Wikipedia talk:Version 1.0 Editorial Team/Assessment/Archive 3

From Wikipedia, the free encyclopedia

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents 1 Set up 2 Policy tag 3 Banners and assessments 4 Featured lists 5 GA/A 6 Short article, not stub or start -- please create a new category 7 WikiProject Banners and assessment templates 8 Council discussion of interest 9 Fewer words? 10 Overhaul and rewrite of the assessment scheme 10.1 Rewriting the criteria 10.2 A class 10.3 Add A Class between B and GA 10.4 Eliminate the List class 10.5 Action plan 10.6 A Proposal 10.7 A vs GA 10.8 Recent related discussion 10.9 What would it take to establish a first-class foundation for Wikipedia standards? 10.9.1 Ideal 10.9.2 Common reaction 10.9.3 Small-scale test 11 Proposed reassessments section 12 Assessment of lead sections? 13 List assessment 14 Printed German Wikipedia 15 Votes on changing the assessment scale 15.1 Comments or suggestions on the voting process 15.2 Option A: No change 15.3 Option B: Put A below GA only 15.4 Option C: Add a new C-Class between Start and B 15.5 Option D: Combine A and GA into GA/A class 15.6 Option E: Replace Start-class with C-Class 15.7 Option F: Rename or remove GA-Class to separate the GA designation from WikiProject ratings 15.8 Option G: Community-granted A and WikiProject-granted A 15.9 Further comments 16 Proposed addition of option F: rename GA-Class as B+ Class 17 Proposed addition of option G: Combine A and GA into GA/A class, process delegated by commmunity in cases of mature WikiProjects 18 Proposed addition of option H: Separate DA/GA/FA (Wikipedia Quality Rating) from A/B/C/D (WikiProject Assessment Rating) 19 Option I - Proposal of C-class usage, with other special usages

Set up

I'm struggling finding a/the page which explains how to set a WikiProject assessment system up. I'm hoping to put one together for the Wikipedia:WikiProject Clans of Scotland. I have a Template:WikiProject Clans of Scotland set up but don't understand how to action the assessment. Anybody able to help? -- Jza84 · (talk) 13:31, 17 January 2008 (UTC)

You should find what you need at Wikipedia:Version 1.0 Editorial Team/Using the bot. :: Kevinalewis : ^{(Talk Page)}/_(Desk) 13:56, 17 January 2008 (UTC)

Thanks! I did! Assessment set up! -- Jza84 · (talk) 15:50, 17 January 2008 (UTC)

That's a great topic for a WikiProject but... erm... why do you need a seperate banner when a) not only does WP Scotland already have one, on which you could be a workgroup, b) {{WP Scotland}}'s documentation already lists a clans=yes parameter (although it doesn't seem to be active in the code)?

Since your project is entirely a subset of the Scotland topic, please strongly consider adding your code and blurb to that banner so that double tagging of talk pages can be reduced. --kingboyk (talk) 13:04, 20 January 2008 (UTC)

Policy tag

I don't believe this should be tagged as a policy, for two reasons. First, we limit the policy tag to a few particularly important documents expressing core community principles. This is not such a document. Second, we avoid the policy tag for pages that are meant to be purely advisory, such as the manual of style. This page is very similar to a style guide (it gives only suggestions and examples, but no requirements). — Carl (CBM · talk) 19:13, 20 January 2008 (UTC)

I don't have any objection to this being classed as only advisory. Walkerma (talk) 19:35, 20 January 2008 (UTC)

Well then how about this one? --Simpsons fan 66 06:27, 29 January 2008 (UTC)

This guideline is a part of the English Wikipedia's Manual of Style. Editors should follow it, except where common sense and the occasional exception will improve an article. Before editing this page, please make sure that your revision reflects consensus.

Er, I think that {{guideline}} would be more appropriate in this case. This isn't really a style guideline. Although I'm really not sure that it should have been tagged as policy either. Titoxd^{(?!? - cool stuff)} 06:37, 12 February 2008 (UTC)

Yes, that sounds good to me. Can I go ahead and add {{guideline}}? Walkerma (talk) 18:43, 12 February 2008 (UTC)

Yeah. I'd say it's widely accepted, as about a million articles use it. Titoxd^{(?!? - cool stuff)} 04:53, 15 February 2008 (UTC)

Banners and assessments

I note that the index page shows that the majority of articles that have a WikiProject banner on them are actually assessed. Of course, this doesn't mean that placing a banner on an article makes it more likely to be assessed, but it might. Perhaps we should try to get a banner on almost every page - it might help us speed up the process of getting virtually everything assessed. Of course, it might not, but there's nothing much to lose. It makes it easier for people to assess an article (and makes it clear that it is not assessed), as well as making it easy to find (present in unassessed articles category). And people know which project(s) to go to if they need input.

One problem is that there are sometimes no projects that are very closely related to the article; I find this often myself. Still, I think it would be a good goal for the project to get banners on all applicable pages, even if we can only use the orphan banner for now. Richard001 (talk) 04:35, 12 February 2008 (UTC)

This is the type of thing that has to be decided at the level of each wikiproject; some may decide to do it, some may decide not to. — Carl (CBM · talk) 12:40, 12 February 2008 (UTC)

This is a perennial proposal; maybe an FAQ should be created to address it. — Carl (CBM · talk) 12:35, 12 February 2008 (UTC)

Featured lists

Does the bot collect data for featured lists? How should the template be coded (I am working on {{WikiProject Dance}}) Paul foord (talk) 13:33, 23 February 2008 (UTC)

In the past we haven't, but I think we may be about to implement the change to allow use of FL-Class. See this discussion, and please add your support for FL-Class if you want to use it! As for the actual coding, that's too hard for me, but I expect early adoption from the Chicago WikiProject, and you may be able to copy things/ask for help from them. Walkerma (talk) 15:13, 23 February 2008 (UTC)

GA/A

I always think it's a bit strange that A is a better grade than GA. It's probably just me but I thought I might-as-well bring it up. George D. Watson (Dendodge).Talk^Help 21:19, 20 March 2008 (UTC)

If you actually look at articles, you'll find you are correct that GAs are generally better than A class articles. This is because the GAs have to go through a process to become GAs, whereas A class just requires one person to decide that an article is 'pretty gud' and slap an A on it (I've seen this many times, and usually remove them). In theory, though, an A is supposed to be a little better - the sort that wouldn't take much to become an FA. I can't stand the A-class myself, but there aren't any good suggestions for what to do about it. Richard001 (talk) 22:41, 24 March 2008 (UTC)

Delete it, perhaps? There's a definite process that is basically working for GA and FA; there isn't one that I know of for A. - Dan Dank55 (talk) 21:06, 4 April 2008 (UTC)

Short article, not stub or start -- please create a new category

Some articles are voluntarily kept short, because they cover a topic closely related to a much larger article. Their purpose is to clarify a concept, then redirect the reader. For example, tumor is kept short to reduce duplication with the huge, encyclopedic cancer article. Tumor will not grow larger. Therefore, it is not a stub nor a start. It is (and will remain) a short article. Please create a new category.

I also noticed you do not have a category for summary articles, as described in the WP:SUMMARY guideline. Emmanuelm (talk) 14:40, 24 March 2008 (UTC)

I think you're confusing length with quality Emmanuelm. If an article is short but good it should have a higher rating than start. Admittedly it's a bit difficult for short articles though. I've never seen a very short article become an FA, for example. They're harder to assess because they look just like a stub even though they may be almost perfect. It might only be the addition of a couple of sentences that would take them from start class to 'perfection', provided the writing is good. I don't think adding a new class is going to help though; I think they're always going to remain a bit awkward to assess.

I'm not sure what you mean by summary articles. Again, that's got nothing to do with quality. Many articles use (or should use) summary style in at least one section, and many are, or at least should be, summarized in another article. We have the template {{summary in}} for such articles, and obviously the template {{main}} (and, to a lesser extent, {{further}}) are used in the opposite cases. Richard001 (talk) 22:49, 24 March 2008 (UTC)

I understand what Emmanuelm is saying - there are certain articles which have limited potential to grow either because they are very specific topics or of low importance. But These articles could also go through stub-start-B-GA life cycle (if not FA). Hence I don't see the need to have a separate class for them. Arman ^(Talk) 10:45, 5 May 2008 (UTC)

WikiProject Banners and assessment templates

Should the general 'editorial team' assessment template be included in the banner shell with other WikiProjects? Richard001 (talk) 20:09, 27 March 2008 (UTC)

The template is a banner of the core topics, WVWP, and orphan assessment projects, among others... so it would be wrapping a banner shell inside a banner shell, which I'm not sure if we want... Titoxd^{(?!? - cool stuff)} 20:11, 27 March 2008 (UTC)

I think this is something we should discuss. I usually place it inside a shell if there is one, since the fact that "nested=yes" is an option suggests that this is reasonable, and I strongly support the effort to minimise template clutter. But if others feel strongly that it should stand alone, we could definitely consider it. Walkerma (talk) 21:22, 27 March 2008 (UTC)

I agree with you, though the problem seems to be in that this isn't technically a WikiProject, hence the need to have some sort of guideline to clear up the uncertainty here. Richard001 (talk) 22:07, 4 April 2008 (UTC)

Council discussion of interest

See Wikipedia talk:WikiProject Council#Non-standard banners Richard001 (talk) 09:51, 27 April 2008 (UTC)

Fewer words?

Discussion moved to Wikipedia talk:Version 1.0 Editorial Team/Assessment rewrite. Please comment there.

Overhaul and rewrite of the assessment scheme

Rewriting the criteria

The early part of this discussion, on rewriting the criteria, has now been moved to Wikipedia talk:Version 1.0 Editorial Team/Assessment rewrite.

There have been two proposals recently relating to assessment, and both seem to be reasonable (IMHO). They would both involve some rewriting and recalibrating, and therefore I think we should consider both proposals at the same time. I'm adding a third proposal, which is in effect how I think the first two would best be implemented together. There's also a fourth, which came up in discussions, and which I'll throw in for good measure. Walkerma (talk) 18:01, 12 April 2008 (UTC). The four proposals:

• Simplifying the descriptions
• Refining the assessment scheme
• Converting the scheme to summary style
• Add an FAQ

Discussion on these has been moved to Wikipedia talk:Version 1.0 Editorial Team/Assessment rewrite. Please comment there, and please sign up to help there also! Another topic (revamping the grading scheme) is described below; this may well be moved to a separate page soon. Walkerma (talk) 17:46, 12 May 2008 (UTC)

A class

A discussion regarding the usefulness of A class can is ongoing here. Just thought it is relevant that I draw your attention to this. Arman ^(Talk) 10:37, 5 May 2008 (UTC)

Add A Class between B and GA

The gap between start class and GA is huge and to cover this wide range B class alone is not sufficient. From a Wikiproject's perspective, we need the B class to specifiy the potential candidates for a GA upgrade. We also need the start class to identify articles that have just started but yet need some work to cover the minimum ground. Between these two stages there are lots of articles which are somewhat complete in the sense they don't have much more to add (mostly due to lack of importance of the topic) but are neither good enough to push for a GA drive. We need a class for these articles. In other words, we need 2 classes between GA and start. This can be achieve by either demoting A-class below GA, or by introducing a new class: B+ like what WikiProject:Mathematics did. Arman ^(Talk) 10:36, 5 May 2008 (UTC)

I pointed out below that if we do, I think it would be better to name the addition as C-Class, instead of B+ class, as the difference between B and B+ is tenuous in ordinary life. Titoxd^{(?!? - cool stuff)} 20:44, 11 May 2008 (UTC)

Eliminate the List class

List is not an article quality class. It is a "Type" of article. A List can grow from Start, gradually move past B-class and eventually become a Featured List. Hence I don't see the justification of retaining a "List-class". Arman ^(Talk) 10:39, 5 May 2008 (UTC)

Agreed. Besides, lists can be graded, as several WikiProjects have shown. However, there were long discussions about this in the past, and many people were in favor of having List-Class as a quality category, alongside FL-Class. Titoxd^{(?!? - cool stuff)} 20:45, 11 May 2008 (UTC)

• Erm, question. If List class was removed, wouldn't FL go as well? Steve Crossin (talk) (review) 22:51, 14 May 2008 (UTC)
  • Not at all. It would just graduate to "FL" instead of "FA". – ClockworkSoul 14:38, 16 May 2008 (UTC)

I have to agree... if I had a choice, I would say that 'list' class should be deprecated and allowed to be graded like and other article. – ClockworkSoul 14:39, 16 May 2008 (UTC)

We debated this extensively last year; the projects who use a lot of lists were very vocal in their requests. One problem is that lists are harder to assess for quality. For example, if you have a List of disciples of Jesus it would only amount to (I think) 13 (Judas had to get replaced); so if all thirteen names are there, should that be a perfect quality article? There are no issues of quality of English, inline citations (usually), aesthetics (usually) to worry about, pretty much all that really matters is completeness. As the disciples example shows, one of the main quality issues with a list is "How wide a scope should a list be?" Should it be "List of known followers of Jesus from the Bible" with a subsection for the disciples?

In summary, then, the main thing we need is simply a tag that says, "This is a list and we want to track it for our project". I accept that the proposal to eliminate list-class has some sound logic; however, until someone writes a separate bot for handling lists, the current compromise seems (to me) to be the best way. Walkerma (talk) 16:12, 16 May 2008 (UTC)

Action plan

OK, hopefully we have enough people interested that we can get something useful done. (One thing - I personally will be mostly offline from May 22 till June 14, though I will have a lot more time for this initiative after June 14.) What should we do first? I would propose the following order as making the most sense:

1. Refine the criteria. Once we've got the detail done we can
2. Write a simplified description (in summary style, with links to the detailed version). After we've gone through all that we should have everything clear in our minds so we can
3. Write an FAQ page.

Does this sound like a plan? Walkerma (talk) 02:56, 8 May 2008 (UTC) If people are OK with this, we can perhaps by agreeing on a new page or pages for this. Walkerma (talk) 03:02, 8 May 2008 (UTC)

Sounds like a plan to me, and logically sequential. How much refinement of criteria do people think is needed? It's a good idea to draw on exemplars when refining criteria, so empirically anchored/based. Keeps checks on the descriptions. I have some comments/suggestions, which I'll make as soon as I can. Holon (talk) 13:56, 8 May 2008 (UTC)

A Proposal

I propose to add a new "who can promote" column to the rating table. My proposal for the new table would be something like this:

Class	Minimum criteria	Who can promote
Stub	A very short article or a rough collection of information that will need much work	Anyone (even a bot)
Start	Any two of the followings: 1) A particularly useful picture or graphic 2) An infobox with useful key information on the topic 3) Multiple links that help explain or illustrate the topic 4) A subheading that fully treats an element of the topic 5) Multiple subheadings with some contents that indicate material that could be added to complete the article.	Any human editor (no bot), a discussion of the assessment on the article talk page is useful but not required.
B	Atleast 4 of the 5 start class criteria. The article may have dispute on factual accuracy or neutrality but is tagged as such if that is the case.	Any user, who is ideally not a major contributor to the article. A discussion of the assessment on article talk page is required if the reviewer is one of the major contributors to the article.
A / B+ (new class)	1) Written in an understandable language, 2) No factual accuracy dispute. At least 60% of the content is verifiable from inline citations. Reference format may need clean-up; 3) Broad in coverage; 4) No neutrality dispute. 5) Stable; 6) Illustrated, where possible. In summary: the article can be Good Aricle candidate with some copyedit and clean-up.	Atleast 2 users - 1 should propose on article's talk page and a second one (who is not a major contributor to the article) should support and upgrade. Request for reasseeement may also be posted on relevant WikiProject Assessment department page.
GA	See GA criteria	GA Nomination process
FA / FL	See FA criteria	FA Nomination process

• Very Strongly Object to any effect to have 'a discussion of the assessment on the article talk page' for anything below A Class. 2 million+ articles to assess, just work out the time it takes! Three man years (24 hours and 7 days a week) of assessments at least. I personally assess about 2500 articles. At a rate of 1 a minute that is a weeks solid work. If you have to discuss each one that will invariably mean that a lot less assessments will occur. I thought the objective of the Editorial Team was to get more articles assessed not less. SunCreator (talk) 23:50, 21 May 2008 (UTC)

• Observations I guess A designation would be good enough, more so if the A criteria is put down for reference. By the design above B+ would not be very necessary. But, it would probably be mighty helpful to have a frame of reference for A criteria (I wouldn't recommend putting down the quantified part, i.e. 60%). Apart from these minor issues the proposal looks quite agreeable. Aditya^{(talk • contribs)} 09:41, 8 May 2008 (UTC)

• I don't see why discussion should be "required" for Start/B/B+ class articles. Those ratings serve mostly as a guide to help people tell which articles to work on, and anyone else can always change the rating later if they disagree with it.
  Also, I think the critera for "Start" above are far too weak. I usually think of start-class as: has some references to print sources and covers at least one aspect of the topic in depth. I don' think that adding just links, images, and infoboxes can turn a stub into a start-class article. — Carl (CBM · talk) 14:03, 8 May 2008 (UTC)
  • Explanation If I understood this correctly, a discussion should not be "required" for start or B class article assessment, though a note/comment to the talk page would be highly useful. Assessing A-class articles may not be as simple (as many of them already has to go through collaborative WikiProject evaluations and discussions). And... well, as start class article should ideally be considered just that — a start, may be a good start too. A lot may not be expected out of one such article.
    • Of course not much can be expected of a start-class article, but it's reasonable to expect it to be more than a stub in terms of the amount of actual text it contains. If links and an infobox were sufficient on their own, many of the bot-generated stubs like PPARG would be 'start-class'. Let's use that article as a starting point - what class should it be? — Carl (CBM · talk) 14:23, 8 May 2008 (UTC)
      • I like this suggestion of a "who can promote" column. I agree with Carl that we don't need to mention discussion as part of the definitions - if it happens, it happens. I disagree with Carl that Start must have print sources - that has NEVER been a requirement for Start; Start-Class covers a wide region from "more content than a Stub" to "getting to be quite a decent article, but either lacking refs or lacking several major content areas. In other words, the main difference at the lower levels is usually in terms of content. Many topics (esp. general ones) may have a decent article with zero refs, but members of the WikiProject (who know the topic well) know that it accurately covers the topic - such an article should never be classed as Stub. Walkerma (talk) 14:46, 8 May 2008 (UTC)
      • I'm going to move much of this discussion over to a new page, at Wikipedia:Version 1.0 Editorial Team/Assessment rewrite. We can put the drafts in the article space, and keep the discussion on the talk page. I'll leave the signup section and pointers here. Walkerma (talk) 14:46, 8 May 2008 (UTC)
      • I think PPARG is a stub! (Not much prose content) We see this a lot at WP:Chem, where we have a lot of data in a box but not a lot of prose content. This was a great article for discussion, Carl! Walkerma (talk) 14:51, 8 May 2008 (UTC)
        • Yes - the grading scheme has never required references for start class. I was only saying that I personally include that in my mental checklist when assessing an article. My work is almost always in the context of mathematics articles, but the math 1.0 grading scheme doesn't require refs for start-class either. But if we are talking about changing the criteria, I would propose that even a start-class article should have at least one reference to a source where readers can learn more about the topic.
          I think that if the phrase "Has a meaningful amount of good prose content" was added to the start-class criterion above, it would cover bot articles like PPARG. — Carl (CBM · talk) 15:05, 8 May 2008 (UTC)

A few observations:

• Explanation of A / B+ (new class): What I meant is, it can be called either "A" or "B+"; certainly not both. We can discuss about this name.
• Discussion requirement: I apprehended that some inexperienced editors may have a tendency to over-rate their own poor contributions. A requirement to discuss the rating (above start class) will at least force them to logically evaluate their own rating. Note: I suggested we require discussion for B class - only if the reviewer is one of the major contributors to the article. However, if majority of you think this will create unnecessary bureaucracy, we can drop this.
• Start vs. Stub: Note: The current proposal makes the start class more stringent than existing requirement (atleast 2 criteria, instead of just 1). the example article PPARG, in my opinion, is a valid candidate for a start class. Why? This article gives an enthusiastic editor enough material, link and reference to work on the topic and expand it. It also provides the editor a structure of the article, a relevant public domain image. The work on the article has already "started". We need to differentiate this class from articles which leave very little clue on how to expand them, e.g. Coffee table book (as of July 2005). Once we add some useful text materials to PPARG it will become a B-Class article based on proposed definition (note - it will yet not the B-class based on current definition). After further expansion, when the coverage of the article will become reasonably broad only then it will move to A or B+ class (equivalent or slightly better than current B-class), which will be the stepping grade for GA nomination. Arman ^(Talk) 02:04, 9 May 2008 (UTC)

• • Some projects (at least math) have both A and B+ classes at the moment. Very roughly, GA is a more bureaucratic equivalent of B/B+, and A has higher standards than GA in projects that have a review process for A-class articles. But there is no assumption that every article will proceed to the GA or FA process; these are completely optional. A lot of very reasonable articles are rated B.
  • The proposal above weakens the requirements for start-class, in that it removes the requirement "The article has a meaningful amount of good content". It is true that if that is ignored then articles like PPARG look like start-class. But really, there is almost no text in that article at all - it even has a stub tag at the bottom. A start-class article needs to have enough text (several paragraphs, at least) to be an "article" rather than just a "stub".
  • I strongly disagree with the proposed requirement for discussion for Stub, Start, B, and B+. These are not meant to by used externally to judge articles, like FA is. And they are not meant to involve any sort of bureaucratic process. They're only meant to be used within a project to keep track of progress and tell which articles need what sort of work. There's no reason not to just let people assign ratings as they wish, since anyone else can always change the rating later if they disagree. The A/GA/FA ratings have approval processes, for people who want to pursue them. — Carl (CBM · talk) 02:21, 9 May 2008 (UTC)

GA doesn't look like a more bureaucratic equivalent of B/B+ nowadays, especially since the new GA reassessment drive. On the contrary a lot of recent GAs look like a less bureaucratic equivalent of older FAs. But, I would agree that "requiring" a discussion for anything lower than a B would be bureaucratic. But, as a lot of inexperienced editors do tend to over-assess their own contribution, I would stand by the proposal that B articles can't be rated so in the absence of an explanatory note on the talk page. That would make over-assessors at least think a bit more. Aditya^{(talk • contribs)} 18:21, 10 May 2008 (UTC)

I really dislike the entire "who can promote" idea, as it makes a lot of assumptions about the ability of WikiProjects to assess articles. There's WikiProjects like MILHIST which have completely developed assessment processes, and for which these guidelines make a lot of senese; on the other hand, there there are nascent projects that would face obstacles with these criteria. For example, if a project only has one major editor acting as a reviewer, and that editor writes an article, who grades it?

Overall, I think this sort of thing is better for WikiProjects to decide for themselves. Titoxd^{(?!? - cool stuff)} 23:49, 10 May 2008 (UTC)

Although I prefer uniformity in these definitions accross Wikipedia, if the decision is to leave the definitions for WikiProjects to decide, I'm fine with that as well. My only request, in that case, would be either to promote GA class above A-class, or create a new B-plus class for auto generation of tables like this. Arman ^(Talk) 02:13, 11 May 2008 (UTC)

I'm not sure the plus-minus scale is such a good idea. I mean, even in academia the difference between a B and a B+ is not that sharp. If there is a need to further discretize the gap between GA and Start-Class articles, I think introducing {{C-Class}} would be more understandable. That would also retain the original meaning of A-Class (articles that are more or less ready for WP:FAC) and give the finer control some projects require. Titoxd^{(?!? - cool stuff)} 03:13, 11 May 2008 (UTC)

I would suggest considering the minumum number of categories required for the purposes for which the assessments are used, then work toward having that many. You can get more sharp distinctions but it takes more up-front work and probably more work for people using the scheme. Getting more genuine distinctions doesn't come just from adding categories. In my opinion, based on experience analysing many data sets from marking rubrics, it is highly likely to come from words alone at all -- rather it comes from the exemplars. For this reason, I'd recommend including the suggestion that new users at the very least briefly familiarize themselves with the exemplars. Having said that, it is possible to have multiple compatible versions for different levels of precision as I mentioned somewhere. I will also comment on this somewhere else. Holon (talk) 07:55, 11 May 2008 (UTC)

I agree, in fact, one of the main benefits IMO of the scale is its simplicity. However, there have been a lot of calls to increase the granularity of that particular aspect into three ranges: Start-Class, not Start-Class, but not close to GA, and close to GA. The latter one would be the homologue of the "close to FA" relationship that A-Class currently holds. Titoxd^{(?!? - cool stuff)} 08:18, 11 May 2008 (UTC)

A vs GA

This subhead is more or less an arbitrary break, but I do have an "A vs GA" question. Suppose someone writes an article on "Edison and the phonograph", applies for GA, and gets rejected. Then they take it to a wikiproject that deals mostly with biographies, and it's rejected for A-class. Then they rewrite it as a history article, and it gets rejected again. Then finally it gets accepted as an A-class technology article. Do we want to consider this an A-class article for purposes of 1.0? We don't have any way to keep track of text that is rejected in one context but accepted in another. I'm very happy that there are many wikiprojects that do such a good job of assessment that there's no need for their articles to be vetted by anyone else. But can just anyone hang up a sign saying "A-class ratings done here"? If so, doesn't that create a weakest link in the process? - Dan Dank55 (talk)(mistakes) 23:20, 10 May 2008 (UTC)

Well, the problem really is in the way articles are counted, rather than in the way they're graded. It is not a problem if two WikiProjects have different opinions on the assessment of different sections of the article, particularly for large multidisciplinary articles like Space or Time. Titoxd^{(?!? - cool stuff)} 23:53, 10 May 2008 (UTC)

This tends to be rarer than it should be - it's always tempting for a reviewer to copy over another WikiProject's assessment! The selection bot takes each WikiProject's assessment separately, and if it passes for one project but not another, it will still make it onto the DVD. I'd like to develop a correction parameter for differences between projects, but thankfully it's not so much of a problem with A-Class. I would say that we have a much bigger problem at the Start/B interface, which is likely to be the deciding point for many important topics. Some projects are really quite liberal with Bs, but others are very strict. Hopefully the new exemplars will help us reduce that variation. As for random/unworthy A-Class assignments, hopefully most WikiProjects will rein in such practices. I'd really like to see peer review for all A-Class eventually, but I don't think we're ready for that just yet. We should say that peer review is recommended, though. See the recent discussion (also mentioned below). Walkerma (talk) 05:04, 11 May 2008 (UTC)

The problem highlighted by walkerma with B-class once again reinforces the need to split class-B into 2 classes. Currently some projects are considering B-class as "close to GA", while others are interpreting it as "somewhat more than a start class". We do need to have two separate classes and probably the stable version will include the upper class, but not the lower class. Arman ^(Talk) 03:25, 12 May 2008 (UTC)

Reading the arguments above (this section and others) I wonder which one is redundant - A or GA? That "near to FA" or "progressing towards FA" thing applies to both. I believe more FAs have evolved directly from GAs than from A-grades, and anyways there are way fewer A-grades than either FAs or GAs, and mighty few WikiProjects can support a proper A-grade evaluation (review, discussion, improvement and all).Why do we have a GA-grade if it's not really important (we already have B-grades and A-grades), or may I ask why do we have A-grades (we already have FAs and GAs)?

One wold expect an article to go through increasingly stringent quality assessments as it goes higher up the ladder: B by individual assessors, GA by a dedicated process, reviewers and criteria, A by a few WikiProjects, and FA by community consensus and dedicated process and criteria. Doesn't the role of A-grades look a bit dubious here?

I'd say the assessment of A-grades higher than GAs represent unfairness and injustice to articles and editors who don't belong to strong WikiProjects like WP:CHEM or WP:MILHIST. Many articles supported by weak or even non-functional WikiProjects have gone on to become FAs, but they never had an opportunity to go through an A-assessment. Aditya^{(talk • contribs)} 05:06, 12 May 2008 (UTC)

The thing is, GA/FA and SSBA are in reality two different scales that are juxtaposed together. The A-Class classification, as well as the rest of the SSBA system, is meant to be internal to WikiProjects only, while both GA and FA are public in nature. The reason both GA and A are needed are because they have completely different purposes. A-Class has no real importance by itself (unlike GA and FA), hence it doesn't actually need the project-wide review that GA and FA have. The class's lightweight assignment process is not a bug, it's a feature by design.

Also, the reason A-Class is sandwiched between the two public processes is three-fold: first, GAs tend to be more focused on stylistic concerns, while A-Class is supposed to be an assessment of the article's content by subject experts. Since in reality, WP:1.0 is more worried about wrong content than the wrong type of dash in the article, we decided to put A-Class first. Another reason is that historically, A-Class has been thought of as "ready for FAC, just needs some sprucing up". Furthermore, there have been somewhat-recent suggestions to remove GA from the scale, but having GA was considered helpful in these previous discussions. Titoxd^{(?!? - cool stuff)} 05:19, 12 May 2008 (UTC)

Well, keep in mind that the converse would be "unfairness and injustice" to said strong WikiProjects which actually try to work with this system. Designing everything around the lowest common denominator isn't really a good idea, in my view, since a vast number of WikiProjects happens to be entirely dysfunctional at the moment. There's no reason why their failures need impact the functioning ones. Kirill ^(prof) 13:17, 12 May 2008 (UTC)

Three questions:

• Do we really have to juxtapose two different streams of assessment lying parallel that doesn't necessarily compliment each other, or can we build it like individual assessment and small-group (i.e. WikiProject-centered sub-communities) assessment, the SSBA part, leading to Wikipedia (i.e. community process/consensus) assessments (Stub > Start > B > A > GA > FA)?
• Is the Wikipedia driving towards an integration of quality and coherence and a convergence of efforts, or is it driving towards a WikiProject dependent fragmentation in quality and coherence and a divergence of small-groups as opposed to individuals contributing to the project and the community?
• Is it unfair to believe that the few strong WikiProjects that can generate A-grades (which may or may not be compatible with GAs) can work on the SSBA module without A-grads popping in between FAs and GAs, and also work on other systems (if I remember right, at least one has A coming lower than GA and another has a B+ to come on top of B) as they see fit? Aditya^{(talk • contribs)} 13:59, 12 May 2008 (UTC)

The issue is that A-class and GA-class aren't really comparable at all. GA review (and, to a significant but lesser extent, FA review) is dominated by stylistic concerns rather than content review. A-Class review typically focuses heavily on content, and less on style issues. By the time an article has reached A-class, it can probably just have the style issues brushed up and then be moved to FA review. GA is much more like "B-Class with fewer MOS issues" than it is like A-Class. — Carl (CBM · talk) 14:04, 12 May 2008 (UTC)

So, based on the explanations given, it now seems to me, for GA-s, it is good enough to have a broad coverage of the topic, not necessarily complete; but for A-class, the coverage needs to be somewhat "complete". If that is the case, I understand A-class being above GA. But this distinction needs to be further clarified in the description of the grading scale. Furthermore, since some wikiprojects already have an A-class assessemnt process, I now recommend to leave it as it is. WikiProjects that don't have a strong A-class assessement framework should just skip it. But to address the need of a separate class between B and GA, I recommend to officially introduce the B-plus class, so that WP 1.0 bot can pick it up. Arman ^(Talk) 02:42, 13 May 2008 (UTC)

Agreed on both counts. The only thing I see a problem with is calling the new class as B+; I'd instead prefer to call it C, for the reasons I postulated before. If projects will be debating whether an article in a newer scale is a B or a B+, it is the same amount of work as to whether the article is a C or a B. Titoxd^{(?!? - cool stuff)} 02:50, 13 May 2008 (UTC)

(Reduced indent) Shouldn't WikiProject dependant assesment (i.e. SSBA) come first followed by community processes (i.e. FA and GA) in the structure? Is it necessary to maintain two different systems overlapping each other? As I understand, GA has come to its current status only lately, undertaking a long journey. But it has done so, and it is only prudent to respond to development. Aditya^{(talk • contribs)} 16:01, 13 May 2008 (UTC)

They could, but I don't see why they should. These are two different assessment purposes, not just processes. Titoxd^{(?!? - cool stuff)} 02:18, 14 May 2008 (UTC)

Agreed. And I believe the proposed option F below to completely separate Good article assessment from the WikiProject scales addresses many of the concerns in this thread by recognising that SSBA and GA-FA have a different nature and purpose. Geometry guy 11:53, 19 May 2008 (UTC)

Recent related discussion

FYI:

• Wikipedia:Village_pump_(proposals)#A_class A proposal to tighten up the A-Class review, which broadened into a discussion of the A/GA order. Some of that discussion has (in effect) spilled over into this discussion, which seems fine to me as we consider how we write things up.
• Wikipedia:Village_pump_(policy)/Archive_46#articles_assessements: My reply includes a complete history of the assessment scheme, with links - which could be very useful background information for those involved in this rewrite.

Walkerma (talk) 05:04, 11 May 2008 (UTC)

What would it take to establish a first-class foundation for Wikipedia standards?

After looking through the discussion here, I think it might be instructive to describe an 'ideal' process, and to work back from there to what's doable. Pretty much every issue that has come up here is fairly common in assessment. I hope it will be easier to see why from the ideal. Please keep in mind that the work put into what I outline to follow overlaps with normal work on articles anyway, and in the long run would likely make that work far easier by helping to identify what needs to be done and when.

It may also turn out something closer to the ideal is achievable with available skills than I realize. With some ingenuity, Wikipedia could be a first for online ratings en mass by developing a top-class process based on solid foundations! OK, not likely, but possible. It's already considerably better than the crude methods normally used, such as ratings of 1-10 plucked out of the air.

Ideal

Given the nature of articles, the following process in the ideal is what I would (and do) recommend.

1. Compare (pairwise) a set or sets of exemplars and scale them.
2. List in order from worst to best (by links). This provides an ordered set analogous to a ruler with many points of possible distinction.

This achieves two things:

1. When the scheme is refined, it can be refined not only based on what 'should' distinguish better from worse, but what is seen to in a carefully calibrated and ordered set of examples.
2. The set of examples can sit in the background and be used whenever anyone wishes to, for greater precision (you can always go from cm to meters).

The last point is key when articles are near a threshold for going from one "grade" to the next.

Probably more important than all else an ordered set of examples provides a clear picture for editors of what it takes for an article to progress toward the highest standard.

Common reaction

A common reaction to this is that it's too time consuming because most are used to easy, but poor, rating processes (e.g. pluck a number from 1 to 10 out of the air or a grade based on best guess).

I understand but my standard response is that the payoffs outway the up-front time, often by a large factor, and of course anything worth doing takes some effort and coordination. The only reason most of us can buy a thermometer and easily, yet precisely, measure temperature at will is that a lot of work lies behind its development and construction. Like anything else, including articles on Wikipedia themselves, quality products require some work.

Good measurement instruments and procedures are a cornerstone of industry and technology -- without common standards, many things are impossible in industry. The same idea applies to Wikipedia as a whole. If editors can quickly, yet precisely, measure against calibrated standards as they work and assess articles, there are similar payoffs. There is a lot more clarity on standards and how to know where you are and what it takes to progress.

I believe around a million have been assessed, is that right?

However, it's like everything, it does take time and coordination. Hopefully though, this helps in explaining various issues and how they all fit together in the bigger picture even if nobody actually ends up participating.

Small-scale test

I can offer to anyone who wishes to do a small scale test in their own project. I don't think I have yet encountered a case in assessment where people have not found the process informative and useful.

Send me a set of article labels, preferably 15 or more, and I'll send back a spreadsheet with a set of pairwise comparisons to be done: each to be compared with each other and a judgment made about which is better. Do these and send me back the results. I will scale it, put them in order, and tell you how consistent you were overall and tell you which articles were anomalous, if any. Include at least two or three of the articles in the scheme so you will be able to see how the rest scale in between. If you can organize more than one judge to make comparisons, even better, and I can give you feedback on each judge's consistency and the agreement between them.

This should be quite quick for someone who is reasonably familiar with the set of articles, if the assessor only needs to refer to them when it's hard to say which is better. Most judgments should be quick and only a portion take more time. The payoff -- for your project you get a much clearer picture of the way articles progress from worse to better quality, and you have a far more precise basis for judging when an article should move up a grade.

This can be extended across projects. This would simply require choosing a number of articles in your project as well as some from another project also doing a calibration exercise. All articles can be scaled jointly and tests conducted to see how successful the exercise was. It's preferable that the assessors have some knowledge of the other articles, but I doubt it would be necessary for them to be experts on the content to get worthwhile results.

Obviously, this requires coordination if it crosses editors and particularly projects. However, the result could be a nice list across projects of articles from the worst to higher quality that everyone can refer to pluse the benefits to the project mentioned.

So to reiterate, this process is beneficial for

• refining the scheme by seeing what actual progression looks like, according to consistent judgments by a methodical process.

• provideing a set of examples (behind the scenes) that includes examples in the scheme, and can be used when the call between one grade and the next is getting difficult, avoiding debate the number of classifications (there is more precision if you want it and editors would know more clearly when an article is getting close to progressing to the next grade).

• giving a clear summary picture of what it takes to progress articles for editors, which would probably also reveal things not anticipated up front.

• founding refinements on the information to make the criteria more accurate, so more efficient to use and more credible.

I know there's a lot, but I hope it gives a clear picture of the ideal, and it might spark ideas even if nobody elects to do a trial.

Don't hesitate to criticize -- believe me it's unlikely you'll raise anything I haven't heard many times, and if you do, I'll be grateful for the challenge.

Cheeers all. Holon (talk) 10:45, 11 May 2008 (UTC)

Sounds excellent to me- sign me up! I have a few questions before I collect my list:

• By "article labels" do you mean things like "Aesthetic appeal," "Verifiability" or what? Or do you just need a list of actual articles? Or both these things?
• If you need a list of actual articles to compare, would you prefer a list of similar topics (with just variation in quality & completeness), or would a broad selection of topics be better? Personally, I'd like to send you two sets - one set that could be all (say) inorganic compounds or cities, and another that could include a wide variety (e.g. Antoine Lavoisier, electrocyclic rearrangement, Claisen condensation, Enthalpy of fusion, stable carbene, HSAB theory, Kraft process). Cheers, Walkerma (talk) 03:17, 13 May 2008 (UTC)

I too was confused by what you mean by labels. I can send you 15 hurricane articles easily... ;) Titoxd^{(?!? - cool stuff)} 03:19, 13 May 2008 (UTC)

Sorry, no I mean the names of the articles. The only thing that matters is that the names/lables allow whoever judges to know which articles, at what time (if relevant), those names/labels refer to. Certainly, if you can, both sets will give you more information; i.e. a similar set and broad set of articles. So long as you can do N(N-1)/2 comparisons for each set, where N is the number of articles. Comparisons will be faster if you're familiar with some or most so that you only need to refer to them sometimes, of course. However, no matter what you select, some comparisons will be so obvious they only take a moment, because the quality is far apart. Closer ones (quality) should take more deliberation. Once you've done this, future assessments against the ordered set can be made simply by choosing the closest match. This is the payoff related to getting precision when you need it, in addition to having a clearer basis for developing criteria etc. As you noted, it is important the articles represent the range of interest. I'd recommend having something very high quality and something very low (toward extremes) for reference points. If anyone wants to do just one set of articles, best is to select a set not too similar, because the results may be artificially consistent, but also not unsually different articles that are especially difficult to compare for some reason. Exemplars should have relatively little noise, without being contrived in terms of being prototypical in all respects. It is best just to choose a slection of articles that you consider reasonably representative of a larger category or population of articles, and to deliberately select a range. Also, without fussing too much, it is best to have exemplars across the range. Just send the list in a column in Excel, or post here for that matter. Then, I'll send back the pairs to de done and brief instructions. Any more questions, let me know. Holon (talk) 14:39, 13 May 2008 (UTC)

OK, I've created a list covering a broad range of topics, importance and quality in the general area of chemistry. I've emailed you an Excel sheet, and also put the list up at User:Walkerma/Sandbox2. I'll try to prepare a "similar topics but differing quality" selection as well, but right now I really should get on with marking lab reports! Thanks a lot, Walkerma (talk) 17:21, 14 May 2008 (UTC)

Here's WP:WPTC's list: User:Titoxd/Sandbox/List. Titoxd^{(?!? - cool stuff)} 01:30, 15 May 2008 (UTC)

Titoxd -- can you select a subset of 20 to 30 articles and I'll send you the set of comparisons. Cheers Holon (talk) 05:21, 17 May 2008 (UTC)

Will do. Titoxd^{(?!? - cool stuff)} 09:07, 17 May 2008 (UTC)

Okay, I updated the sandbox link above with twice the articles. Titoxd^{(?!? - cool stuff)} 09:15, 17 May 2008 (UTC)

Sorry, have had no time lately, I'll get the list to you soon. Cheers Holon (talk) 11:41, 29 May 2008 (UTC)

Proposed reassessments section

I'm not really sure how we ended up with this section of the page, but people keep adding entries. Is anybody actually reviewing these pages, and should we even have that section to begin with? Titoxd^{(?!? - cool stuff)} 05:46, 30 April 2008 (UTC)

Yes, I was wondering that myself! Walkerma (talk) 02:45, 8 May 2008 (UTC)

I have flagged the section (lamely) for removal. If somebody is actually using these entries, please provide usage instructions on the page and remove the {{off-topic}} template. If somebody can state definitively that the material is being ignored, then please be bold and jettison the section. Trevor Hanson (talk) 05:42, 28 May 2008 (UTC)

I cannot state definitely that the section is being ignored, but I'm quite certain that it doesn't belong there. Requests for reassessments should be brought to the assessing project(s), and not here. We don't do that here. – ClockworkSoul 06:01, 28 May 2008 (UTC)

Thanks for taking action. I was pretty sure this was the right step, but thought there was a small chance this was an informal mechanism being used by...well, by somebody. – Trevor Hanson (talk) 21:54, 28 May 2008 (UTC)

No problem. I've never been accused of not being bold enough. :) Seriously, though, it was added by a new user who probably just wasn't aware of how assessments work, and in no time it started to attract a small amount of attention. It was better to just remove it now rather than let it get too big to easily excise. – ClockworkSoul 23:00, 28 May 2008 (UTC)

Assessment of lead sections?

WMF got $3 million in March from the Sloan Foundation to promote stable and written versions. I hope that means that there is money available in support of Version 1.0. There are a variety of organizations that would probably jump at the chance to be hired to help help with vetting, fact-checking and copyediting, because it would increase their own visibility. If we could get some serious help, I'm wondering if we could dramatically increase the number of lead sections, or lead plus first sections, that make it into some version of Version 1.0. (If the only two options for inclusion in 1.0 are "lead" and "the whole article", that will tempt people to cram stuff into the lead that really doesn't belong there; better to offer a third option of "lead plus".) Should there be a separate assessment for "lead" and "lead plus", or should the quality of the lead and the first section simply weigh heavily in the assessment of the entire article? - Dan Dank55 (talk)(mistakes) 21:06, 10 May 2008 (UTC)

I'm not so sure about the lead+first section approach. Many articles are organized chronologically, so taking only the first section would actually remove most of the coverage inherent in them. Titoxd^{(?!? - cool stuff)} 21:51, 11 May 2008 (UTC)

Okay, what do you think of the idea that "lead only" (across an entire wikiproject, say) is probably a bad idea, because people will respond by pushing things into the lead that shouldn't be there? Perhaps, if people like the idea of throwing in the lead sections from 100 articles, we could use summary-style in reverse: figure out why we're attracted to that idea, what it is about those 100 articles that makes the leads worth including, and then write an article that incorporates material from those leads, that links to those articles as "main articles". - Dan Dank55 (talk)(mistakes) 16:25, 12 May 2008 (UTC)

It's just that I don't know where we're using lede-only articles for inclusion. I thought that was an idea suggested a while ago, but that never really got off the ground. Titoxd^{(?!? - cool stuff)} 19:55, 12 May 2008 (UTC)

I'm quoting Martin, but I don't know whether the idea has gotten off the ground, either. - Dan Dank55 (talk)(mistakes) 20:58, 12 May 2008 (UTC)

My understanding was that it didn't get off the ground simply because we haven't done it yet, not that we'd abandoned the idea! Sites like Wichempedia (which is a mirror of WP chemicals lead sections + chemical data, see this example) manage to do this very easily (I know the guy running the site, and it was an hour or two for him to add the leads).

I've often heard that many times people use WP just to find a basic definition, or basic piece of info that is in the lead. We could use this "leads only" approach to produce a concise-form encyclopedia, that could have a million text-only leads on a single DVD, flash drive or easy download, or perhaps 40,000 on a CD. I agree with Tito about the "first section" - that section tends to be no more important than later sections IMHO. I think we can worry about people bloating the leads if and when it happens - personally I don't think it would be a major problem. Assessing leads - that's an interesting idea, but it could get complicated!

I'll try and talk to the relevant people to find out how this project could benefit from the financial support. Thanks a lot, Walkerma (talk) 02:35, 13 May 2008 (UTC)

I doubt that people would start cramming stuff into the lead, but if anyone does that sort of thing just tell them it's inappropriate. A lead is just a short (but not too short, which most of them are) summary which shouldn't go into too much detail. Richard001 (talk) 01:47, 14 May 2008 (UTC)

List assessment

I can't find any guidelines on assessing the quality of list articles. Are there any? -Freekee (talk) 16:06, 11 May 2008 (UTC)

I think it's either a list or it's not, and if it is it's either a featured list or it's not! However, now that lists are "official" (they are read by the bot) we should at least mention something in the guidelines/criteria - good idea. Thanks, Walkerma (talk) 17:24, 12 May 2008 (UTC)

Printed German Wikipedia

I get a little more worried every month. Other organizations are spending lots of money on article creation, and we're not, and we're getting closer and closer to being toppled from our perch. While we have generated a little over 6000 FAs and GAs, the German Wikipedians are getting ready to publish a 50000-article, 1000-printed-page encyclopedia, with generous financial support for article proof-reading and fact-checking from their publisher, Bertelsmann. Bertelsmann owns Random House, and if our publisher can't come up with similar financial support, I would support approaching Random House for help. - Dan Dank55 (talk)(mistakes) 16:31, 12 May 2008 (UTC)

The Germans had their first offline release in 2004, ours was in 2007. They did a lot of things right, particularly in just getting on with the job instead of just talking about it! However, we are on track for a 30,000 article release on DVD by late summer, which will be a big step forward. The Polish WP followed the German model, and although they produced a very nice DVD they found they were at the mercy of the publisher (which had paid for article checking etc). We have several things the German's don't:

• Fully open software; the German offline reader can only be used by the publisher, but we have an open source reader written by our publisher, Linterweb. This in turn allows our output to be used for free giveaways like One Laptop Per Child.
• A very powerful assessment scheme covering all corners of the project. We can filter out the dross far more effectively - so our selection can be much higher in quality than the German one (assuming equal quality of overall content). We don't need a publisher to control our article selection process.

We would dearly love to produce our first printed materials, but my personal attempts didn't work out. There are all sorts of issues that lie behind these decisions, and the spotlight is on us far more than with the German Wikipedia (so we can NOT screw things up!) With the Sloan money mentioned above, maybe now is the time to consider an initiative. Walkerma (talk) 17:22, 12 May 2008 (UTC)

Thanks Martin, that was very helpful. I'm still worried that we will eventually lose the race we started, but as long as we're at the top of the Google rankings, there's plenty of time to fix things. - Dan Dank55 (talk)(mistakes) 19:52, 12 May 2008 (UTC)

Votes on changing the assessment scale

Proposed addition of option F: rename GA-Class as B+ Class

With apologies to Walkerma, I'm afraid I really must propose the addition of an option F. I'm proposing it in a new section per request. The purpose of this option is to clear up a long-standing headache in the assessment system, which is the widespread confusion between GA-Class and Good article. The Good article process and the WikiProject assessment systems are totally different content review processes with different goals and different methodologies. The current scale attempts to shoe-horn a community-wide assessment process which focuses on style and policy-compliance issues into an assessment scale that varies from project to project and is managed by content experts. This doesn't wash and causes problem after problem. That this is an issue can already be seen in the fact that an article can be A-Class without being a Good article, and of course a Good article need not be GA-Class (with this again varying from WikiProject to WikiProject). Thus the current scale looks something like:

Stub - Start - B - A
               |   |
              GA -FA

Further, the current system leads to Good article reviewers adjusting ratings without any sort of WikiProject consent. When I delist a Good article which is A-Class, do I drop it to B? Good article reviewers also often do not have the content expertise to judge the place of an article on a WikiProject assessment scale.

What I am proposing is that we remove the Good article issue entirely from WikiProject assessment scales by renaming GA-Class as B+ Class. This recognises the fact that Good article is a community-wide (or at least, project independent) seal of approval that the article meets basic standards, and allows Good article status to be awarded to articles of any class except FA-Class. The green dot could be added as a symbol to the rating if desired. For the most part, this would mean B+ and A articles have a green dot. Good article reviewers could add and remove the dot without affecting WikiProject ratings.

If we could sort out this headache, it would make my day. Geometry guy 10:42, 19 May 2008 (UTC)

• I support this proposal--or indeed, any variant; for instance the projects could work according to an A/B/C/D scale, while FA and GA would remain community-wide assessments. I note that the above discussion seems mainly to come from people involved in the projects. It would be good to drop a note with those involved in GA and FA, to make them aware. --jbmurray (talk • contribs) 10:50, 19 May 2008 (UTC)

• I support this change, but not the renaming part. We don't want to get a green dot in mainspace at the expense of further demoting ourselves (Or the FA crowd will have one more reason to oppose the green dot proposal: they will say "it was a GA class but now it's only a B+ class, not even reaching A." OhanaUnited^{Talk page} 17:44, 19 May 2008 (UTC)

• I'm not sure how this separates GA from the project assessment scale - renaming GA to B+ seems to me to be wedging it firmly into the scale:

Stub - Start - B - B+ - A
                        |
                       FA

Have I misunderstood? For the record, like jb I'm all for anything that will help to make things clearer ;) EyeSerene^talk 16:59, 19 May 2008 (UTC)

Yes, you both have. Here is a better diagram, which might help clarify these misunderstandings

Stub -> Start -> B -> B+ -> A -> FA Class (WikiProject class assessments)
                                 ||
                           GA -> FA       (Community wide/independent content review and FA-Class = Featured article)

The purpose is not to rename Good articles as B+ !! It is to separate completely the Good article designation from the WikiProject assessment scheme. Your misunderstanding demonstrates the huge problem with referring to articles as GA-Class.

I would also note that this issue has received some attention on this page recently: see Wikipedia_talk:Version_1.0_Editorial_Team/Assessment#A vs GA. Geometry guy 19:11, 19 May 2008 (UTC)

Thanks for the clarification - the similar nomenclature is a recipe for confusion, and I'm still not sure I understand the difference between a Good Article and a GA-class article. However, could your diagram be incorporated with the Stub > Start > C > B > A scheme that's been proposed above? It may result in exisiting 'B' class articles being reclassified as 'C' class, but it would avoid 'B+' as a somewhat awkward designation. EyeSerene^talk 19:31, 19 May 2008 (UTC)

If both this proposal and C-Class were accepted, the diagram would become

Stub -> Start -> C -> B -> B+ -> A -> FA Class (WikiProject class assessments)
                                      ||
                                GA -> FA       (Community wide/independent content review and FA-Class = Featured article)

for reasons of independence. I agree there is an opportunity to eliminate B+ by using the new B for the best current B-Class articles, and the new C for the rest. This is why I haven't yet supported the C-Class proposal. Geometry guy 19:56, 19 May 2008 (UTC)

I agree with most of what you say, but for simplicitys sake I think we should go with A B C instead of A B- B+.. I do support the separation of community evaluations: GA FA from wikiproject evaluations: A B and hopefully C Acer (talk) 20:01, 19 May 2008 (UTC)

Okay. In view of these remarks, option H below, and remarks made by editors in the other options above, I've added a more generic option F to separate the community GA evaluation from the WikiProject assessment scheme. Geometry guy 10:17, 20 May 2008 (UTC)

• Strong oppose The solution should not be to remove GA from the assessment scale. The solution should be a tighter integration of all review & assessment options across all levels (FA, GA, PR, and the wikiproject assessment scales). We need to work together here; don't isolate one group and leave them out. Dr. Cash (talk) 14:59, 20 May 2008 (UTC)

• Strong Oppose Absolutely not. This proposed rename would be totally confusing, and it would be a watering down of what GA means. Dr. Cash hits it on the head, leaving out GA is big mistake, considering we have assessed thousands of quality articles. Quite a few more than FA, actually. VanTucky ^{Vote in my weird poll!} 02:55, 21 May 2008 (UTC)

I think this is a further illustration of the confusion created by having two assessment schemes with similar names, and proof of the need for Gguy's proposal to be implemented. WikiProject GA-class is not the same as Good Article status, as awarded by WP:WPGA. EyeSerene^talk 07:42, 21 May 2008 (UTC)

Umm, actually GA-class IS the same thing as Good Article Status. When articles are promoted to GA, they are assigned GA-class in the assessments. I am still waiting to hear from someone regarding my request for an actual wikiproject that assigns a 'GA-class' assessment without going through WP:GAN. All that's been offered so far is one instance of one user assigning it to a wikiproject, which was later reverted, with an actual nomination and review taking place several weeks later. Dr. Cash (talk) 04:06, 22 May 2008 (UTC)

I apologise for my lack of clarity ;) Let's see if I can explain myself better: I agree, the guidelines on the V1.0 grading scheme currently state that WikiProject GA-class should be awarded by taking an article through the Good Article assessment process; thus a Good Article award also translates into a GA-class award. However, this leads to contradictions later on - for example, if a WikiProject then reassesses the article as A-class... Clearly it cannot be A-class and GA-class at the same time, and the A replaces the GA in the WikiProject template, but the article is not removed from WP:GA - it retains its Good Article status, despite no longer being a GA-class article. This seems completely illogical to me, and suggests that we should either:

• fully integrate Good Article status into the hierarchy, so an A-class award would automatically replace GA status (can't see this happening for all sorts of reasons!)

or

• remove the GA-class from the WikiProject hierarchy, so GA status is completely separated and no longer translates into an equivalent class on the hierarchy (which is, I think, what Gguy's proposing)

I hope that's clearer! EyeSerene^talk 11:47, 22 May 2008 (UTC)

I don't really see the issue here. The way I see it, A-class articles being above GA means that they should adhere to the GA criteria, and then go beyond. I read that as A-class being a subset of GA, sort of like a GA+. And I have, and will continue to, delist A-class articles that are not GA, especially if there is no A-class review appearing anywhere in article talk or archives. If you are going to remove GA from the assessments scale, then we should remove FA as well. Let the wikiprojects do their own thing from stub-->start-->B-->A, and then let the community assign a Featured or Good status to it. The same arguments for having A below GA, also apply to FA -- how can an A-class article be "below" FA, if A is supposed to be the top? But removing GA from assessments scale on its own is only devaluing the project, which is something that I simply cannot agree to. Dr. Cash (talk) 15:20, 22 May 2008 (UTC)

Actually, I think that suggestion ("Let the wikiprojects do their own thing from stub-->start-->B-->A, and then let the community assign a Featured or Good status to it.) has a lot to recommend it. GA and FA have to serve the entire community, not just those articles that also happen to come under a WikiProject, so why not make the distinction clear? As you say, A-class articles exist that would not qualify as Good Articles, and although according to the V1.0 assessment scale that shouldn't really be the case, barring FA and GA WikiProjects are ultimately free to classify their articles in whatever way they see fit. The assessment scale is not, despite appearances, unified or even consistent. I do understand your objections and see the sense behind them (and it should go without saying that I certainly don't want to devalue WPGA!), but from other comments on this page the way you regard A-class (as a subset of GA) clearly isn't the way some others regard it. Doesn't this indicate, if we're not even all on the same page, that the whole thing needs deconstructing? EyeSerene^talk 17:00, 22 May 2008 (UTC)

I could support decoupling BOTH GA AND FA from the assessment scale, since they are separate processes and not tied to the wikiprojects. This might actually make sense; so that we could essentially read it like, "This article is a 'A-class Featured Article', or a 'A-class Good Article', or a 'B-class Good article'. It might even strengthen the system a bit by creating somewhat of an unwritten rule of requiring wikiproject A-class prior to FA nomination, or wikiproject B-class prior to GA nomination. Although I don't think this should be a written rule, since an article could certainly be reassessed on the fly during FA or GA reviewing. Although theoretically, all FAs should be A-class upon initial listing at FA, separating FA as well as GA out of the assessment scheme would give the wikiprojects another avenue to review older FAs themselves prior to listing at WP:FAR, which would make it possible to have a 'B-class Featured Article'. It might also help to alert Raul & Sandy (or whoever runs WP:FAR) about the quality of the older FAs, if a wikiproject reassessed the article at a lower class for some reason; they could then take appropriate action if necessary.

However, I still think that decoupling GA alone without taking action regarding FA is the wrong approach, and will continue to oppose that avenue. Dr. Cash (talk) 18:03, 22 May 2008 (UTC)

Proposed addition of option G: Combine A and GA into GA/A class, process delegated by commmunity in cases of mature WikiProjects

I think that this can address the problem of variability of quality and maturity among WikiProjects. The community-wide GA process would have the role of giving "A"s to articles, except in selected topic areas where a mature WikiProject supersedes this role.

So, we have a vote, say, that establishes that WikiProject Military history is mature enough, and has a thorough enough process, to handle "A"-granting in its military history topic area. The designation of "A-granting WikiProject" can also be removed by community decision, if standards fall in the future. This would be like educational accreditation.

And maybe an immature WikiProject could mark articles "B+" (which would have the same rank as "B") in preparation for future "A" accreditation.

So, the default would be the community-wide GA process, with the participation of maybe three or four "A-granting WikiProjects" in selected topic areas to start.--Pharos (talk) 17:09, 19 May 2008 (UTC)

Question: What if an article is part of more than one WikiProject? More interestingly, what if there are fundamental philosophical differences between the WikiProjects? And finally, can article editors "forum shop" by changing the WikiProject templates on an article's talk page? 69.140.152.55 (talk) 08:28, 23 May 2008 (UTC)

Again, this seems to be a tacit recognition that (whatever the original intention) there is no longer any meaningful distinction between GA- and A-class articles. I'm not entirely sure this is actually the case though - certainly for MilHist, who are blessed with knowledgable and prolific editors, expert content reviews are no problem (and beyond the current remit of GA). Are you suggesting that GA take on content reviews in addition to the current criteria for articles that belong to less-'professional' projects (or no project at all)? EyeSerene^talk 19:15, 19 May 2008 (UTC)

There are quite significant distinctions at present. For the great majority of WikiProjects, A-class has no meaningful review process, and no real enforceable standards at all (and so quite below GA). For WikiProjects like MilHist there is a great review process, and A-class may be considered equivalent or even better than GA. I am essentially proposing that we take away "A-granting powers" from immature WikiProjects, so that "A" is more meaningful. Possibly we should still distinguish between "Community-granted A" and "WikiProject-granted A", but these should still be relatively similar standards (with "WikiProject-granted A" perhaps considered a higher standard, but not explicitly).--Pharos (talk) 19:26, 19 May 2008 (UTC)

Yes, I think that unless GA criteria are expanded and the GA WikiProject can find subject experts able to take on content reviews (and so assess to the same standard as the MilHist A-class, which I believe is generally regarded as above GA) there should still be distinctions between community and project A-class. This would be a significant quality-control wake-up call for many WikiProjects though - not necessarily a bad thing IMO ;) EyeSerene^talk 19:53, 19 May 2008 (UTC)

Indeed, not only would this provide a wake-up call to the WikiProjects, it would also set a clear boundary between "Community-granted A" as a catchall for uncovered subjects, and "WikiProject-granted A" for specific covered subject areas. There would be no more conflicting jurisdiction, and as more WikiProjects adopt MilHist-like practices, new subject areas would one by one come under this more rigorous scrutiny.--Pharos (talk) 23:09, 19 May 2008 (UTC)

I would name it "A" class, though, essentially droping "GA-class", thus simplifying:

Stub - Start - C - B - A - FA

Kevin Baas^talk 20:29, 19 May 2008 (UTC)

Yes, the idea would be to name it "A" class, but we would still distinguish between "Community-granted A" and "WikiProject-granted A" on the talk page box.--Pharos (talk) 22:17, 19 May 2008 (UTC)

• Strong Oppose Again, too similar to be anything but confusing and detrimental to the good work that GA does. VanTucky ^{Vote in my weird poll!} 02:56, 21 May 2008 (UTC)
  • Perhaps my wording has not been clear. This proposal would leave the GA process to the exact same "good work that [it] does" now, except that for certain subject areas where highly mature WikiProjects (like MilHist) have developed alternate processes.--Pharos (talk) 04:37, 21 May 2008 (UTC)

I hate to point this out, but I think wikiproject military history is in the serious minority here -- most wikiprojects don't have very clear and developed processes for reviews. Plus, specific projects aside, the criteria for 'A-class' varies quite widely among all wikiprojects. Dr. Cash (talk) 13:42, 21 May 2008 (UTC)

That's the whole point of this proposal. MilHist is in the minority now, but it shouldn't be. The proposal would strongly encourage MilHist-like review departments at other WikiProjects, and only those with clear and developed processes would get "A-granting authority" at all. In the beginning, there might only be three or four WikiProjects that fit these criteria, but I think eventually most WikiProjects that have a decent-sized membership will develop in this direction.--Pharos (talk) 17:04, 21 May 2008 (UTC)

I'd actually be in favor of removing A-class altogether, just have GA and FA. The gap really isn't that large between them. Wizardman 00:00, 8 June 2008 (UTC)

Proposed addition of option H: Separate DA/GA/FA (Wikipedia Quality Rating) from A/B/C/D (WikiProject Assessment Rating)

I think that these two systems need to be separated from one another entirely. WikiProject Assessment should be carried out within the projects which treat a particular article, granting a 'grade' A/B/C/D. An article should reach A-Grade according to that project's assessment standard prior to becoming eligible for a Wikipedia Quality Rating. If it belongs to more than one WikiProject, it needs to satisfy the A-Grade requirements of all those projects. (This should be intuitive: an article that brings together knowledge from two or more fields needs to be A-Grade in respect to both of those fields before it can be considered 'Good' from Wikipedia's 'global' perspective.) Wikipedia Quality Rating, on the other hand, looks at the article from the perspective of the encyclopaedia as a whole, and therefore cannot possibly use the same kinds of standards as a WikiProject Assessment. An article that is still undergoing development (i.e. D/C/B) should receive a Wikipedia Quality Rating of DA for 'Developing Article'. A WikiProject is responsible for ensuring that an article is detailed and accurate regarding its area of expertise. Its members, due to the nature of the work they do here, often lack the broad vision that is needed to rate articles for the encyclopaedia. Experienced reviewers, however, while they probably lack expertise in the particular field, are capable of viewing the project as a whole and rating the article accordingly. Thus:

(1) Separate the two systems entirely.

(2) Make it incumbent upon all WikiProjects to develop detailed Assessment guidelines for articles within their scope.

(3) Make A-Grade WikiProject Assessment required for nomination to Wikipedia Quality Rating.

—Aryaman (talk) 20:12, 19 May 2008 (UTC)

The problem with that is that it introduces a lot of unnecessary red tape as a requirement for GA and FA, which is not necessarily a good idea. Currently, there is no requirement for an article to be a GA before an FA; there's not even a requirement that the article be peer reviewed before it can be submitted at FAC. Additionally, many WikProjects are not capable of producing the assessment guidelines you propose, be it because they're nascent, or just not very active. Titoxd^{(?!? - cool stuff)} 22:50, 19 May 2008 (UTC)

Yes, this system has its benefits but it relies on active and interested WikiProjects. That is simply not the reality at the moment. Most projects simply haven't the interest in creating comprehensive assessment departments. Woody (talk) 22:54, 19 May 2008 (UTC)

I agree with (1) (at least in part), but not with the rest, for the reasons mentioned by Titoxd and Woody. Geometry guy 10:26, 20 May 2008 (UTC)

'Developing article' is a horrible idea, probably far worse than the "Jump to Conclusions" mat. Seems like it's adding more bureaucracy and process, and in the end it won't be much different than a 'start' class article. Dr. Cash (talk) 15:02, 20 May 2008 (UTC)

I read this proposal very differently. Sorry for the long comment, but I think this idea deserves more attention. Throughout these discussions, I see two separate recurring issues:

1. Defining a useful taxonomy for articles, incorporating the distinct perspectives of:

• Subject matter expertise, and
• Encyclopaedic integration.

2. Creating a useful family of processes for classifying articles within that taxonomy, processes that accommodate both:

• Mature WikiProjects, and
• Nascent projects that sometimes can produce FA articles, despite a lack of broad editorial resources and consensus.

I of course agree that a more-bureaucratic process is doomed to failure. However, I don't see that creating intuitively meaningful taxonomic pigeonholes must require red tape. Thus:

• I think the proposed DA/GA/FA and A/B/C/D schemes do a very good job at addressing the first issue, by recognizing that subject matter coverage and encyclopaedic quality are orthogonal issues (to an extent). The problem remains of whether such a system could be practical.
• Clearly, the requirement for a WikiProject A-Grade before a GA rating won't work in today's environment, unless we wanted to restrict GA/FA ratings to mature WikiProjects. But the obvious solution is to allow individual WikiProjects to establish their own GA threshold (A-grade, B-grade, or some magic internal classification – essentially an automatic "no consensus" for promoting certain articles). For nascent WikiProjects, such thresholds would not apply. Red tape would be an internal WikiProject decision.
• I don't find the idea of a DA-rating as a horrible idea per se. This rating would simply define an explicit default stratum for non-rated articles – classifying them by using a named placeholder, rather than by the absence of a GA/FA rating. Isn't "DA" more meaningful than "non-GA/non-FA"? This separate rating scale makes it clear that Stub>Start>C>... are not intrinsically part of the progression to GA/FA, but an independent quality axis.

Trying to bronze-plate the diverse rating schemata already in use guarantees that the system can never make sense. Trevor Hanson (talk) 02:17, 21 May 2008 (UTC)

• Oppose the requirement that "If it belongs to more than one WikiProject, it needs to satisfy the A-Grade requirements of all those projects." This would allow one WikiProject to impose its POV on an article before it can even make GA. For FA, maybe a good gauge of consensus, but you are driving the standard for GA too high. 69.140.152.55 (talk) 10:15, 4 June 2008 (UTC)

Option I - Proposal of C-class usage, with other special usages

Eh, never mind - since it seems that no one agrees - just ignore this.^Mitch32^contribs 23:06, 19 May 2008 (UTC) Since it seems this discussion is going nowhere I have a new proposal:

1. Add C-class - C-class would benefit those B-class articles not ready for GA. I know that's what B+ is for, but adding a + sign does not define much. I do feel C-class then B-class would do much better.
2. Use Developing article and redefine Stub - I feel this could be shoved between Start and Stub. A definition of a stub, at least in my eyes would be something like. A developing article to me would be like K-232 (Kansas highway) - which has prose and the right templates included. A stub would be something like K-9 (Kansas highway) which is not complete with the infobox and has minor prose. Start would be something like New York State Route 443, which has a certain amount of prose and is detailed, but not complete.
3. GAN becoming a little more strict - Per discussions on WT:GAN, I feel GAN should be a little stricter on what it accepts. However, I do not believe what was proposed about turning it into FAC-type is a good idea. Instead, I feel a full review of the review by an experinced editor, such as User:Derek.cashman, would help benefit what's added. If the article doesn't have a real full review, the article can go either to GAN, or be failed by the 2nd reviewer. Something like at Talk:New York State Route 9L would be in my eyes, a good review.
4. Merge A-class into GA under certain conditions - As I see this is a highly good idea, I feel there are some drawbacks and there are some advantages. One major drawback is that without A-class, FAC would be loaded with people saying they are ready when they are clearly not. If it were to go through, I'd say make peer review given more attention and make sure its ready for FAC, not just GAN passed straight to FAC. This'll cause stress on reviewers and the nominators as well.
5. Redefine the B-class - To see that an article is ready for GAN, I suggest that redifining the B-class would be in order. This is sort of like FAC and the A-class thing above, it would as well help the backlog at GAN with less articles that are really ready. Such example would be New York State Route 343, which I have in nomination. I will regret that I have 11 GANs at the moment, and is causing a backlog, but I can slow down and this may help. No B+ class in this case.

I am not saying that all will go through, but this is, in my opinion, a twice better option then some I've seen.^Mitch32^contribs 22:47, 19 May 2008 (UTC)

Comment

• Frankly, I oppose all of this. It seems to be orientating the assessment scale towards GA, something which I wholeheartedly disagree with. Woody (talk) 22:52, 19 May 2008 (UTC)
• I'm afraid I must agree. I think GAN is better than it ever was, and I believe it should be free to stay its own entity. For various reasons, I am also somewhat opposed to your other proposals, as well. While good ideas, I feel the risks outweigh the benefits. Juliancolton ^{Tropical Cyclone} 22:58, 19 May 2008 (UTC)