Wikipedia talk:Good article criteria

This page, a part of the Good article talk page collection, is archived by MiszaBot II. If your discussion was mistakenly archived feel free to go retrieve it. Current Archive location: Wikipedia talk:What is a good article?/Archive 3

Proposal to amend criteria

I think that point 3 of the GAR criteria defeats the object of encouraging the nomination of short articles. The criterion presently states:

Broad in its coverage:

(a) it addresses the main aspects of the topic;[3] and

(b) it stays focused on the topic without going into unnecessary detail (see summary style).

Although the footnote clarifies that short articles are welcome, it is a footnote and these are not always read. The main problem is the title "broad in its coverage". It is clear that some people who undertake reviews think this means the article must be lengthy. At the same time, broadness contradicts the stated aim of focus. If you are focused, you are not looking at the big picture.

I would replace "broad in its coverage" with something like "focused upon the subject" and bring the salient points in the footnote up into the criterion itself.

The criteria must make clear that GAR caters for short articles. At present, the process fails miserably. BlackJack | ^{talk page} 20:57, 18 October 2008 (UTC)

I think I disagree with your use of the word "scope". I have never seen a GA reviewer criticise an article because of its limited scope. But whatever an article's scope is, its major topics must be covered; focus has nothing whatsoever to do with it. Take a look at Space Science Fiction Magazine for instance, an article that has failed to reach FA, but would, in my opinion easily pass at GAN. Or, to put it another way, if an article like that one couldn't pass at GAN, then the GA vision has gone horribly wrong. --Malleus Fatuorum (talk) 22:29, 18 October 2008 (UTC)

I've removed my reference to scope as it is misleading. An article should cover all of its scope topics but, if the scope is limited by lack of surviving or available data, then there will be few topics and a reviewer must not expect to read a long article. The criterion must express the idea that the subject's major topics, as you say, have been covered in full. There is a big difference between sufficient coverage and broad coverage. BlackJack | ^{talk page} 05:18, 19 October 2008 (UTC)

There doesn't seem to be much interest in this although I could take the silence to mean that no one objects to the proposal. If there is no opposition by next weekend, I will make the proposed change. ---BlackJack | ^{talk page} 06:47, 4 November 2008 (UTC)

Note. The discussion to this point was on Wikipedia talk:WikiProject Good articles but I think it is more appropriate to this page so I have copied it here. I haven't yet edited criterion #3 as I would like to see more comments by GA reviewers. ---BlackJack | ^{talk page} 10:35, 6 December 2008 (UTC)

I agree with Malleus: the current broadness criterion is fine. It's long-standing and most experienced reviewers know what it means by now. If there are editors who misunderstand it, they should be politely corrected. Geometry guy 11:55, 6 December 2008 (UTC)

This contradicts the site's disambiguation policy. We should not be reactively correcting editors who misunderstand something, we should be proactively seeking to disambiguate so that the risk of misunderstanding is reduced. And just because something is "long-standing" does not mean it is right. ---BlackJack | ^{talk page} 07:55, 7 December 2008 (UTC)

No, disambiguation policy refers to the main article space, not project space, where consensus is the modus vivendi. Geometry guy 00:11, 5 January 2009 (UTC)

The ambiguity in the criterion is sometimes troublesome, but (1) I don't think "Focused upon the subject" is any "better" than the current verbiage in addressing issues of article length, and (2) the problem exists because some reviewers pass judgment by only reading the criterion title ("Broad in its coverage") without reading what that means (i.e., "(1) it addresses the main aspects of the topic, and (2) it stays focused on the topic without going into unnecessary detail, and (footnote) This requirement is significantly weaker than the "comprehensiveness" required by WP:FAC; it allows shorter articles, articles that do not necessarily outline every part of the topic, and broad overviews of large topics." Best regards --Eustress (talk) 14:32, 6 December 2008 (UTC) (note: copied from WT:GA).

Okay, I'll put it to the test but I'm not convinced. I've re-nominated an article that was failed previously simply because it was deemed too short. I don't mind if it fails because it needs more citations or, indeed, if it does not cover something that clearly is in scope; but the GAR process is supposed to welcome articles that are short by necessity because of unavailable data. I still think that some reviewers will be misled by a phrase like "(a good article is) broad in its coverage" to expect a lengthy article. I accept that the wording of points (a) and (b) within the criterion are adequate but I believe the title of the criterion needs to be amended. I agree with Malleus' statement that "its major topics must be covered" and I think we should reword the title of the criterion along those lines: e.g., "(a good article is) sufficient in its coverage". ---BlackJack | ^{talk page} 14:20, 6 December 2008 (UTC)

I have read this topic with interest as I have just been reading the criteria page and it occurred to me then that the word "broad" is somewhat dubious. I would change it to "sufficient" in its sense of "ample" or "adequate". An article that is narrow in coverage can have ample content and meet the requirements of addressing all main topics and remaining focused. I should point out that I'm a new member and so not up to speed yet, but it does seem sensible to me to make this change. --KenKt (talk) 14:38, 7 December 2008 (UTC)

Ample content is what I was trying to get across. We need to ensure that the article covers its scope, which may be very narrow, without bringing in unnecessary padding. Hence a short article will have as much chance at GA review as a long one. The other point I made was that the bit in the footnote ought to be part of the criterion, perhaps as a sub-section (c). Any thoughts on that? ---BlackJack | ^{talk page} 20:21, 7 December 2008 (UTC)

I think the footnote does make clear that length does not disqualify any article. The problem is the word "broad", in my opinion. By the way, which is the article that you have re-nominated? Although I am a new member I am not without review experience and I will look at it if you will permit me. But I am reviewing another article and that must take priority. --KenKt (talk) 21:50, 7 December 2008 (UTC)

Fair enough. I'll have to nominate that article again so I'll let you know about it, but thanks for your offer. ---BlackJack | ^{talk page} 07:50, 8 December 2008 (UTC)

I understand the difficulties with the term "broad" - though "ample" holds pretty much the same meaning. Perhaps "appropriate" is a more helpful term? I tend to feel that the more narrow the intersection of the topic, the greater the specifics of the details should be. When contestants for Mastermind (TV series) select broad topics (World War 1), they get broad questions ("What date did the USA declare war on Germany?"), but if they select a narrow topic (The 369th Infantry Regiment in World War 1) they get more specific questions (what colour were the regiments belts?).

Narrow topics demand greater focus on small details. It is a different type of coverage, and both the editors and the reviewers should be paying attention to that.

I am not convinced that selecting a narrow intersection for which there is not much available material, and then producing a very short article and saying "There is no more information to be found so this is broad enough," is a genuine rationale for a Good Article. If there is not enough material to create a detailed article that satisfies the interest of the general reader, then perhaps the article will never become a Good Article - and if the article is very short, it might be more appropriate for it to be taken back into the most appropriate parent article. I'm not against short articles, but I am against flimsy articles that don't answer the basic questions of What and Why and When and How and Where and Who. The example given above of Space Science Fiction Magazine is interesting as it is certainly quite borderline in that it barely manages to answer all these questions, and fails in some respects. I would expect from such a narrow topic to be at least told in which of the two issues Clarke's "Critical Mass" was published. If you image somebody doing some research on Clarke and wishing to find out the first American publication of the story, they might expect to discover that fact in a Wikipedia article on the magazine in which the story appeared (regardless of the article being GA status or not). I don't think it is clear that Space Science Fiction Magazine is a GA, and is perhaps not a good example to use of a short article fulfilling the requirements of "broad coverage" (or "appropriate coverage"). SilkTork *^YES! 13:07, 4 January 2009 (UTC)

What and Why and When and How and Where and Who is appropriate for a newpaper article, but not necessarily for an encyclopedia article, at least in my opinion. It's very easy to imagine unanswered questions after reading any article, but it's not the purpose of an encyclopedia to answer every conceivable question. I still think that "broad" is clear enough, but interestingly the FA criteria were recently changed slightly to include "... and places the subject in context", which I think is maybe at the heart of what you're saying Silktork. That might make a useful addition to the GA criteria as well. --Malleus Fatuorum 13:41, 4 January 2009 (UTC)

I would agree that an encyclopedia is a general reference work and does not seek to be a textbook on a subject. However, definitions of encyclopedia generally use the word "comprehensive". It would normally be considered that an encyclopedia article on a topic would answer more questions than a newspaper article, and would aim to give a rounded knowledge. The actual depth of detail would depend on the exact focus of the article. History of the world would have a broader less detailed focus than S. A. Andrée's Arctic balloon expedition of 1897, which is a very specific topic, and the article goes into admirable detail and answers all the questions in What and Why and When and How and Where and Who. I wouldn't expect a GA article to have that level of coverage, but I would expect a GA article to deliver a bit.

I like "places the subject in context", but that is perhaps a demand too far for a GA article. That requires a higher level of writing, research and organisation. I wouldn't wish to add more demands on GA. I paused to consider that perhaps "shows an awareness of context" might be less demanding, and would add value, but I think it is already demanding enough that we are asking people to address the main aspects of the topic. And I don't think that people are understanding the demands of that requirement. Some people are interpreting "main aspects of the topic" as "what can be found during a Google search on a wet Wednesday afternoon" - such as a search for information on Matchbooks for a new article History of match books, and being pleased that so little can be found, as that means that all the main aspects of the topic are covered! A lack of information is simply a lack of information, it's not a free punt into the end zone. SilkTork *^YES! 23:51, 4 January 2009 (UTC)

"Comprehensive" for encyclopedias refers to the coverage of every branch of knowledge, not the comprehensive nature of each article. However, I agree with you that "addresses the main aspects of the topic" is a demanding requirement. However, the main requirement that GA reviewers miss is that the material in the article can be attributed to a reliable source. That isn't a matter of looking for uncited sentences, but checking (oh my god) that the sources actually say the same things as the article. That is really hard work. I'd welcome any proposal to improve the criteria to encourage it. Geometry guy 00:11, 5 January 2009 (UTC)

As the one who "reframed" another editor's suggestions for modification of the FA criteria into "places the subject in context", my meaning was to present the major aspects of a topic and not spend major article space on "offshoots". I actually got the idea for the FA modification from the GA criteria which I thought were clear in a way that the FA criteria were not. How it is interpreted by FA evaluators is another issue. I personally thing the GA criteria are fine when followed. —Mattisse (Talk) 00:21, 7 July 2009 (UTC)

Proposal: criticism and controversies sections

Let me just express my personal belief that the criticism and controversies sections that pervade Wikipedia (especially biographies) are unencyclopedic. Try reading Britannica, Encarta, or World Book sometime. They never have criticism and controversies sections. Only Wikipedia does. Why? I believe they exist in Wikipedia because they attract editors who either (a) love scandal, or (b) have an agenda (often political) they want to push in an encyclopedia. Quite often, criticism and controversies sections take up a disproportionate portion of an article. I propose minimal or non-existent criticism and controversies sections as a criteria for good article status. By having minimal or non-existent criticism and controversies sections as a criteria, it would allow such sections to exist in those rare articles where they are truly warranted, but would give editors a strong incentive to remove frivolous criticism. Note that this proposal goes beyond just having WP:RS and WP:NPOV as criteria. Frivolous criticism and controversies are often backed up by reliable sources. That's because frivolity and reliability are entirely different concepts. NPOV, meanwhile, is often used as an excuse to add a criticism and controversy section. The existence of the section then entices editors to expand it, in order to fill in any missing criticism and controversies. Discouraging, but not banning, these sections are a great way to trim the fat. --JHP (talk) 06:47, 26 June 2009 (UTC)

Using an old book as a reference

We're working on wikipedia-spotlight on Marco Polo but we don't have many reliable source on the web. We have however access to the 1947 edition of Venetian adventurer By Henry Hersch Hart which describes Marco's life in a very scientific and scholarly level. It has also received a good review and is cited by many other books. The problem is that the book was written in 1947 and I was wondering if that would pose a problem when nominating the article for GA or FA, thank you.--Diaa abdelmoneim (talk) 22:48, 6 July 2009 (UTC)

For a subject like that, I wouldn't think so. The question is essentially how much scholarship on a given subject has advanced in the intervening years. If you were using a 1947 book to write an article on, say, homosexuality, you'd probably be in some trouble. But without knowing much about Marco Polo, my initial reaction is that it shouldn't be a problem there. Steve Smith (talk) (formerly Sarcasticidealist) 22:50, 6 July 2009 (UTC)

Alt tags

FA and FL have gone to requiring alt tag use for images. GA may as well too. — Rlevse • Talk • 23:44, 15 July 2009 (UTC)

Why? As opposed to encouraging them? --Malleus Fatuorum 23:46, 15 July 2009 (UTC)

The alt tags show up as a jumble of too much information on my browser (Firefox 3). I am hoping that it will be considered unnecessary to "require" them for GAs. For one thing, I will be unable to review that aspect when I review articles. —Mattisse (Talk) 23:56, 15 July 2009 (UTC)

I'm using Firefox as well. You don't see the alt text unless you go to your preferences and switch off Load images automatically, which is what I do if I want to check it. I agree with you that I don't think it should be added as a requirement though. GA is not FA. --Malleus Fatuorum 00:01, 16 July 2009 (UTC)

Switch of "Load images"? Ok. Didn't know about that. Thanks. (Where is "Load images"?) —Mattisse (Talk) 00:33, 16 July 2009 (UTC)

On my version, 3.0.11, it's under Edit/Preferences. --Malleus Fatuorum 00:57, 16 July 2009 (UTC)

One easy way to review alt text, which should work with any graphical browser, is Altviewer. This is a Wikimedia Toolserver application that displays all the images in a Wikipedia page, next to their alt text. I've found it much better than IE or Firefox to review alt text. Eubulides (talk) 06:33, 21 July 2009 (UTC)

I don't think alt text should be a criterion for GA; GAs aren't held to the same standards as FAs. Though it should be encouraged in as much much as all articles are encouraged to be better and to aspire to FA standards. Perhaps it would be helpful to add a note to reviews about alt text to raise awareness about it. It is after all intended to help the readers. Nev1 (talk) 01:45, 16 July 2009 (UTC)

Precisely, if we can make them better, why not? — Rlevse • Talk • 19:05, 18 July 2009 (UTC)

That GAs can always better is self-evident, as they are not FAs, and are not held to the same standard. Adding a note during the review is something that is often done anyway. --Malleus Fatuorum 23:38, 18 July 2009 (UTC)

We just started checking alt text for FAs and FLs within the past few weeks. It would be helpful, I think, to debug the process there a while more, before inflicting it on GAs. Often we still run into templates that aren't alt text aware, for example. But once things settle down I don't think it'd be onerous to require alt text for GAs, any more than it's onerous to require captions. Alt text isn't that hard; it's easier than captions, in my experience. Eubulides (talk) 06:33, 21 July 2009 (UTC)

Since it goes to accessibility I think it should probably be encouraged as good article-writing practice. However, I'd be wary of introducing it as a requirement per Mattisse, Malleus and Nev1. EyeSerene^talk 16:03, 21 July 2009 (UTC)

I agree with EyeSerene, Mattisse, Malleus and Nev1. Even if Alt text isn't that hard, there are hundreds of similar fine points that would improve articles, and the GA process doesn't cover them: it does not need to duplicate the role of FA and peer review; those processes do a fine job. Instead it aims to focus editors' attentions on the most important issues of verifiability, neutrality and coverage. With perhaps over 2 million articles not meeting basic policy, we need at least one review process that concentrates on this major issue. Geometry guy 21:13, 21 July 2009 (UTC)

Proposed changes to 2b

GAC2b currently says:

it provides in-line citations from reliable sources for direct quotations, statistics, published opinion, counter-intuitive or controversial statements that are challenged or likely to be challenged, and contentious material relating to living persons

I think that this would be better expressed as a list, and with slightly different wording:

it provides in-line citations from reliable sources for:

• direct quotations,
• statistics,
• published opinion,
• counter-intuitive or controversial statements whose factual accuracy is likely to be challenged, and
• contentious material relating to living persons

The point behind making this a list is to emphasize that WP:V -- and therefore WP:GA -- does not actually require an "ideal" level of sourcing, and certainly not a one-to-one ratio between inline sources and sentences. We have five specific kinds of statements that require citations, and everything else is technically optional.

The point behind the change in the wording is to emphasize that {{fact}} is about factual accuracy, not "let's make editors find sources to support that the Earth is basically round, despite the fact that it's not required by any policy" WP:OVERTAGGING. Additionally, it seems kind of stupid to specifically name the need for sources to support a claim that the reviewer believes is false. WhatamIdoing (talk) 02:30, 24 July 2009 (UTC)

"... whose factual accuracy is likely to be challenged". Sloppy. What other kind of accuracy is there? Non-factual accuracy? --Malleus Fatuorum 02:45, 24 July 2009 (UTC)

Well, there's the "I'm challenging this unsourced statement because I'm being WP:POINTy or want to disrupt work on the article" kind of challenge. Although you should assume good faith initially, bad faith does happen, and GA shouldn't hinge on resolving challenges of agreed-to-be-accurate facts that were done merely for the sake of "forcing" someone to add a ref, or to push the mainstream POV out of an article. Accidental idiocy also happens: Last month, someone actually fact-tagged a statement that said that people that cannot pass a practical driver's license test cannot legally drive. WhatamIdoing (talk) 23:03, 4 August 2009 (UTC)

Just to compare and contrast this, here is the featured article criterion pertaining to citations: "(c) well-researched: it is a thorough and representative survey of the relevant literature on the topic. Claims are verifiable against high-quality reliable sources and are supported with citations; this requires a "References" section that lists these sources, complemented by inline citations where appropriate;"

It seems to me like the proposed list above for GA is meant to basically cover the critical stuff, like major issues in BLP articles, quotes, and other things. But it leaves it open that GAs don't necessarily have to have every little item cited. It might increase the number of math & science articles, which have a lot of good general information that is well accepted and pretty obvious that any possible vandalism on it would be caught rather quickly. Dr. Cash (talk) 20:08, 4 August 2009 (UTC)

More proposed changes

While the two-sentence GAR story is unfolding, let's fix that problem quickly by adding Wikipedia:SCG to 1(b). It only targets Mathematics, Physics, Molecular and cellular biology and Chemistry articles, which arguably requires a higher degree of precision yet without going over the edge on references. (P.S. For those who're lost, read Wikipedia talk:WikiProject Mathematics#Mathematics GA status) OhanaUnited^{Talk page} 20:21, 4 August 2009 (UTC)

It is not a good idea to modify long-standing guidelines in the heat of the moment. WP:SCG is already in footnote 2 to 2b, and that is where it is needed. It isn't primarily a MoS issue, but a verifiability issue. Geometry guy 20:35, 4 August 2009 (UTC)

My concern is often ensuring that there is no original research. However, editors can stage citations in such a way that the result is OR. Only by actually checking the sources can this be discovered. —Mattisse (Talk) 20:40, 4 August 2009 (UTC)

Part of the problem here is that some reviewers have not heard of WP:SCG, so presumably they have not noticed it in its footnote. I support giving it more visibility. Septentrionalis PMAnderson 22:18, 4 August 2009 (UTC)

While I think highlighting this further is a good idea, and will probably help in the long run, I am still concerned that this isn't really addressing the problem, and might be kind of like putting a band-aid on a gushing flesh wound. Part of the problem is that GA (and to some extent, FA) is not really accepted and/or used by wikiprojects in the more scientific articles -- e.g. Wikipedia:WikiProject Chemicals doesn't support GA, either. And if you look at the vast majority of where GA activity is, it's more in the pop culture and non-scientific articles. We really need to look deeper at ways to make GA work better with science articles in general, and listen to those involved in particular wikiprojects of how to make GA work better for their project. After all, in the end, the main goal of GA, FA, and all of the wikiprojects, is to produce the best article content possible. Right? Dr. Cash (talk) 01:34, 5 August 2009 (UTC)

Agree 100% with this statement. —Mattisse (Talk) 01:38, 5 August 2009 (UTC)

The only problem I see with "science articles in general", and this also applies to maths articles, is the citation requirement. It does appear to me that too many reviewers demand citations on a pound-for-pound basis, without looking at whether the material actually needs to be cited. My favourite example (can't find the link right now) was a request for citation that "most human beings have five fingers". --Malleus Fatuorum 01:45, 5 August 2009 (UTC)

You bring up that example every time this issue comes up. —Mattisse (Talk) 01:51, 5 August 2009 (UTC)

Does that invalidate my point in some way? Or does it just make me boring? To be clear, I have a science degree myself, and the issue I see here is the distinction between what is accepted as being almost axiomatic to those working in, or familiar with, the field, and mysteriously unexplained to those who aren't. --Malleus Fatuorum 01:58, 5 August 2009 (UTC)

Yes. If after all these years you can only come up with the same example over and over (and no link). —Mattisse (Talk) 02:06, 5 August 2009 (UTC)

Well, it is a particularly appealing example. On a different note, thanks very much for your feedback on the character GAN I'd posted about, Malleus. I did notice it, and genuinely appreciated your improvements and views. I'm still catching up with wikitasks, so hadn't quite replied yet. So, merci. –Whitehorse1 02:13, 5 August 2009 (UTC)

IMO in science articles we should not interpret "obvious" any differently from in other subjects - i.e. "obvious" should mean "... to non-specialist readers". I'd accept "most human beings have five fingers" without a citation, but e.g. I'd expect citations for most of the deductions in Exponential growth. Reducing the citation requirement for science topics assumes that the editor adding / changing the unsourced content has got it right, has correctly judged what is common knowledge in the field and is not POV-pushing. If we relax the citation requirements for science, we'll get an increase in errors (to be publicised in Nature), WP:FRINGE "science" and POV-pushing. I understand that finding citations for "basics" can be harder than finding them for advanced concepts, because the journals take basics for granted so Google is not so helpful in finding refs for basics. But there's a simple solution - get a couple of 101-level textbooks from a library. --Philcha (talk) 06:12, 10 August 2009 (UTC)

Proposed restructuring

I would like to propose a slight restructuring of criteria 6 and 1. At the moment criterion 6 combines two logically distinct notions: that an article should have interesting images, and that it should be free (e.g. of copyright violations). The first point is really a readability issue: it is often the case that without images, an article is opaque. We should encourage editors to create articles which convey their concept not only using prose, but also images. This is all part of criterion 1.

The second point is a serious problem. Many of our articles, including some GAs, contain copyright violations. The GA criteria are explicit about this in terms of images, but not in terms of prose. If the encouragement of images is adopted into criterion 1, then criterion 6 can be more clearly defined as the requirement that GAs do not contain copyright violations. Such a restructuring will be easier for new editors to understand, and for experienced editors to quote. Geometry guy 22:37, 9 August 2009 (UTC)

I think that #6 needs to stay focused on the specific issue of images. (IMO, the major point to be made here is that images are not required for GA.)

I also worry about dramatically changing the contents of a long-standing numbering scheme: Old comments like "GA6 does not require [the presence of images]" will make no sense if GA6 becomes "Copyright violations are evil" instead of "You may not fail a GA nomination simply because no images are present."

If we want to directly address the issue of WP:COPYVIO in text, then perhaps it should be part of "well-written" -- or perhaps we just need to have seven criteria instead of six. WhatamIdoing (talk) 05:52, 10 August 2009 (UTC)

Just change it to something like "The article should be free of any issues related to copyright violation. OhanaUnited^{Talk page} 06:10, 10 August 2009 (UTC)

WhatamIdoing, I share your desire to maintain stability in the criteria and their numbering, so I am reluctant to propose changes. However, in this case, I believe there is a good case, with minimal drawbacks. As you point out, criterion 6 is often quoted for what GA doesn't require, i.e., images. This is actually footnote 6, which could be attached both to criteria 1 and 6, and quotes of GA6 would still make sense in that context. Criterion 6 is rarely quoted for the use of captions. It is this aspect of the criterion which sits more naturally with criterion 1. Geometry guy 10:51, 10 August 2009 (UTC)

I have no objections to changing what is currently footnote #6 to anything that you like, as I doubt that anyone would expect those numbers to remain consistent indefinitely. WhatamIdoing (talk) 07:39, 12 August 2009 (UTC)

C'mon guys, real reviewers don't use the criteria numbers :-) --Philcha (talk) 12:11, 10 August 2009 (UTC)

They often use {{GAList}} and its relatives.

Let me be plain here. The concern which prompted this suggestion is that GA doesn't have a clear focus. Many view it as a stepping stone, a poor man's FA, or just an unreliable check by an independent editor. By and large the criteria are not to blame for this, but if you look through them, you will find that the obvious wet blanket is the part of criterion 6 which says "images are nice, but not required". My view on this is similar to my view on the idea that "MoS compliance is recommended but required for GA": it is wishy-washy. Each criterion should have a clear mission to improve the encyclopedia by eradicating a problem, one article at a time. No time should be given to things which are recommended but not required: that destroys focus. Criterion 6 does have a mission (eliminate image copyright violations) embedded in it, but by confusing this with a separate half-hearted mission (make Wikipedia articles interesting, easy to read) it loses its focus. My proposal is that the mission to make Wikipedia articles interesting for readers should be part of criterion 1. Geometry guy 20:33, 10 August 2009 (UTC)

I like your concept. My reaction to the "recommended but not required" bits is that it's an effort to raise the bar against consensus. No experienced editor would (e.g.,) oppose compliance with MoS, so everyone more or less agrees that it's nice and "recommended", and then it turns into "Well, yes, the article technically meets the criteria, but I expect you to jump through these extra hoops, because I personally don't pass articles that don't comply with my pet sections of the MoS -- and it's 'recommended', after all..." WhatamIdoing (talk) 07:39, 12 August 2009 (UTC)

Thanks. I agree and am against raising the bar for GAs. In regard to the criteria, for me the key is to articulate the bare minimum requirements for GAs in a way that every editor and reviewer can easily relate to, so that substandard articles are not listed and acceptable articles are, with much more consistency (and fewer iterations) than at present. The criteria should be summarizable in single adjectives, along the lines of "Is it readable, verifiable, broad, neutral, stable, and free?" (these may not be the best adjectives, but you get the idea), so that new editors and reviewers pick up the idea quickly. Geometry guy 21:02, 12 August 2009 (UTC)

It turns out that the citation requirements for FAC are exactly the same as for GA. Just whatever it takes to satisfy WP:V. So that should be good enough for GA also. Plus, medical articles should follow Reliable sources (medicine-related articles) in addition to Scientific citation guidelines. —mattisse (Talk) 21:12, 12 August 2009 (UTC)

WP:V is open to interpretation, especially when it comes to inline citations (what does "challenged or likely to be challenged" mean in practice? Should a frivolous or ill-informed challenge count and what is a "likely" challenge?). GA2 attempts to capture the bare minimum acceptable (but still strict and non-negotiable) interpretation of verifiability. WP:MEDRS isn't explicitly part of the GA criteria, although it no doubt contains good advice. However, the message in criterion 2 has been strong and stable for nearly 2 years, and I'm happy with it. I'd like all the criteria to be similarly compelling. Geometry guy 22:47, 12 August 2009 (UTC)