Wikiquote talk:Sourcing

Why are we not using the references function as used on Wikipedia? This method of citation is cleaner than adding a bullet underneath the quotation followed by citation text; although, citing numerous quotations from the same source could be cumbersome? Perhaps there's another way? Adraeus 21:31, 18 September 2006 (UTC)


 * My personal preference is to keep things the way they are. Wikiquote was started before cite.php was supported, but we certainly could use it now if that were the community consensus.  I feel that the current way is vastly preferable to endnotes in a compendium of quotations, where nearly every line should ideally have a citation.  121a0012 00:29, 19 September 2006 (UTC)

Changes on the WP side
We should take note of the changes in WP's policies in this area, particularly the merger of several related policies into the new w:Wikipedia:Attribution. I've looked the new page over briefly but haven't thought too much about how this page needs to change in its comparison of WP and WQ requirements. 121a0012 17:43, 21 February 2007 (UTC)

Thanks and some concerns
I'm still not ready to do a full review of 121a0012's excellent effort to get this policy defined, but I've got enough time to express some concerns. &#9786; ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * You're welcome. My comments interspersed.  121a0012 04:20, 31 July 2007 (UTC)


 * Screenplays are called "primary sources". Where screenplays themselves are published works, this is arguable, but screenplays are often treated as frameworks for the actual work of art; i.e., the performed play, televised show, or screened film. Current Wikiquote practice (at least for A/V works) treats these latter works as the "primary source", and the screenplay as an arguable backup source to decipher unclear dialog. ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * My view, although perhaps the minority one, is that WQ being a collection of written quotations, should always give preference to a written source as being authoritative over a nonwritten one; even if it is not available to us now, it is at least potentially available and should be recognized as the definitive record of what the screenwriter actually wrote. This is of course distinct from the definitive record of a particular performance, which is its own, distinct, primary source material.  It may be that I'm trying to draw a finer distinction for the purposes of a guideline than most editors can be expected to implement.  In any event, I invite a subject-matter expert to rewrite the guideline in accordance with the community consensus.  121a0012 04:20, 31 July 2007 (UTC)


 * Encouraging the use of unpublished primary sources can give editors the incorrect idea that speeches are primary sources and can be quoted without a published source. Primary sources suggests that the primary source for a speech is a document that records it, which agrees with Wikimedia's general requirement for physical instead of ephemeral "sources". Clarifying this would also address the question about radio programs and improvisation — if it's not published, it's not sourceable. (The justification is that if it's really memorable, it will get published somehow.) ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * I'm not sure I understand your comment here. A speech, unless it was made to an entirely private audience, is by definition published.  The only question is whether there is a reliable source for its text.  For speeches of public figures, the text of what the speaker intended to say is usually released to the press (often prior to the actual performance), and these days is usually made available on the speaker's Web site as well.  Speeches by entertainers and other non-official figures are often impromptu and have no reliable source for their text, but they are no less published for that.  The lack of a reliable source for the text does not mean that there is no reliable source for the speech itself!  The question for WQ to decide is how much leeway should be given to editors to supply their own transcriptions (or translations) when the published reliable source is not textual, and that's the same issue whether we're talking about what an actor said while acting a part on stage or what she said while accepting the Tony award for that performance.  121a0012 04:20, 31 July 2007 (UTC)


 * "Wikiquote's analogous practice to Wikipedia's no original research policy is the prohibition of vanity pages." Besides the non-PC use of "vanity pages" (which 121a0012 can't be faulted for, as we were still commonly using that phrase when he wrote this), this isn't really accurate. "Original research" in Wikimedia is opinions, synthesis, extrapolation, deduction, etc., from editors instead of from reliable sources. Self-quoting is more analogous to self-authored articles on Wikipedia, which was w:WP:VANITY but is now w:WP:COI (conflict of interest). Our ban of original research is much simpler because we don't accept commentary anywhere. Self-authored quote pages, on the other hand, are at least theoretically acceptable if scrupulously sourced. (It's just that we've almost never had such a siutation — one only that I can recall offhand.) ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * Feel free to recast. It's import to emphasize that we don't prohibit original research as it is generally understood in academia (regardless of how WP may use the phrase); what we prohibit is drawing conclusions that are unsupported by third-party sources.  (I would not say that WQ probits "commentary" generally, unless things have drastically changed in the past six months, although perhaps it should -- that would get rid of most of the proverbs pages, all of the boldfaced quotations, and most of the pointless images.)  121a0012 04:20, 31 July 2007 (UTC)


 * "Another collection of quotations" as a low-preference source can be read to mean that other quote websites are minimally acceptable sources. It should be made clear that they aren't, and that this refers to publications like Barlett's Famous Quotations or The Oxford Dictionary of Quotations. ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * Agreed -- but how to phrase this? Even Simpson's, probably the best source for contemporary quotations before Oxford was published, has very poor sourcing for most of its material.  That's a real, printed-on-dead-trees Reliable Source by WP's way of thinking, but not what I would consider a WQ-reliable source.  (I must continually emphasize the necessity of recognizing that just as WP and WQ have different purposes, so must they have different criteria for sourcing -- that was what drove me to create this proposed guideline in the first place.)  121a0012 04:20, 31 July 2007 (UTC)


 * "Essentials of a good citation": this section will have to be carefully synchronized with still-developing standards for citing sources. ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * Actually, the intent was to replace citing sources. I don't see the need for two separate articles.  121a0012 04:20, 31 July 2007 (UTC)


 * I'd like to add some examples about why we want all the sourcing details; e.g., if you don't include the article title and date for a news item, we can't reasonably expect someone to be able to find the quote so they can verify it. Many quote sources, including professional publications, don't seem to care about this, but we have to. ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * I have added one example for news sources, making sure to name a defunct newspaper that will be difficult to search online for quite some time to come. 121a0012 04:20, 31 July 2007 (UTC)


 * The Web section needs much more detail because it can be a ready substitute for any of the other categories (books, news, audiovisual material, correspondence) and is therefore frequently used instead of those more tangible sources. ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * We need to emphasize that the Web is not a substitute for the primary source, both because of the ephemeral nature of Web pages and URLs (despite Tim's pleas of ten years ago) and because Web presentations may not be available for access by the entire public. (Consider subscriber-only newspaper and magazine archives.)  121a0012


 * Our standard song sourcing format does not include lyricists. We need to figure out how to give reasonable credits to the performers and the lyricists, especially given the difficulty in sourcing such information. (Most musical crediting I've seen, outside album liner notes which aren't readily available for most fact-checkers, don't distinguish between lyricist and composer, if they even mention such information at all.) ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * Someone who is fact-checking a quotation from a musical performance will of necessity have to obtain a copy of that performance somehow, so I don't think reference to the liner notes is a horrible imposition. However, the real intent of this section (which I only added because nobody else with more subject-matter expertise stepped up to the plate) is to make sure that song lyrics are not erroneously credited to the performer.  121a0012 04:20, 31 July 2007 (UTC)


 * The Usenet information implies that it may be used for sourcing quotes, even though Usenet alone cannot be used to confirm the identity of the author of a post. The current warning is extremely unlikely to dissuade anyone from citing Usenet posts as sole sources. ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)
 * By intention: I was attempting to combat the prejudice seen among many WP and WQ editors, you included, against properly used Usenet sources. One can infer the in-real-life identity of a Usenet poster (assuming for the moment that it is even relevant) in exactly the same way as one evaluates a blog or Web site (both of which are effectively anonymous).  Just to give some more motivation for this: it might at some point be worth having a people page for the English blogger "Bystander", a notable (but anonymous) magistrate in London.  He is scrupulous in maintaining his anonymity, to the point of requiring radio interviewers to disguise his voice and attending awards banquets incognito, so there can be no reliable "real life" identity to associate to.  Similarly with Usenet, we may consider two groups of people: those who are notable for their Usenet presence (e.g., James Nicoll), for whom Usenet articles are the authoritative source, and then there are those who are notable outside of Usenet where an editor should have the same burden linking the fellow who posts as "Charles Stross" in rec.arts.sf.written to the writer Charles Stross as they do in linking the author of Charlie's Diary to that same person.  121a0012 04:20, 31 July 2007 (UTC)

Despite these concerns, I think the overall structure and content is a great beginning for this sorely needed policy, and I thank 121a0012 for putting so much effort into it. I hope we can get some community support to address my and any other concerns, make revisions, and ratify the policy in the foreseeable future. ~ Jeff Q (talk) 05:16, 30 July 2007 (UTC)

Brief, specific source lines with detailed footnotes
The following is a modified copy of a post I just made to Talk:Sam Harris, with some generalizations about author citation. I'm posting it in a hurry right now, making no attempt to integrate it or have it reflect the current WQ:SOURCE or WQ:CITE drafts, or the information above, because I think it's a valuable practice that can solve some problems, and if I wait until I have a chance to flesh it out properly, it'll probably be another year in the works. &#9786; I invite discussion, but please do not intersperse comments in my post. (Just quote a relevant line in a separate post if necessary.) ~ Jeff Q (talk) 08:24, 17 December 2007 (UTC)

Specific source information, including page numbers, is extremely important, as this makes it reasonably possible for our readers and editors (who are essential our editorial board) to verify quotes whenever they wish. (There is no such thing on Wikiquote as a confirmed "verification", so material must be reasonably verifiable at all times.) For the foreseeable future, Wikiquote's standard source formatting uses a sub-bulleted source line, not Wikipedia's footnoting system, to make each source explicit and tied immediately to each quote. On the other hand, it is certainly wasteful to repeat detailed source information on each line when the only thing that changes is the page number.

I've seen (and used) two related methods have been used to reduce the repetition while retaining the immediate and specific source line. Both use the footnoting system. First, when just a few quotes come from a single work, one can provide a brief description with the page number, then add a footnote for the detailed description. (I'm showing only the source lines and footnotes for clarity and brevity here.)
 * Sam Harris, The End of Faith (2004), page 651
 * 1

For ease of sorting and verification, the source line needs the author, title, publication year, and page number. (The author may be omitted when the article subject is clearly the author, but be careful. In a source line, "author" means the person who wrote the cited work, who is not necessarily the quotee.) The full citation goes into the footnote.

With this method, subsequent quotes could arguably be as simple as this:
 * Sam Harris, End of Faith, page 1231

This isn't completely safe, as there is no guarantee that someone won't add a quote with the simplified format but pull the quote from a different edition with different pagination. But it's an arguable compromise, and is subject to verification (and fixing) at any time, just like any fully sourced quote.

Notice the novel use of the "pages" parameter of the cite book template ("336 pages" instead of "page X"). Wikipedia's use always demands the page of the citation. However, since Wikiquote needs to have individual pages cited just below the quotes, but can benefit from having the common details collected in a single footnote at the bottom, I've "misused" the pages parameter to cite the total page count. In this way, anyone wishing to verify a quote but having a different edition can note the length of their copy and interpolate the approximate page number of the quote they're looking for. Remember that the entire reason for this source information is to make it easy for people to verify quotes.

The second method is for citations of quotes that are already collected by publication. In a section titled "The End of Faith (2004)", one need only cite the page number:
 * Page 651
 * 1 Harris, Sam (2004). The End of Faith: Religion, Terror, and the Future of Reason, 336 pages, W.W. Norton & Company. ISBN 0739453793.

These organization-by-work sections, to the best of my knowledge, are always within articles that imply the quotee as author, either as the explicit subject of the article (like Sam Harris) or as the author of a series (like J. R. R. Tolkien for The Lord of the Rings). This makes for a very brief, non-distracting source line.

In summary, I would recommend, in those circumstances where the editors are not happy with the excessive repetition of full source information for each quote, that they use this combination of source line and footnotes to move much of the distracting details to the bottom while retaining the essential verification information next to each quote. ~ Jeff Q (talk) 08:24, 17 December 2007 (UTC)

Adopt as policy
This proposed policy has not been amended significantly since December 2010. I move that it should now become an established policy.--Collingwood (talk) 11:53, 7 June 2012 (UTC)
 * Support -Tryst (talk) 13:55, 9 June 2012 (UTC)
 * Support. This has long been de facto policy. Although it could benefit from some copyediting where it rambles a bit, and the lede paragraph seems more appropriate for an essay than a policy, these issues should not override the need to make citing sources a de jure requirement. ~ Ningauble (talk) 14:39, 9 June 2012 (UTC)
 * Comment: this page still needs work to better document current practice for non-written sources. I sprinkled comments here and there inviting editors who know something about these sources to say something appropriate, but nobody has ever done so.  It should probably be split into two separate pages as a policy: one (the original intent of this page) to talk about what is an appropriate source, and the other explaining how to cite such a source.  I originally wrote it because I was not happy with Citing sources, which was (in 2006) a direct import from Wikipedia and not adapted to the more specialized needs of Wikiquote; as a result, editors were often citing WP policy documents as justification for some edit when a WQ-appropriate policy would have counseled a different result.  As those were the formative days of WQ, I felt it was important to draw that fundamental distinction before bad habits were set in stone.  121a0012 (talk) 04:59, 15 June 2012 (UTC)
 * And I see that most of Citing sources is still Wikipedia-specific gunge. 121a0012 (talk) 05:02, 15 June 2012 (UTC)
 * Support; a nicely outlined policy including a useful list of preferences. BD2412 T 17:33, 17 June 2012 (UTC)