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

Proposed retargeting of WP:GAC

FYI: It has been proposed that "WP:GAC" be retargeted to this page. Feel free to contribute to the discussion. Thanks! Swarm ^X 18:08, 21 April 2011 (UTC)

Scope and scale

I've started my first GA review and have been looking at the criteria for guidance. Unless I'm missing something, there doesn't seem to be anything about defining scope or measuring scale.

I would expect an article to have a clearly defined scope and remain within it whilst providing a full scope coverage. One of the criteria is "broad in coverage" but that could imply that it is in order to discuss associated topics which are not necessarily within the scope of the article. Should I therefore assume that scope includes anything that is either directly or indirectly relevant; or should I take a value judgment and in effect define the scope myself?

"Broad in coverage" also implicates the scale of the article which might be too long (as some articles unquestionably are) or simply not long enough. Obviously, a stub with just a couple of paragraphs can never be a good article but what about an article that is of necessity short, relatively speaking, perhaps because of limited scope? Equally, at what point should I consider an article too long (e.g., 100kb)? I realise it would be impossible to define minimum and maximum limits in terms of kb but I think some broad guideline is needed to make clear that articles which are very short or too long cannot qualify. Again, is the question of scale one for the reviewer's own judgment?

Any useful advice or pointers about other aspects of GA reviewing would be appreciated. I do have review experience in other spheres but this site has its idiosyncracies and I will need to be aware of any special considerations. --Mykleavens (talk) 07:08, 8 May 2011 (UTC)

• The requirements are given in WP:WIAGA and that is what you should be reviewing them against, not your own individually-imposed idiosyncrasies. Looking at your talkpage, you seem to have been editing of about one month on wikipedia and don't appear to have have personal experience of the GA review process, i.e. you do not seem to have submitted an article at WP:GAN (nor appear to have contributed to "fixing problems" of an article under view). As an editor with a username you are entitled to do reviews; but is it wise/fair be doing them if you are unsure of what the requirements are: scope, measuring scale, and length are not requirements of a GA? Pyrotec (talk) 10:40, 8 May 2011 (UTC)

I understand you are short of reviewers with a resultant backlog. I review articles and specifications on a weekly if not daily basis in my profession and so I wouldn't have thought reviewing articles on a leisure site like this would be any big deal. I was merely asking a question about aspects that are not clear in WIAGA which, for your information, I had already studied. In professional reviews, it is very important indeed to understand the scope (size is not very important) but you do not appear to share that view and I wonder how you decide what is relevant to an article and what is not? Fortunately I am not easily discouraged. --Mykleavens (talk) 19:40, 8 May 2011 (UTC)

• OK, lets start again. Hi Mykleavens, We are short of reviewers, we also suffer occasionally from poor reviewers. I can only go on the information that you provide. There is little mention until now that you review on "a weekly if not daily basis in my profession" and perhaps your choice of the statement "I do have review experience in other spheres but this site has its idiosyncracies" was open to misinterpretation - and I fell into that trap. I have no wish to discourage "good reviewers", but I hope that your use of the term "leisure site" was not intended to be provocative. No, if you wish to do "leisure reviews", hopefully that is reviewing nominations of leisure articles, that is fine. As I said before we do get poor reviewers, articles that would be expected to fail do get passed and articles that should perhaps not pass do sometimes get passed. A "bad" review can be reversed by means of another review, so it is no big deal; and there are various degrees of badness. Idiosyncratic views being imposed by reviewers (sorry, these are your words) is one. There is no upper or lower limit on length. Pyrotec (talk) 20:33, 8 May 2011 (UTC)

• I think the words or concepts of "scope" and "scale" is causing difficulties. In a wikipedia article, the first part of the article before the contents, a few paragraphs, is known as the WP:Lead. The Lead has two functions, it introduces the article and summarises the main points. So, what I think you are calling "scope" is defined by the Lead (and the title of the article): if a summarised subtopic is in the lead then the subtopic should be in the article, the lead should not "tease" by including info/facts that are not in the article. So I tend to review an candidate WP:GAN section by section, but do the Lead last. I probably understand scope creep: possibly "your" approach would be: (1) is the scope of the Lead adequate, (2) has scope creep occurred between the Lead and the article. The criteria, however, has to be WP:WIAGA. Are we talking the same tale (I won't say the same language)? Pyrotec (talk) 20:33, 8 May 2011 (UTC)

Hi Pyrotec. I think you've answered it very well as the lead should be the place where the scope is outlined, although not in a direct way. I presume the reviewer can decide if the article is too short or too long so I'm fine with WIAGA. Thanks very much. --Mykleavens (talk) 21:53, 9 May 2011 (UTC)

• Hi, Mykleavens.WP:WIAGA says "Broad in its coverage: (a) it addresses the main aspects of the topic; it stays focused on the topic without going into unnecessary detail (see summary style)." Part (b) is relatively easy: if an article gets too long or covers too many topics, put the excess in one or more smaller articles, give the smaller ones give links to the main one with {{main}} and make the main articles link to the smaller article(s) with wiki-links in the text and/or {{see}}. The "coverage" criterion is the most difficult to interpret, we've had discussions about it with no conclusion. My own unofficial criterion of "Broad in its coverage" is "are there any obvious gaps on a quick read-through?" I hope that helps. ----Philcha (talk) 13:26, 8 May 2011 (UTC)
• And yes, you should get about 4 your" articles of through GA reviews. After that, we always need more reviewers, welcome onboard on the good ship WP:GAN. --Philcha (talk) 13:26, 8 May 2011 (UTC)

Hi Philcha. Thanks for giving me a positive response and I agree with you that looking for gaps is a good technique but it is equally important, in my view, to guard against the dreaded "scope creep". I'm afraid you've lost me a bit with the first sentence of your second post. Do you mean I should be aiming to promote four articles I personally have developed? If so, I am some way off because I am only working on one article at present although it has led me to make a few minor fixes in related articles. --Mykleavens (talk) 19:40, 8 May 2011 (UTC)

Yes, I believe that Philcha's suggestion involved the concept of "walking before you run", e.g. by submitting some articles to WP:GAN to gain experience before GA reviewing is undertaken. On a positive note (I don't wish to be negative) I did not; if you wish to constructively critice my reviews that is OK by me. I suspect that you will find it easier to review first and submit later. Pyrotec (talk) 20:43, 8 May 2011 (UTC)

On balance, I think reviewing will help me to write articles as it will show me what to look for and what to do (and not to do), especially around site compliance. Thanks again. --Mykleavens (talk) 21:53, 9 May 2011 (UTC)

Have you read WP:Reviewing Good articles and What the Good article criteria are not or any of the other advice pages?

For some of your questions (e.g., the scope/subject of the article), the correct solution is to ask the editors at the article. For others, there are some guidelines elsewhere (e.g., WP:SIZE), but please remember that if it's not actually in the criteria, then compliance isn't actually mandatory.

Overall, reviewing an article isn't especially difficult. I don't think that it's necessary to have had articles promoted to GA to be able to do it. In fact, I wouldn't claim any GA as my own work, and yet I doubt that anyone here would tell me that I'm an incompetent reviewer. WhatamIdoing (talk) 05:07, 12 May 2011 (UTC)

I believe that scope is probably the most subjective criteria. It depends on three factors - the article title, article length (see WP:SIZE), and the scope of related articles. If some aspect of the topic that comes to mind when you think of the article title and you find it missing from the nominated article, you should raise that problem in the review. However, if the nominated article is a survey article with many daughter articles, then the omission of a topic would be more understandable. As a practical matter, the scope becomes a matter of negotiation between the nominator(s) and the reviewer. There is no single "right" answer, and it involves a great deal of judgment. Racepacket (talk) 20:13, 16 June 2011 (UTC)

The nominator posted a request for further info on the 15 June and I posted a request for further info on the 19 June and both have not yet received a reply Talk:Klemens_von_Metternich/GA1. I have asked directly now, on your talk page Mykleavens, as there are some major concerns that the article was not really properly assessed. I understand that it was your first review but putting a fail for stability, when there were no edit wars or content dispute, was the final straw which persuaded me that perhaps the review was less than satisfactory.

If I have not heard anything in another few days I will put it back up for community review. Chaosdruid (talk) 13:29, 28 June 2011 (UTC)

Clarity of Note 2

Criteria 1(b) is "it complies with the manual of style guidelines for lead sections, layout, words to watch, fiction, and list incorporation.[2]" Note 2 reads: "Compliance with other aspects of the Manual of Style, or the Manual of Style mainpage, is not required for good articles."

What is not clear is if the expanded details on a sub-page are included. For example - WP:Layout has a section on linking - Wikipedia:Layout#Links - which directs people to Wikipedia:Linking for further information, and that page includes advice on overlinking. I have regarded overlinking as outside the GA criteria, though I have noticed that it does get mentioned in GA reviews.

Some clarity regarding the subpages would be useful. Either:

1) "Compliance with other aspects of the Manual of Style, including the Manual of Style mainpage or subpages of the guides listed, is not required for good articles."

or

2) "Compliance with other aspects of the Manual of Style, or the Manual of Style mainpage, is not required for good articles. Though subpages of the guides listed, such as WP:Linking, a subpage of WP:Layout, are included."

Thoughts? SilkTork *^{Tea time} 09:08, 2 June 2011 (UTC)

• I propose adding "or subpages of the guides listed" as I feel the spirit of the criteria is to limit the focus of attention on the listed guidelines. SilkTork *^{Tea time} 16:46, 11 June 2011 (UTC)

GA and bulleted lists

I am writing a historical article for which dates of certain events are not really known. There exists however an old historical chronicle which assigns dates to events. Since those dates are possibly not always correct, I would like to leave them outside of the main text, but put them into a kind of timeline section linked as footnotes. The timeline section will have an introduction discussing the correctness of dates. Is this a good idea, i.e. does it satisfy good article criteria, MOS, etc; or is there a better way to provide the same information? bamse (talk) 09:51, 15 July 2011 (UTC)

Yes, this is a tricky area. It comes under Wikipedia:No original research which is a GA criteria. It says: "All interpretive claims, analyses, or synthetic claims about primary sources must be referenced to a secondary source, rather than original analysis of the primary-source material by Wikipedia editors." Your approach is reasonable. Have a section on the dates noting what Shoku Nihongi says, but without giving your opinion or views or interpretation - so "Shoku Nihongi gives the date as xxxx" but not "Shoku Nihongi gives the date as xxxx, but this may not be correct" and not "The date is xxxx". Questionable dates should not be presented as facts, but given as "source says". Ideally having a secondary source commenting on the dates would be the way to go. SilkTork ^{✔Tea time} 10:37, 15 July 2011 (UTC)

There seem to be a number of sources which are indicating dates. Is it the precise months you are after? If the bulk of the sources are satisfied with using just the year, and are not getting into the detail of the months, and only one source, which you indicate is dubious, is giving the months, then perhaps it might be better to mention the Shoku Nihongi source with its breakdown in months entirely in a footnote, and if you have another source which questions the reliability of the dates, that could be used to cite a statement that the dates are unreliable. So: Footnote - "Shoku Nihongi gives monthly dates for these events, however Brian Foo and other scholars have indicated concern about the reliability of these dates. [source cite]." SilkTork ^{✔Tea time} 10:50, 15 July 2011 (UTC)

Thanks for replies. I don't think there is an issue with "No original research". All I wanted to do is to present the dates (including months and day) from the Shoku Nihongi in a timeline section (source=Shoku Nihongi) and write a short paragraph at the start of that timeline section discussing what historians think about those dates (source=some history books). Since I remember from previous GA nominations that bulleted lists are generally discouraged in GA, I was wondering whether such a timeline/footnote section would be admitted, since it is kind of a bulleted list as well. bamse (talk) 10:58, 15 July 2011 (UTC)

Well, it sounds as though you are dealing with the OR issue correctly. Are you wondering if the presentation of the material will contradict any of the appropriate uses given in Wikipedia:Manual of Style (embedded lists)? SilkTork ^{✔Tea time} 20:14, 15 July 2011 (UTC)

Timelines and chronologies weren't explicitly covered in WP:EMBED. I've now added some comments and links. They are in effect serving the same purpose as a list of works, in that they are graphically summing up and giving a quick overview of some essential facts, and that the data will be supported by prose analysis of the main points - either within the timeline itself or elsewhere in the article. From what you have described, that is what will be happening in the Fujiwara no Hirotsugu Rebellion article: the dates will be discussed as part of the main text, and the timeline will collect the dates together in a graphic representation. Does that help? SilkTork ^{✔Tea time} 23:26, 17 July 2011 (UTC)

Yes it does. Thank you. bamse (talk) 06:58, 18 July 2011 (UTC)

WP:COPYVIO not mentioned in WIAGA

• I've been here just days short of five years, and never noticed that WP:COPYVIO, a Wikipedia policy with legal considerations, is not explicitly mentioned anywhere in WIAGA. It is mentioned in Wikipedia:What the Good article criteria are not, but buried at the bottom, and as a minuscule exception in the "Beyond the Scope" section.
• Read that last bit again, please, mentally underlining the words "a Wikipedia policy with legal considerations". I urge the addition of 2(d): It contains no copyright violations.  – Ling.Nut (talk) 12:46, 29 July 2011 (UTC)

• It's never going to possible to guarantee that an article has no copyright violations, so I'm not sure I see the value of such an addition. The best one can ever say is that no copyright violations have been found. Malleus Fatuorum 15:22, 29 July 2011 (UTC)

• Hmm, quite. Maybe some woolly wording like "no readily apparent copyright violations". --Demiurge1000 (talk) 15:27, 29 July 2011 (UTC)

• If no wording is included, no one will even bother to check. The point of WIAGA is not to say that the resulting GA is free of copyvio; it is to say that the reviewer has looked and did not find any. I think that is understood. I don't even think any "no readily apparent" hedging is necessary.  – Ling.Nut (talk) 15:57, 29 July 2011 (UTC)

• But your wording is quite explicit: "It [the article] contains no copyright violations", a guarantee that's impossible to give. The GA reviewing guidelines already say that "The article should not copy text from sources without quotation or in text attribution"; that not all reviewers always at least do a spot-check, examine the sources for suspicious phrasings, or have even read the guidelines at all, is not likely to be addressed by adding anything to the GA citeria. Malleus Fatuorum 16:28, 29 July 2011 (UTC)Malleus Fatuorum 16:21, 29 July 2011 (UTC)
• (edit conflict) Agree with Malleus here. re: If no wording is included, no one will even bother to check.; while I understand the great desire here to always ensure that we produce the very best article(s) we can, I can't agree with that. MANY of those at GA and FA do extensive reviewing before making a determination. Naturally there's going to be a higher level of "eyes on" in FA than GA (I won't even get into DYK), but that doesn't mean that GA just gives a "free pass" on COPYVIO or PLAG. — Ched :  ?  16:33, 29 July 2011 (UTC)

• The "except copyvios" language was added by Geometry guy about a year ago. I have no objection to mentioning copyright issues, although I think it's more an issue of "Well-written" than "Factually accurate and verifiable". I'd also be willing to make it a separate, seventh criteria. Perhaps something like "it contains no obvious copyright violations" would be adequate. WhatamIdoing (talk) 17:40, 29 July 2011 (UTC)

• The first comment is not entirely accurate: Copyright/Copyvio can refer to both text and images. Copyright of images is specifically covered by 6 (a) and that is grounds for not "listing" an GAN; copyvios in text is (partially) covered by the 2 (a)/(b) verifiability checks, but these checks do not necessary go as far as finding & "proving" Copyvios, which is what the DYKers have started doing on existing GAs. Don't forget that these checks would cover both electronic and printed sources; so if implemented reviewers would not be able to check any text that was covered by book and accademic journal (both printed and electronic on subscription-only sites) citations unless they had access. How many reveiwers have to hand, for instance any or all of the books cited in the nominations they review, and unless you have a readers ticket for a university library many journals are unaccessible; also what about foreign language texts, I've taken on trust Crilic font references since I can't translate them (Norwegian, Spanish, etc, for instance I can check with a combination of Google and a dictionary, even if I can't say the words out loud). Pyrotec (talk) 18:44, 29 July 2011 (UTC)
• My first comment was entirely accurate. Copyright of images is in an entirely different mental space (apples and oranges) than copyvio of text when the reader processes that section of WIAGA. Your argument that one covers the other is abstruse and unrealistic.As for the second argument, "Who has journals?", that is a dodgy tack as well. Just because something is difficult to do doesn't give us license to say "F*ck it, it's too darn difficult. Screw WP:COPYVIO." That is just an irresponsible attitude. Moreover, I have also addressed this concern from another direction in an earlier post (above): no one should thiink that aa GAN is guaranteed to be free from copyvio, but GAN reviewers should be required to make a Good faith full effort to check for it. It's policy, after all... Let's be clear: no one is forcing anyone to do anything, but if you do nothing about copyvio, then you shout "Meh!" at the heart of the 5 Pillars.  – Ling.Nut (talk) 03:05, 30 July 2011 (UTC)

• Ling Nut, you are writting Bullshit and deliberately misrepresenting what I have stated above and selectively using Good faith when to suits your purposes. Images are covered by copyright (it's also made explict in Wikipedia:Featured article criteria, 3) and they are covered in WP:WIAGA 6(a), so that leaves text. As Malleus has already pointed out your proposal is requiring a guarantee of no copyright violation - nowhere does your proposal mention Good faith full effort. I was merely expanding on Malleus' comment that your proposed 2(d) requirement in practice means that everyone reviewing a GAN has to have access to the full set of references used in that nomination. You've obvious had little or no experience of GAs with a comment like "Who has journals?", but you use big words such as abstruse and unrealistic. What is " "Meh!", is this some of the bullshit that you routinely use. Pyrotec (talk) 08:05, 31 July 2011 (UTC)

• There's a big difference between doing nothing and doing what can reasonably be done. Malleus Fatuorum 04:29, 30 July 2011 (UTC)

• That is one of my points. You recall that you and I have talked about "eating my own dog food" with respect to the comment on my user page that GA is the place where writers should be trained. Well, GA is not FA, and FA is not dissertation, and dissertation is not publication by CUP or OUP. But even at the GA level (the bottom rung of that ladder, if everyone will forgive me for putting it that way), academic values need to be taught and propagated and supported, even if they are not perfected. I think we need to codify the idea that GA reviewers need to make a full and good faith effort to verify that there is no copyvio. The only place to codify such an idea is WIAGA.  – Ling.Nut (talk) 04:47, 30 July 2011 (UTC)

• I agree with this. I think spotchecking is adequate - a lot can be learned, both by the person doing the spotchecking and the person whose work is checked. But on a more fundamental level, the problem is that many editors don't seem to understand that plopping in a copy/paste bit of work and slapping a ref tag after it constitutes copyvio. This kind of writing is propagated in schools in the US, though with antiplagiarism software that's beginning to change, but defining the mere existence of copyvio seems to be a big hurdle, in my view. Many either don't want to hear, don't believe it, or simply don't understand - this applies to reviewers as well as nominators. Truthkeeper88 (talk) 04:54, 30 July 2011 (UTC)

• You may well be right, but I hope you aren't. When I was at primary school many, many years ago, we had regular lessons in what was then called English comprehension. Basically it consisted of either listening to a piece of text or reading it and then coming up with a precis in your own words. Malleus Fatuorum 05:04, 30 July 2011 (UTC)

• I'm an English as a Foreign Language teacher in a university in an Asian country, so this particular problem is my daily breakfast of toad (when school is in session). But that's neither here nor there. I think GA can be a training ground for academic values, and I think we owe it Wikipedia's readership (and by extension, our collective reputation) to at least make good faith efforts in this direction.  – Ling.Nut (talk) 05:23, 30 July 2011 (UTC)

• No argument from me there. Malleus Fatuorum 05:29, 30 July 2011 (UTC)

• Volunteer Marek has made some constructive comments on User talk:SandyGeorgia#Misallocation of effort on how Duplication detector might be used on online sources to find "duplicated texts" an how visual comparison is needed for Amazon and PFD files. Howwever that is not the full story, those on this page who selective use WP:AFG when it suits them need to recognise that not all such copyright violation is on wikipedia (wikipeida text and/or text on mirror sites) can appear in electronic and printed form and the false postive results can be obtained. Pyrotec (talk) 17:45, 31 July 2011 (UTC)
• There is still much "holier than thou" attitude on this page: having done two Masters in this decade: respecting copyright, avoiding pliagarism (putting short quotations in "quotes"), summarising and quoting sources were taught to me. I also have to assume (WP:AGF) that graduate students working on wikipedia have had similar training. I recognise that things can go wrong through no fault of the editor: perhaps the short quote was quoted in "quotes", but the next editor along removes the "quotes", someone else tries to use a longer quotation but forgets the "quotes",etc, etc. I've reviewed wikipedia educational assignments at GA (in 2009): my university lecturers did day-courses on copyvio and all student submission had to be through a anti-copyright violation package. Nevertheless, as a GAN reviewer, when initial evidence of copyright violations comes to light it can be quite difficult to know what to do with it. Various editors have and do assist in reveiwing educational assignment GANs. Ling.Nut (a teacher) makes the valid point "I think GA can be a training ground for academic values, and I think we owe it Wikipedia's readership (and by extension, our collective reputation) to at least make good faith efforts in this direction". Yes certainly, but I've not seen any evidence of these teachers, apart from one, making any constuctive effort in providing training and instruction to GAN reviewers, let alone GAN reviewers reviewing education assignment GANs particularly on instructing their students. They also seem entirely detached from reality: In one place it is being said by Ling in defence of DYK: "Copyvio/plagiarism is occasionally easy to spot (e.g., a sudden gem of brilliant prose parachuted into general mass of crap). However, more often than not, copyvio is quite difficult to catch. As we have discussed ad nauseum, downstream Wikifluvia makes determining "who said what, when" a very, very time-consuming and labor-intensive process. Given that, missing instances of copyvio is simply to be expected from any one reviewer, and even (to a lesser degree) from a group of reviewers. Simply put, it is a cost of doing information business on the Internet." and later at GAN is selectively feeding misquotations to SilkTork (who can't be bother to check the source) and comes up with I think Ling has a point. Awareness of copyvio is already implicit in criteria 2, and only poor reviewers will pass a GAN without looking at sources, in which an ordinary copyright violation would be observed. For someone who has done 87 GAN reviews that is somewhat a cavalier attitude: I very much doubt that he has checked all the sources, printed as well as electronic for copyright violations. Pyrotec (talk) 17:45, 31 July 2011 (UTC)

(←) Adding 2(d): It contains no copyright violations is easy enough; a second question is, do we consider whether to add something about "full and good faith efforts of the reviewer" in the introductory text, or do we just let it remain unspoken but assumed?  – Ling.Nut (talk) 05:47, 30 July 2011 (UTC)

• I think Ling has a point. Awareness of copyvio is already implicit in criteria 2, and only poor reviewers will pass a GAN without looking at sources, in which an ordinary copyright violation would be observed. However, some violations are hard to judge, and it would take an expert to make a decision, or even to spot the violation. Copying data from lists is typically done. While we accept that copying map data is a violation, copying financial data such as sales of films or albums, or the assets of a businessman, is routinely done, and even defended because "you can't copyright facts"! Wikipedia's approach to lists that might be intellectual property, is to only use a sample of the list (I think that 10% may be standard - and that is usually the top 10 of a list of "100 Best Foo"). Because of the intricacies and difficulties of identifying some cases of copyvio, the phrase "contains no copyright violations" might be a bit too ambitious for a GAN. The 5 pillars says: "Respect copyright laws", and that is a more reasonable expectation. I also agree with WhatamIdoing that copyvio is part of "Well written" - as Malleus says, summarising text or data in your own words is a precis exercise. And precis is the heart of what writing a good Wikipedia article is about. Perhaps there could be an addition to criteria 1 (a): "the prose is clear and concise, respects copyright laws, and the spelling and grammar are correct;". This also skips around the tricky area of data copyright violations, dealing with the easy to spot cut and paste jobs. SilkTork ^{✔Tea time} 06:33, 30 July 2011 (UTC)
• (ec) That was waffle. Try again. Agree with Ling that looking for obvious copyvio is already implicit in criteria, and is done by all but the poorest reviewers. Making it explicit will be helpful to new reviewers and a reminder to all, including those nominating, to check for duplicated text. Agree with WhatamIdoing that copyvio is part of "Well written", and placing it there avoids the more tricky data copyvio situations. Suggest amending criteria 1 (a): "the prose is clear and concise, respects copyright laws, and the spelling and grammar are correct;". SilkTork ^{✔Tea time} 09:58, 30 July 2011 (UTC)
  • Dude, you rock – despite losing style points, because the painting on your user page makes me mildly seasick.  – Ling.Nut (talk) 09:50, 30 July 2011 (UTC)
    • Well considering this thread [1] its somewhat of an improvement, but in response to your answered question Ling is deliberably selectively taking Pyrotec's statement out of context under the label "poor reviewers". I will try and correct your prosse: "That was waffle. Try again. Agree with Ling that looking for obvious copyvio is already implicit in criteria, and is done by all but the poorest reviewers. Looking for obvious copyvio is already implicit in criteria, and is relatively easy to do when websites are being used as citations, but is far more difficult for off-line sources. Making it explicit will be helpful to new reviewers and a reminder to all, including those nominating, to check for duplicated text. Agree with WhatamIdoing that copyvio is part of "Well written", and placing it there avoids the more tricky data copyvio situations. Suggest amending criteria 1 (a): "the prose is clear and concise, respects copyright laws, and the spelling and grammar are correct;". Pyrotec (talk) 14:00, 31 July 2011 (UTC)

(←) 1. A new clause 2 (d) was suggested with good indentions but it was badly drafted with little appreciation of reviewing at GAN and is at worst "box ticking". Adding a new clause 2(d): It contains no copyright violations is easy enough to write on the page but it is entirely unenforceable, where any editor with a username can review any GAnomination, (see below), so anyone putting a tick in that box has probably not carried out sufficient checks that would stand up in a court of law. Let's be clear here Ling Nut specifically states: Read that last bit again, please, mentally underlining the words "a Wikipedia policy with legal considerations". On the other hand, making a "good faith effort" to check for copyvio is to be welcomed. That could be considered an adduct of the checks that are carried out to try and verify that what is claimed in the article is supported by the citations, but it could also be covered seperately. Perhaps SilkTork's suggestion above is a better way forward than the 2(d) clause.

2. Unfortunately, Copyright violation is not such a simple subject as some editors are trying to present it here. Practically, Copyvio can be carried out electonically against sources in those article such as websites, journals that are on open access in electronic form, some books that are out of copyright and on googlebooks and extracts of books that are visible via Googlebooks and/or Amazon. So those checks could be done by any reviewer in any part of the world. It cannot be done in that manner when books and journals are used as citations. Checks can only be done on these when the books and journals are available to the reviewer. A copyvio check will provide a list of lines or phrases of text where those in the wikpedia article and those in, say, "web X" match; but that is only part of the story - who wrote the text first needs to be considered, it not unknown for websites (and newspapers) to copy wholesale from wikipedia and not acknowledge that fact. If and when wikipedia articles are shown to have breached the copyrights of valid copyright owners, the article should be handled in accordance with wikipedia copyvio procedures, and that could/should include failing the GAN nomination. In some cases, the wikipeida editor might not have breach the copyright: but I'm not sure that good faith is being consider and/or extend to them.

• I suppect that, this will be dismissed by those that have no real interest in improving wikipedia; and are therefore unwilling to discuss it. At the very worst, we have editors doing GAN reviews (as they are entitled to do) with no concept of doing reviews and certainly none of copyright. I'm not against adding another box, per se, but what difference does it make to those reviews that don't review against WP:WIAGA and/or use {{GAList}}, {{GAList2}}. At the other end of the scale there are good reviewers who will make an effort to chech the article against copying and pasting for the web-based sources, but may not go much beyond that. I did find what I suspected to be serious copyvios in two out of the three educational assignments I reviewed concurrently at Wikipedia:School and university projects/User:Piotrus/Fall 2009 and those three rewiews took over one month to complete and I needed the assistance of a wikipeida Copyright expert. That was an educational assignement and "copyright training", to summarise a phrase used above, came from wikipedia (me, and an expert): I don't remember signing up to do that when I started doing GAN reviews; and I not been trained to teach it or detect it. Over the weekend, I got a DYK reviewer to run a copyvio check on one of the articles that I am currently reviewing at GAN: the check for matches against the "electonic" soures took less that 13 minutes; none were found. The books and the (non-electronic) journals have not been checked, yet. So "good faith effort" checks appears to be a realistic option. Pyrotec (talk) 09:54, 31 July 2011 (UTC)

3. Finally, I would caution SilkTork (but more importantly Ling) to stop making disparaging remarks about reviewers not finding copyvios, as yet none appear to have been found in nominations that editor has reviewed. DYK has been under attack for copyright violations and as a matter of interest they were/are doing copyvio checks on Geography and places GAs. So far copyvios have been found in two GA articles reviewed by me (one of which was already a DYK and had those copyvios present at that stage) and in other several other GAs passed by editor(s) who I regard as outstanding reviewers. Pyrotec (talk) 09:54, 31 July 2011 (UTC)

During my many years as a professional software developer I learned one lesson that's very relevant to this discussion, one that FAC has already largely adopted. When you're developing a piece of software it's literally impossible to test everything, so you have to target your tests to areas of potential vulnerability. That may be particularly complex bits of code or unusual constructions, but it may equally well be code written by developers of unknown or dubious ability. FAC has an institutional memory in its delegates, so it has an advantage over DYK and GA, but I'm quite sure that you, Pyrotec, at least sometimes do what I often do; look for articles nominated by editors you're come to trust. I wouldn't dream, for instance, of checking every single source used by Ealdgyth in one of her bishop articles even if I had access to them. It would be a misapplication of effort. On the other hand, if I'm looking at a nomination by an editor I'm unfamiliar with on some kind of pop culture article largely backed up by online sources, for instance, I'll certainly check out at least a few of them. But to return to my software analogy, no company would claim that its software contains no bugs, just as we can't claim that no GA contains copyright violations, which is why I object to Ling's proposed addition in the form he suggests it. Malleus Fatuorum 18:07, 31 July 2011 (UTC)

Well put Malleus. SilkTork's suggestion does have some merit. But for someone who has done 87 GA reviews (according to his userpage) he is somewhat out of touch to come up with a statement "Awareness of copyvio is already implicit in criteria 2, and only poor reviewers will pass a GAN without looking at sources, in which an ordinary copyright violation would be observed.". It is made clear that I am the "poor reviewer", by reference to his talkpage: selective misquoting by Ling.nut and too lazy to check the source. Well he (ST) is an Admin and presummably is getting round no Personal Attacks that way (so I might be banned, but you can guarantee that he will not be). Perhaps I aught to check for copyvios on his 87 GAN reviews. I could do that as an IP user during my ban - that would be fun. Pyrotec (talk) 18:36, 31 July 2011 (UTC)

I'd be prepared to make a small bet that you'd be certain to find copyright violations among those 87 articles. But I really don't think we (at GA) need to beat ourselves up too much over this. When problems are brought to our attention we have processes in place to deal with them. What does DYK have? Malleus Fatuorum 18:56, 31 July 2011 (UTC)

Firstly, I found this Wikipedia talk:Did you know#Helpful tools to identify plagiarism/copyvios, so new DYKs could be better, but like GA there will be skeletons in the cupboards; and I've had some training (see below). Secondly: I have no wish to beat myself up: I'd much rather do it to another(s). There are 10/11/12/13 year-olds doing GAN reviews; a typical review: Yes, cool!!! Really Great!!! GA. (P.S. can you done one for me now) and no mention of WP:WIAGA - there is some poetic licence here but not too much. They will not do these checks: and some people are shouting: you've got a GA now do one review yourself (if the GA was awarded by a child editor then the awardee will also do a "child review"). Jez gets a barnstar for 500 reviews, which is really great I've only done 401 (& 46 GA sweeps); but having done 58 reviews myself in one month during the April 2010 GAN backlog drive and another editor (hate, hate) does twice that in 2011 I can guarantee that Copyright was not checked in any detail during the various backlog drives. I'll quite happily run DD on "free to air" web sources to check for copyvios on all current and future GANs I review (I've had a training session, see User talk:Pyrotec#Nibiru collision) but I can't see any practical way of doing it on printed sources without having the sources to hand. My local library changes money for an inter library loan (70p, not a lot, but why should I pay), buying a photocopy of a journal article from the British Library is daylight robbery for an individual, and I have access to a university library (hard copies only) but that is a three-hour trip by car and "170 miles of fuel" for each trip. Even SikTork states (on his talkpage) I also have no interest in doing any form of copyvio check that would involve having a copy of Hansard's Copyright Law Essentials, a magnifying glass, and a sheet of litmus paper by my side. We are talking about the basic "material copied from sources", where the sources have a copyright claim. If the Wikipedia text says "Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua", and the copyrighted source text says "Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua", then that would be flagged as a concern with the nominator. If the issue is not resolved after reasonable time has been given, then the article is not listed as a GA, which is not fully compliant with the Wikipedia process for Copyvios Wikipedia:Copyright violations; but he comes out on this page with inuendo about poor reviewers/poorest of reviewers (two revisions). So far the GA copyvios were found (in my case) in two 2009 GAN reviews, those of us still doing GA reviews are (I hope) much better at reviewing: but I supect the standard goes downhill sharply everytime there is a backlog drive. Also, some of us are/were doing education GAN reviews which can take month(s): and there is more bullshit about journals ("no one uses journals") and books (see above). I don't see how GAN reviewers can be expected to provide free lessions on copyright during GAN unless those teachers who are slagging us off provide practical support as opposed to character assassination. As a student I was taught how to avoid (legally) potential claims of copyright violation and plagiarism but training in how to detect it (them) in students/editors GANs was never given. It can take me upto a month to do a GAN on a long article (for WP:WIAGA) but I have done (as you state) Ealdgyth's bishop articles in half an hour, in my last MSc we had some classes of 80 students (joint MSc/MBA classes), the professor's turnround on assignments was about one week or so for the course as a whole (I've never done 80 GANs in one/two week(s), neither has Jez, yet). So, "Assuming Good Faith" is easiest, "making reasonable efforts" is acheived and giving a "written guarantee (even a tick in the box and a digital signature)" is probbaly only possible by teachers (who don't do GANs, and don't have journals nor books as sources).Pyrotec (talk) 21:23, 31 July 2011 (UTC)

• I'm really sorry if any of my remarks have been offensive in any way. I had no intention at all of criticising Pyrotec in my comments here or on my talkpage. I was not aware of Pyrotec's involvement in this issue when I made the comments above, and in the comments on my talkpage I thought that I was expressing support for Pyrotec's view that we should not be required to take out a law degree or employ sophisticated text analysis techniques. I do, however, stand by my statement that it is a poor reviewer who doesn't do ANY spot checks of sources or fails to notice a copy and paste. I feel that a decent GAN involves a bit more than a quick copyedit, and a lot of the GA criteria require some looking at sources and a bit of background reading. That doesn't mean that I would expect every source to be checked, nor that I would expect expert legal analysis of text and data in order to uncover potential copyright theft; but reasonable and obvious checks to be done, which anyone who is going to be on this talkpage talking about this issue is going to do. We all know what a poor review is: there'll be a comment saying - "I'll look at this", and a half hour later the comment "Looks good. Two spelling mistakes, but otherwise fine. Sources accepted in good faith. Well done. Pass."

Looking at a source and noting that the statement in the Wikipedia article is worded exactly the same as the source is an easy check. I think checking for something that the average reader would pick up, is something we should be doing. Anything more than that probably belongs to the Wikimedia Foundation lawyers, or at least some of the more experienced Wikipedia editors who specialise in copyright. I think that checking for neutral balance, no original research and broad coverage are slightly harder tasks than checking for text copying, and we accept those as standard. We do not, however, assume that by doing checks for neutral balance in a BLP article that we would be expected to commit ourselves or the Foundation to stand by that check in a court of law. We are not doing a legal check, we are doing an informal peer review against some minimal standards that we have set up ourselves to encourage improvement of articles on Wikipedia, motivate editors, and to give some reassurance to readers that a GA listed article has some basic standards. A GA article is not expected to be of a finished or exceptional standard. It is just a decent article that should have no glaring errors, such as text copied word by word from a copyrighted book. SilkTork ^{✔Tea time} 21:45, 31 July 2011 (UTC)

Thanks SilkTork for your kind works, above. Pyrotec (talk) 09:32, 1 August 2011 (UTC)

Would all GAN reviewers, please consider that copyvio is not limited to the declared sources. As an editor (hypothetically) I might have lifted some of my words from Googlebooks, Amazon, a library book, Scientific American, British Medical Journal, etc; of course I have cited quite a few of my references, but not all, I forgot to note down the source. Those unacknowledged authors have copyrights that "my" hypothetical article has breached, so when you do your GAN on my nomination and you certify that it is free of copy violations: did you check those sources that I "forgot" to declare? Some universities have software that does those types checks and I've seen "hits" on my submissions against sources that I know that I did not use. Are you are claiming that this is do-able at GAN and easy - I think it is not. This is not an argument against doing checks, we should be doing them; it is more a warning that some demands being placed on reviewers are unrealistic and unachievable (and that some of you are not even prepared to even listen). Pyrotec (talk) 09:32, 1 August 2011 (UTC)

I really think this should only be part of the reviewing instructions - something like "Ensure that the article is checked for copy right violations and plagiarism."

I have only done one GAR, but it had lots of copyvios in it. I brought these to the attention of the nominator and they were changed within 48 hours. I use Duplication Detector to check the article against a few often quoted refs, Google searches (by placing a few long sentences into the search box) and Google books to see if any of the books referenced can be found and checked. I realise it takes a while, probably around half an hour if three or four are found, but I really think it is a necessary part of the process. Chaosdruid (talk) 18:14, 1 August 2011 (UTC)

I've had a tutorial on the use of Duplication Detector on one of the GA nominations that I'm currently reviewing (its on my talkpage) and some of my reviews can take several weeks, or as little as half an hour, it depends on what I consider is needed, but to return to the main point. WP:WIAGA is not a set of reviewing instructions, its more a set of quality standards, i.e. "this" is what is required of GAs. The point has been made that WP:WIAGA does not specifically include copyright considerations of text, so this section is to determine what sort of changes (if any) are needed. Whether Duplication Detector, other software, or Mk. I eyeballs, are used to do some of these checks is part of the process of reviewing, rarther than the quality requirements. Pyrotec (talk) 21:01, 1 August 2011 (UTC)

Status check

Before the above discussion gets any further into the weeds, can we have a quick 'show of hands' to make sure that we all actually agree on what we seem to agree on?

1. Does anybody object to mentioning something about copyvios in the criteria? Anyone at all?
2. Does anybody actually object to listing it (somewhere, somehow) under "well-written"? (The other options suggested so far are "verifiable" (all copyvios are 100% verifiable by their nature) or as a separate, seventh criteria.)

If you don't object, then you don't need to reply. But if you do object, please sing out ASAP; we don't want to overlook your concerns in a wall o' text. WhatamIdoing (talk) 23:18, 31 July 2011 (UTC)

I'd be for adding a couple of words in Criterion 1, but I don't think we need to add a new criterion just for that. To me, it does seem common sense that GA reviewers should be checking for that, anyways. However, given the history of FAs and plagiarism the past couple of years, I fear that I'll be eating my own words here. –MuZemike 00:10, 1 August 2011 (UTC)

• I will object to any words that link or appear to link lack (or apparent lack) of copyright checks of text to poor GAN viewers. I will object to any suggestions that doing copyright checks of text is easy (I'm going to be brutal here: if copyright violation is shown to exist in a GAN review the appropriate action is not "failure" of the GAN, the article is immediately removed from wikipedia by blanking with the copyvio template and it stays blanked whilst a formal investigation is carried out - the article could be reinstated, reinstated with the offending material removed, or lost entirely, but that is not the decision of the GAN reviewer). I will violently object to adding a statement that all copyvios are 100% verifiable:- under the current system of viewing a reviewer has to have access to every source used in the article and all the "sources" that are not declared in order to do these checks. It also has to be shown that the "matching text(s)" was copied by wikipedia from the valid copyright owner, not some website has copied without acknowledgement copyfree text from wikipedia and/or a mirror site of wikipedia. Is WhatamIdoing willing to certify that every declared source was checked in full against the article for each and every article reviewed; also that each and every article was checked against all possible undeclared sources. (P.S. Remember: copyvio checks could be done on your reviews, and I expect the some would be appear). Finally, I am not against some form of words such as: 1 (a): "the prose is clear and concise, respects copyright laws, and the spelling and grammar are correct (made by SilkTork, a foot or so (30 cm, or more) above). Pyrotec (talk) 09:07, 1 August 2011 (UTC)

Since I spend far more time supporting other people's reviews, the list of "my" reviews is actually quite short, and you are entirely welcome to double check them. I am apparently unusually thorough in source checking, so I doubt that you will find either NOR or COPYVIO problems.

But this is all off-topic at this moment: We're nowhere near the point of sorting out wording. I believe that when we reach that point, you may safely rely on my firm conviction that the primary description of a poor GA reviewer is somebody who makes up fake criteria, not someone who fails to execute a complex task perfectly on the first try. WhatamIdoing (talk) 16:31, 1 August 2011 (UTC)

Yes, I knew that, but did not wish to bring it up first; and there are reviewers who don't appear to use any criteria, fake or otherwise. Pyrotec (talk) 20:35, 1 August 2011 (UTC)

Let me get this straight. Pyrotec and I have probably done something like 1000 GA reviews between us. Yet you, who by your own admission have done rather few, and have an inflated sense of your own competence ("I am apparently unusually thorough in source checking") knows better than we do what is and isn't practical? How does that work exactly? Malleus Fatuorum 23:50, 1 August 2011 (UTC)

No, I don't have an inflated sense of my own competence. I do have a good sense of the proportion of sources that the typical reviewer checks, and I know how that compares to the proportion of sources that I check. I do not pretend that either of you are typical reviewers. For example, I believe that I would be far from the only person who would rate you as being more thorough, more proficient with text, and more needlessly insulting than the median GA reviewer. That I am more thorough in my source review than the typical GA reviewer does not mean that I am more thorough than every single GA reviewer.

Conducting reviews does not provide one with any information at all about how the typical GA reviewers behaves. Conducting hundreds of GA reviews tells you only how you personally conduct a GA review. Reading hundreds of GA reviews by other people (and I have) tells you quite a lot about about how the typical GA review proceeds. WhatamIdoing (talk) 00:04, 2 August 2011 (UTC)

Comments and a proposal

I read the above with interest and have some comments and a proposal.

1. An article containing plagiarism is not well written, and so does not satisfy criterion 1, but 1(a) does not at present make this explicit and perhaps it should. Wikipedia has been "the encyclopedia anyone can edit" for 10 years, and cut'n'paste text can be deeply buried in the edit history of an article. This serious problem certainly does affect GAs: I've encountered many examples such as Joseph Moir, Dancing the Dream, Peak water, and Frank Barson and David Yates.
2. There seems to be some confusion above about the role of this guideline. It is not an instruction list for reviewers, but a set of standards for articles. The responsibility for ensuring a good article meets these standards is a collective one, shared not only by the entire reviewing community, but by nominators and article editors (who are most familiar with the content and sources). It isn't the job of an individual reviewer to validate or certify that every sentence of the article meets every criterion. They should instead, at minimum, make sufficiently many checks to be reasonably confident that a listed article has no manifest shortcomings with respect to any of the criteria. I'm not in favour of adding comments ("good faith full efforts" etc.) to this guideline: WP:Reviewing good articles is the place to describe what reviewers should do, not here.
3. It is practically impossible to guarantee that an article meets all the GA criteria, so GA status is more like a warranty or a vehicle safety test than a guarantee: "this article has been tested and checked; if you find any problems, please send it back immediately for reinspection". Consequently the criteria are in part aspirational, and should be inspiring: it is nonsense to state that GAs should "contain no apparent copyright violations"! Does that mean well-concealed copyright violations are okay? No, any copyright violation is grounds for delisting, and the criteria should be unequivocal on each and every point.
4. In order to be inspirational and educational, the GA criteria need clarity of purpose, so that editors and reviewers understand what each criterion is for. There is a lot of confusion concerning the distinction between copyright violation, plagiarism and close paraphrasing. The last is (or can be) a prose issue, while the first is a legal and free content issue. In-text copying without quotation and/or attribution is primarily plagiarism (evidently it is additionally a copyright violation only if the source is under copyright). Adding "it contains no copyright violations" to criterion 1 does not clarify the matter! Criterion 6 is also mixed up in this way, conflating a content quality issue (using images and other media effectively to communicate content) with a copyright issue (fair use rationales and all that jazz).

In my view, a criterion expressing the aspiration to provide free content is long overdue. That includes "no copyright violations", but also issues such as attribution, licensing information, fair use rationales etc., so that content can be reused under CC/GFDL without violating copyright or license agreements.

If it were entirely up to me, I would:

• Add "not plagiarized" (or something similar) to 1(a);
• Move 6(b) to 1(c) "it is illustrated, if possible, by images relevant to the topic with suitable captions";
• Incorporate 6(a) into a new criterion that a good article is "Free reusable content, without copyright violations..."

In that way, every criterion would have a clear purpose, with all copyright issues fully covered by just one of them. Geometry guy 10:33, 1 August 2011 (UTC)

I disagree with your first assumption that an article with plagiarism is poorly written. It is possible that the copyright violations and original text are joined together very coherently, giving the reader a clear understanding of the topic. I am not advocating plagiarism (!!), I am just declaring that there could an exception to your assumption. Binksternet (talk) 11:53, 1 August 2011 (UTC)

• Considering the justification first:

1. I agree with Binksternet. I could for instance take a whole chunk of copyrighted text from The Oxford Companion to British Railway History (for example) and integrate it into a preferred railway article without any acknowledgement, and one that does not use that reference already. Being "Oxford" it's well written, it could appear to be my text to a reviewer, its not (for this example) controversal, likely to be challenged or a BLP, so arguably a citation is not needed. It's definitely Copyvio (the reviewer may not know that) but poorly written? No.! I could do the same approach with some high quality text that is free of copyright. In this case it's plagiarism, not copyright violation, it is still not poorly written in the usual sense (grammar, etc), but I have failed to provide proper acknowledgement of someone else's work.
2. Good point: In a "prefect world" editors have responsibility to produce articles to the standard before submitting, reviewer has the responsibility to review aginst the standard and award GA only if the article is compliant. If the article is not up standard a good reviewer can help bring it up / or withhold GA. A "poor" reviewer could unjustly withhold GA, and nothing further happens, it could go to GAR, or just be resumitted at GAN (mostly OK: this proposal does not change this situation). A poor article could be passed by a poor reviewer, nothing further happens or it goes to GAR. (Note: this proposal does not change this situation).
3. I mostly agree. But Copyright violation (text and/or images: yes I know 6a, but removing the image from the article and copyvio-ing the image will solve that) is NOT necessarily grounds for delisting an article at GAN/R or a GA-delisting the correct approach. The only applicable process is Wikipedia:Copyright violations. Under your system, you find a text "copyvio" so you delist it at GA and presummably remove the affending text(s). A couple of edits or so later some vandalism occurs so a rollback or "undo" brings the article back the the GA-state with copyvios (but maybe without the green star). Under Wikipedia:Copyright violations proper checks will be made, your copyvio might be a false positive, the owner may be willing to give permission, etc. I've done reviews (educational assigments) where copyvios were found during my reviews: the articles (two or three, I forget) were blanked with Copyvio templates but Moonriddengirl provided "rolled back" versions for the review/corrective actions processes to continue: one at least made GA in the end.

• The proposal has some merit, but I would like it to emphasise both text and images, So: "Free reusable content, text and images (where used) without copyright violations...". I'm happy for my grammar to be improved/changed but I would like to be clear that both must be considered. Pyrotec (talk) 13:51, 1 August 2011 (UTC)

  I agree and that is what I mean by the "...": a criterion 6(a) for images and 6(b) for text, perhaps?

  I also broadly agree with the other remarks. For me, a plagiarized article is poorly written ("well copied" perhaps!), and plagiarism is bad writing (I've used this to delist articles), but I don't want to argue a point where we are basically in agreement: 1(a) should be clear about plagiarism, and it currently isn't. Concerning point 3, I was not intending to propose any system: by "grounds for delisting" I simply meant that if a listed article does not meet one of the criteria, there is a prima facie case for doing something about it - such as fixing the problem. This guideline is about what the GA criteria are, not about what happens when they are not met. Geometry guy 15:29, 1 August 2011 (UTC)

• Thanks for clariying that. I'm happy with criterion 6(a) for images and 6(b) for text. I had made the assumption that there was only a new 6(a) Free reusable content, without copyright violations..."; and I'd seen the move of 6(b) to 6(c) but somehow failed to see that there was a gap at 6(b). I must be burning out. In terms of Copyvio at GA, I'd probably use the copyvio process, put the review On Hold and if there was reinstatment but wholesale cuts consider failing on scope/broadness (if applicable); but I'm happy not to discuss this aspect any further. Pyrotec (talk) 15:55, 1 August 2011 (UTC)

I'd rather keep text issues under Criterion 1, and images isolated at Criterion 6. Because images aren't actually required (unlike an absence of copyvios), I think that we're better off addressing copyvios in an always-required criterion.

Given the (lack of) skill level in the community for evaluating plagiarism, I'm nervous about directly naming plagiarism as an issue. That standard is mushy, context-sensitive, and subject-dependent. We have people who believe that nearly everything without an WP:INTEXT attribution is plagiarism. Someone a while ago was arguing that incorporating public domain text is always plagiarism unless you begin each sentence with "According to the 1911 Encyclopedia Brittanica..." By that standard, Wikipedia is known to have tens of thousands of significant articles with major (so-called) "plagiarism" problems.

I think that the community is (barely) capable of identifying tolerably obvious copyrights at a reasonably consistent level. I don't want to introduce a standard that we can reasonably expect to be interpreted in significantly different fashions by different experienced editors. WhatamIdoing (talk) 19:26, 1 August 2011 (UTC)

Again I don't have any strong objections. It would be useful if you could summarise what changes if any you would make/suggest to the existing Criterion 1 to include copyright aspects of text. Pyrotec (talk) 20:43, 1 August 2011 (UTC)