discussing the nature of A-class reviews, which I believe is tomorrow's topic in the general scheme of trying to get ourselves to discuss constructively instead of just bickering ;) Physchim62 (talk) 01:10, 2 March 2009 (UTC)[reply]
  • Support: This is what I was already considering proposing to my project, essentially. Though, of course, there doesn't have to be a direct correlation between Project-Class and Wikipedia-Class. FA- and GA-Classes are already recorded elsewhere and don't need the project banner to help categorize them. Though we could create subclasses to denote this for Editorial, if they wish; Featured A-Class ??? articles, Good C-Class ??? articles, etc. hornoir (talk) 00:30, 2 March 2009 (UTC)[reply]
  • Support This is probably the easiest path to go down. It leaves how to assess "A" to individual wikiprojects who may choose not to have any process or have guidelines to follow. maclean 00:36, 2 March 2009 (UTC)[reply]
  • Oppose This ironically manages to increase the bureaucracy yet decrease the transparency - not enough scrutiny and no paper trail. The good thing about FA and GA is we know how they got there, you can follow that process and challenge it if necessary. I can foresee significant edit warring between vested contributors and their opponents over ratings were A to replace GA and FA. Orderinchaos 00:44, 2 March 2009 (UTC)[reply]
  • I don't understand your comments. "Not enough scrutiny" – surely there is more scrutiny than under the current system. "No paper trail" – a problem that could easily be solved for a small project by doing its review at WP:PR, and hardly insurmountable for a project large enough to have an internal review system. Physchim62 (talk) 01:10, 2 March 2009 (UTC)[reply]
  • WP:PR has been basically dead for years, so that is not an alternative. The fact is, at GA and FA, there is a centralised paper trail. With this "A-class", there is absolutely none. There's no central idea as to what it means, or how it would be determined or enforced, and keep in mind we're trying to preach to a media who already has a rather bemused or negative view of us. Orderinchaos 01:33, 2 March 2009 (UTC)[reply]
  • And so "central" is supposed to mean "scrutiny"; "paper trail" is supposed to mean "transparency" is it? Don't make me laugh! The best way to "preach to a media" is not to pretend to do things of which we're patently not capable, and the centralised structures are at least as guilty of that as the WikiProjects. If your only objection is that the WikiProjects can't be trusted even with assessing the articles in their own field and even when centralised mechanisms are retained as an 'external' opinion as to article quality, then that objection seems not only utterly, utterly ridiculous to me, but quite frankly offensive. Physchim62 (talk) 02:28, 2 March 2009 (UTC)[reply]
  • For one thing, let's get one thing straight - WikiProjects do not own articles. Secondly, while some projects have stable and capable review staff, by far the majority do not - many are struggling to attract volunteers just to edit articles, and review processes are stuck in an eternal backlog. Therefore such "review" becomes the opinion of one person with more time, rather than a considered decision coming out of a process. This is fine for anything up to say C- or B-class - but for our highest end of articles, the group we effectively say "these are our best, or pretty damn close", I think there should be project-wide scrutiny, a paper trail, and an opportunity to review at any time. Encouraging ownership by a small group of people who may be time-limited, overworked and otherwise committed, damages these aims. Orderinchaos 03:27, 2 March 2009 (UTC)[reply]
  • Well, I don't see why the poor state of many current WikiProjects should be taken as a fixed assumption. There's no fundamental reason why we can't have more capable projects, if people are willing to make the necessary structural adjustments. Kirill [pf] 03:47, 2 March 2009 (UTC)[reply]
  • Comment: I don't see why you systematiaclly oppose everything that is related to this page. There is no "secret discussion". It's on {{cent}}, a crapload of wikiprojects have been notified, etc... I don't see how bureaucracy would increase. It's not "more" bureaucracy, it's simply realizing that the way things are not is not very structured, and very confusing. Headbomb {ταλκκοντριβς – WP Physics} 02:00, 2 March 2009 (UTC)[reply]
    • I don't "systematically oppose" anything. I support common sense, and that common sense tells me that increasing bureaucracy (creating a "WikiProject" layer as an official, rather than unofficial, mode) and reducing transparency (giving decision making powers to those WikiProjects rather than to the community) are bad things. WikiProjects can only work when they're simply projects with an aim - when they start assuming trappings of government, we end up having "article tsars" and "my project is better than your project" wars and internal politics within previously harmonious WikiProjects and secession issues. The notion that things *need* to be centrally structured is a bit of a fallacy anyway - we need central institutions, but not a central government. Orderinchaos 03:33, 2 March 2009 (UTC)[reply]
  • Oppose. Too much screwing around with something that isn't so broken. I'm not convinced that we really do need to divorce GA/FA/FL from the assessment scale, but if we do then there's no real need for the two to have a direct correlation. And the last thing we need to be thinking about is adding a |type= parameter. PC78 (talk) 00:45, 2 March 2009 (UTC)[reply]
  • I'm not focusing on the "technical deatils", I just find this whole proposal highly unnecessary. Our assessment scale is not so fundamentally flawed that it requires this level of restructuring. PC78 (talk) 11:52, 2 March 2009 (UTC)[reply]
  • Weak Support. I think this approach at a minimum lays out that FA and GA are related to the other classes but also different. I think that the devil with this approach would lie in the details. Changing to this model would require a lot of updating, but that should not be a reason to oppose. Vegaswikian (talk) 01:17, 2 March 2009 (UTC)[reply]
  • Yes the details could be problematic, but it's nothing a good coordination between a)the metabanner coders, b) bot coders, c) WikiProjects can't prevent. I have extensive experience with templates on wikipedia and I can personally vouch for the non-disruptiveness of a well-made transition. For example, and I'm giving this as an possible way to go at things, we could set up a "metabanner2" which supports the new wikiclass/class/type thing (facilitate category handling etc...). WikiProjects could then migrate to the new banner simply by switching "metabanner" to "metabanner2", then request bots to do the dirty work (which would simply be making the change |class=GA → |wikiclass=Good |class=, and |class=FA → |wikiclass=Featured |class=A (or |wikiclass=Featured |class= if the projects doesn't want A-class). We're talking about 2,500 FA and 6,000GA which is ridiculously low as far as bots are concerned. Wikiprojects who uses special banners could choose to either go with the new metabanners, or to adapt their existing banners, which amounts to rewriting 25 lines of codes at worse. I know plenty of people who would be more than glad to help adapt the WikiProjects banners so that WikiProjects who, for whatever reason, don't want the extended system can keep doing business as usual, including myself.Headbomb {ταλκκοντριβς – WP Physics} 02:00, 2 March 2009 (UTC)[reply]
  • That's not the only possible solution to the technical difficulties. A bot operator could have the new "wikiclasses" on all the necessary templates in less than an hour, without them (the codes) doing anything for the moment: after that, projects could simply opt in by setting a usewikiclasses parameter. On the other hand, I think we've still got a fair bit of discussion to go before getting to that point! Physchim62 (talk) 02:14, 2 March 2009 (UTC)[reply]
  • Yes that is what I mean by not getting bogged down in details. There are dozens of ways to do this, I was just listing one of them to show that this really isn't that big a deal in term of implementations.Headbomb {ταλκκοντριβς – WP Physics} 02:27, 2 March 2009 (UTC)[reply]
  • I would put it slightly differently! The technical problems are minor if there is consensus that this is a good road to go down for at least some (note plural) WikiProjects. I'd like to see a little more consensus (or at least approximation of opinions) before suggesting it outside of WP:MEASURE (where I'm the only person currently doing assessments) or the chemistry projects (where I'd need the consensus of the projects, but where we already have a very similar system). Physchim62 (talk) 02:52, 2 March 2009 (UTC)[reply]
  • We're not doing anything soon. If there is consensus here for such a change, we'll probably create a CENT subpage dedicated solely to it to ensure that there is the consensus of the entire community and not just the users watching this page. Right now we are suggesting it for use outside of WP:MEASURE, but aren't going to implement anything until there's been plenty of input. -Drilnoth (talk) 02:54, 2 March 2009 (UTC)[reply]
  • Yes, we'll need consensus for this, but consensus isn't a technical problem. I'm just saying that we shouldn't let the fear of technical problems overwhelm us, as there would be little.Headbomb {ταλκκοντριβς – WP Physics} 03:12, 2 March 2009 (UTC)[reply]
  • Support That seems perfectly logical. A long-neeed change to a system that is trying to stay up-to-date using older mechanisms than what it should be using. §hepTalk 02:30, 2 March 2009 (UTC)[reply]
  • Oppose - Featured status should always be the be-all and end all. There is no reason that an article should need approval from the wikiprojects, which operate under a much less scrutinized system, to be considered Wikipedia's best work if it has already received approval from FAC. If an editor has concerns about a process, the very clear FARC is also there. A-class should most definitely be subordinate the FA. Also, give me one instance where a project would rate something as C class, and the current system would rate it as GA. If you can't, then this process is simply process creep.
    • On that note, I'm still not sure why you all want to increase the prominence of Wikiprojects. While they are nice and useful for finding other users who like to write about what you do, why are they so special that the entire community shouldn't be joining in on their discussions? NuclearWarfare (Talk) 03:05, 2 March 2009 (UTC)[reply]
How is this process creep/buro? A-class is not proposed to be required for FA to be awarded. If anything your proposal is the creepy thing, with its "A class as "a subordinate" of FA-class" thing. Headbomb {ταλκκοντριβς – WP Physics} 03:16, 2 March 2009 (UTC)[reply]
  • Oppose, simplicity is best. Less classes, less confusion. There should be a linear slope of quality everyone can use as a standard, no segregation involved. I know many others have the experiences of dealing with invaluable editors who just can't wrap their heads around the intricacies of Wikipedia. Why confuse these wonderful contributors even more? I'd love to drop A-Class altogether! --pashtun ismailiyya 03:17, 2 March 2009 (UTC)[reply]
Don't use things? I'm not concerned with types which are simple as well as needed, but I'll take your comment and apply it to classes. If this is put into action, we have two standards with which to measure articles. Should each project then choose which (if not both) standard to go along with? Those who choose the ABC-Class route don't shoot for global Wikipedia standards, those who choose the GAFA-Class route abandon the measurement of articles underneath those levels, and those who choose both will end up spending time explaining to countless users what could have very well been a self-explanatory system. This organizing of article classes will make things even more complex than they already are. Can I abandon the usage of A-Class articles in WikiProjects I am coordinating? Sure, but abandoning something that exists across most WikiProjects creates issues for the minority. --pashtun ismailiyya 06:48, 2 March 2009 (UTC)[reply]
  • (ec)Strong Oppose - I happen to think that this has the potential of being far more unnecessarily confusing than our current system (similar to the oppose above by Pashtun Ismailiyya). And sorry Headbomb, but some of the technical details do matter; we should've learnt by now what happens when we don't pay those details enough attention from the beginning. Community assessment processes need to be as much part of the WikiProject assessment processes to maintain some sense of consistency and to maintain a sense of high standards, at least for those articles that are at the top of the spectrum. If WikiProjects are going to award grades that do not reflect those that were given by the wider community (at the very least, for those articles at the top of the spectrum of the quality scale), then not only does this limit the useful purpose of having wider-community assessment processes, but it will result in chaos. Although I disagreed with Orderinchaos comments at the MFD, his reasons for opposing are quite in line with the practical implications of enacting this proposal. Ncmvocalist (talk) 03:23, 2 March 2009 (UTC)[reply]
  • The technical details do matter when it's time to implement things, not when it's time to discuss what we want to do. There are plenty of ways to make sure this is near-problem free. I listed one, someone else suggested an alternative. Getting bogged down in technical details at this point is like have a debate about whether or not you need a steel structure or a wooden structure, but don't know what you'll build. Headbomb {ταλκκοντριβς – WP Physics} 03:43, 2 March 2009 (UTC)[reply]
  • Well here's where we have to respectfully agree to disagree; while some of the technical details can wait until implementation time, my position is that some do matter in initial stages as I've noted above. The ones that do matter go hand in hand with the debate. Getting bogged down with those technical details is important even on this proposal; we can see what is trying to be built in this proposal, but we can also see what structure this proposal uses - it's just wrong. Let's avoid demolishing something after spending so much time building it, and instead, make a better effort of getting it right the first time. Cheers, Ncmvocalist (talk) 07:00, 2 March 2009 (UTC)[reply]
  • Regardless of what's done to FA and GA as it relates to their position in the SSCBA system (I'm opposed to separation of the scales, but I'm warming up to the idea), conflating the addition of a type parameter on top of moving GA and FA is too much for one discussion. But now that we're at it, I'm generally opposed to implementing it, as its usefulness is limited. Titoxd(?!? - cool stuff) 10:22, 2 March 2009 (UTC)[reply]
  • Oppose, and this is exactly what I've been arguing against. This is imposing a certain viewpoint on how assessments should happen when that isn't either the consensus view, the consensus practise or the best way of doing it. The current system actually works very well. Any WikiProject who doesn't think their FA articles are A-Class should delist them, because completeness is one of the criteria required for an FA. And at WP:COMICS we would be very reluctant to equate C-Class articles with GA, and if any were brought to our attention we'd ask for them to be reviewed in that light. Hiding T 11:10, 2 March 2009 (UTC)[reply]
  • Support, and not before time ;) Separating FA and GA and removing GA-Class will visibly set apart the community-wide standardised assessments from the various project assessment scales, and although I appreciate some of the objections raised, I think the benefits greatly outweigh the disadvantages. The bulk of the confusion with the current system seems to arise from GA's place in the assessment hierarchy, and while removing it altogether creates a new debate over where the C/B/A equivalences should lie, I believe this can be left to individual projects to decide. EyeSerenetalk 12:54, 2 March 2009 (UTC)[reply]
  • Support - To me this is a clearer and better system than before Cabe6403 (TalkSign!) 14:49, 2 March 2009 (UTC)[reply]
  • Oppose - Current one-tier system works OK. I cannot see how this will help to improve a single article. (I'd prefer dropping A-class) Djanga 15:10, 2 March 2009 (UTC)[reply]
The current system does not work A-OK. The problem is that we are using one class for two distinct things, "wikipedia-wide" assessment, and wikiproject assessment. If a project rates an article B-class, and then passes GA, it looks like GA supercedes B, which is does not. Same goes for an article that is A-class then goes to GA. If A-class is kept, the GA assessment is lost (in terms of wikibanners and categories). If the GA assessment is chosen, then A-class is lost. There's also the problem of not being able to assess lists. Right now we have "list" and "featured list". If you want to assess a list as B-class, then you lose the list class. If you want to tag a B-class list as a list, then you lose the B-class. This is not so much about improving articles as it is about making things clearer, but this has the added consequence of finally allowing people to assess lists, projects, topics, etc... Currently you can have good topic and featured topic, but other than that you can't rate your topics.Headbomb {ταλκκοντριβς – WP Physics} 18:02, 2 March 2009 (UTC)[reply]
  • No objection - but I wouldn't have an objection if GA and FA were slotted into the WikiProject assessment scheme either. I would support any solution that clarified the existing system and made it clear to people where GA and FA fitted in with the WikiProject assessment scheme. What is important here is that an assessment scheme helps drive progress on our articles. The GA and FA schemes are very successful, and while the WikiProject assessment scheme is patchy, it can work well in some WikiProjects. Where I am more concerned is that the WikiProject assessment scheme is set up under the WikiProject tag, and so individual contributors may hestitate to assess the article when reading it if they haven't signed up to the WikiProject(s) who have tagged the article. Any critical assessment of any article is helpful and should be encouraged. Giving that some thought, it may even be helpful to have a Reader's Assessment template in which any reader of any article can tick a simple assessment box for grading an article on readability and satisfaction. It could be a culmative assessment giving an averaged rating based on each reader's assessment. This would be a confidence booster for the editors (and WikiProjects) involved as they see audience response to articles they have worked on, as well as useful feedback on which articles need attention and which ones are working. Currently our only assessment is "inhouse", so it might be useful to get a third stream alongside the GA/FA stream, and the WikiProject Assessment stream. SilkTork *YES! 16:02, 2 March 2009 (UTC)[reply]
  • This proposal seems two include two changes (separating GA/FA from SSCBA, establishing a "type" parameter), and I generally support both of them. My one concern is that I'd be a bit skeptical of C-class articles being Good Articles; I suspect that in most cases a Good Article will be B-class quality or better. cmadler (talk) 18:50, 2 March 2009 (UTC)[reply]
    • That's a point a like about this proposal: it provides an internal check! If we have C-class Good articles or B-class (new) Featured articles, it tells us immediately that there's a problem with one of the two assessment schemes. The scheme at fault might be the central one (GA/FA) or the decentralised one (the WikiProjects) but the important point for the Community as a whole is to know where these problems are, not to pretend that they don't exist as is done with the current system of conflating the grades and then continually arguing as to whther A-class is better than GA or vice versa. Physchim62 (talk) 19:25, 2 March 2009 (UTC)[reply]
      • I think that, absent an article-specific rationale, a Good Article should be presumed to be at least B-class, and a Featured Article presumed to be A-class. That's not to say that a lower-classed article couldn't achieve one of those designations, but that 1) I'd expect that to be a rarity and 2) in those cases, I'd expect to find a specific explanation of the discrepancy documented on the article talk page (or article talk archive). cmadler (talk) 21:25, 2 March 2009 (UTC)[reply]
  • Oppose The system is good as is. --Rschen7754 (T C) 20:11, 2 March 2009 (UTC)[reply]
  • Oppose. I don't really see the point in splitting out the two systems. That seems to be putting a little extra burden on wikiprojects. I'm practically the only one in WP:WikiProject Texas, which has a large membership, that does any assessments at all. None of the wikiprojects I belong to have any sort of scheme for identifying articles as A-class, so I suspect that they'll likely just auto-label an FA or even GA as A-class. If a large number of wikiprojects end up using the more community-wide processes to assign A-class (or even B-class), then what is the point of separating the two assessment ranking systems? It will just muddy the waters more - did the project actually review to see if the article was good, or are they just accepting the other assessment level? It doesn't appear that any problem will be solved by this, and more confusion might be created. Karanacs (talk) 21:27, 2 March 2009 (UTC)[reply]
  • Slightly surprising myself here, but I must oppose this. The smaller WikiProjects - and many of the larger ones like TEXAS above - will just end up using the community-wide assessment schemes because of a few number of reviewers/assessors...so what is the point of splitting these out? The only projects that could be possibly be positively affected are the dedicated people in MILHIST and maybe a couple others who ensure that all of their assessments are current and up-to-date. While I'm a coordinator in MILHIST, I can't in good conscience support this when it would be detrimental to so many other projects.
  • Having said that, I think that this would be a good time to discuss where, exactly, does GA fit into a project assessment scheme? Also, I would support the introduction of a |type=. —Ed 17 (Talk / Contribs) 21:39, 2 March 2009 (UTC)[reply]
  • Surely you'll see the advantage of splitting the "feature and good" from the SSCBA scheme when introducing the "type" parameter. Other wise you'll end up with class=GA, type=List or FA-Class, type=Portal, or a slew of new classes such as Stub-T, Start-T, CT, BT, GT, AT, FT, Stub-P, Start-P, CP, BP, GP, AP, FP, etc...Headbomb {ταλκκοντριβς – WP Physics} 00:22, 3 March 2009 (UTC)[reply]
  • From the perspective of a project assessment scheme, GA is an "outside check", not a grade in itself. From the perspective of the Wikipedia community as a whole, GA is an "outside check" on the article, and maybe also on the project assessment scheme if things are going badly (eg A-class article fails miserably at GAN). If you keep GA within the project assessment scheme, you lose this checking tool. Physchim62 (talk) 01:18, 3 March 2009 (UTC)[reply]
Why would an individual WikiProject's opinion matter? I don't know any WikiProject that wants to get an article up to their personal standards, rather they want to shoot for a community wide standard. Not only does this segregation seem weird, it is implying that WikiProjects either have, or should have, some separate standard that is separate from the entire goals and community of Wikipedia. That's not what WikiProjects are for, and have never been for that. They work on a selection of articles towards an encyclopedic standard and not for a personal view of how articles should be produced, at least that has been my understanding of it. --pashtun ismailiyya 07:32, 3 March 2009 (UTC)[reply]
I think that what Physchim62 is trying to say is that the community has essentially given WikiProjects the task of assessing articles using the community-developed SSCBA system. (After all, SSCBA is used as a common grading rubric for WP:1.0 and other assorted purposes.) GA and FA serve as vectors through which the community can verify that the WikiProject is not being too lenient in its assessments, or being too strict. This provides an additional tool which allows us to be more confident that the comparison between articles' qualities is more of an apples-to-apples comparison, instead of an apples-to-oranges verification. But ultimately, you're right: WikiProjects strive for community-wide recognition of the quality of their articles. That's not in disagreement with the WikiProject-based distributed assessment system. Titoxd(?!? - cool stuff) 08:12, 3 March 2009 (UTC)[reply]
Yeah we're in agreement, it wasn't so much directed at Physchim62, rather the people who are defending the issue. --pashtun ismailiyya 08:19, 3 March 2009 (UTC)[reply]
Oppose I think both individual projects and the overall project are all working toward improving quality of individual articles. I think separating the efforts may discourages smaller projects that need the project wide asseessment assistance.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTM) 04:19, 4 March 2009 (UTC)[reply]
Oppose After reading all of this (and I do mean all of it), I have to say that ultimately, the system isn't broken. So there's no need to "fix it".--Kung Fu Man (talk) 18:00, 4 March 2009 (UTC)[reply]
  • Oppose well intended but eye-rollingly complex when we already have enough issues. In one project we simply don't have the people power to maintain bumping articles up and down by scale. In part this is because it's not readily apparent what the criteria are and who cares. I suggest pushing for a universal scale to be adopted and phased in with consultation from the good folks at Featured Articles on what would serve our readers and editors best. The system should be intuitive and easy to use. -- Banjeboi 02:38, 5 March 2009 (UTC)[reply]
Okay, I really don't get what's complex about this. Is is it a good article? If yes, then wikiclass=Good? Is is a featured article? If yes, then wikiclass=Featured. Is the page a list? Then type=list. Etc... That's all there is to is. What's so hard to understand about this? What's the confusion?Headbomb {ταλκκοντριβς – WP Physics} 08:07, 5 March 2009 (UTC)[reply]
Alright, I can sort of get behind this. Let me take this back to WP:COMICS. Although, I think images can be featured, and thinking about it, it might be an idea to assess images. I mean, some images are poor quality, but when they're free, you take what you can get. It might be a way of letting people know, hey, this image of Zach Snyder is pretty poor, can you get a better one? Hiding T 10:10, 5 March 2009 (UTC)[reply]
Having FA, GA as well as ABC is confusing. Maybe not to you but personally I avoid assigning these parameters because I don't want to mess them up. -- Banjeboi 10:48, 5 March 2009 (UTC)[reply]
I'm not sure I understand you. Are you saying you don't assess articles currently? Hiding T 10:58, 5 March 2009 (UTC)[reply]
Exactly, others do but I do not. I also steer away from categorizing for similar reasons. I'm ok adding them but in an effort to not screw something up I avoid deleting them unless it's obvious. -- Banjeboi 11:01, 5 March 2009 (UTC)[reply]
Okay. So, and correct me if I am wrong, but from your point of view it wouldn't actually matter what the system used was, because you wouldn;t use it? Would that be correct? Hiding T 11:18, 5 March 2009 (UTC)[reply]
No. I'd like a system that was simplified and intuitive so even I would have to work hard at screwing up an assessment. My hunch is that it should be across all projects. It may even make sense that use of ratings within a project's template is unneeded as each article would simply have a standard stand-alone assessment and bots could pull from it instead. It's never made sense to me why different project would have different standards for the same article. We should work together to improve articles and when the article advances we all share that credit. If the system were understandable I would likely use it and appreciate it. -- Banjeboi 11:41, 5 March 2009 (UTC)[reply]
I'm not sure how we don't all share the credit at the moment. The reason different projects have different standards is because they look for different things. Each project is making sure its field has been covered adequately. I'm also unsure how much simpler the current system could be. All you need to do is add something like |class=A to a template. Anything else would still involve editing a template. I don't think you can ever get a one size fits all assessment, nor that such a thing should be a goal. Hiding T 12:51, 5 March 2009 (UTC)[reply]

Proposal to add Reader Assessment

[edit]

Following my comments above in response to Headbomb's proposal, I've added the Reader Assessment into the scheme to see how it would look.

Wikipedia
(|wikiclass=)
Project
(|class=)
Reader - readability
(|readability=)
Reader - information
(|information=)
Type
(|type=)
Featured A* Professional Professional Article (default), List, Topic, Portal
Good** A*** Enjoyable Expert Article (default), List, Topic, Portal
B Readable Excellent Article (default), List, Topic, Portal
C Readable Satisfactory Article (default), List, Topic, Portal
Start Unclear Poor Article (default), List, Topic, Portal
Stub Unclear Not enough Article (default), List, Topic, Portal
NA NA NA Template, Category, Image, Disambig,
Redirect, Project, Future, Merge, Needed ...
*A-class isn't a prerequisite for Featured, I simply mean that Featured implies an A-Class at the project level.
**Good is still given after a GAN and is independant of project class
***What A-Class means in this case is left to the project

I copy my comments from above: Where I am [] concerned is that the WikiProject assessment scheme is set up under the WikiProject tag, and so individual contributors may hestitate to assess the article when reading it if they haven't signed up to the WikiProject(s) who have tagged the article. Any critical assessment of any article is helpful and should be encouraged. Giving that some thought, it may even be helpful to have a Reader's Assessment template in which any reader of any article can tick a simple assessment box for grading an article on readability and satisfaction. It could be a culmative assessment giving an averaged rating based on each reader's assessment. This would be a confidence booster for the editors (and WikiProjects) involved as they see audience response to articles they have worked on, as well as useful feedback on which articles need attention and which ones are working. Currently our only assessment is "inhouse", so it might be useful to get a third stream alongside the GA/FA stream, and the WikiProject Assessment stream.

I'm not sure how the culmative scores would be managed within the template - or even if it is possible - but I feel that would be an important aspect of the scheme. SilkTork *YES! 16:38, 2 March 2009 (UTC)[reply]

That is way too complicated. Some editors find it hard to understand the system as it is. --Rschen7754 (T C) 20:10, 2 March 2009 (UTC)[reply]
I think that it's a great concept, but it would probably be much too confusing for both readers and experienced editors. I could see having a single "Rate this article"-type thing that readers could use, which might be helpful, but not this whole system. -Drilnoth (talk) 22:34, 2 March 2009 (UTC)[reply]
I see that as rather unnecessary, and can't think of any project that would use it. If the interest is there, the option could be made availible as an "opt-in" thing on the metabanner. But as far as reader are concerned, very few would even be aware of this. If you want to improve reader feedback, changing the "talk" button to "Feedback" would probably do wonders.Headbomb {ταλκκοντριβς – WP Physics} 02:52, 3 March 2009 (UTC)[reply]
Suffers from tl;dr, unfortunately. This is way too complex. Titoxd(?!? - cool stuff) 07:01, 3 March 2009 (UTC)[reply]
I think the idea behind this is good, but I think we'd need to find a way to implement it simply and efficiently. There is also a risk of "drive-by" assessments, which are fine for stubs and starts, but not for anything beyond that. I think, though, that a certain amount of this sort of thing already occurs between wikiprojects already - I tag something as B-Class for WP:CHEMS, and someone from WP:Food and Drink or WP:Physics sees their project's article assessed by an outsider. Walkerma (talk) 07:31, 3 March 2009 (UTC)[reply]
But at the same time, it gets rid of the main benefit of the current scheme: It is linear, and hence it is simple (maybe too simple, or otherwise we wouldn't be writing kilobytes upon kilobytes of virtual ink here...) It also runs into the problem FlaggedRevs stumbled into, which is of having multiple, conflicting reader assessments, as well as the multiple dimensions of assessment. Overall, this seems like something which should be dealt with something other than a template, if done at all. Titoxd(?!? - cool stuff) 08:05, 3 March 2009 (UTC)[reply]

If some form of Reader Assessment were to be implemented it would be separate to WikiProject assessment. It would be as Drilnoth says, a "Rate this article" thing. A simple click-in assessment on either readability, usefulness or both. SilkTork *YES! 08:25, 3 March 2009 (UTC)[reply]

Essentially, what you're describing is the feedback right on Extension:FlaggedRevs, as is used on Wikinews. Titoxd(?!? - cool stuff) 08:32, 3 March 2009 (UTC)[reply]

As others have said I feel this is way too complex to be of use to the project. The previous table further up the page is much better imo -- Cabe6403 (TalkSign!) 12:52, 3 March 2009 (UTC)[reply]

My head is spinning

[edit]

If you look at the way the quality scale is laid out, it goes Stub, Start, C, B, GA, A, FA. It also states that to reach A quality you do not need to be GA (yes, I know that this is not actually a linear scale). The difference between B and A appears to be minor. As an example, the A article should have no copyright problems. Duh, this should be picked up earlier. So while there is some benefit to being A in quality, do we need this if, for the vast majority of articles and projects, it causes unnecessary confusion and work?

From a simplistic point of view, I would rather see Stub, Start, C, B, A, GA, FA as the order and allow Jumping over some ratings in the process. This would make things less confusing for most editors.

How does this relate to the need for an A class review? Well, how much more work is needed for a B or GA class article to reach A class? Well from B class it is more background work. More references and maybe some other cleanup. Would the encyclopedia be hurt if B class criteria were modified to include most of what is needed in reaching A or GA? There is a suggestion above about meeting WP:MOSNUM (currently in dispute) at A class. We should be working all long to be in compliance with WP:MOS at every stage of the article. Clearly by the time it leaves C class there should be few if any MOS issues.

So maybe my question is, why focus on the need for A class and instead ask why are some of these items being checked so late in the process? After B class we have the need for independent reviewers. So the criteria for leaving B class should be such that surprises like not having enough references should not be revealed after reaching B class. Would the process be better served by tagging missing references in C class articles? After all, the further along we are from the time the text was added without a reference, the more difficult it is to find the source. I found this out trying to get a GA article to A.

While MoS issues are difficult to tag, lack of sources are not. If rating an article as C class called for tagging missing references that will be needed for GA or FA status, would we be doing harm? If we did that would we still need A class? If we did that and made some small changes to B class criteria, would we streamline the process and maybe reduce the work to convert an article to GA OR FA status?

I'm not the assessment process expert, just a confused participant in the process. Vegaswikian (talk) 21:58, 1 March 2009 (UTC)[reply]

Our experience at WP:CHEMS when we set up the assessment system was that it took at least as long to get an article from B-class to A-class as it took to get it from zero to B-class. Yes, you're right that things like copyright issues and major style faults are important for every single one of the 2 million+ articles on Wikipedia, but you also have to balance out the effort put into article assessment and the effort put into article improvement: if article assessment (the proces of reviewing, not the process of article writing) is too onerous, especially for the more numerous lower-class articles, no one will do it and we'll never find out that we have all these huge problems hidden in the encyclopedia. Physchim62 (talk) 22:09, 1 March 2009 (UTC)[reply]
I'm not from an assessment department and i still participate this discussion so you are warmly welcome to give more input ;)
Most projects that don't use A Class, are using a stricter B Class. WP:Anime has an impressive 0.68% of B Class articles. A key feature of A Class is the article completeness that can't be evaluated during GA & FA review if you can't check that feature then doing A Class review is pointless. The questions here are: Do we need always A Class review to check completeness, Can we do it with our limited resource, Are the gain versus the benefit of a such review worthwhile... --KrebMarkt 22:18, 1 March 2009 (UTC)[reply]
It appears to me that WikProjects are evolving separate tracks along the same path, which has been allowed because of the way the assessment structure evolved. The approach you suggest os the approach we're trying to implement at WP:COMICS, and we haven;t really had the issue other projects are having regarding the completedness of articles. I don't know what to put this down to. The thing that I'm finding puzzling is that there doesn't seem to be consideration of the idea that if the article doesn't meet the FA or GA requirements, the concerned projects don't just delist it rather than attempt to alter an assessment process used by a wide number of people in a disparate number of ways. I'm perfectly happy that people be free to use the tools provided as best they can in an attempt to reach a shared goal, and I'm happy not to insist that my way of doing it is correct. But I'm not too happy if people insist that we unify approaches, tell me my approach is wrong and that I have to adopt another approach. Logically, my reaction will be to insist that if we're going to pick one method, we should pick mine. The preferred option would be to carry on as normal, tweaking it to our own benefits and concentrate on improving the articles, rather than the processes, which aren't working too badly truth be told. Unless I am missing something? Hiding T 22:46, 1 March 2009 (UTC)[reply]
I think if WikiProjects have articles which they don't think merit their community grades, the first thing they should try to do is to improve them so that they do think they meet those grades! After all, that is the current philosophy at both WP:FAR and WP:GAR. Delisting should be the last resort, if improvement has been shown to be impractical for whatever reason.
In general, we need to remember that article assessment, be it by the WikiProjects or by centralised processes, is not the end in itself: it is meerly a tool to developing a better encyclopedia. So yes, we should be careful of not breaking tools which others are finding useful. However A-class (and FA for that matter, but let's concentrate on A-class here) is a tool which some people seem to think is essential, some people seem to think is broken and some people seem to think is useless from the very start. At the very least, the different actors in the Wikipedia process could do with bringing their opinions on the matter a little closer together. If that means tweaking some central parameters a little bit, sobeit. After all, the MoS is changed just about every single day of the year: project assessment is a paradigm of stability in comparison! Physchim62 (talk) 00:53, 2 March 2009 (UTC)[reply]
Sorry, I had thought it would be taken for granted that you should try to improve the article first, my bad. I don't disagree overly much with any of your points, but I think the MoS is a perfect example of why there needs to be a better idea of what the problem is and what we're hoping to achieve before we bring in a top-down solution. The MoS is bogged down in so many disputes it is becoming a hinderance, in my opinion. I think that the assessment scales greater flexibility is a boon to be exploited. I've proposed above the idea that we look at trying to integrate A-Class into the tree better, so that the tree still goes B-GA-A-FA, so that there are no changes, no matter how cosmetic, that cause too many issues. I think the current structure has to remain simply because it's ingrained too far, and that that in itself is a bonus, because it allows changes under the bonnet to bring the same result. From where I'm sitting, what's being proposed is Peer Review, so it makes the most sense to see if we can't get our aims met through an existing practise. If we want FA's to be our best articles, which is what they are taken to be, then we have to get expert review into that chain. Which I guess is what people think is being missed when they don;t see many A-Class articles. But which may not be the right conclusion to draw. Hiding T 11:36, 2 March 2009 (UTC)[reply]
I think it goes without saying that every single person here is trying to improve the encyclopedia, even if we don't agree quite how to go about it. When I (personally) have spent a certain amount of time on an article, I am blind to its defaults: I know what I'm trying to put into it, but is that really what the project in general wants? I take that on as a personal default, but one which I think is shared by many editors! One way to answer the question is to ask a centralised body such as GAN. Another way to find out is to ask editors who are working in the same field as me (ie, a WikiProject). Both are valid approaches, but they may give different answers. Is Planck constant (one of mine, deliberately chosen as in need of improvement) a good article? What needs to be done to improve it? I will get different answers from physicists or from general readers, but who is to say that the comments of the physicists are "better" or "worse" than the comments of a general audience. Both need to be recognised and addressed, as far as the project and its volunteers are capable. Physchim62 (talk) 15:36, 2 March 2009 (UTC)[reply]
I agree entirely, sorry. Got locked into one path and didn't consider the ramifications of what I was saying. Hiding T 10:06, 5 March 2009 (UTC)[reply]

Peer review, my thought

[edit]

In reply to a message left on the peer review talk page, I just wanted to say that peer review typically does not get involved in assessment as such. Sometimes people will request assessment as part of a peer review, but they are typically referred to their WikiProject. It is more typical for a peer review to say this is (choose one) ready for / close to / far from either GAN or FAC or FLC.

Finally it seems to me that part of the problem is that there are two parallel systems here: those with formal review (GAN, FAC or FLC), and those that are less formal (in the worst case just putting a letter into a template on a talk page, i.e. A, B, C, stub). Most wikiprojects are not functional enough to have their own formal (and parallel) A class assessment. I did not wade through all of the discussions above - apologies if I repeated anything. Ruhrfisch ><>°° 03:19, 2 March 2009 (UTC)[reply]

  • I'm a bit confused. It looks like we have a huge disconnect here between the peer review process and what a peer review means. Surely WikiProjects should be involved in the peer review for it to have any meaning as a peer review? Hiding T 11:38, 2 March 2009 (UTC)[reply]
    • Yes, I think there is some misunderstanding here between WP:PR as the mechanism for gaining peer reviews on articles, the peer reviews which it mostly produces at present (geared towards FAC or GAN with generalist reviewers) and the A-class reviews which could also be hosted there (open to all comments, but geared towards editors with a special interest or knowledge in the topic and the criterion of "completeness"). Physchim62 (talk) 12:20, 2 March 2009 (UTC)[reply]
      • WP:PR has a fairly constant backlog as it is, so to try and place more on them would probably be a bad idea. As it stands, when WikiProjects do their own internal peer reviews it helps take some of the pressure off of WP:PR. Also, WP:PR does NOT issue assessments; merely comments on ways to improve an article. I think A-Class review would be a completely separate beast; much like GA- and FA- are. hornoir (talk) 12:39, 2 March 2009 (UTC)[reply]
        • But if you had an influx of editors from WikiProjects helping out with peer review, wouldn't that help with the backlog? And with regards issuing assessments, what difference is there between saying, "I think you need to do this, this and this to get to this level", and "I think you need to do this, this and this to get to this level, but you've certainly passed this other level"? Hiding T 12:43, 2 March 2009 (UTC)[reply]
          • Forgetting the issue of page size (since the PR page constantly needs to be trimmed due to size), the entire idea behind WikiProject internal peer reviews (the project running a peer review within the project) was to get a PR without placing more of a backlog on the WP:PR. There are people who request peer reviews for Start-Class articles, just to know where they need to work on the article some more. Its a completely voluntary procedure and not required for progression in the quality assessment system. hornoir (talk) 13:18, 2 March 2009 (UTC)[reply]

Let me just quote some statistics here (from Tom B, not myself): of the Featured article candidatures posted in January 2009, only 40% has gone through WP:PR, and the fact of having gone through PR or not made no difference at all to their success rate at FAC. It's all very well to say "this isn't what WP:PR does", but does WP:PR even do what it pretends to do. Either way, would that be a reason for not bringing on another function to the existing machinery? Physchim62 (talk) 14:44, 2 March 2009 (UTC)[reply]

  • I think looking at WP:PR as strictly a tool towards FA-Classification is a problem, since there are many review requests which simply want to know how to improve their articles. There certainly are many review requests that state they wish their article to get to FA- level, but most are not even at GA-Class when submitted. It is a tool for helping editors that might be the sole person working on an article to receive feedback on what they've done. It certainly does work fine as is, but it is not a stepping stone process to FA rating as some view it. I review a bit over there and, for the most part, the articles listed are on the lower end of the quality scale and the editor just wants suggestions to improve it but can't get those suggestions elsewhere. hornoir (talk) 10:48, 3 March 2009 (UTC)[reply]

Second task

[edit]

Initial discussion on A-Class

[edit]

The second part of the discussion continued here, on "getting a system that works for all". The third part of the discussion is now taking place here, on "expanding the appropriate use of A-Class."

Can someone list the (fewer than 10) currently active A-Class reviews here. I think they were enumerated at the first discussion. Also, we need a link to the first discussion here.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTM) 16:21, 1 March 2009 (UTC)[reply]

Yes; see here (permalink to preserve status even when I change my sandbox). -Drilnoth (talk) 19:52, 1 March 2009 (UTC)[reply]
Updated that link: see [1]Ed 17 (Talk / Contribs) 21:43, 1 March 2009 (UTC)[reply]
Comment

All Wikipedians are welcome to comment here, as long as you wish to contribute something helpful to the discussion. This is simply a discussion, and there is no authority attached to this group other than the usual - the influence of hard-working Wikipedians on fellow members of their WikiProjects. Walkerma (talk) 07:04, 1 March 2009 (UTC)[reply]

Overall discussion strategy

I (Walkerma) will post a followup question each night, dependent on how the discussion develops. Below I've described how it might go.

  1. Establish a consensus that there is a need - or not - for A-Class. This will be started on March 1st.
  2. If such a need is established, should it require WikiProject-based peer review? If so, what sort of review would be appropriate? This would start March 2nd, if desired. This has been started below here.
  3. How can we get WikiProjects to use A-Class in a way that is useful & productive to them and the community? This would start on March 3rd, if desired.
  • Alternately, if the consensus from point 1 is to ditch A-Class, then the followup question would become, "How can we ensure completeness in our articles?"

Walkerma (talk) 07:04, 1 March 2009 (UTC)[reply]

Fourth discussion point: Where next?

[edit]

What action should be taken now? Please comment above on the following earlier proposals, and also post any new proposals that may move the work forward.

Earlier proposals

[edit]

Please post comments above on the following.

I picked these two, because these two seem to be the ones with the most support, though not necessarily enough for consensus support.

I think I'd like to make this the last of my "daily discussion points", though I will try to make sure that we move from talk into action. I think an open thread on this page for comments and ideas is probably appropriate now that we have picked over the main issues in detail. I also want to raise the issue of controlling how A-Class is posted, and review of older A-Class content, but that can wait for a few days, I think. Walkerma (talk) 09:41, 4 March 2009 (UTC)[reply]

I'm not sure I've got much to add to the above two debates. The consensus at WP:COMICS is quite clearly not to separate the two scales because we utilise the system as is rather well and it doesn't cause us any problems. The only proposal I can offer is to allow forking of the scale, but I'm not sure how that would work. It's quite clear there are two opposing views at loggerheads if you ask me. As to teh rewriting of the A-Class review standards, I think what would make A-Class work is a more centralised structure, but again, the consensus appears to be against that, so I can't see the value in discussing the issue. It's hard when your viewpoint isn't really being considered to keep restating it. I think I've stated the case editors at WP:COMICS want me to state, and all I can ask is that it is taken on board. I'll try and keep tabs on the conversation, but what value is there in simply restating a position? If anyone has any sort of compromise, I'm all ears. Hiding T 10:45, 4 March 2009 (UTC)[reply]
In which case, maybe we should be discussing options for centralisation. I'm generally against it, but if editors like Hiding think we should be discussing it then we ought to discuss it! Physchim62 (talk) 13:44, 4 March 2009 (UTC)[reply]
It's not so much that I'm generally for it, it's just that that strikes me as the best method for increasing participation. But I think the feeling at WP:COMICS is that we'll introduce it, probably using the peer review system whether they like it or not. At least that way you get external and internal input, so the three of us carrying the assessment structure are kept an eye on, as it were. I mean, by centralised I guess I mean coming up with a checklist or something, that would be a great idea, and who knows, one of those little images like the FA star and teh GA, um, whatever that is, they always seems to go down well. Hiding T 14:41, 4 March 2009 (UTC)[reply]
Heh, please don't let's get into the GA 'green dot' debate :P Just a quick request for clarification though - are you suggesting a centralised reviewing apparatus (ie pool of reviewers and tracking mechanisms etc), or a single unified set of A-Class assessment criteria that all projects work from (or both!)? EyeSerenetalk 15:54, 4 March 2009 (UTC)[reply]
Well, I think the mood is against a centralised reviewing apparatus, but I can see a value in a single unified set of A-Class assessment criteria. So I'm sort of just suggesting the latter, because I don't think the former would fly, but the latter would be useful however loosely they are framed. Does that clarify? Hiding T 16:43, 4 March 2009 (UTC)[reply]
That was how I read your post, but I wanted to be sure. Thanks for taking the time to reply ;) EyeSerenetalk 18:53, 4 March 2009 (UTC)[reply]
No problem. Hiding T 09:58, 5 March 2009 (UTC)[reply]

Getting help from "drive-by" reviewers

[edit]

I'd also like to request comments on the idea given above by TRS-80:

"As a drive-by reviewer (see below) I think the thing that would make A-Class reviews easier is to reduce the coordination required. Rather than going to some central WikiProject review page, a template that can be dropped into the talk page that can be filled out by three reviewers in turn, over any amount of time allows the load to be spread across anyone who happens across that page. Having to coordinate a review with other people interactively over a set period of time (as FAC requires) reduces the chances of interested parties contributing. I'm sure I'm not alone in contributing to Wikipedia in spurts, and it's not fair that an A-Class review should rely upon casual WikiProject members being constantly available."

  • This idea needs a lot of fleshing out, but I think it could provide the basis of a workable scheme for smaller WikiProjects to solicit external help in reviewing. Could we make this idea (or some variation) work? I note that some projects like WP:MILHIST already have a flag in the article talk page template indicating "This article is undergoing A-Class review" - could this concept be extended? Could A-Class reviews for small projects be handled on the project talk page, or on the article talk page? The project would have to coordinate the review and "award" the A-Class designation, but perhaps this approach could help draw in some fresh pairs of eyes? One needs to remember that the reviewers may not be seeing the same version of the article, unless one particular version is "set" as the one for review. Thoughts and ideas please! Walkerma (talk) 09:41, 4 March 2009 (UTC)[reply]
    • We have a similar flag at WP:CHEMS, see Talk:Hydrochloric acid for a current example. To generalise it, we would probably need to set two parameters into the project banner: ACR_PAGE to show where the review is going on (or went on) and ACR_CURRENT which is set to yes whenever the article is undergoing review. The banner would then display "This article has been reviewed as A-class on the quality scale." (instead of "rated" as at present) with a link to the review page, or "… is being reviewed for A-class…" if the ACR_CURRENT parameter is set. Physchim62 (talk) 10:37, 4 March 2009 (UTC)[reply]
    • This would also provide some reply to the WikiJacobins who think that a centralised community process must be better, just because it's centralised. If we provide a link the the A-class review page, anyone could see what had been considered during the A-class review. I don't think it's important that reviewers are seeing different versions of the page: not even FAC pretends that all reviewers should be looking at the same version! Physchim62 (talk) 10:56, 4 March 2009 (UTC)[reply]
    • Having the review on the article talk page would be a good idea. I'm not overly worried about which version is looked at, there shouldn't be much article drift in a B-Class and above article, really, and what drift there is should be upwards in quality. I think it would be helpful to have a checklist in the banner, like the B-Class checklist. We've got a hook in the comics banner that I copied from somewhere, but milhist don't think it was them, so it might have been either India, Africa, Australia or Film. Hiding T 11:40, 4 March 2009 (UTC)[reply]
    • I like this proposal... maybe we can have a three-point checklist on the banners of B and GA-Class articles and people can just fill them in. If a person thinks that the article should be A-Class, then they just check one of the statements. If they oppose the review, they can fill in an "X" mark and, until the article has been approved, it can't be assessed to A-Class. Of course, this would have the effect of making some GA-reviewed and some B-Class articles all be A-Class, but that's an issue to be discussed elsewhere. -Drilnoth (talk) 14:10, 4 March 2009 (UTC)[reply]
      • I dislike Drilnoth's suggestion as it seems a little too drive-by ;) Also, it doesn0t specifically involve the WikiProject (whatever that might be) in the assessment, which is one part of A-class which I think is essential. However a similar option would be the following: if any editor thinks an article is A-class, they assess it as A-class and start the review at the same time. If the review finds that there are significant elements missing from the article, either those elements are included or the rating is downgraded again. This is pretty much the system at WP:MEASURE/A: it saves on bureaucracy (because the review machinery is only invoked when there's a need for it), and has the same medium-term safeguards as a more cumbersome process such as that used at WP:CHEMS. The downside, of course, is that a small number of articles will spend some time wrongly assessed as A-class, but I think this is a fairly minor price to pay. No assessment system has 100% accuracy all the time, and we shouldn't pretend that Wikipedia processes are perfect, merely the best we can do at any given moment ;) Physchim62 (talk) 14:29, 4 March 2009 (UTC)[reply]
        • Where do you do the review and what notification system is there that an article is up for review? That's another perceived issue with A-Class, and a reason peer review never took off at WP:COMICS, there's so much bureaucracy involved. Hiding T 14:43, 4 March 2009 (UTC)[reply]
          • The review is at WP:PR (here to be precise, all comments welcome if you happen to know anything about the subject). Six other WikiProjects have been notified of the review (as well as WP:MEASURE, of course). The article talk page has the normal peer review notice. Physchim62 (talk) 14:56, 4 March 2009 (UTC)[reply]
            • Is all the notification done by hand? Hiding T 15:57, 4 March 2009 (UTC)[reply]
              • Yes, I just spammed the WikiProject talk pages a couple of days after I'd posted the note at WT:MEASURE, the pasted the diffs into the peer review so there was a record that I done it. The hardest part was thinking which projects might have members who could help: with a fairly transversal project like MEASURE, the list would probably be diferent for each article, but I'm not too worried as I don't expect to have to do this very often ;) There are probably about a thousand articles about measurement topics, most of which are pretty dire it has to be said (hence the WikiProject). As I'm tagging by hand, the rate at which I find potential A-class articles is going to be pretty low. Physchim62 (talk) 16:07, 4 March 2009 (UTC)[reply]
                • Cool. I think that's probably the approach we're building to at WP:COMICS. I can certainly sell that. Does that get us anywhere further forwards? Hiding T 16:44, 4 March 2009 (UTC)[reply]
                  • It's a practical example which could be adopted or adapted by other small WikiProjects should they wish. I don't think it's ideal for larger WikiProjects (eg MILHIST or CHEM) because of the disruption that that might cause at WP:PR, but I've no qualms about using WP:PR for a little project like MEASURE when it solves a practical problem. As I've mentioned above, the key point about the review process for me is the "asking around" bit: if I've asked around and nobody has found any major omissions in the article, then de facto it is as "essentially complete" as WP can manage or judge at the moment. Physchim62 (talk) 17:00, 4 March 2009 (UTC)[reply]
                    • I've responded a couple of sections down, my thoughts meandered from snippiness at tl;dr and I realise I've addressed your comments a bit down there. Basically, I think there's a need for a very loose central structure where we sort of say, this is what A-Class should be, and this is how you can go about doing A-Class reviews, and then list the methods we've got to get there, from the larger sized projects maintaining their own, through medium projects buddying up if that's how we go, to smaller projects using peer review. So we have a go to page we can link to in banners and the like, but the methods themselves aren't centralised. Does that make sense? Hiding T 10:02, 5 March 2009 (UTC)[reply]

New proposals and action points?

[edit]
  • I'd like us to look at the central A-class criteria, but not today, let's give people the chance to look over and digest the discussion above. In particular, I am thinking of proposing a specific "completeness" criterion to replace some of what is on that page at the moment. Physchim62 (talk) 17:00, 4 March 2009 (UTC)[reply]
    • Awaiting your proposal. I stayed silent mostly because my input isn't much required anymore. I feel that a decentralized, more serene A Class review process will be a great gain for the community as the GA & FA are kind of find mistakes in the picture games and are not apt to look for missing thing in the picture. The A Class review notice system discussed above will also bring the needed flexibility --KrebMarkt 21:04, 4 March 2009 (UTC)[reply]
    • I think perhaps discussing making A-Class reviews inter-WikiProject related whereas GA/FA articles are as a whole and B-class is just 1 Wikiproject. Some will still only have 1 Wikiproject associated with them, but a number have 2+ Wikiprojects.じんない 05:47, 6 March 2009 (UTC)[reply]
      • Yea, a notice cross-posted in relevant projects and maybe anywhere good reviewers could be found at the discretion of the person requesting the A Class review (No spaming thought). --KrebMarkt 09:46, 6 March 2009 (UTC)[reply]

A and GA class, are both really needed?

[edit]

I'm assuming you're devising a system like this in the above discussion, but it's pretty much tl;dr territory. I'd just like to make known that introducing an "A" class in front of the "GA" class seems too confusing and decreases the status of Good Articles. As it stands, it seems better to actually eliminate GA's in favor of A-class, than introduce a whole extra class and a lot more work for many Wikipedia editors who have to review articles. This is an example of feature creep and doesn't make sense in the context of the KISS principle.--ZXCVBNM (TALK) 00:03, 5 March 2009 (UTC)[reply]

  • We're not introducing anything: A-class predates Good Articles by a fair stretch, which is why many of the projects which have been doing article assessment the longest don't include "GA-class" inn their assessment scales. I don't see any attempt to knock Good Articles in the discussion above, not even from myself with my well known aversion to centralised WikiProcesses. Quite the contrary, many editors have noted the useful role that GA plays in the current system, even when they don't quite agree what that role is exactly! What we are (I think) trying to do is to make sure there is no unnecessary overlap between the different jobs that people are doing, and also to promote the use of the "tool" of A-class among WikiProjects that don't currently use it for whatever reason. Physchim62 (talk) 00:28, 5 March 2009 (UTC)[reply]
    • I don't really see how GA-class actually makes articles better - it's just the B+ class, yet it's nominated the same way as a FA. Wouldn't it be the same if A-class articles were advertised as such? Personally, I oppose the idea of separate nominations for anything other than FA articles; A-class would have to be discussed on the talk page. If you want projects to start using A class AND GA-class, you have to get rid of the GA-class again.--ZXCVBNM (TALK) 00:37, 5 March 2009 (UTC)[reply]
  • GA is now firmly established in the system. An A-class review could consist of 2 criteria: (a) it meets the GA criteria (ie. let the GA people do that work) and (b) it is comprehensive. maclean 00:50, 5 March 2009 (UTC)[reply]
    • "Well structured and essentially complete" are the basic criteria for A-class and, personally, I completely agree that my idea of "well structured" is very similar to what is assessed at WP:GAN. Projects might want to include some project-specific guidelines – WP:CHEMS does, for example – but I don't see any major or serious disagreement on that point. I don't think projects should be forced to go through GAN before awarding A-class, that would be unnecessary bureaucracy for those projects which are already capable of assessing against the Good Article criteria, but GAN is useful to all projects (even large ones such as WP:CHEMS) as an outside check on that particular criterion, to make sure that the project assessments are going roughly OK. Then there is the criterion of "essentially complete", which, in my view, only a project-based review can assess. Physchim62 (talk) 01:07, 5 March 2009 (UTC)[reply]
      • Right, GA-status shouldn't be required, but the article should meet the GA criteria. Whether or not a project/individual wants to use the GA infrastructure, is up to them. --maclean 01:15, 5 March 2009 (UTC)[reply]
        • Exactly, but we shouldn't throw our arms up in horror if a mistake is made once in a while. Nor should we overburden GAN with articles which can be easily be assessed within a project: let's leave GAN for test cases and articles which are hard to assess (for whatever reason) by an individual project. 2008–2012 Icelandic financial crisis is on that I've done a lot of work on an which springs to mind here: it's a GA, on an ongoing event/subject which is not devoid of political and nationalistic controversy, tagged by eight different WikiProjects none of which seem to have lifted a finger to help out! (not that I'm bitter or anything! Luckily we have had many, many good editors on that article) Physchim62 (talk) 01:25, 5 March 2009 (UTC)[reply]
        • "For whatever reason" would include articles on subject areas where there's no active WikiProject, or where the editor doesn't "feel at home" in the WikiProject, or where the editor doesn't even know that the WikiProject exists, etc etc Physchim62 (talk) 01:32, 5 March 2009 (UTC)[reply]

I am very supportive of the GA class. In WP:WPIN we have no A class review process established, primarily due to us only have about six active editors and the workload is more than we could accommodate given our several thousand articles. GA serves as our "A class" and is usually a very helpful process for developing articles and getting outside input. I have always liked the idea of separating the community ratings from the project ratings, personally. But keeping both assessment ratings is a must, projects like ours rely heavily on GA, while others rely heavily on A class, and others use both. It would be a big deal to either type project to loose and entire class, IMHO. Charles Edward (Talk) 12:30, 25 March 2009 (UTC)[reply]

"quality" and "quality assurance"

[edit]

This page is already TLDR, but for what its worth, here's my outsider's view of what the problem really is:

People expect the level of scrutiny to match the assertion of quality. Or, to put it another way, the higher the assertion of quality, the higher expectation of quality assurance. Consider:

  • Stub and start are weak assertions of quality, so we are happy for anyone to perform such assessments, even the person who wrote the article;
  • C is a slightly stronger assertion of quality, so we expect the assessor to have some clue what they are doing; and there may be some people who would frown on C-class self-assessements;
  • B is a strong assertion of quality, so we have high expectations of the diligence of the assessment and the independence of the assessor;
  • GA is stronger still, so we maintain formal processes, but we still leave it in the hands of a single reviewer;
  • FA is the strongest assertion of quality, so there is such a strong expectation that we get it right that it is put in the hands of the community as a whole, rather than a single reviewer.

When you consider this pattern, it is easy to see why people aren't using A-class. They don't feel comfortable with it because it is an assertion of quality without the proportional quality assurance to back it up.

Personally I think there is no room for another formal process between GA and FA, and this idea of running parallel assessment streams violates the KISS principle. The only way forward is to abandon A-class. It won't hurt a bit, because people aren't using it anyhow. And to those who complain that they want their articles to be recognised as attaining some standard, but don't want to go through these processes, I would say "Stiff shit. Recognition of quality is meaningless if there is no assurance that the recognition is deserved."

With A-class gone, we have a single simple quality hierarchy, with quality assurance increasing as the assertion of quality increases.

Hesperian 03:20, 5 March 2009 (UTC)[reply]

That probably would be the best solution for most projects (and those that still want A-Class reviews can just rename them). Possibly then, GA should be renamed to A so as to keep a C-B-A pattern rather than having C and B but no A. -–Drilnoth (TC) 03:25, 5 March 2009 (UTC)[reply]
I like it. - Peregrine Fisher (talk) (contribs) 03:42, 5 March 2009 (UTC)[reply]
I don't understand - A isn't "between" GA and FA, and it isn't running parallel to it. I thought that was the whole point of much of the above discussion. You're quite right to raise the quality assurance aspect, it's what we're now discussing - how to make A-Class represent a stamp of content quality, something that GA can never be, and which FA can only aspire to be. Walkerma (talk) 03:49, 5 March 2009 (UTC)[reply]
I think the average person thinks A is "between" GA and FA, so for all effective purposes it is. Hesperian 03:57, 5 March 2009 (UTC)[reply]

Featured Articles! It's strange that this solution rarely gets mentioned. (<=) Question, how are you going to enforce renaming of A-class if all of that is up to the WikiProjects? ;)
I don't know if the average person thinks that. Actually, I don't know if many people outside of MILHIST normally think about A-class at all! If a 'pedia-wide A-class review system was instituted, I think that it would just become an additional review that helps improve the article. I don't see how that is detrimental to the project or a violation of KISS, considering the notable shortcomings that GA can have. —Ed 17 (Talk / Contribs) 04:02, 5 March 2009 (UTC)[reply]

Of course, if A was just like GA and FA, then it would be redundant. But it isn't. That's why many believe A-Class to be useful. If the average person thinks it is "just like GA", then clearly we need to fix the perception, not abolish A-Class because people don't understand it! Although a few believe that GAN and FAC are adequate for reviewing content, many do not, and the latter believe that content can only be checked by subject experts, aka the WikiProject. If it's a topic that doesn't need such input, it can ignore the A-Class procedure. Many projects should and will ignore A-Class, that's something I've come to accept through these discussions. But for a subject like chemistry (my field), FAC is a waste of time without input from WP:CHEMISTRY people being involved. As Krebmarkt said above (edited), "GA & FA are "find mistakes in the picture" games, and are not apt to look for the missing thing in the picture" - I couldn't have found a better metaphor. A-Class is there to find the "missing thing", not to fix mdashes. Walkerma (talk) 04:36, 5 March 2009 (UTC)[reply]
To further i will hurt some people by saying that:
B Class article is complete enough to fool non-expert passerby readers.
GA & FA article is complete enough to fool non-expert reviewers.
--KrebMarkt 07:30, 5 March 2009 (UTC)[reply]
Comment It seems to me that we have just had three days where half the dozens of respondents have said they like the current system to an agreement among five people to completely obliterate A-class. Did I miss something?--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTM) 08:22, 5 March 2009 (UTC)[reply]
Yea there is also those who could find some use for A Class review but gave it up due to lacks of qualified manpower and time. --KrebMarkt 09:41, 5 March 2009 (UTC)[reply]
Sometimes tl;dr is a bad thing which needs to be addressed. It's kind of like turning up at your exams having done no revision and expecting an A+. Especially considering Walerma has gone to the lengths to provide bite sized summaries to avoid tl;dr. I think the idea of buddy projects ois one to take forwards, for example WP;Comics, could team up with WP:FILM and WP:ANIME, maybe D&D, Star Wars, Television, the arts projects if they're still functioning... there's scope to buddy up where there's shared conventions or crossover source material. And okay, there's the potential for Peer Review to get flooded, but I'm not someone who sees that as a bad thing if it gets more people using the process. Surely more articles going through peer review makes us better, not worse? Maybe it's worth setting up a hierarchy, I recall that was suggested way back when, with the top level project "sponsoring" the A-Class review. Hiding T 09:57, 5 March 2009 (UTC)[reply]

TLDR is a problem with articles as well. When you look at some of the FAs on the main page, it's perfectly obvious that no one read them before they were approved. It's simply unthinkable that committed and competent editors such as you usually find on an FAC review would have missed the simple faults had they actually read the article. The problem gets even worse because of the tendency for FAs to be written specifically with FAC in mind. There is even a template, {{FAPath}}, which has as step #1 to a featured article "start a new article". So if the community wants to have a strong assertion of quality, it has to find people who might actually read and understand some of the material we have in the encyclopedia. The best place to find them is in the WikiProjects, rather than simply opening a centralised page and hoping that reviewers will magically show up. So I disagree that there's a fatal lack of quality assurance with project A-classes: project A-classess are all about quality control. Physchim62 (talk) 12:18, 5 March 2009 (UTC)[reply]

I might even go as far as looking in fan sites and forums for knowleagable persons on a given manga/anime to check completeness.--KrebMarkt 12:30, 5 March 2009 (UTC)[reply]

It seems to me that we can do a pretty thorough job of assessing completeness - even as non-experts - by following up the sources used for an article. Granted, not every reviewer has access to those, especially where dead-tree references are cited, but surely that's the best indicator we have of the coverage of article content. If I find something in a source that's not in the article, I can reliably conclude there are gaps in its coverage; if an expert tells me there are gaps, but fails to provide sources for them, their expert opinion is irrelevant. Essentially what I'm saying is that articles can only ever be complete to the extent we can source their content per WP:RS and WP:V, and this can and should be checked at FAC (and to a lesser extent at GAN). Because an article hasn't been assessed for A-Class, it doesn't follow that it's incomplete. EyeSerenetalk 12:46, 5 March 2009 (UTC)[reply]

  • I'm not sure I'd follow that definition for completeness. You can't expect to summarise every fact. I suspect I'm misreading you. For instance, I'm currently working on the Fantastic Four, on and off, and you can't summarise everything Jack Kirby and Stan Lee had to say on the comic in the article. I'd always thought the idea was to summarise the positions. For example, Kirby asserts The Thing was based on him, something I need to add to the article. I don't really need to summarise all the examples Kirby gives, do I? I just need to say something like, "Kirby believed The Thing shared a similar temperament to him, and pointed to similarities in The Thing's background and his own. Ben Grimm had grown up in a Jewish household in a rough neighbourhood in New York, a background very much in keeping with Kirby's own." I don't need to get into the Yancy Street Gang, and Aunt Petunia, do I? I'd always thought that in summarising sources, we provide the overview. For the detail, you go to the sources. I'd be fearful of plagiarism if I summarised too much, to be honest. Hiding T 13:01, 5 March 2009 (UTC)[reply]
    • I detest EyeSerene's criterion. It's nothing personal, EyeSerene holds it in perfect good faith and there are perfectly logical and reasonable arguments for supporting it. Many other editors hold a similar criterion. The criterion is so apparently logical and sensible that the fact that it is a Bad Idea had to be put into WP policy very early on in the project: Wikipedia is not an indiscriminate collection of information. Physchim62 (talk) 13:18, 5 March 2009 (UTC)[reply]
      • While EyeSerene's first point, "If I find something in a source that's not in the article, I can reliably conclude there are gaps in its coverage" overstates the matter, I think if you add a caveat about relevance, undue weight, etc. it gives a reasonable starting point. EyeSerene's second point, "I'm saying is that articles can only ever be complete to the extent we can source their content per WP:RS and WP:V" is well-taken; a topic may have things that we'd like to write about, that we believe should be written about, but for which information (or at least appropriately sourcable information) just isn't available. For example, I'm currently working on a biography. I've found that the individual reached a fairly high military rank during a war, but as of yet, I've found absolutely no reference to which campaigns, battles, etc. this individual was involved in, or what the person's role was. I'm not ready to say that the information doesn't exist, but that's a possibility, and at that point the article would just say that "In WAR, PERSON attained the rank of RANK." It's such a big gap that without it, I don't think you can really call the biography "complete", but the information simply might not exist. cmadler (talk) 13:53, 5 March 2009 (UTC)[reply]
This is a very real problem, and one that can only be addressed by a group of editors, not a single editor on their own. My quick answer is that, if you've really asked around and tried to get the sources, but they still can't be found, the article is as good as we can get it at the moment, and ranks as "essentially complete" on a practical scale. WP:FAC is no better than this for "completeness", before anyone makes the comparison.
At WP:CHEMS, we have two types of these problems, which can each be summarized with an example:
  1. "What can we write about radium chloride?" is the CHEM-equivalent of your problem above, something about which everyone (at WP:CHEMS) agrees that we should have an article, but about which there is remarkably little available data from reliable sources.
  2. "What is the melting point of cyclohexanone?" is the problem of conflicting data from reliable sources, of which, to make a choice, we have to use expert knowledge and fly very close to WP:OR. My criteria at Talk:Cyclohexanone have not been challenged by chemists, as they are something that most chemists will recognise immediately, but I do wonder if I could ever find an external source that spells them out clearly and consisely. Chem-students learn this type of thing in lab-classes, and by trial-and-error, it's not really a textbook topic, but it is still an essential part of chemistry. All the same, I made my comments at Talk:Cyclohexanone in 2005 and yet the problem still comes up in discussions between WikiChemists: we don't know what to do, so we do the best we can.
Physchim62 (talk) 14:47, 5 March 2009 (UTC)[reply]
        • I should probably have written "If I find something significant in a source that's not in the article, I can reliably conclude there are gaps in its coverage" - I apologise for giving the impression that I meant we should be including every trivial tidbit ;) I had something more like the example above in mind - say, a source cited in a chemical article tells me that it's used in important industrial process X, but this isn't in the article, or a source for a comic article says that there was a lawsuit over rights to the royalties, but the article doesn't mention it. I absolutely agree that we should be distilling source material, not reproducing it. I hope that makes my observation less detestable :D EyeSerenetalk 14:34, 5 March 2009 (UTC)[reply]
          • I usually left the probably true but need references stuff into the discussion page of the article so people working on it know that the information exist and but still need a reliable source to go from discussion page to main article. --KrebMarkt 14:41, 5 March 2009 (UTC)[reply]
            • Good idea. I've often wondered if we should be more actively educating readers to read the talk-pages too, as that's often where one picks up the real issues with an article subject. EyeSerenetalk 15:51, 5 March 2009 (UTC)[reply]

Completeness criterion

[edit]

When I was drafting the assessment scheme for WP:MEASURE, I wrote this as a gloss on the "essentially complete" criterion for A-class:

"Essentially complete" doesn't mean "exhaustive". There are several issues which need to be considered, such as article length, the likely audience of the article and the appropriate weight to be given to different aspects of the article topic.

Surely we can do a bit better than that! What have I missed? Is there a better way of wording it to give clearer guidance to reviewers? Am I barking up entirely the wrong tree? All comments are welcome, especially polite ones! Physchim62 (talk) 13:07, 5 March 2009 (UTC)[reply]

As mentioned above, availability of information from reliable sources needs to be considered. There may be aspects of an article topic that, in an ideal world, would be discussed in more depth but for which sources can't be found. I'll play Devil's advocate against myself, however, and point out that the fact that a source hasn't yet been found doesn't mean that one can't be found; it's probably not possible to prove that no source exists, and in turn, Wikipedia can only comment on the absence of a source to the extent that others have already written about that absence. cmadler (talk) 14:00, 5 March 2009 (UTC)[reply]
Reviewing articles for V0.5 and V0.7, I found that articles on ancient people were often sketchy, because the knowledge was sketchy. Often it is known that the ancient Greek philosopher XXX existed, but perhaps nothing is known from reliable sources about his life before the age of 30. (Even Jesus almost fits that description...!) So that question becomes - how do we handle obvious gaps for which there IS no reliable source, and probably never will be? Walkerma (talk) 14:55, 5 March 2009 (UTC)[reply]
I think you are too short in your comparison: the "reliable sources" for the the lives of Abraham, Jesus or Muhammad would be laughed at if applied to individuals without a religious connexion, yet there are entire university departments devoted to the study of these individuals and their legacy, and have been ever since universities have existed. We cannot simply dodge the question of "what should we include in an article about Pythagoras" just because it doesn't fit nicely with Goggle searches and accessible sources. Physchim62 (talk) 15:18, 5 March 2009 (UTC)[reply]
I don't have a quick-and-easy answer for the problem that cmadler raises, but I don't agree with the analysis either. If someone writes a book suggesting that 1+1=3, then that book might be notable: in socio-biological terms, it's true in many cases! Who decides if this concept is relevant in mathematics? You would struggle to find a "reliable source" that 1+1=2, but you wouldn't doubt that it's true. It's so true that only an expert in the philosophy of mathematics could point you to the place where it's proved to be true. I'm not such an expert, I just accept the fact because other people have told me so. I don't doubt that there's such a proof somewhere, and that someone else calls it a "proof" because they don't agree with it. If anyone thinks this is a silly example, they might like to look at the list of references and further reading for Addition! Physchim62 (talk) 15:44, 5 March 2009 (UTC)[reply]
We shouldn't be insisting on sources for self-evident material anyway; that's different to gaping holes in the article coverage. To be honest, I think lack of coverage is the one area where there's an argument for allowing a little editor commentary (perhaps even bordering on OR). Some articles will always have obvious gaps in their subject coverage due to lack of source material, and as Cmadler has noted, we can only currently comment on the gaps to the extent that other published sources have done so. I think there's an argument for permitting editors to note that lack of information, as a service to the reader. The source would be the lack of sources, if that makes sense ;) EyeSerenetalk 15:48, 5 March 2009 (UTC)[reply]
This is becoming a circular argument, but what happens when you have gaps caused by lack of knowledge or unreliable sources? I guess here it becomes a point of whether that gap constitutes a critical omission or not: If it is a relatively minor gap, let it be; if not, we may have to use the sketchy reference. Titoxd(?!? - cool stuff) 16:06, 5 March 2009 (UTC)[reply]

(outdent) I think the important point is that someone has to decide what to do. In practice, it seems to lead to a slight relaxation of either WP:RS or WP:OR, and it's the article editor who decides. The talk page and the WikiProject (where it exists and is active) provide a first line check on this, the community as a whole (through the policy pages or through processes such as FAC) provides a second line check. It's pointless to pretend that this oesn't go on – we're here to write an encyclopedia, not to blindly follow rules that we ourselves invented! For the completeness criterion at A-class, I think there needs to be a rough consensus among the reviewers that the article's approach to gaps in the knowledge or the literature is appropriate for the field of study and the exact subject of the article. Physchim62 (talk) 12:05, 6 March 2009 (UTC)[reply]

Absolutely - WP:IAR in fact :) EyeSerenetalk 17:59, 6 March 2009 (UTC)[reply]
I was trying to avoid citing that particular policy page, and instead try to convince people that it's no big deal but rather something we have to deal with anyway ;) At least an A-class review should guarantee a review of the decision, and a review by editors who know something of the problems of the subject area. Physchim62 (talk) 19:54, 6 March 2009 (UTC)[reply]
WPMED has addressed this issue in part through WP:MEDMOS#Sections, which lists the expected content for articles on several common medicine-related subject areas (like articles about drugs or diseases). While we don't have an A-class procedure in place at the moment, we do seem to have a definition of completeness: an article is probably complete if it includes the relevant material for each of the suggested sections. WhatamIdoing (talk) 20:32, 9 March 2009 (UTC)[reply]
  • There may be some ideas at WP:PERFECT, especially Larry Sanger's first draft, [2]. I know most of it is better espoused in long policies, but I think Sanger gets into the nuts and bolts with his little bullet points. The fourth and fifth look like they might have legs, as well as the first three. But see what you think. Hiding T 23:59, 9 March 2009 (UTC)[reply]

Summary so far

[edit]

I was getting lost in the discussions above so I just wanted to see where we're at. It looks like most discussion is centred on getting a central definition of what A-Class means, rolling that out and encouraging WikiProjects to start A-Class assessments using one of three methods: the MilHist approach, the buddy system or the WP:MEASURE approach. There's also spirited discussion continuing on whether to split the scale into two. Am I missing anything? Hiding T 12:27, 6 March 2009 (UTC)[reply]

There was a fairly heated discussion at the start about whether it was necessary to merge some smaller WikiProjects into larger projects (view held by several editors) or whether this was impractical, impossible or downright undesirable (views held by a different group of several editors). So there are really five options which have been discussed:
  1. A-class review on a dedicated page of a large project, such as is done at WP:MILHIST (not discussed much, but no one has suggested that MILHIST shouldn't be able to keep on doing what they're doing now)
  2. Merging projects to get a critical mass for MILHIST-type systems to be set up (proved controversial, but nothing should stop projects merging if they want to)
  3. Projects forming joint A-class review systems (no one has suggested this is a bad idea, but no obvious new candidates have come forward)
  4. Projects using ad hoc A-class review systems, perhaps with the help of the machinery at WP:PR (a small amount of opposition to this idea, but candidate projects willing to try it out)
  5. Projects simply not using A-class at all (quite a lot of discusson on this, with no consensus to abolish A-class and no consensus to make it "compulsary". A signficant number of editors felt that A-class is redundant to GA and FA, while a significant number felt that it was a useful and separate quality class)
Along with some more 'technical' discussion on the criteria for A-class and the way that classes are displyed in project banners, as Hiding mentions. Have we missed anything? Physchim62 (talk) 13:11, 6 March 2009 (UTC)[reply]
That sounds about right. –Drilnoth (TC) 13:19, 6 March 2009 (UTC)[reply]
Thanks for this! I can only manage for so many days on four hours a night of sleep! I appreciate this summary, PC. Let's try to start putting things put into action this weekend. Meanwhile I have 90 minutes to get a poster written & ready... Cheers, Walkerma (talk) 13:32, 6 March 2009 (UTC)[reply]
The mess above is tl;dr for me, but going off Physchim's summary, I don't understand why simply allowing WikiProjects to use it at their discretion doesn't work. Now, there probably should be a set A-class criteria—that's sufficiently ambiguous to allow for project-specific requirements—but otherwise, if a project feels it has enough members with decent enough reviewing skills, then go ahead. If a WikiProject doesn't have enough quality review, then they simply don't have to use it. At WP:ANIME, where I participate a lot, there's maybe three editors (including myself) that consistently bring forth quality content, so asking other editors to review them on the way to FA/FL is a bit pointless considering that the editors asking for the assessment are the ones capable of giving a thorough review. The WikiProject idea is nice, but when you go too far up the project hierachy, you can get a set of editors reviewing the article that aren't really familiar with the type of article; for instance, bringing an anime or manga article for review at WP:JAPAN wouldn't do a whole lot of good considering how focused the anime and manga project is relative to the Japan project. If there's a correlation between projects that works, then great, but they can implement them on a case-to-case basis. — sephiroth bcr (converse) 10:18, 8 March 2009 (UTC)[reply]
It sounds like we have the basis of a consensus in some areas. I'll try to list some proposed action points late today. Walkerma (talk) 10:46, 9 March 2009 (UTC)[reply]

Comment on "Summary so far"

[edit]

This says there were 328 A-class articles across wikipedia as of 8 March 2009, with apparently 161 in WP:MILHIST and 113 in WP:SPID. It is no longer a question of deciding whether or not to abandon A-class. It is already abandoned for all but a very small minority of projects. Hence, "there is no consensus to abolish A-class" is incorrect. There is consensus not to use it for the vast majority of projects.

Reviewers have abandoned A-class reviews in favour of the community-wide processes, as seen by the rise in GA. Claims that the community wide processes are used by ignorant morons who don't know anything about the subjects they're reviewing are insulting and plain wrong cannot identify incomplete or non-comprehensive articles do not stand up to scrutinyAmended DrKiernan (talk) 09:20, 18 March 2009 (UTC). The editors there are exactly the same people who edit for the WikiProjects. Every editor here is equal, and there is no division between "experts" in projects and everybody else.[reply]

Trying to make out that the "completeness" criteria is somehow special to A-class are bogus. The completeness criteria is not special to A-class.Amended. DrKiernan (talk) 09:20, 18 March 2009 (UTC) FA-class specifically says that articles must be comprehensive, and GA-class specifically says that articles must address all the main aspects of a topic. DrKiernan (talk) 11:11, 9 March 2009 (UTC)[reply]

Reviewers have not "abandoned" A-class in the slightest, what makes you suggest that? Most projects have never had an A-class review, nor an A-class article for that matter, and many of the "project featured articles" have been adopted by the projects rather than them having any input whatsoever into the review (or the editing). Much of the above debate centered around how to get a workable A-class review which smaller projects could adopt for the first time, not convincing them to come back to the system.
As for FAC reviewing completeness, it is very true that criterion 1b states that the article "neglects no major facts or details and places the subject in context". That's all very well in principle, but does FAC actually have a means of controlling this? Indeed, is an FAC-type review the best way to control completeness? Wouldn't FAC be better off catching errors that other reviewers have missed, rather than trying to review everything from scratch for the first time?
I could go on about the backlogs at FAC and GAN, and the impossibility for a centralised process to review the more than 2 million articles on Wikipedia. As for "all editors are equal", how come one-fifth of FAs are written by just ten editors? Are these really the ten most valuable editors on WP, or is it simply that they have a hand in shaping the criteria for everyone else who chooses to use them? Let's not forget that many editors choose not to use FA as simply not worth the bother for the gain in article quality. Physchim62 (talk) 13:03, 9 March 2009 (UTC)[reply]
Biography was abandoned in April 2008. History was abandoned almost as soon as it was set up. Chemicals was abandoned in April 2008; your attempt to restart it is ignored.
It is controlled by voting "support" if one believes the article comprehensive, and "oppose" if one does not, noting the missing details alongside the "!vote".
There is no backlog at FAC. The number of nominations is constant (see for example Wikipedia:Wikipedia Signpost/2009-02-16/Dispatches). DrKiernan (talk) 13:53, 9 March 2009 (UTC)[reply]
I think you need to review the many thoughtful comments over the last week or o, for and against, and then you will see where we gathered the consensus from. But don't worry - there is a clear consensus to make it clear that A-Class is optional. So you can carry on ignoring it, just like List-Class or B+ Class, if you wish. Walkerma (talk) 15:08, 9 March 2009 (UTC)[reply]
Or my point (not universally shared) that there's no point in reviewing at B+-class or "strong B"-class when you could be reviewing at A-class. As for the backlogs at GAN and FAC, they are freely admitted on the talk pages of the processes concerned. Physchim62 (talk) 15:20, 9 March 2009 (UTC)[reply]
If there is a backlog, let's examine the reason why: there is a shortage of reviewers. And yet, you insist on creating new review processes. Your own evidence argues against the foundation of new review processes, since there are not enough reviewers anyway, even in the existing ones. DrKiernan (talk) 07:53, 10 March 2009 (UTC)[reply]

DrKiernan gives us an interesting statistic in that there are only 328 A-class articles in the March 2009 statistics: after all, there were 518 in February 2009 and 615 in December 2008. What can be going on? Surely they're not all being "promoted" to FA!

The answer lies in this history file. The same DrKiernan who says that projects are "abandoning" A-class for the "Community" processes of FA and GA has also gone through the entire list of A-class articles, and downgraded all those which don't fit his/her personal A-class criteria, without (as far as can be seen) any consultation with the WikiProjects concerned. This is a short list of diffs from early February, but there are probably nearly 300 similar ones: [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. Centralization appears, for this editor, to be at the whim of a single person: so much for consensus! What use is an assessment without a discussion of the faults among editors who might just be able to help to improve the article? Why did DrKiernan concentrate on A-class articles and not simply delist the many featured articles that would not meet current criteria? Such is the kind of assessment DrKiernan has been undertaking over the last six weeks, and it is useless to the encyclopedia. Physchim62 (talk) 16:17, 9 March 2009 (UTC)[reply]

This completes the above list from 1–14 February:
[22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78].
An interesting case is this one (not in the main list), where DrKiernan actually decided to remit the aricle to a WikiProject (as it happens, MILHIST). Why shouldn't all the others have been remitted to an A-class review if there were problems? Physchim62 (talk) 18:12, 9 March 2009 (UTC)[reply]
Here are the ones from 15–21 February:
[79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111], [112], [113], [114], [115]
These are the ones from 22–23 February. Again, one was sent back to MILHIST (not included in the main list), the others were listed as minor edits (sic).
[116], [117], [118], [119], [120], [121], [122], [123], [124], [125], [126], [127], [128], [129], [130], [131], [132], [133], [134], [135], [136], [137], [138], [139], [140], [141], [142], [143], [144], [145], [146], [147], [148].
These are mostly from 24 February, when DrKiernan's eclassification rate reached peaks of six articles a minute, but continue up to yesterday.
[149], [150], [151], [152], [153], [154], [155], [156], [157], [158], [159], [160], [161], [162], [163], [164], [165], [166], [167], [168], [169], [170], [171], [172], [173], [174], [175], [176], [177], [178], [179], [180], [181], [182], [183], [184], [185], [186], [187], [188], [189], [190], [191], [192], [193], [194], [195], [196], [197], [198], [199], [200], [201], [202], [203], [204], [205], [206], [207], [208], [209], [210], [211], [212], [213], [214], [215], [216], [217], [218], [219], [220], [221], [222], [223], [224], [225], [226], [227], [228], [229], [230], [231], [232], [233], [234], [235], [236], [237], [238], [239], [240], [241], [242], [243], [244], [245], [246], [247], [248], [249], [250], [251], [252], [253], [254], [255], [256], [257], [258], [259], [260], [261], [262], [263], [264], [265], [266], [267], [268], [269]

It is obvious that those edits were accompanied by reviews; the reasons for the re-assessment are described in the edit summary, and the procedures, processes and grading schemes of each individual wikiproject were followed. There is no basis for the complaint that the edits are against consensus or useless to the encyclopedia. DrKiernan (talk) 12:34, 10 March 2009 (UTC)[reply]

  • Would you mind awfully toning it down just a little, DrKiernan? As to your points, I've made the exact same points myself over the last couple of weeks, but I'm more interested in working towards a consensus than defending an absolute position. The points you raise have been discussed and as people have already indicated, are understood and being worked around. Hope that clarifies, Hiding T 16:48, 9 March 2009 (UTC)[reply]
    • I just looked at a few of that user's recent edits and saw articles for many Congressional sessions moved from "A-class" to "list-class", which technically, they probably are. I think this illustrates the argument made elsewhere in this discussion that "type" should be separated from "quality"; a list-type could be A-quality (or "A-class" if you stick with current terminology). cmadler (talk) 19:03, 9 March 2009 (UTC)[reply]
      • There was a point raised above from an editor from WP:PHILLIES that they would like to have A-class lists because some of their lists are complete but too short to be considered at WP:FLC. I have no problem with either solution: I have difficulty assessing lists because I tend to write prose but, as lists are very much accepted as useful to the community, why shouldn't we allow projects to asses their quality? If the centre doesn't "allow" this, the projects will do it anyway, just informally and with the consequent loss of information for everyone else. Physchim62 (talk) 20:25, 9 March 2009 (UTC)[reply]
        • Now that you mention it, DrKiernan was the very reason WP:VG became interested in this discussion, because he changed the class of several video game articles from A after they had been approved within the project. — Levi van Tine (tc) 12:15, 10 March 2009 (UTC)[reply]
I believe by "several" you actually mean "one", which I asked you to link to the review discussion: [270]. DrKiernan (talk) 12:25, 10 March 2009 (UTC)[reply]
I meant two. The other was Rock Band. But, I'm splitting hairs. — Levi van Tine (tc) 15:05, 10 March 2009 (UTC)[reply]

It is obvious that those articles were not A-class. Articles which are not A-class which had WikiProject A-class article review processes were remitted to those processes. You're attempt to smear me is a personal attack. You should be concentrating on the arguments presented not commenting on the contributors. The fact that you have religiously and deliberately reviewed every one of my edits for more than the last month, demonstrates stalking. Such behaviours are frowned upon. DrKiernan (talk) 07:53, 10 March 2009 (UTC)[reply]

Besides which you have chosen to ignore my promotions to A-class (understandably because they are from when A-class was still active): [271] [272] [273] [274] [275] [276]. You've assumed that my edits are attacking A-class without even looking into them. The edits are in defense of A-class, ensuring that the A-class quality standard is adhered to. DrKiernan (talk) 13:07, 10 March 2009 (UTC)[reply]

You've unilaterally downgraded 268 articles without any attempt to contact the projects concerned or to improve the articles. You did this at a time when there was project-wide discussion going on about A-class, and yet the majority of your downgrades are marked as "minor edits". You then tried to cite the low number of A-class articles (due in no small proportion to your own recent actions) as proof that the majority of WikiProjects had abandoned A-class. If that is not disrupting Wikipedia to make a point, I don't know what is.
The A-class quality standard you claim to adhere to is purely your own, as shown by your lack of consultation in performing the downgrades. I note as well that you don't seem to have attacked any of the articles on Wikipedia:Featured articles/Cleanup listing, despite them having the same if not worse faults than those which lead you to summarily downgrade A-class articles.
All of this was done, not only at a time when there were widespread discussion going on, which you yourself contributed to, but also just before the selection procedure for Version 0.7: as such, given the scale on which it occurred and your lack of consultation with other editors, it is arguably vandalism. After having been asked to "lower the tone", you then claim that I am "slurring" you and that I am somehow "stalking" you, when all have done is collect the evidence of your own actions. Rest assured, I only needed to check your article talk page contributions, and almost all of those since 1 February have been to remove A-class status from articles (usually with an edit marked as "minor"), as any editor can verify.
I ask that you reverse your unilateral downgrades without delay, and engage in constructive discussion with other editors as to what can be done to address your perceived problems with these articles. Physchim62 (talk) 13:30, 10 March 2009 (UTC)[reply]
Excuse me? You are so far off the mark, it is unbelievable.Struck. DrKiernan (talk) 09:20, 18 March 2009 (UTC) I am one of the most active editors at Featured article clean-up. For example:[reply]
George III Improving edit
History of evolutionary thought & Solar system Improvingedits
Warsaw Uprising Improving edits
Aaron Sorkin Improving edits
There are many, many such examples, both from FA clean-up and Featured article review, where I am also one of the most active contributors.
Changing a single letter is a minor edit.
You calling me a vandal is grotesquely insulting. DrKiernan (talk) 13:50, 10 March 2009 (UTC)[reply]
You've already been asked to tone down your remarks once during this discussion: I'll ask you to do it again. Physchim62 (talk) 14:04, 10 March 2009 (UTC)[reply]
I'm not going to get involved in this discussion, but I'd just like to remind everyone to be civil... no matter what happens here, it's not going to delete the main page, so you probably shouldn't stress it to much. –Drilnoth (TC) 14:12, 10 March 2009 (UTC)[reply]
OK, I think it's clear that DrKiernan has jumped the gun before re-assessing so many articles from A to B or other things; that is something which should have been agreed here before being done. I checked just a couple of the examples, and I note that DrKiernan is not listed as an active member (or even inactive) at WP:INDIA, yet one of the WP: India articles was reassessed; such "drive-by" reassessments are strongly discouraged (unless WP:INDIA has requested help from outside). It may be that we will need someone with AWB to revert those edits, if that's what this group thinks is best. However, I think that it is completely inappropriate to call DrKiernan a vandal, and such name calling; I believe that he/she acted in a way that he/she believed was best for the encyclopedia. So, now we need to focus instead on what action points are appropriate for moving things forward. I'm working on something right now, to post today. Walkerma (talk) 15:18, 10 March 2009 (UTC)[reply]
That's just the sort "You're not a member, so you can't participate" elitism that started the MfD. Anyone can be a member of any project. It is not for us to decide who should or should not be a member. DrKiernan (talk) 15:23, 10 March 2009 (UTC)[reply]
It's not elitism when clearly anyone can join. Only people actively working on Indian articles can really assess that article as A-Class; that's why it has a WikiProject tag on it. By editing the banner, you're saying "The India WikiProject has reassessed this A-Class article as B-Class. Can you honestly say that? Did you discuss the reassessment at WT:India first? Walkerma (talk) 15:47, 10 March 2009 (UTC)[reply]
Recent edits to a South Asian article. Editors in good standing are not required to ask permission before editing. DrKiernan (talk) 15:58, 10 March 2009 (UTC)[reply]
We're not talking about editing an article - clearly WP:OWN:no one owns an article, even a project. But a WP:INDIA project banner, that represents the project! We're not talking about stubs and starts, and we're not talking about formatting issue, we're talking about an A-Class assessment, which only a subject-expert is qualified to judge. That's the whole point of A-Class! Walkerma (talk) 16:08, 10 March 2009 (UTC)[reply]
How do you know that I am not an expert? DrKiernan (talk) 16:12, 10 March 2009 (UTC)[reply]
I'm assuming that you aren't a subject expert in every subject area where you demoted an article. I don't deny that you will be qualified in some areas. I am probably qualified to contribute to a review on an electric guitar, even though I'm not a member of any music project, but (a) I would NEVER unilaterally demote any article from A-Class without raising it first with the authority that assessed it, and (b) I wouldn't claim to have such expertise in more than a handful of subjects. Are you claiming that you're an expert in everything? More knowledgeable than the WikiProject? Walkerma (talk) 16:20, 10 March 2009 (UTC)[reply]
No. I'm claiming I'm as knowledgable. Since it is impossible to know whether I am a precocious eleven-year-old Indian girl or a Professor of Nuclear Physics at Harvard. All editors must be treated equally. DrKiernan (talk) 16:24, 10 March 2009 (UTC)[reply]
Well, on a topic relating to India I would always defer to someone actively editing in that area, as being more knowledgeable than me. Apparently you would not, which surprises me greatly! Walkerma (talk) 16:32, 10 March 2009 (UTC)[reply]
I would agree that altering a single letter would generally constitute a minor edit. However, in this case, that single letter represents a much larger concept, the class of an article. That could hardly be considered “minor”.
When you found those video game articles whose A-class reviews you considered suspect, you reverted their class with complete disregard for consensus. If you had merely wished that the pertinent discussion had been linked to the article’s talk page or wherever (which, I might add, your edit summary did not properly explain), you could have left the rating alone and done that bureaucratic formality yourself, or left a message on WT:VG and one of us would have done it. If instead you had honestly thought that the article didn’t merit an A-class, and you felt strongly enough to challenge it, you could have also left the rating alone and opened a discussion on the article’s talk page or at the aforementioned WT:VG, and we could have achieved a compromise based on consensus. There was no need for you to run rampant over perfectly legitimate A-class reviews. — Levi van Tine (tc) 15:54, 10 March 2009 (UTC)[reply]
  • I'm a little dubious about this debate and the utility of it. The consensus is quite clear that assessing an article is not a minor edit, so much so that I'm staggered that anyone could seek to defend it as such. The rest of the discussion seems to be centring on how and by whom A-Class assessments are done, something I guess we should really hash out here and now. I'm sure DrKiernan would be happy to defer to whatever consensus emerges on that score. There's certainly a valid point here that's perhaps being missed. What happens if an article is thought by someone not to be A-Class? It's a tricky one. DrKiernan removed the A-Class rating from a comics project because it had two requests for issue numbers to verify plot points. Is this enough to remove A-Class status unilaterally? I guess it comes down to what we're looking for in A-Class. If editors are going to insist on their interpretation, we'll definitely have to consider mediation and other avenues. But let's not filibuster our way into stasis, please. Hiding T 17:31, 10 March 2009 (UTC)[reply]
If I may correct you, the revision of the article to which I believe you refer had 10 citation needed tags at the time that I reviewed it [277]. The tagged material has since been removed. The issue numbers are a trivial point. DrKiernan (talk) 18:02, 10 March 2009 (UTC)[reply]
Feel free to correct me. What would have made this conversation a whole lot easier is if you'd posted your concerns on the talk page and engaged in discussion, rather than using a truncated edit summary, wouldn't you agree? I'm intrigued that your solution was to reassess the article. Any thoughts why that was your first and only action? I can't see your name listed in the article history. Hiding T 18:31, 10 March 2009 (UTC)[reply]
See Wikipedia:WikiProject Comics/Assessment#Frequently asked questions: "Who can assess articles?" and "Why didn't the reviewer leave any comments?" I also notice that you assessed a very large number of articles yesterday. I don't see any written review comments. DrKiernan (talk) 08:12, 11 March 2009 (UTC)[reply]
And I'm intrigued as to why DrKiernan did the same thing more than 250 more times than the individual cases discussed here. During February, virtually all of DrKiernan's time in article talk space was spent on downgrading A-class articles: like I said, any editor can check that. And then why were the results of this unilateral cull presented as evidence that WikiProjects have abandoned A-class? If anything, DrKiernan has proved that A-class is a much more flexible assessment tool than FA: had he/she listed all the articles on Wikipedia:Featured articles/Cleanup listing at WP:FAR, the whole featured article system would have collapsed for lack of reviewers! Physchim62 (talk) 00:16, 11 March 2009 (UTC)[reply]
  • I'm intrigued you chose to answer just the one question and then turned this into a confrontation. Is this really going to require mediation or are we able to sort this out amicably? Are you willing to explore and discuss alternatives, or is it a case of your way or the highway? I mean, if we want to quote Wikipedia:WikiProject Comics/Assessment#Frequently asked questions: "What if I don't agree with a rating?" But where does that get us? It gets us into a confrontation, doesn't it. And who does that help? Is it implausible that we can simply discuss the issue without any need to assume bad faith? Hiding T 11:04, 11 March 2009 (UTC)[reply]

This section should be archived. It is superceded by #Action below. DrKiernan (talk) 12:06, 11 March 2009 (UTC)[reply]

More to the point, it is superceded by Wikipedia:Requests for comment/DrKiernan. Physchim62 (talk) 14:24, 11 March 2009 (UTC)[reply]

I apologise for any offense caused by inappropriately worded comments. DrKiernan (talk) 09:20, 18 March 2009 (UTC)[reply]

Thank you, DrKiernan! And thank you, also, for reviews like this one, which I've asked WP:Germany to address. (I hope to work through all of the list, once Version 0.7 is released.) Hopefully the "demoted" articles will be reviewed and improved in the coming months, and things will turn out for the good. Walkerma (talk) 16:04, 18 March 2009 (UTC)[reply]

New discussion point: WikiProject banner names

[edit]

Right now the names of WikiProject banners is a tangled mess... {{WPBiography}}, {{WPMILHIST}}, {{Engineering}}, {{Comicsproj}}, {{WP India}}, {{AfricaProject}}, {{WikiProject Scouting}} and {{WikiProject Idaho}} all use different naming conventions for their banners, and new banners are being created based off of different ones of these all the time. This large number of different names makes it difficult for someone not already involved in a WikiProject to assess an article for that project, because they need to locate the name of the WikiProject banner if they want to add it. Therefore, I propose that we standardize all WikiProject Banner names, so that all you need to remember to tag an article for a project is the project's (usually intuitive) name. Naturally, such a change would leave redirects from the old names. –Drilnoth (TC) 12:34, 9 March 2009 (UTC)[reply]

Comments

[edit]

Comment. There is a precedent for this sort of idea in Wikipedia:Infobox standardization and, of course, its sister page, Wikipedia:No infobox standardization… In 3½ years, we've made great moves towards infobox standardization through discussion in creating tools that editors can use and adapt to their specific needs. No centralizing force has been needed. Physchim62 (talk) 15:12, 9 March 2009 (UTC)[reply]

Agreed... I'm not asking for any sudden change, but just a gradual establishment of a single type of name. –Drilnoth (TC) 15:23, 9 March 2009 (UTC)[reply]

Comment: My reasoning behind this is that, if I see an article which is related to mathematics, I want to be able to easily assess it for WikiProject IMathematics. Right now, if I didn't know what the template was, I'd have to go to WP:WPM and find wherever they list their template's name, which has historically discouraged me from adding articles to projects I'm not completely familiar with... why should we need to do the extra research? I don't know whether the template for WP:WPM is at {{WPMathematics}}, {{WP Mathematics}}, {{Mathematics}}, {{WikiProject Mathematics}}, {{Mathemeticsproj}}, {{MathematicsProject}}, {{MathematicsWikiProject}}, or what, so when I see an math-related article without the banner, I need to dig around to figure out which banner is correct. This has lead me, personally, to not add project banners to many articles even when they'd be needed because the research would be excessive. An alternative option would be to, over time, create redirects to existing ones from other naming conventions, rather than moving the template themselves. –Drilnoth (TC) 15:23, 9 March 2009 (UTC)[reply]

  • Would it at least be valid to create redirects to the templates from various other formats to allow ease of tagging, even if the actual banners aren't moved? –Drilnoth (TC) 21:01, 9 March 2009 (UTC)[reply]

What do people think about tackling as a separate issue those banners that are at entirely gratuitously-obscure names, where it actually makes it genuinely difficult to guess from the list of transcluded templates which is the project banner? I'm thinking about names like {{csbir}}, {{ASUE project}}, {{cnproj}}, {{AARTalk}}, etc. Go on, I challenge anyone to correctly guess which projects those banners correspond to, without looking. This is a rather different kettle of fish to "WPBiography" vs "WikiProject Biography". Happymelon 17:52, 10 March 2009 (UTC)[reply]

  • I don't think the first and third ones are active. As to all of them, if someone really is that bothered, move requests on their talk and project pages might be an idea. But I think you'd need the consensus of an active project to move it. I'm afraid I'm not much help, I'm kind of in the "just create the redirects you need" camp. Hiding T 18:10, 10 March 2009 (UTC)[reply]
    • I think that moves to standardize those would be incredibly helpful, although I'd still rather see even more standardization. Redirects can always be created, though, if a large-scale move isn't wanted. –Drilnoth (TC) 18:32, 10 March 2009 (UTC)[reply]

Action

[edit]

OK, I think we've had plenty of time to mull over things, so we need to move on to doing things. It seems clear that A-Class will remain, but it's also clear that not everyone will want to use it. Looking over Physchim62's summary for about the tenth time, here are some thoughts gathered over recent days:

  1. We need to rewrite the A-Class criteria, and clarify what sort of process is reasonable and what is not. This was discussed here recently. We need to indicate what sorts of WikiProject-peer-review meet the standards, what is a "quorum" of reviewers, and clarify that unilateral reassessments by a single editor are unacceptable. (PS: We should make these criteria more visible, too!)
  2. We need to look over the complete hierarchy, and find some clusters of WikiProjects that have some common ground. The most effective way to do this (IMHO) would be to ask our delegates/coordinators/users/editors/interested users (I'll call them the last) individually about this. What projects do you think your project could work with? Is there a viable consortium of projects that could work together on A-Class review? Should some related projects be merged or become task forces? Every case will be different, but that's what our "interested users" are here to help with.
  3. We should set up a list of floating reviewers and their corresponding interests, as suggested by KrebMarkt. If a project needs more reviewers, they could look there to get another fresh perspective yet with some (presumed) expertise.
  4. We need to design an ad hoc system, using WP:PR as one option. This could work via a project's talk page, or via an article talk page - I would suggest that a prototype for both be developed. If we do this right, this could become the most common form of A-Class review on WP. So a lot of thought and discussion will need to go into this before it goes "live".
  5. We need a formal system for revisiting older A-Class assessments, to make sure that standards are upheld and they match the goals of the WikiProject. This will need to be done at the WikiProject level, and could probably be done on the same page as the promotions.
  6. We need a log tracking all promotions to A-Class across all Wikipedia (and demotions). I will talk to User:CBM about this, it should be very easy to get the 1.0 bot to do it for us. This is something really needed for the 1.0 project anyway, but this group can watch the list to make sure that we don't get "drive-by" reassessments by a single person. This has been on my non-urgent list for some weeks now, but I think the need has become more apparent recently.
  7. We need a "help desk" of people experienced with A-Class review to help (only if asked!) with projects setting up new review systems.
  8. Once we have got the system sorted, we need to revisit how the system is presented to the community - things like the scale, the templates, etc. I think that can wait until the above issues are clarified, but we will want to start thinking about them again, because they always cause a lot of debate.

I don't think these proposals are perfect - they don't provide the rigor we might want, but they provide a complementary and more reliable system compared to the content review systems currently used for most articles. Any additional work will largely fall onto this group, but it shouldn't be too onerous once the system is in place; any additional work at the project level is work that should be done for any top-quality article anyway. Have I missed anything? Walkerma (talk) 10:09, 11 March 2009 (UTC)[reply]

    1. I think that we need to clarify that A-Class assessments are the impression of a WikiProject as a whole, and as such, non-trivial changes (trivial being A->FA promotion, for example) should be raised with the WikiProject, especially if there's an A-Class review involved.
    2. A good starting point for this would be the Council's directory, I think.
    3. There's like forty of these lists already, so we need to be careful of Yet Another List Syndrome. See for example the Wikipedia:Peer review/volunteers listing.
    4. Maybe use PR with an added note, "This review will also work as an A-Class review for X, Y and Z WikiProjects"?
    5. That's dependent on the formalized standards. :P
    6. It's technically speaking possible with wp1.0v2 but it is slow as hell right now (hence I'm not linking to it). There would need to be some optimization tweaks, but all the data for this is available.
    7. Maybe using Wikipedia talk:Version 1.0 Editorial Team/Assessment/A-Class criteria, once it's defined. That'll probably be the most natural place for it. Titoxd(?!? - cool stuff) 10:51, 11 March 2009 (UTC)[reply]

In response to Walkerama's point 5, MILHIST already has a way to demote A-Class articles that don't meet the standards: Wikipedia:WikiProject Military history/Review/A-Class reappraisal review instructions -MBK004 21:14, 11 March 2009 (UTC)[reply]

  • Thanks for comments. Point taken about the "Yet Another List Syndrome" - let's try to add to an existing list. Does anyone know what might be best? Walkerma (talk) 05:35, 12 March 2009 (UTC)[reply]
    • Yea, you have my thanks too for pointing that issue. My main objective was to have places where knowledgable non-project members could be easily reached for input. I think we should first use the already available lists of users and then crate new ones on case by case basis. --KrebMarkt 08:08, 12 March 2009 (UTC)[reply]

OK, let's do some of the work! I'd like to propose a hiatus in discussion for a few days (if possible!), so we can work on the above. I will start contacting people and projects via talk pages. Drilnoth, will you be around to help with some of that? Physchim62 and others, can you let us know when you have some new criteria that you're all happy with, and then we'll perhaps try to get agreement to ratify these. I don't think we need to spam all the WikiProject pages again when we do that- I'd rather just contact the main projects that are currently using A-Class and get their opinions. And we'll also have to think about how to handle the mass demotion of A-Class articles that occurred - I may get the opinions of some of the affected WikiProjects that have delegates here. We can all post the results and feedback from our efforts on this page. Thanks, Walkerma (talk) 03:41, 14 March 2009 (UTC)[reply]

Who will do the work?

[edit]

Please sign up if you can contribute some time in the next couple of months.

  • On that note, did you get a chance to look at WP:PERFECT, especially Sanger's initial draft at [278]. I think there's stuff there that would help expand on the definition of complete, for example "acknowledges and explores, in depth, the different ways there are of approaching the same topic", "present each of various competing views on controversial subjects fairly and sympathetically. Not only are the competing views sympathetically presented, they are organized logically, so that the reader has a clear understanding of the 'lay of the land.' Of course, the most popular views, whatever they are, are emphasized, and the extreme minority views are underemphasized" and "reflects expert knowledge of a sort that is careful and precise. While it is clear and accessible, it is not full of the sorts of vague generalities and half-truths that almost inevitably crop up when nonexperts try to write articles on stuff that they only think they understand." Hiding T 15:01, 12 March 2009 (UTC)[reply]

Action #1: Criteria

[edit]

So, the first thing on Walkerma's excellent list of actions is to revise and expand the A-Class criteria, and to make them more visible. To this effect, I will shortly begin a discussion at the talk page to discuss these various topics; all additional input would be much appreciated. –Drilnoth (TC) 13:01, 25 March 2009 (UTC)[reply]

Inline coordinates and A-Class articles

[edit]

Thought you guys might be interested in this discussion at WT:MILHIST. §hepTalk 20:55, 11 March 2009 (UTC)[reply]

Conflicting ratings

[edit]

One item no one has raised, as far as I can tell, is what to do in cases of conflicting article ratings from different WikiProjects: if this hasn't happened yet, it will. Many of these conflicts will simply be due to the vagarities of volunteering -- an article may be improved by a person, one WikiProject will notice this & upgrade their rating while one or more other WikiProjects will not. However there will be some whose members may still think that the article hasn't been improved enough to warrant an upgrade; these are the instances we need to know about, especially for articles which are being considered for A, GA, or FA class.

What I'd like to propose (& unless there is serious objection here will likely do myself) is the following:

  • No article proposed for an A, GA or FA review should have any WikiProject ratings lower than "B". In other words, stubs & start articles can't be approved for those ratings.
  • Unless there is a written explanation why an article has a different class rating from the others, then any editor can modify the rating to make it match the rest of the ratings.

In other words, if there is one or more reasons why an article should still be considered a stub or start when everyone else thinks it is better than that, then they need to be shared so they can be fixed. Otherwise, it's a matter of being disruptive.

Any comments? -- llywrch (talk) 17:43, 13 March 2009 (UTC)[reply]

95% of wikiprojects (at least) don't regularly check up on their ratings. When I update ratings i do them for all wikiprojects, because the other ones aren't going to catch up. I've seen things at GA and FA that were marked as stubs by wikiprojects, but that's because they weren't updated. That's the fault of assessment updaters more than anything. Wizardman 19:34, 13 March 2009 (UTC)[reply]
Let's not pretend that FA and GA are immune from the same problem! FA keeps this page of articles, for example. Physchim62 (talk) 19:42, 13 March 2009 (UTC)[reply]
  • Strongly oppose. This infringes on the rights of the projects to control their own ratings, imposes an unnecessary layer of buraeucracy, and works from the assumption that lower ratings are current.
    • Some projects don't use some of the rating levels. Some have different standards: What's B-class for a few projects (those not adopting the new standards) is C-class for others. What's B-class for most projects is C-class for others (those that require specific information in each article [e.g., WPMED is not going to count any disease-related article as B-class if there's no information about symptoms or treatment]). Some projects have structured approaches to assessment, and actually do not want you to bypass their assessment process. You can legitimately have different assessments.
    • The fact that some largely inactive project thinks that an article is still a stub should never prevent me from nominating an article for GA/FA/FL. Consider, too, those WikiProjects that want their assessments to reflect the views of their own members: Should my work be put on hold while I try to convince them to find a member that will bother with a reassessment? Should I have to put a note on thousands of articles to explain the rating according to our project's standards, just to keep someone else from saying "Well, I want to rate it ____ class instead"? I think what's actually wanted here is a simple note that says, "If you don't meet the GA/FA/FL criteria, then please don't bother applying for it!"
    • If one project has a lower quality rating than what you think the article is now, the likely issue is that the project rating hasn't been updated. You shouldn't write rules that assume that the project is hiding secret information about why a five-screen-long, fully referenced article is "still a stub". The much safer assumption is that the project hasn't looked at the article for months or years. Editors should ignore project ratings when they're working on articles. WhatamIdoing (talk) 22:38, 13 March 2009 (UTC)[reply]
      • Is there a simple way to update the banners to display the date/time that an assessment was done, maybe even a link to the diff? That way it would be very clear whether it was a recent assessment that might reflect project-specific criteria or an old assessment that may need to be updated. cmadler (talk) 09:01, 14 March 2009 (UTC)[reply]
        • The WikiProject tables such as this example show when the assessment was last done. Walkerma (talk) 14:19, 14 March 2009 (UTC)[reply]
        • It is a possibility, though for the banner to display assessment date on the page, where all users would see it. I'm don't have strong views either way on that idea. Other opinions? Walkerma (talk) 15:12, 14 March 2009 (UTC)[reply]
          • That would be the best solution. How many time i wondered when an article was last assessed. However we should propose and every projects should dispose. --KrebMarkt 15:31, 14 March 2009 (UTC)[reply]
            • And, um, you didn't just look at the talk page's history? I can usually find the answer to that question in less than fifteen seconds. WhatamIdoing (talk) 18:00, 14 March 2009 (UTC)[reply]
              • Yea, you are right. I still believe that it should more accessible than digging the discussion page history especially on a rather active one covered by multiple projects. I think that having a visible assessment time stamp would increase the number of re-assessment requests but exchange projects assessments tables would be more accurate. --KrebMarkt 18:48, 14 March 2009 (UTC)[reply]
          • It's a technical nightmare. AFAIK, all of the "last update" templates that exist need manual input to update the date (e.g. {{In the news/Next update/Time}}). But going back to the original proposal, I also object to the requirement that any articles proposed for GAN, ACR or FAC be required to be B. WikiProjects sometimes improve Start- or Stub-Class articles significantly in one shot, and send them to GAC directly in order to avoid spending time assessing an article that will be quickly reassessed after the GAC. Titoxd(?!? - cool stuff) 18:19, 14 March 2009 (UTC)[reply]
  • Oppose Apart from the reasons given above, there may be perfectly good reasons why an article has conflicting ratings if the subject is well covered from one aspect, but not from another. For example, x ray (which I haven't looked at) is relevant to medicine and physics. The article might all be about one and neglect the other. Johnbod (talk) 18:16, 14 March 2009 (UTC)[reply]
Well, then people defending a lower rating ought to add their reasons per the link in every rating banner. Otherwise, there is nothing stopping an interested party from adjusting the other ratings accordingly. -- llywrch (talk) 07:22, 15 March 2009 (UTC)[reply]
I once asked re-assessment to have harmonized rating from concerned projects for Torikaebaya Monogatari [279] so asking related project does work and it should be the proper way to do it especially if we are not sure of a project criterion for a given class rating. --KrebMarkt 08:22, 15 March 2009 (UTC)[reply]
Llywrch, would you have people leave ratings for thousands of articles "on spec", because a few of them might have conflicting ratings? Why not just drop a quick note to the project when you think their assessment is out of date? Every project I've seen that participates in 1.0 ratings has a designated section expressly dedicated to that purpose (e.g., WP:MEDA#Requesting_an_assessment_or_re-assessment. WhatamIdoing (talk) 04:29, 16 March 2009 (UTC)[reply]
  • Comment I think there's already a bot either updating conflicting ratings or merely adding a rating when only one project has one listed. Either way, a simple timestamp in the banner (updated by bot?) wouldn't be a bad thing since it could remind people to update the rating occasionally. As for conflicting ratings, usually there's no reason for them - but if a project does give a reason, it should probably be let stand (provided the rating doesn't conflict with GA/FA class). --Philosopher Let us reason together. 13:51, 15 March 2009 (UTC)[reply]
    • I think the bot is only adding a rating when only one project has done assessment; I'm pretty sure it doesn't change existing assessments. At least, that's what I've seen. cmadler (talk) 15:38, 15 March 2009 (UTC)[reply]
  • Oppose if the article is good enough to be proposed for FA or GA then it should be considered on its merits and not on a assessment from the past. If the assessor who gave the low grade cares they can contribute to the discussion on whether or not the article should have its rating upgraded. Of course that depends on whether the user has it in their watch list or someone is watching for the project. For many of the generic projects such as WPBiography the workers may be too thin on the ground, so they may appreciate any help. Once a GA or FA is achieved then all projects should accept that, otherwise send it for review. Graeme Bartlett (talk) 06:12, 16 March 2009 (UTC)[reply]
  • Oppose Per Graeme and others' comments - also let's say I am working on an article and wish to nominate it for review, but I'm in fact the only person in my area and really shouldn't be rating my own work above Start (or maybe C). Should I not be able to submit to a review process simply because of low visibility? Orderinchaos 04:07, 25 March 2009 (UTC)[reply]

Conflicting ratings (more comments)

[edit]

I think there are several points which people are trying to discuss all at once here.

  1. The "problem" of different WikiProjects giving different ratings to the same article. I'm not sure that this is a real problem. It merely reflects the fact that article quality is not completely objective, that different groups of editors will have different opinions. A second cause might be a multidisciplinary article which is very good in some aspects of its coverage, but not as good in others. The Version 1.0 Editorial Team doesn't seem to be too worried about conflicting ratings, and I would have thought that they would be the ones who are most affected. The "solution" to the "problem" surely has to be discussion between the projects involved, especially at the higher levels of grading.
  • Comment: From the perspective of the 1.0 team, conflicting ratings don't matter at all. Every article is checked for inclusion under every WikiProject that included it. If most articles are seen as improving, then an updated assessment from Project X will get the article into the release, even if an out-of-date lower assessment from Project Y caused it to fall short of inclusion. Walkerma (talk) 15:30, 17 March 2009 (UTC)[reply]
  • Comment: I also don't think this is a problem. In fact, sometimes it helps point up the fact that an article ought to be split. Example: An article discusses a novel and the film adaptation of the novel. The discussion of the novel is more complete, and WikiProject Novels rates it B-class, while the Film project rates it Start. This helps us realize that the film article ought to be split off so it can expand and organize the article about the film according to its own project guidelines. -- Ssilvers (talk) 01:32, 28 March 2009 (UTC)[reply]
  1. Recording the basis for an assessment. I proposed this a couple of hundred kBs ago for A-class articles, where the banner could contain a link to the A-class review. I don't think it's useful for the lower classes, as it just adds to the bureacracy of assessment without really giving us information that we would use in the future.
  1. "Drive-by" assessments, including bot-based assessments. These are certainly useful for the lower grades, but I think they ought to be discouraged for B-class and higher. There is a very real benefit from having the assessment done by the people who are most likely to improve the article, ie the WikiProject(s) concerned. I think this is a key to article improvement across the whole encyclopedia, which is the ultimate goal of the whole exercise.
  2. Making a certain project grade a prerequisite for GA and/or FA. This is a perennial proposal at WT:FAC, and there is considerable opposition to changing the current system. At the end of the day, such a change only requires a change in attitude of the reveiewers at the two Community-wide processes, which is not something which can just be ordered upon them.

Apologies if I've missed anyone's point: feel free to include it below if you think there are other points that need discussion here. Physchim62 (talk) 12:59, 17 March 2009 (UTC)[reply]

  • Comment. Personally, I think that GA should be a prerequisite to nomination for FA. The GA process is very efficient: One reviewer gives the article a high-quality review that touches on essential criteria. If you can't pass GA, you surely can't pass FA; and if you can pass GA, then you've at least got an article that won't waste everyone's time at FA. When I bring an article to FA, I usually do GA, then peer review, so that I really have a high-quality product going in to FA and I try to keep it efficient for the very busy FA reviewers. -- Ssilvers (talk) 01:32, 28 March 2009 (UTC)[reply]
If FAC wants to recommend that, or to give review priority to GAs, then they should feel free to do that. But I strongly believe that this group is the wrong place to suggest such a sweeping rule. Similarly, my personal complaint with GA is that the standards go up every month, when it was (once) intended to identify "good enough" articles instead of "practically FA" articles. (That's why they're good articles, not great ones.) But even though this is my long-standing opinion, this group really does not have the standing to tell all of Wikipedia how GA ought to be run. Instead, you should probably take your idea to FAC, and I should probably go promote a bunch of "good enough" GA nominees instead of whinging that most reviewers don't do it my way. WhatamIdoing (talk) 04:43, 28 March 2009 (UTC)[reply]

Why is there a need for a seprate WPFD?

[edit]

I think that MFD is more than capable of handling wikiproject deletions. Esperanza, the largest and most beuracratic of all the projects, was brought to its knees at MFD (even though the discussion had to have an archive). It is usually better to rehash or tweak an existing process or policy than to make an etirely new one.--Ipatrol (talk) 22:12, 13 March 2009 (UTC)[reply]

Ummm... there isn't a WPFD. The only thing that I found (just by going to the shortcut) was a userspace draft which has never been used (User:SatyrTN/WPfD). Could you please clarify your comment? –Drilnoth (TC) 22:57, 13 March 2009 (UTC)[reply]
I never heard of that one until you mentioned it. Deleting project should go by MFD there is no need of more cumbersome administrative process. --KrebMarkt 07:31, 14 March 2009 (UTC)[reply]
Does any one think we should delete WP:WPFD? (which is a redirect only).
It could just be redirected to MfD. EyeSerenetalk 09:53, 17 March 2009 (UTC)[reply]
 Done Agathoclea (talk) 11:07, 17 March 2009 (UTC)[reply]

Holy bandwidth Batman!

[edit]

Could those more familiar with the threads and issues please boldly archive some of this? It will help keep energy directed at current issues. Thank you! -- Banjeboi 17:28, 28 March 2009 (UTC)[reply]

Question

[edit]

Does this supersede or in some way resolve the status of Wikipedia:WikiProject/Naming convention? MBisanz talk 08:10, 7 April 2009 (UTC)[reply]

Importance vs Priority

[edit]

I have inexperienced editor pitching a fit because a rare cause of death has been tagged "importance=low". Supposedly, saying that having a good article about a rare cause of death is not particularly high on WPMED's list is "pejorative" and "insulting" (even though the editor asserts that it's more of a law enforcement issue than a medical one).

Do projects that use "priority" for this field get the same stupid complaints? Or do you trade these complaints for people changing the ratings in an effort to "force" or "trick" editors into working on pet articles? WhatamIdoing (talk) 16:26, 10 June 2009 (UTC)[reply]

I'd be in favour of abolishing the Importance scale. As far as I'm aware, it was created to assist Projects in focusing collaborative efforts to working on Top priority articles first. However, it has been my experience that people work on what interests them, not what is deemed Top priority. And Collaborations of the Months (the few that have been successful) have not been led by this importance scale, but rather someone who has a random interest and wants some help. The distinctions between High, Medium, Low are often very subjective. Do any projects actually use this importance-scale for something? --maclean 17:22, 10 June 2009 (UTC)[reply]
The WP:1.0 team uses it to decide which articles should be included in releases of Wikipedia.
WPMED has used it occasionally to drive article improvement, e.g., we decided that it was inappropriate to have "top" and "high" importance subjects be stubs, so we improved those articles. WhatamIdoing (talk) 17:35, 10 June 2009 (UTC)[reply]
In my experience, such complaints are rare these days, though they are commoner for Importance than for Quality. In some subject areas, like Military History, they would re-start World War II if they attached importance to their articles - even if it was called priority! As a result, many such projects don't use Importance at all; some, like Films or Biographies come up with a list to mark as "Top" (by discussion & consensus) and then they ignore all lower levels. However, in some subject areas like chemistry, the importance is rarely controversial, and so it's not an issue. Indeed, at WP:ELEMENTS it has become enshrined in the project's goals, and they've achieved a very impressive record by using that approach!
If the editor is otherwise a good editor, I don't think it's worth fighting about - maybe they could agree to "Mid" as a compromise? I'm sure you'd much rather be working on articles instead of arguing over metadata! If you need to take a stand over it, then find the statistics (August 2008) for the article by browsing Medicine articles on the Version 0.7 list; you can get information on no. of hits, links-in and interwiki links. If there are 200 hits per month, zero interwikis and 2 links-in, it is clearly Low. If there are 5000, 5 and 25, then "Mid" may well be appropriate. Note that we use these three parameters as well as the WikiProject importance parameter to judge articles for inclusion in WP1.0, so tagging an obscure article as "High" is probably not enough to get an article into the collection. Hope this helps, Walkerma (talk) 19:10, 10 June 2009 (UTC)[reply]
Here are the numbers:
  • Page links: 15
  • Interwiki links: 1
  • Hit count: 3,390
  • Importance score: 511
  • Overall score: 661
It's an anon who claims to have written most of the article. ("Almost half" is perhaps plausible.)
Perhaps more importantly, it's of borderline importance to WPMED. The anon (incorrectly) asserts that the article has no medical content at all and that WPMED shouldn't be allowed to consider the article to be within its scope at all. It's importance to WP:LE might be as high as Mid, but WPMED is probably going to stick with Low. WhatamIdoing (talk) 21:08, 10 June 2009 (UTC)[reply]

At WP:WPM, we partially switched from "importance" to "priority" to address this issue. "Importance" is ambiguous, as it could refer to the importance of the topic, or the importance of the article to Wikipedia and/or the WikiProject. "Priority" makes it much clearer that the issue is how high a priority should it be for Wikipedia and/or the WikiProject to have an excellent article on the topic. One example I often refer to is Srinivasa Ramanujan, whom I would rate as High importance as a mathematician and no more. However, his article is Top priority to the Mathematics WikiProject, because of its additional cultural significance and public interest in him. There are also articles where the balance goes the other way. See also (the historically named) WP:WikiProject Mathematics/Wikipedia_1.0/Importance Geometry guy 22:17, 10 June 2009 (UTC)[reply]

Those stats suggest an article that would most likely receive a "Low", but which might possibly receive a "Mid" rating from a WikiProject of fairly narrow scope (FYI- WP1.0 has a weighting for WikiProject scope to allow for this). So a "Mid" from Wikipedia:WikiProject Law Enforcement may be OK.
You should make it clear that the importance rating is the importance TO THAT PROJECT. My favourite example is at Talk:Albert Einstein - this article is ranked #7 among biography articles with a massive importance score of 1488 (so it would be included in V0.7 even if a stub!), and of course WP:Physics ranks it as "Top", but WP:New Jersey only ranks him as "Mid". Clearly his presence at Princeton is significant to NJ, but clearly not of top importance to the state. This is not "insulting" to Einstein or his legacy, it's a very appropriate rating for the scope of that project. Good luck! Walkerma (talk) 04:02, 11 June 2009 (UTC)[reply]
Indeed - we get similar issues in the Australian project. Orderinchaos 09:12, 11 June 2009 (UTC)[reply]
Barack Obama is a good illustration: that article gets all the possible importance ratings from Low (WikiProject Kenya) to Top (WikiProject Barack Obama, among others). Physchim62 (talk) 09:29, 11 June 2009 (UTC)[reply]

No recent activity on South Carolina Project

[edit]

I posted a question in the talk page of the project 2 days ago and got no answer so far. There has been no activity on the page for a few months but some people are editing the pages covered by the Project. How can I find out who the coordinators are for this project to try to get it moving again? Marine79 (talk) 13:00, 31 January 2010 (UTC)[reply]

Well look at the project members, see if the original or biggest contributors are still active, and then ask them. You may have to wait for weeks to get an answer, depending on when people log in. This page is not very active either. Graeme Bartlett (talk) 07:19, 29 November 2011 (UTC)[reply]