User talk:LeadSongDog/Archives/2010/November

Inline citations
About this tag, I think you need to read the text of Template:Inline. The article clearly contains (many) inline citations, but nothing that looks like the "list of references, related reading or external links" that the template is about. WhatamIdoing (talk) 03:39, 7 October 2010 (UTC)
 * You're right, I suppose I should have used . I'll fix it. Thank you. LeadSongDog come howl!  03:48, 7 October 2010 (UTC)


 * I've been working on WPMED's Clean up listing, and that tag was one of my recent targets. As far as I can tell, it's almost never used correctly.
 * Can you tell me what it is that you actually want to see? Is there anything in particular that you thought was unverifi able ?  Is this a request for a greater number of sources?  Or a footnote in every paragraph, just because I can?  Or something else?  WhatamIdoing (talk) 04:54, 7 October 2010 (UTC)
 * Reliance on just two sources is an issue, as is the inclusion of numerous uncited statements. Readers should not have to guess where to find the supporting source for a statement. I've no real doubt they can be cited, it just needs doing. LeadSongDog come howl!  05:34, 7 October 2010 (UTC)


 * I apologize for not getting back to you sooner.
 * This is a serious question: Why is relying on "just" two top-quality sources a problem?
 * The thing is, I tend to agree with you. My gut-level response is that well-written articles name more than two sources.  But is more than two sources actually an essential property of a decent article?  Or is it just a marker for a decent article?
 * I know what our policy-based reason for not relying on one source is: You can't really demonstrate compliance with DUE if you've only used a single source.  (The article might actually be perfectly balanced, but you haven't shown that balance, and showing the balance is the editor's version of defensive medicine.)  But that leaves me wondering -- what's the minimum number (if it's not two)?
 * This question is mostly motivated by my recent work on misunderstanding the GA criteria, combined with the belief that you're a reasonable, analytical editor who might be able to help me identify and articulate the issues. WhatamIdoing (talk) 23:51, 13 October 2010 (UTC)
 * Thank you. That is an interesting question. My personal take is that wp:GACR is rather too vague about this. To establish that an article has "broad" coverage, wp:NPOV, and is wp:V through wp:RS can only be made possible by the application of informed editorial judgement. As editors we should be able not only to choose suitable sources but defend the choice of those sources as sufficient to meet those goals. To be able to show they are "broad" we of course prefer the use of secondary sources, but even then, we need some evidence that these sources actually are "broad" and "neutral" (or at least "balanced") in their treatment of the subject. In a few cases we can refer to high quality sources for which there are reviews in other high quality sources, but this is so rarely possible that it hardly matters. For medical topics we place heavy reliance on recent systematic reviews, such as the Cochrane Database provides, but even these are inadequate in some regards. They hardly ever provide a wp:Worldwide perspective, and they are frequently written in a fashion that borders on being unintelligible to the lay reader, but they are certainly excellent quality sources. Textbooks, particularly evergreen texts, are often useful, but even they will often tend to lag in their coverage of recent developments. By using several diverse high quality sources we avoid most of the traps. Two can't do that, but I doubt there is a general answer to the question of "how many is enough". One might suffice to establish notability or support uncontroversial simple assertions of fact, but for some topics such as abortion dozens are needed. LeadSongDog come howl!  13:22, 14 October 2010 (UTC)


 * Thanks for your thoughtful response.
 * So consider the first source in this article. It's the current (2010) formal standard-of-care-establishing practice guideline.  In terms of source quality, I don't think it gets any better than this.  I think we can tick off "neutral", "high-quality", "secondary", "independent", and "current" with that source alone.  I'm not sure that we really care whether a single source is, itself, "broad" in coverage (that's a goal for the article, measured against every source that exists, not merely those that are cited), but this one appears to cover everything except the (unencyclopedic) human-interest story that a newspaper or magazine article would lead with.
 * In this particular niche, there doesn't seem to be much regional difference: Wealthy countries pretty much do the same things, and poor countries almost don't bother treating cancer, because of resource constraints, and thus don't bother writing about this aspect of its treatment.
 * So if those are the goals, then why isn't this one source enough? (Or these two, since IMO naming two usually prevents inappropriate AFDs and NPOV challenges.)
 * Put another way: If I named a medical textbook at the end of some sentence that currently doesn't have an inline citation, would that actually improve the article's quality?  WhatamIdoing (talk) 00:14, 15 October 2010 (UTC)
 * My response above was not really addressing the specific article, but responding to your later, more general question of 13 October. In the specific case I don't diminish the quality of that source, but a practice guideline is written for practitioners, not for a general wp:audience. Re your last point: yes. By providing that inline cite you prevent subsequent good-faith editors from deleting the otherwise unsupported assertion.

LeadSongDog come howl!  05:26, 15 October 2010 (UTC)


 * I know your comments were directed towards the general case, and the general case is really what's on my mind.
 * I'm thinking about these "subsequent good-faith editors": Does "immunizing" an article against these editors actually make the current article any better?  That is, imagine that you write an article today, based on half a dozen pretty good sources.  Imagine that I read it tomorrow, and that no one else has edited it before I read it.
 * In this universe, you provide inline citations at the end of 50% of the sentences.  In an alternate universe, you produce an absolutely letter-for-letter identical article, except that you name an inline citation at the end of 100% of the sentences (without introducing any new sources).
 * Is the article I'm actually reading in this universe actually worse—from the perspective of the reader, not from the perspective of staving off well-meaning but misguided future editors—than a version containing exactly the same words, in exactly the same order, but with twice as many inline citations? WhatamIdoing (talk) 00:40, 19 October 2010 (UTC)
 * The point really isn't whether the inline citation is at the end of every sentence, but rather whether a reader can consistently tell which cited source backs up a given sentence. In the usual usage, a single citation at the end of a para is taken as meaning that the entire para is supported by that citation. If citations appear within the para then that assumption is invalidated and subtler reading is needed.
 * Thus the better universe would be the one where there are twice as many quality sources cited, not one where there are twice as many citations of the same few sources. LeadSongDog  come howl!  18:15, 19 October 2010 (UTC)


 * I'm not sure that "more separate sources" = "better". Schizophrenia has more than 200, and I seriously consider that to be an indication of poor sourcing (despite its FA status).
 * To give a trivial example, do you think that these sentences have materially different quality?
 * "Motor vehicle accidents may be caused by impaired drivers. They can also be caused by mechanical failures, poor skill, and road conditions.[1]"
 * "Motor vehicle accidents may be caused by impaired drivers. They can also be caused by mechanical failures, poor skill, and road conditions.[1][2]"
 * "Motor vehicle accidents may be caused by impaired drivers.[1] They can also be caused by mechanical failures, poor skill, and road conditions.[2]"
 * If we can safely assume that the material is verifi able (that is, some source, somewhere in the world, thinks that having a wheel fall off a car might cause an MVA, etc.), then I'm not sure that the reader really gets more out of any of these than s/he would get out of a completely uncited pair of sentences. WhatamIdoing (talk) 20:30, 19 October 2010 (UTC)
 * Given that the statements are contestible the number and quality of the sources is relevant. An official position of a national highway safety authority would likely rank as more significant than a rant by the third guy from the left at Moe's Tavern. However, if [1] and [2] each support the statement equally well and are both sources of the best quality there is little extra value in citing both. Those preconditions are not always pertinent. More often the best source is not the one that most literally supports the statement. A lower quality source which says the same thing in a way that is more comprehensibly or more accessibly supportive of the statement can be a good reason to cite both. LeadSongDog come howl!  21:21, 19 October 2010 (UTC)
 * (off-topic):Some writers would, with some justification, opine that impaired drivers don't have accidents, that their collisions are predictable. Pilots and race drivers are routinely instructed that it isn't speed that kills, but lack of control.


 * Overall, it sounds like one of your significant goals for inline citations is to point the reader at sources for independent study/WP:FURTHERREADING, rather than to provide some (as-yet unidentified) direct benefit to the reader while the reader is actually reading the article.
 * Does that sound like a fair description of this purpose? WhatamIdoing (talk) 20:49, 2 November 2010 (UTC)
 * That's more a secondary benefit, as is the fact that subsequent editors can extract additional content from the identified sources. Mainly, though, we cite to meet WP:V, which is a core policy. As pseudonymous editors, we have no independent credibility. The reader can only trust the article if it can be verified in reliable sources. Sometimes we get so caught up in details we forget that. LeadSongDog come howl!  02:26, 3 November 2010 (UTC)


 * WP:V doesn't require citations for my sample sentences (which we are pretending are accurate): It's not a direct quotation, they have not been challenged, and there's no reason to think that they are WP:LIKELY to be challenged.
 * So that can't really be our primary reason for wanting inline citations in (more or less) every paragraph. WhatamIdoing (talk) 04:29, 3 November 2010 (UTC)
 * Even if the samples were accurate, they are POV. The contrary POV can be seen here to support a challenge. Editors are human, and often fail to see that LIKELY doesn't apply to their statements. LeadSongDog come howl!  13:11, 3 November 2010 (UTC)


 * How does naming a source show original editor that other sources disagree with it? It seems to me that the act of naming a source does not help the original editor that he's presenting a POV.  I don't understand how naming a source would tell the editor that the source is LIKELY to be challenged.  (I very much understand how naming a source might confer an unwarranted aura of accuracy on a dubious statement.)
 * The operative word in my question is how: In what manner, by what mechanism, does naming a source tell the editor who is naming it that the source is accurate/unbiased/etc?  WhatamIdoing (talk) 20:54, 4 November 2010 (UTC)
 * Err, it doesn't. It simply says "this is the source on which I (the inserting editor) base this assertion". Then he, or later editors, or reviewers, or readers can read that source and make up their own minds as to its support for the assertion, accuracy, bias, etc. My experience has been that practically anything worth writing can and will be challenged by the time an article reaches FA, short of the classic example "The usual human hand has five digits" (which, perhaps, fails the "worth writing" test). LeadSongDog come howl!  21:21, 4 November 2010 (UTC)

Did you know canola oil can kill you?
You might want to check out User:RiceMilk's edits on canola and Brassica napus‎. --sciencewatcher (talk) 01:47, 5 November 2010 (UTC)

He has also been similarly active on erucic acid - Weetoddid (talk) 22:25, 5 November 2010 (UTC)

Robert Koch "He is the innovator of the uppercut"
http://en.wikipedia.org/w/index.php?title=Robert_Koch&action=history

You reversed my deletion of an obviously nonsensical comment on the scientist Robert Koch, "He is the innovator of the uppercut" which links to an article on boxing and is unrelated to the reference that follows it. Revision mentions "Cydebot using pop ups". Just curious how and why the Cydebot reversed my edit. Thanks Sdward40 (talk) 13:39, 7 November 2010 (UTC)
 * My error. I must have been cross-eyed, thinking your edit was the vandalism, when it was the earlier IP. Cydebot didn't reverse your edit, but was the editor immediately prior to you. When I used pop-ups to revert, it went to Cydebot's version. Again, I apologize for my haste. LeadSongDog come howl!  13:56, 7 November 2010 (UTC)

Rendering h on ‘Kilogram’
I responded to you there. Greg L (talk) 18:44, 10 November 2010 (UTC)

Referenced, or not?
(I'll get back to pestering you about inline citations later; this is a new question.)

Imagine that I have decided to cite some eMedicine publication in an article.

If I list the publication in tags, it's an inline citation. If I supply a bare URL at the end of a sentence, it's an inline citation. If I list the paper at the end of the page, it's a general reference (even if I incorrectly name that section heading "External links").

Right?

And if you encountered any of these in an article that contained the unref tag, you would remove the unref tag (and perhaps replace it with nofootnotes or refimprove, if that seemed appropriate).

So what if the eMedicine article is (only) linked in the infobox? Is the article still minimally "referenced"? Or not? WhatamIdoing (talk) 01:11, 12 November 2010 (UTC)
 * I wouldn't agree entirely. The purpose of citations is twofold: wp:Verifiability and academic honesty (which of course is a tighter requirement than copyright compliance). When you list something under "External links" you are saying that you haven't used it as a reference. The criterion for unref should be "no assertion in the article can be attributed to a cited source". If there is something in the article that the eMedicine link supports, then yes it is (however well or poorly) referenced. But just making the link to that article doesn't mean it is relevant to anything said in the article, though in the usual case it will be. It could in theory say something that is the direct opposite of what the article says. So this is something of a grey area. It comes up fairly often at wp:WikiProject Academic Journals, where linked parameters in Infobox journal are often the only references for stubs, but they are generally accepted as sufficient to establish wp:N of the subject. LeadSongDog come howl!  14:44, 12 November 2010 (UTC)


 * About the first bit: You're right that when I put something in ==External links==, it means that I didn't use it as a reference (even if it's usable as a ref/contains information that the article also contains).  But I'm not the median editor; I'm one of the folks at WP:EL who explains that EL is not RS, just like the EL guideline says repeatedly, sometimes in bold-faced text.
 * In the real-wiki-world, we get people "helpfully" changing the ==References== into ==External links== if it only contains URL-containing general references. It's not difficult to imagine someone listing eMedicine as a general ref in a tiny stub, someone else renaming the section, and a third editor helpfully moving it to the infobox.
 * But more often, I suspect that these really are unref'd articles, in the sense of "none of the people writing sentences in the article actually read the eMedicine source", even if the eMedicine source actually does contain all the information.
 * I think I'm going to err on the side of losing unref tags for stubs that contain a "possible" source in the infobox.
 * (About WPAJ, notability never requires sources to be named in the article. Subjects are notable if the sources exist, anywhere in the world, regardless of whether they are named in the article, or even if the editor who wrote the article believes the sources to exist.  But I'm not convinced that I'd accept the publisher's own website and routine inclusion in enormous databases as an indication of notability.)  WhatamIdoing (talk) 00:18, 15 November 2010 (UTC)
 * One reason I don't like the "External links" header is that it is intrinsically biased in favour of online sources. "Further reading" for the same purpose would be more neutral. But I don't think the MOS is going to change in regard of this matter, because we routinely use that header for "Related articles". (insert sigh here)
 * Regarding WPAJ, there has been prior discussion of the whole notability question, but in practice if the journal publishes notable content then the journal will also be subjected to reviews, letters will be sent to the journal's editor by experts in its discipline, and services like Thomson ISI will rate it. While there are few areas in which I would support automatic notability, this is one that comes very close. Still, there's a difference between a subject being notable and being established as being notable. It takes references to establish wp:notability. Until this is done, it remains subject to challenge. This is the real meaning of "notability is not temporary". One might rephrase it as "once noted, always noted". We've never really bought in to the idea that "wp:notability" is anything more than "noted by a wp:RS". If it hasn't been noted, we usually wait until it has, even though it serves no good purpose to do so. Sometimes though, we sensibly just IAR in the knowledge that, if challenged, we can readily find suitable sources.
 * So far as removing the tags, how tough is it to click on a link and see if the e-medicine article is about the same topic as ours? If our lede isn't in approximate agreement, it shouldn't be linked. If it is, then it's safe to remove the tag. LeadSongDog come howl!  17:48, 15 November 2010 (UTC)


 * Yes, I think that's the right approach with the eMedicine links. I'm doing the same with OMIM links.  More commonly, however, the unref-tagged article has only ICD codes in the infobox, and I think that probably counts as truly being unref'd (the ICD links pretty much contain the number, the name, and nothing else).
 * For WPAJ, I'm not convinced that these articles are notable under the GNG, but I think they often help the encyclopedia (by providing blue links for citations whenever one is wanted) and should normally be included if a paper in the journal has been cited in a Wikipedia article. I have the impression that some of the publishing houses send helpful worker bees around to create stubs.  I wish that we could convince them to add citations to any third-party attention the journals have received, even if it's just in an industry rag.
 * Of course, I'm also hoping that someone will write some papers that compare and contrast the journals within different fields, but that seems less likely. (Hmm... perhaps some poor grad student...)  WhatamIdoing (talk) 20:10, 15 November 2010 (UTC)

The Bugle: Issue LVI, October 2010
To stop receiving this newsletter, please list yourself in the appropriate section here. To assist with preparing the newsletter, please visit the newsroom. BrownBot (talk) 23:18, 21 November 2010 (UTC)

Prolog
Hello, you recently added a tag to Prolog saying "Please improve this article by introducing more precise citations where appropriate." Could you suggest where it might be appropriate? The whole Syntax and Semantics section could be attributed to (LLoyd, 1984), but is it good practice to tag each sentence with the same ref? pgr94 (talk) 10:37, 26 November 2010 (UTC)
 * There's various advice on this. Inline_citation leaves considerable leaway, but I'd suggest the end of each major paragraph, which includes any quotations or potentially controversial assertions. Readers really should not have to guess which source pertains. LeadSongDog come howl!  14:02, 26 November 2010 (UTC)

References on EgyptAir Flight 763
Hi LeadSongDog I noticed you've added a few references to EgyptAir Flight 763 which is currently up for deletion. It's great that you're trying to reference the article, and this might well change the outcome of the AfD, which is always cool. However, I just wanted to mention that the references you've added so far don't do any more than verify the facts of the article. This can be helpful in some circumstances, but when an article is up for deletion as being non-notable, the references need to do more than simply establish the facts (see WP:ENN). They need to establish notability. You're probably aware of all this already but, if not, I hope this advice is helpful.-- K orr u ski Talk 17:35, 26 November 2010 (UTC)
 * That's true, but until someone actually looks for sources, there's no way to tell if they exist. I found all those in a short time, though the article has been needing sources for many months. That tells me nobody has really looked. The Flight International source, for instance tells me that it wasn't an aircraft owned by Egyptair, but by Inex-Adria and that Egyptair's Misrair division was the operator. These details provide additional things to search for. It's a fairly safe bet that a trip to the library will yield more. Given that deletion doesn't build content, I prefer to see that the effort has been made before assuming it can't be done. LeadSongDog come howl!  18:06, 26 November 2010 (UTC)
 * I've made a few suggestions for giving the article a bit fo structure on the talk page. Mjroots (talk) 05:14, 27 November 2010 (UTC)