Wikipedia talk:Featured article criteria

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
          A Wikipedia ad has been created for this project page

location[edit]

When citing books, is adding the location necessary? It seems tedious and adds nothing of value to the article. I've got an article where one citation includes location and about 40 other sources to manually look up and add location as a result. Seems easier to just remove the one location that to add the other approx. 40. SpartaN (talk) 17:37, 26 July 2017 (UTC)

Presumably, you mean the location of the publisher; the FA criteria neither require nor forbid this location. What is demanded by the FA criteria is consistency: so either all of the refs should have the publisher's location, or none should. --Redrose64 🌹 (talk) 22:02, 26 July 2017 (UTC)
what Redrose said...I won't demand locations unless there are problems with inconsistency. I will say that as someone who checks sources, locations do help, as anything out of the ordinary will often help flag up an unreliable/self-published source...but they are not required Ealdgyth - Talk 22:13, 26 July 2017 (UTC)
I couldn't disagree more strongly. Style should never be allowed to triumph over verifiability. The "consistent" requirement is intended to avoid a mix of citation styles (e.g. parenthetic vs <ref></ref> tags). It is not to ensure that bibliographic information is minimized. Since not every source cited in an article will have the same set of bibliographic information available to relate @Redrose64:'s interpretation would imply citing to the least common denominator: a very poor policy. In the particular, a location is sometimes necessary to distinguishing similarly named publishers or publications. If a different unique identifier is provided then the location may not be useful to a reader, but it never hurts to provide the location when it is available. In the spirit of wp:SAYWHEREYOUGOTIT, it's certainly possible for different printings to vary even if they bear the same year and publisher. LeadSongDog come howl! 14:04, 27 July 2017 (UTC)
I'm not aware of any policy or guideline that requires locations for references, however. The problem of which edition/printing a work is can be solved by using ISBN/OCLCs to distinguish. I always use locations, but absent a requirement for them somewhere, we can't require them at FAC. Ealdgyth - Talk 14:27, 27 July 2017 (UTC)
I am totally in agreement with LeadSongDog on this. Excessive forced consistency does nothing for quality, it is purely decorative, and can do harm.· · · Peter (Southwood) (talk): 10:59, 14 December 2017 (UTC)

Criterion 1c[edit]

Criterion 1c requires featured articles to be based on "a thorough and representative survey of the relevant literature". But can a survey be both "thorough" and "representative" at the same time? The two terms seem to be mutually exclusive; "thorough" means complete in every detail, "representative" means a sample. The point has arisen from a current FAC review, but to my mind the wording presents a general problem, and perhaps requires reconsideration. Does anyone agree? Brianboulton (talk) 17:44, 10 November 2017 (UTC)

Brian, I understand "representative" to refer to the sources a subject-matter expert would expect to see, so "thorough and representative" means "don't leave anything out that an expert would expect to see used as a source in an article of this length." SarahSV (talk) 18:13, 10 November 2017 (UTC)
I agree (with Brian), and if you have long memories you may remember my complaining about the "thorough" at the time. For any broad topic, a "thorough survey of the relevant literature" is impossible (do you really think the authors of Sea or Jesus conducted a thorough review of every relevant book on the topic?). Consequently, by having a criterion that's literally impossible to satisfy, WP:FA has developed an unhealthy culture of turning a blind eye and of "the criteria mean whatever I want them to mean", which in turn feeds into its at least partly deserved reputation as a clique of insiders slapping each other on the backs. ‑ Iridescent 18:55, 10 November 2017 (UTC)
"Thorough" does seem impractical for some subjects. But before changing the criterion, I would ask: are there problematic cases that "thorough" weeds out that would not be stopped by "representative" in conjunction with the remaining criteria? If there are articles that should not be considered FA that are only stopped by the "thorough" part of the criterion, then before changing it we should at least think of some alternative wording to cover those cases. --RL0919 (talk) 19:50, 10 November 2017 (UTC)
"Representative" should be sufficient, provided that the full scope of the subject is represented, and all due and relevant aspects are represented. If a reviewer is aware of something that should be covered, but is not, then that aspect has not been represented. "Thorough" is probably impracticable for a large proportion of major topics. The "criteria mean whatever I want them to mean" is a real problem and is applied all the time, it seems that many people will submit to extortion as it is easier than insisting on following published criteria. There seem to be more unwritten rules than written ones in both GA and FAC. · · · Peter (Southwood) (talk): 10:53, 14 December 2017 (UTC)
I'll make the change. If there are any issues, that's what the talkpage is for. I can't think of anything off the top of my head; a representative survey of the relevant literature sounds like the article is well researched, but is not pulling in ridiculous amounts of new sources for broad topics, which can lead to serious bloat or unrealistic expectations. The only issue that comes to mind is, say, the omission of a major detail found in a couple reliable sources but not in the rest of the literature; but in that case, is that content even necessarily notable enough for inclusion? Hmm. I think it's fine to remove as redundant and/or overly vague. —Deckiller (t-c-l) 15:20, 4 February 2018 (UTC)
  • Thorough does not mean exhaustive. It can be and should be both thorough and representative. Representative is enough for GA, but shouldn't be for FA, in my opinion. If I were to remove one term I would remove "representative" and not "thorough". The problem is that one cannot know if the chosen literature is representative if one has not first done a thorough review.·maunus · snunɐɯ· 15:37, 4 February 2018 (UTC)
Perhaps the issue is that the criteria are implying that the article/reference section is thorough, when it really means the research process? —Deckiller (t-c-l) 15:46, 4 February 2018 (UTC)
Something along the lines of "it has been subjected to a thorough and representative survey of the relevant literature"? —Deckiller (t-c-l) 15:47, 4 February 2018 (UTC)
Or, perhaps: "it has received a thorough and representative survey of the relevant literature"? —Deckiller (t-c-l) 15:58, 4 February 2018 (UTC)
Yes, something like that, or maybe that references included are representative based on a thorough survey of the relevant literature.·maunus · snunɐɯ· 16:06, 4 February 2018 (UTC)
I think we're on to something: "its references represent a thorough survey of the relevant literature"? —Deckiller (t-c-l) 16:09, 4 February 2018 (UTC)
I would buy that.·maunus · snunɐɯ· 16:13, 4 February 2018 (UTC)
We may indeed be on to something but the wording seems to have moved beyond what was initially discussed so I think we need further comment before considering a change. While that goes on I'd like to return the wording to what we started with. Cheers, Ian Rose (talk) 18:52, 4 February 2018 (UTC)
Works for me. Is anyone else interested in commenting? —Deckiller (t-c-l) 13:22, 5 February 2018 (UTC)

Request for clarification[edit]

2.c. consistent citations: where required by criterion 1c, consistently formatted inline citations using either footnotes (<ref>Smith 2007, p. 1.</ref>) or Harvard referencing (Smith 2007, p. 1)—see citing sources for suggestions on formatting references. The use of citation templates is not required.

Is this requirement for consistent formatting intended to imply that all citations in an article must either display full first names only, or all must display first name initials only, and that a combination of initials where provided by the source, and full first names where provided by the source is not acceptable?

I have just finished reading through the archives of this talk page, and cannot find anything definitive. · · · Peter (Southwood) (talk): 16:05, 15 December 2017 (UTC)

I would think that providing the names as they appear in the source is a consistent system. That isn't about formatting so much as honest representation, in the spirit of WP:SAYWHERE. --RL0919 (talk) 16:52, 15 December 2017 (UTC)
An article looks polished when the citations are consistently formatted using the same scheme. Saying that, if a few sources didn't give full first names, I wouldn't say anything if we had a few initials in a sea of citations with full first name, while the reverse wouldn't be very polished. Imzadi 1979  02:56, 16 December 2017 (UTC)
There was a discussion on this very matter a year or two back, I'll try to find it. One person held that if just one source provided only initials, then all of the other sources must be altered to also show only initials.
My own view is that you should give the credit according to what the source actually says, omitting any titles or honorifics. That said, if I have two books which are definitely written by the same person, and one is credited to "J. Smith" and the other to "John Smith", I will use |first=John |last=Smith for both of them. --Redrose64 🌹 (talk) 10:11, 16 December 2017 (UTC)
These suggestions appear entirely reasonable to me. The incident for which I am trying to find guidance is exactly as Redrose64 describes, where one reviewer contends that all sources must be either initials only, or full names only, which implies that if only initials are available for a single case, the entire list must be altered to be initials only. · · · Peter (Southwood) (talk): 13:15, 16 December 2017 (UTC)
I concern myself first with the reliablity of sources used. Then with making sure the same information is presented in the references. At some point, however, if we start demanding consistency in this form, we're descending into madness. As long as the references all contain author information and it's all either "first last" or "last first", we should not care if they are all initials or all first names. We should use what the source itself uses - which abides by WP:SAYWHEREYOUGOTIT, which has always struck me as a good general guideline to follow. Do what the sources do as far as providing information. Thus, I don't insist that ISBNs be provided for all references, if some sources predate the ISBN system. Same for how the names are presented for authors - go with what the source uses. Ealdgyth - Talk 13:38, 16 December 2017 (UTC)
As per Ealdgyth and Redrose. Not least, there are US v. UK variations in this (the UK tradition being to use initials, the US to use full first names), which mainstream academia has no problem with (you follow the way it was presented in the original publication). It can be very hard, in fact, to identify the first name associated with some UK published works, particularly if the writer was not well known. Hchc2009 (talk) 14:15, 16 December 2017 (UTC)
This is the kind of decision we ought to leave to the writers. Insisting on consistency would mean that all citations would have to use initials if the full name of just one author was unknown. SarahSV (talk) 17:04, 16 December 2017 (UTC)
@Raul654, SandyGeorgia, Ian Rose, and Sarastro1: If any of you who have experience in closing FACs would like to comment, it would be appreciated. · · · Peter (Southwood) (talk): 08:56, 18 December 2017 (UTC)
The most important principle here is that inline citations must be unambiguous. If the article cites two books by the same author, then the inline citations must explicitly call out the title of the book being referenced; if there's two authors cited by an article who have the same last name/initials, the article should cite them by their full names.
Beyond that, I'd support either (1) citing authors using whatever names that they publish under, or (2) citing them by their full names, as long as we are consist about it within an article. Raul654 (talk) 18:10, 23 December 2017 (UTC)
No, a title is not necessary in either harvard referencing or shortened footnotes if the author and year taken together are unambiguous. --Redrose64 🌹 (talk) 19:17, 23 December 2017 (UTC)

I have a different question about this requirement. Is it a requirement to use Harvard-style or shortened footnotes? I've seen a number of FAs with regular citation templates--is that allowed? Rachel Helps (BYU) (talk) 19:53, 26 March 2018 (UTC)

No specific system for formatting the references is required. If you want to use Harvard refs, that's fine (I like them myself), but if you prefer full citations, that's fine also. Templates can be used or not, as you prefer. Just be consistent with what you do -- for example, don't use "Cite book" some of the time and "Citation" other times in the same article. --RL0919 (talk) 22:16, 26 March 2018 (UTC)
Thank you. Rachel Helps (BYU) (talk) 15:27, 28 March 2018 (UTC)

FA without GA[edit]

Does a Featured Article have to be previously designated as a Good Article? Are there cases when an article was directly given the "featured" status without first being given the "good" status? hujiTALK 00:36, 9 January 2018 (UTC)

Hi, there's no requirement for an article to have passed GAN (or any other assessment) before being nominated at FAC. That said, I'd always encourage anyone to get as many eyes as possible on an article before nominating at FAC, e.g. at GAN, Peer Review, and A-Class Review if the article belongs to a project that employs A-Class assessments. There's also a mentoring scheme for first-time FAC nominators. Cheers, Ian Rose (talk) 00:50, 9 January 2018 (UTC)

Consistency[edit]

Given the comments above timed at 19:53, 26 March 2018 (UTC) and 22:16, 26 March 2018 (UTC), also the somewhat lengthy threads beginning at Help talk:Citation Style 1#Purpose of Cite web, perhaps we should define exactly what criterion 2c means by "consistent". We could provide some extra examples.

Inconsistency means:
  • the use of both parenthetical referencing and inline <ref>...</ref> tags
  • the use of Shortened footnotes for some references and full citations within <ref>...</ref> tags for others
  • using templates for some references and plain text for others of similar type
  • using one or more of the Citation Style 1 templates and also {{citation}}
  • using Chicago style for some refs and Vancouver style for others

We might even note that inconsistency does not mean that {{cite web}}, {{cite journal}} and {{cite news}} cannot be used in the same article - they can, and they must be if one source is found only on the internet, and the other two are found only in the form of printed matter.

At WP:WIAGA, several of the criteria are annotated with <ref>...</ref>, perhaps we could do the same. --Redrose64 🌹 (talk) 20:50, 27 March 2018 (UTC)