Template talk:Citation needed/Archive 2

A possible means of categorisation
I notice the old category got deleted, leaving pages with this tag on currently unsorted. One solution I've been toying with is since the old category was deleted largely because it was growing too large, perhaps it might be an idea to create sub-categories according to subject area, so people with kowledge and expertise in that area can look into referencing the articles. The way this could be achieved is by adding the following code to the template:

What this would do is create an optional cat= parameter that people could use, so would add the relevant page to Category:Physics articles needing references. The #if function means that currently existing tags, and any tags added without the cat= paramater, remain unsorted while the #ifexists paramater means that only categories that already exist would be populated, so the categories would have to be created beforehand (I added the paramter in to guard against categories being created because of typos). Ideally the relevant categories would be created and monitored by appropriate WikiProjects, and the list of possible values ("physics", "football" etc) and the categories they relate to could be kept and updated here. The same sort of code could also be added to other similar templates, of course, if it's seen as being useful. Anyway, I thought I'd put it up here first so I can see whether there's a consensus that it would be useful, and see if there's any potential problems I haven't thought of (I've tested the code and it works fine but you never know). -- Daduzi  talk  21:03, 4 August 2006 (UTC)


 * Argh, so I missed the category link at the bottom of the template, not sure how I managed that. Still, the above could still be inserted as an optional, supplementary means of categorisation. Does anyone have any strong feelings (or feelings of any description) either way? -- Daduzi  talk  12:55, 8 August 2006 (UTC)


 * I think the best way to create topic-specific lists would be through the PNA mechanism, which uses a bot to keep updated per-topic lists of articles with various problems. See e.g. Pages needing attention/Genetics. I will add this to my todo list for the bot. -- Beland 18:35, 30 November 2006 (UTC)

Category:Articles with unsourced statements
Could someone please explain to me what we gain having a category with 21,008 pages that miss citations? --Ligulem 23:41, 20 August 2006 (UTC)


 * Please see Administrators%27_noticeboard/Incidents. Dragons flight 00:26, 21 August 2006 (UTC)


 * Thanks for pointing this out. I was not aware of this until now. Amazing what you can miss if you don't go looking. Carcharoth 23:54, 21 August 2006 (UTC)


 * Incident in question has been archived. New location: Administrators%27_noticeboard/IncidentArchive130

This template needs editing
This template has a misfeature that, if it is next before an end-of-line, makes it swallow all immediately following whitespace, thus making the next paragraph amalgamate with the previous paragraph. Anthony Appleyard 22:14, 27 September 2006 (UTC)


 * From what I have seen, this happens only when the next paragraph starts with a link. I have been getting around it by replacing the newlines in those situations with BR tags.  I put the BR just after the template call and then begin a newline normally, but with only one newline character. Will (Talk - contribs) 04:41, 4 November 2006 (UTC)


 * This has happened to me in two articles today. An admin with good knowledge of Wikicode should check this template's code out and try to fix the problem. – Lantoka  (talk) 00:40, 1 February 2007 (UTC)

Shorten
I think it can and should be shortend to [need cite]. - RoyBoy 800 03:13, 28 September 2006 (UTC)
 * Nah, too informal. It would have to be [needs citation] at least. Which is 1 letter shorter. --tjstrf 04:40, 28 September 2006 (UTC)

how about [citation?] Zunaid 09:56, 5 October 2006 (UTC)
 * why shorten it? I say make it blink, in an 18pt font. This is our last line of defence against the ever-looming "Wikipedia is unreliable" verdict, so make sure readers see it and understand it. dab (&#5839;) 17:27, 24 October 2006 (UTC)
 * Wow, I had no idea that this template which has stood for at least several months is being edited. BTW, also works. Here's what it looks like: citation needed or . AstroHurricane001 23:50, 12 November 2006 (UTC)

Template code needs an edit
Old:

New:

Why: 1) The HTML is invalid; "title" does not apply to "sup". Does not work in all browsers in its present invalid state (Safari, for one).  2) This tag can be applied to other things than "text", such as a chart or graph in image form; thus "material" instead of "text". 3) 'title' is useless if it simply reiterates the link text without explicating it. 4) Easier to read/understand this template if category stuff is put at the end. 5) Using  to prevent linebreaking is klugey; do it with CSS instead. 6) The category code should be includeonly'd.

Cf. all of the other templates in this series; Fact is the only one that isn't consistent. —The preceding unsigned comment was added by SMcCandlish (talk • contribs).
 * I can confirm what SMcCandlish is saying about the  code in the   tag not working universally.  03:37, 11 December 2006 (UTC)


 * Edit made. Being a Safari man myself made that particularly easy to verify. ;-) EVula // talk // &#9775;  // 04:01, 11 December 2006 (UTC)


 * Title is a core attribute, allowed in almost any element, including sup in HTML 4.01 and XHTML. It is possible that it doesn't work on all browsers, because I understand MSIE 6 has poor support for title attributes which are not on a elements. —Michael Z. 2006-12-11 17:22 Z 


 * My bad; should have verified when someone else told me that it was an invalidity in the code. At any rate, the original code definitely did not work in Safari (which probably means it also won't work in Konqueror; same core codebase.  The new version is valid code, so no biggie. :-)  &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 22:24, 11 December 2006 (UTC)

No superscript
The superscript stuff makes lines in a paragraph oddly spaced. Its really bad style, and should be fixed. 24.254.87.52 03:22, 12 December 2006 (UTC)
 * You'd have to take that up at a much higher level. Loads of things on Wikipedia use  , including all of the inline templates of this sort, footnotes, reference citations, etc.  The   features exist in HTML for a reason (and no it's not "bad style"; it's quite common, and has been since long before the Web, for styling annotations of precisely these sorts.  Sorry that your browser doesn't provide enough default line spacing to make   fit in seamlessly (mine, Safari, doesn't either), but that's a browser issue (or, in truth, it could be a Wikipedia stylesheet issue, I think.)  Maybe bring it up at the Village Pump?  &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 08:28, 12 December 2006 (UTC)


 * Formatting a short phrase, in italics, in brackets, as a superscript is exactly bad style. It hasn't been used for styling anything of precisely this sort, ever.  There is no professional typographic precedent for this, none at all, and for good reason.  It makes running text look like shit at first glance, and distracts the reader's eye to make it frustrating to read.  The blink "feature" exists in HTML too, but painting a serious article with it using the 40-gallon drum and the wide roller isn't the reason it exists in HTML either. —Michael Z. 2006-12-12 10:36 Z 


 * Mzajac, do you have some alternative in mind? I'm inclined think the superscript code is perfect myself. 10:49, 12 December 2006 (UTC)


 * Re: &gt;Formatting a short phrase, in italics, in brackets, as a superscript is exactly bad style.
 * Actually, superscripted annotations have been done for centuries, since long before printing even existed (i.e. in the manuscript age). And please watch your language per WP:COOL, WP:CIVIL, WP:AGF, WP:NPA, WP:CONSENSUS, etc.  NB: Actually,   does not exist in HTML, other that as a "Deprecated" reference, for many years now.  Let's not get silly.  Lastly: Again, If you have a problem with the handling of super- and sub-scipts in Wikipedia, take it up with at Village Pump, don't yell at bystanders on a random template's talk page. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 13:11, 12 December 2006 (UTC)


 * As an editorial annotation, this should be simply put on the line, at body-text size, in square brackets. The brackets clearly indicate its nature, and the blue link text makes it stand apart from the text.


 * Blink exists, as a deprecated reference. But that distracts you from my point, then how about &lt;strong>?  It exists in all versions of HTML, but peppering an article's text with dozens of extra phrases in bold font, which are not part of the text, will make it less readable.  Because it exists is no reason to misuse it.


 * Manuscript annotations were in the margins and between the lines of carefully spaced letters, not floating above whitespace like a bloody lip over a gap-toothed smile. But that is irrelevant, since this encyclopedia is not a manuscript. There is no precedent for the style of this template in competent typography since Johannes Gutenberg.  None.  It is a ham-fisted example of poor design. —Michael Z. 2006-12-12 20:55 Z 
 * Well if ham figured in my diet I'd be searching for some sandwich bread right about now. 21:04, 12 December 2006 (UTC)
 * I find the superscripting nicely non-obtrusive compared with the alternatives. Putting it on level with the main text is potentially confusing to readers. As for what print encyclopedias do, Wikipedia is not paper so why limit ourselves to the constraints and style of the print medium? Traditional print encyclopedias don't use navboxes or include external link sections or accept reader submissions either. --tjstrf talk 21:23, 12 December 2006 (UTC)
 * Zzzzzzzzzzzzz... &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 21:29, 12 December 2006 (UTC)


 * The main goal of a typographer setting running text is to make reading smooth and natural by avoiding blobs of dark type and gaps of white-space; it's called preserving the colour of the page. In good-quality type, for example, text figures (e.g. 12345 if you have Georgia font installed) are used to avoid the darker rectangle of titling figures (e.g. 12345, in Times).  Likewise, superscripts are only used for short reference numbers, or asterisks and daggers, to pull them out of the flow of the text, but without leaving a large white gap.


 * This template goes against all of this. It draws the eye from all over the page, and makes the page spotty and pock-marked if used several times.  And it's a poor substitute for using the talk page.


 * Superscripts are not some new element of the web medium—what's not being paper have to do with this? That's a strawman argument for abandoning the principals of good typography. —Michael Z. 2006-12-12 21:45 Z 
 * A long-delayed post-script: I do have the relevant fonts installed, and I have to say that I find your Georgia example hideous, annoying, pretentious and so old-fashioned as to border on antique. I'm sorry that your design education has led you to believe that such weird mis-sizing and horizontal mis-spacing of numerals is objectively "good" design, but I assure you that not everyone agrees with this in fact very subjective viewpoint.  I would certainly use the typesetting you illustrate if I were attempting to immitate the appearance of a church flyer from 1903, however.  &gt;;-)  Or to be more plain about it, people who grew up in the modern era are generally entirely used to characters aside from g, p, q, etc., all being on the same line, and find it very jarring to see 3 treated as if it were a g.  Which (to get back to the topic of this thread) also means we don't find super- or sub-scripts annoying - they serve precisely the purpose they are intended to, namely they set themselves off from the main text and say "I'm something different".  If Wikipedia were intentionally done in the Georgia font, I do understand how your view could make sense, mind you.  I don't think you're crazy, I just think you're arguing a point that is not relevant (to Wikipedia, and online media for that matter, as a whole, not just to this template.) &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 05:45, 16 January 2007 (UTC)
 * You are bordering on trolling. Please stop.  For the third time, this is not the forum for the issue you are raising.  If you have a problem with Wikipedia usage of super- and/or subscript, bring it up at the Village Pump. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 21:54, 12 December 2006 (UTC)

Ugly
This template uses the "sup" tag, which makes its effects on the page extremely ugly. Perhaps it should simply be set to "small" or something. --Tony Sidaway 22:57, 19 March 2007 (UTC)
 * I prefer the sup tag in that it makes the notice rather difficult to miss. If the tag is "ugly" then that is a good thing as unsourced passages are rather "ugly" in a textural sense.  23:02, 19 March 2007 (UTC)
 * Absolute agree! Other than I don't think it is ugly at all but signifies "this is an annotation of some kind; if you are not interested in annotation right now, ignore and move on". I can read entire articles without ever once noticing one of these superscripts if I put myself in the frame of mind to ignore them (like when I'm genuinely reading Clouded leopard because I saw it on the news and want to know more about the animal; when I'm not being an encylcopedia user but an editor I pay quite a lot of attention to the superscripts. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 00:01, 20 March 2007 (UTC)
 * All inline cleanup/dispute templates, as well as other annotation templates such a footnotes and ref citations use  .  This is a rehashed topic (see above), and I'm marking it "Resolved" here because this is emphatically not the venue for this discussion.  If someone wants to do away with this use of superscripting, that is a far-ranging Wikipedia-wide issue with numerous ramifications that must be brought up the Village Pump, and cannot possibly be resolved on the talk page of some random template that happens to be following the already-established convention. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 00:01, 20 March 2007 (UTC)

Extraneous space at end
this edit introduced a space between the and tags. Since the space is not inside any div or span set within this template, it doesn't appear to be necessary, and it has undesireable consequences on at least one page. I'm asking that the space between the and tags be removed. Gimmetrow 03:39, 28 December 2006 (UTC)
 * Done. Kirill Lokshin 06:33, 29 December 2006 (UTC)
 * See . This change broke something.  If the change was made to other similar templates, will need reversion there, too. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 09:48, 18 February 2007 (UTC)

Formatting
Please include the brackets, at least the first bracket, in the mask of the link if at all possible. As it is, they not only show up black instead of blue, they can display with a line break between the first bracket and the rest of the text, thus: [ citation needed]

On my computer, this happens on this version of Stephen A. Douglas. Septentrionalis PMAnderson 04:59, 11 January 2007 (UTC)
 * Is this related to the issue this edit tried to address, but was self-reverted shortly after? Gimmetrow 05:32, 11 January 2007 (UTC)


 * WRT those two edits just mentioned, the self-revert was done at my behest, after someone reverted an identical edit by me to a related template, and explained why. It turns out that the nowrap must be done with a span contained by the sup, rather than with the sup itself, or the intended effect fails in at least one major browser (let's see if anyone can guess who it's made by), despite the more efficient intended code being valid CSS/XHTML.  It's just one of those things...
 * Also, if the formatting is changed in this template, is should probably be made in all relatedly-formatted inline templates, for consistency purposes. Last month I did a bunch of work (though incomplete as it turns out - they weren't all listed together on the same Wikipedia:-namespace template documentation page, alas) to consistency-fix these templates and would not like to see that work undone.
 * PS: If it is true that not including the square brackets inside this or that &#91;I'm not sure what Gimmetrow means by "mask" in this context&#93; does in fact consistently produce weird wrapping, and that including them fixes this problem, I'm all for the fix, provided it is done to every template in this class, and the change doesn't have other negative consequences.
 * &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 05:32, 16 January 2007 (UTC)
 * What action if any was taken with regard to this topic's edit suggestion? &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 12:20, 4 February 2007 (UTC)

Addition [of date] to template
Is there any way to format this template so that the date it was added is included. I've run across this template tons of times and had no idea how long it has been there. I know I could go through the history and find when it was added but that can prove difficult, especially if it was added with no edit summary. This could also allow us to find old templates quickly and remove the unsourced statement. I don't know how practical this is but it seems to make sense.~ Joe  Jklin  ( T C )  06:34, 23 January 2007 (UTC)
 * Well, you could probably have an invisible tag, one that shows up in the actual tag itself, but not in the article (ie: ). I know some bots go around and date Unreferenced tags and the like, so it wouldn't be too hard to do, I don't think... EVula // talk // &#9775;  // 06:39, 23 January 2007 (UTC)
 * Thats what I was thinking. If something like this could be added automatically and keep it invisible, I think that would be the easiest but I don't know if that is possible. You could even do that using AWB but since there are around 48,000 pages that use this template I don't know if that would be the best option (although people with serious editcountitis would love that). It seems bot should probably do it.~  Joe  Jklin  ( T C )  09:05, 24 January 2007 (UTC)
 * It could be added to the link title (that's the yellow tag which appears if you hover over the link, it currently says: The material in the vicinity of this tag needs references to reliable sources. ).~ trialsanderrors 01:35, 25 January 2007 (UTC)
 * That seems a little obtrusive, but no big objection. I'd personally rather see it in the source code only, instead. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 05:19, 28 January 2007 (UTC)
 * I'd rather have it in the tag so it's easy to check if the tag is old enough that the claim may be removed. The text itself could be shortened to This claim needs references to reliable sources (tagged 2007-02-02) . ~ trialsanderrors 08:41, 2 February 2007 (UTC)
 * Works for me! &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 10:36, 2 February 2007 (UTC)
 * I can be bold and change to "Source?" and shorten the yellow tag, but adding the date to the yellow tag is beyond my pay grade. Anyone with more expertise? ~ trialsanderrors 21:58, 2 February 2007 (UTC)
 * Sounds like a good start. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 00:45, 3 February 2007 (UTC)
 * Done. let's wait for the firestorm... ~ trialsanderrors 00:56, 3 February 2007 (UTC)

(unindent) New version looks OK to me. The date idea could be useful too if someone can work out how to make it work automatically. --Scott Davis Talk 02:56, 3 February 2007 (UTC)


 * Well, I guess I'll see if Smackbot (aka Rich Farmbrough) would be willing to add this to his current jobs. I will also try and find someone who could add it to the code so that it also goes to categories by date, like Template:wikify.~ Joe  Jklin  ( T C )  16:24, 5 February 2007 (UTC)
 * I've actually started dating tags as SB comes across them. Since the date parameter is undefined in the template it has no effect yet. I can start on the bulk of them and set up the template and category structure, provided people don't mind a "lumpy" start - i.e. lot gets put into Feb 2007. Otherwise I can do the same but only date as I come across them.  Getting the date right is not mega hard, but I don't have resource to do it now.  Rich Farmbrough, 16:44 5  February 2007 (GMT).
 * Re: "lumpy" &mdash; no objection here - we have to start somewhere. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 19:25, 5 February 2007 (UTC)
 * I've put the logic in the template and created the first month cats, I'll run the beginning of the collation sequence so that the subcats show on the main cat page. Rich Farmbrough, 17:24 5  February 2007 (GMT).
 * This is now included in SmackBot's standard run, but will take some time to catch up. Rich Farmbrough, 23:12 5 February 2007 (GMT).
 * I must say, this is a fantastic idea and the fact that it's now automated is great. W3stfa11/Talk to me 02:38, 9 February 2007 (UTC)

(undent) Respectfully, I must disagree. This is a horrendous idea, and I am concerned that there is no indication that anyone thinks this needs to be discussed on the pump (or similar venue) to gain input from the wider community before proceeding. KillerChihuahua?!? 13:05, 16 February 2007 (UTC)
 * Why do you think it's a bad idea? Garion96 (talk) 13:15, 16 February 2007 (UTC)
 * I started typing a long explanation of Why It Is a Bad Idea, but lets go simpler: why would anyone think it is a good idea? What "problem" does it solve or address? We can all agree it adds verbosity to the tag, yes? So there needs to be a very good reason for doing this. KillerChihuahua?!? 13:44, 16 February 2007 (UTC)
 * I think adding the date is quite helpful as that way we know how long the the disputed sentence has been uncited and it can be removed if it isn't cited in a reasonable time. And especially as it doesn't show up in the final article I have hardly any concerns about verbosity.--snowolfD4( talk / @ ) 16:48, 18 February 2007 (UTC)

Syntax of date addition
I don't see any place in the main Template page or at the top of the Talk page which specifies what the syntax is for the date. Is there any value name, like "date ="? Do we specify "YYYY-MM-DD", or "MMMM D, YYYY"? Can we include or remove commas at our pleasure? Given that a robot is now busily populating these fields on "fact" tags all over Wikipedia, it would be nice to document what the proper syntax is. I just passed up a chance to manually apply a date to a "fact" tag I was putting in an article because I didn't know the right syntax. --Jdlh | Talk 20:43, 17 February 2007 (UTC)
 * Hi, the syntax is "|date=Monthname YYYY", however I spend a bit of time clearing up mis-applied dates, mis-spelled months, etc. (and since these are often manual fixes, I know I can make these mistakes too ), so I am quite happy for people to leave it to the bot. Rich Farmbrough, 00:28 24  February 2007 (GMT).

Tooltip
I have an update to the tooltip for including the date, with an implementation at User:Sigma_7/Sandbox1. Any comments or objections to this update?


 * Comment: At the same time as we update the tooltip, could we also fix the wrapping problem? I have an implementation at User:Kevinkor2/Fact. If you need to test these changes, look at User:Kevinkor2/FactTest. --Kevinkor2 23:13, 26 March 2007 (UTC)
 * Regarding the wrapping - does it make any difference if the span is opened and closed as part of the link cover text, or whether the span includes the base link as well? Gimmetrow 23:22, 26 March 2007 (UTC)
 * Thank you for noticing that, Gimmetrow. When the span includes the base link, the span's tooltip is overwritten by the base link tooltip. In my case, all the links were getting "Wikipedia:Citing sources" as their tooltip. --Kevinkor2 23:42, 26 March 2007 (UTC)

I wonder if it would be worth asking the developers for some kind of built-in user-editable javascript pop-up functionality? — Omegatron 23:28, 26 March 2007 (UTC)

editprotected done, to version at [User:Kevinkor2/Fact]]. CMummert · talk 18:46, 27 March 2007 (UTC)

Line break issue
In line breaks, the tag currently gets split into two lines either between the open bracket and the citation or between the needed and the closing bracket. The brackets could be moved into the no-wrap span or the formatting could be changed to citation needed. Opinions? ~ trialsanderrors 01:39, 25 January 2007 (UTC)
 * Agreed (with regard to moving, not removing, the brackets), but should be done to every template in this "class" (similarly-formatted inline templates). We need to keep them as consistent as possible. At some point, we should probably simply replace all of them with a meta-template, so that they can all be maintained at once. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 05:21, 28 January 2007 (UTC)

Less obtrusive?
On the deletion debate, someone suggested that it should be replaced with a simple superscript question-mark to make it less obtrusive. I'm not sure this isn't an avenue worth pursuing. —Random832 20070125T022608UTC(01/24 21:26EST)
 * Ick! I don't have a problem with a) making it shorter, and b) using a question mark, but only a question mark is a bit much (or a bit too little, rather).  I could probably get behind something like "source?".  Several of these similar inline templates could probably be improved this way.  I think that the one about naming specific sources as opposed to weasel-words things like "some scientists" (I forget the name of the template) was recently shortened in precisely this way (or there is ongoing and largely favorable discussion on its talk page to do so). &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 05:24, 28 January 2007 (UTC)
 * Why not &#91;?&#93; , to match &lt;ref> references? —Random832 2007-01-28 06:07 UTC (01/28 01:07 EST)
 * Because it doesn't stick out as something that needs citation. the point of this tag is to draw attention to the fact that the piece of information is questioned and lacking citation. It is meant to be obtrusive.--Crossmr 06:30, 28 January 2007 (UTC)
 * Also, little question mark links like that are already used for something; I just saw one next to a Japanese word; didn't follow it, but I'm guessing it is to an audio file of the pronunciation. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 10:16, 28 January 2007 (UTC)
 * I'm in favor of &#91;source?&#93; . ~ trialsanderrors 08:45, 2 February 2007 (UTC)
 * Works for me! &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 10:36, 2 February 2007 (UTC)
 * That isn't too bad, though I wonder if its even necessary since WP:V seems to have been rewritten to give me the impression that any unsourced info should just be nuked until a source has been provided.--Crossmr 18:22, 2 February 2007 (UTC)
 * In that case, WP:V needs a revert; guidelines should not be prescriptive as to practice, but descriptive, and deleting all non-sourced material on sight is not in fact actual consensus practice on WP (except as regards material under the purview of WP:BLP and even that assumption is debatable, Jimbo statements on the matter notwithstanding.) As Fact is in extremely broad actual usage, I think the point kind of settles itself. :-) &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 19:15, 2 February 2007 (UTC)
 * That would be a disaster, on non-material claims we can assume good faith and allow the editor some time to provide the necessary verification. I'm all in favor of pushing for good sourcing, but we should avoid sourcing paranoia. The hierarchy established on this page works fine and is practicable. ~ trialsanderrors 21:53, 2 February 2007 (UTC)


 * When a new reader(not an editor) comes to Wikipedia, the citation needed tag explain exactly what the issue is to them, I think it is good as is. HighInBC(Need help? Ask me) 23:45, 12 April 2007 (UTC)

Add class attribute
Could someone add a class attribute to the span tag? Something like class="fact" or class="citation needed", so that it can be customized or highlighted.— miketm - Queen WikiProject - 14:54, 27 January 2007 (UTC)
 * Agreed, but should be done to every template in this "class" (similarly-formatted inline templates). We need to keep them as consistent as possible. That is, for any Template:Foo of this sort, it should have a  in it, in the same position (note I say "template-foo", not "foo"; any given "foo" might already be in use as a class.  Better yet:  ; the 2nd class would allow all of them to be styled the same, while the first would allow them to be styled independently.  Or perhaps the "inline-template" class should be applied at a higher level.  But anyway, you get the idea. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 05:27, 28 January 2007 (UTC)
 * Like,   and so on. I think that would work best.— miketm - Queen WikiProject - 20:05, 28 January 2007 (UTC)
 * Works for me. It strikes me that we should probably set up a WikiProject to identify and maintain all of these like-formatted templates (or, rather almost like-formatted - they keep diverging, and in fact replace as much of them as possible with single meta-template. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 08:13, 1 February 2007 (UTC)

Source vs. citation needed
Source looks more like it is directed at the person who orinigaly added the text rather than people in generaly. It is also rather agressive. Citation needed looks more like the request is universal and there is much less room for missunderstanding as to what it means.Geni 03:49, 3 February 2007 (UTC)
 * I agree with Geni. 1ne 03:50, 3 February 2007 (UTC)
 * Disagree on both points. The first one makes no sense to me whatsoever; Wikipedia articles do not individually credit their editors, never have, never will, and even most brand-new editors understand that within moments.  "Agressive"?  It's simply not excessively long-winded.  "Unsourced!" would be "agressive".  These tags are dispute tags.  They are not supposed to be mollycoddling; they are to indicate that an article has a serious problem and that it needs to be fixed.  A possible compromise might be "cite source".  Cf. Who, or, Dubious (which needs its parenthetical moved into a tool tip, but is otherwise in the same vein).  They are all very concise, and they don't say "please" or "wouldn't it be nice if" or "in my personal opinion".  If that's "aggressive" then so be it.  :-)  &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 06:49, 3 February 2007 (UTC)
 * I don't follow Geni's argument. Should we consider this a quasi-page move request, since this template is really about nothing but this/these word/s? ~ trialsanderrors 09:32, 3 February 2007 (UTC)
 * Wouldn't worry about it for now, until after the dust settles; then the page can be moved to where it needs to live at, and we can juggle the redirs. Already made one from Template:Sourceme (Template:Source is already used by something else) to Template:Fact, and imagine this relationship would be reversed pretty soon.&mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 12:05, 4 February 2007 (UTC)
 * Req. editprotected revert of last edit by Geni, on the basis that it is supported by 2 (Geni, 1ne), and opposed by 3 (Tacitly: SMcCandlish, Trialsanderrors; implicity: Scott Davis &mdash; pre-Geni ver. "looks OK to me", above), so no consensus to revert, and the pre-Geni version was not opposed by anyone when it was discussed prior to implementation. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 12:05, 4 February 2007 (UTC)
 * And there's also no consensus to have it say 'sources?'. 1ne 16:01, 4 February 2007 (UTC)
 * Irrelevant. There was consensus at the time that edit was made, to the extent that people following this template even cared enough about the issue to comment.  After-the-fact disagreement by latecomers to the discussion shouldn't result in a revert that 50%+ of interested edit here thus far are disputing.  If you want to propose an alternative to the "sources?" text, by all means do so here (as proponents of "sources?" did), but the good-faith edit should be restored until there's agreement that whatever you propose (something new or a revert to "citation needed", whatever) is the way to go. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 16:44, 4 February 2007 (UTC)
 * where did you advertise the debate?Geni 02:11, 5 February 2007 (UTC)
 * There's no need to advertise trivial debates, or ones that are active. If people cared about this template they'd be paying attention to it.  And it's not like this is a big sudden change; the talk pages of many of the templates in this class have been, for quite some time, consensusizing overall on making them more concise and (with the tooltips) at the same time more informative, meanwhile the principal and quite common complaint is that they are too large/long and interruptive/distracting.  Just read this page for examples of those arguments, and see especially the TfD archive linked to in the tag at the top of this page.  This isnt' a sudden coup, it's actually been a rather slow-moving, deliberative process; it's a debate that's been self-advertising for a rather long time.  All that said, I'm not even insisting on the source? wording being kept, just restored and re-debated to the extent anyone has issues, and I even suggested compromise wording, above, which, quite notably, hasn't been responded to at all.  Anyway, my Editprotected isn't about my preferred language in the template, it's about process.  Your reversion wasn't discussed, and did not reflect consensus-at-that-time.  I'm not arguing that "source?" has consensus now (though it did when implemented), however it is clearly closer to consensus than the reversion to "citation needed". You've walked into a party, all ready to roll, but so late that everyone else is already getting their coats on to going home, if you get my analogy.  :-)  &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 04:21, 5 February 2007 (UTC)
 * Guys, for such a widely used template, I think it'll be best if we can get a wider consensus before changing it. As far as I can see just a few users have discussed it on this page and maybe wider community involvement should be sought before changing it.
 * Personally I agree with Geni that when you say "Source?" its like asking the guy it added the text where he got it from. "Citation needed" is kind of too long, but usually its only added to one or 2 sentences. If more than that need citations the sources tag should be added to the whole article/section. --snowolfD4( <b style="color:#339966; font-family:Constantia;">talk</b> / <b style="color:#CC0099; font-family:Constantia;">@</b> ) 06:24, 5 February 2007 (UTC)
 * Keep it at "Citation needed". "Source?" looks too unprofessional. And that's beside the point that having to use this template is unprofessional, but it simply is necessary. Garion96 (talk) 13:15, 5 February 2007 (UTC)
 * I also don't like "Source?". "Citation needed" is okay, but indeed rather long. An shorter alternative that works for me is "unsourced", or "unreferenced" (to take Geni's concern into account) or "uncited" (though I have my doubt that the last one is proper English). -- Jitse Niesen (talk) 02:11, 7 February 2007 (UTC)
 * Hmm... "unsourced" would work for me, but there's already a template called that, and I fear it would thus be confusing. "Unreferenced" also conveys the message but is still a bit.  Making it a new subtopic. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 06:24, 7 February 2007 (UTC)
 * (deindent) Editprotected change not carried out at this time - there's no clear consensus to do so. Proto ::  ►  18:26, 5 February 2007 (UTC)
 * Fine, really. I was making a point about process, and no one seems to get it, so I'll just drop the issue. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 19:26, 5 February 2007 (UTC)
 * Fine, really. I was making a point about process, and no one seems to get it, so I'll just drop the issue. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 19:26, 5 February 2007 (UTC)

What about "unreferenced"?
OK, so Jitse proposes "unreferenced" or "unsourced". I really like "unsourced" better, but it would be confusing and would almost certainly lead to mistemplating, because of the existence of Unsourced already. Meanwhile "unreferenced" gets the point across, but in a pretty long word. I could live with it, given than it's shorter than "citation needed", and I can't right now think of anything better (other than "source?"; heh.) Hmm... "add cite" or "cite fact", or "cite source"? I really don't understand what the objection to "source?" is - seems perfect to me. But on the assumption that it's going to meet resistance, I'm trying to help find non-Quixotic alternatives... &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 06:24, 7 February 2007 (UTC)
 * "Need cite" or "need ref" are probably the shortest we can get. Rich Farmbrough, 10:09 8 February 2007 (GMT).
 * At this point, I'd be happy(-ish) with "cite source", "needs source", "add souce", "needs cite", in descending order. —The preceding unsigned comment was added by SMcCandlish (talk • contribs) 13:42, 8 February 2007 (UTC).
 * "cite source" still looks like it is directed at the intial editor. Personaly I fail to see anything wrong with citation needed.Geni 17:34, 10 February 2007 (UTC)
 * I still don't understand this "directed at the initial editor" bit. Nothing anywhere on Wikipedia, template or otherwise, that appears inside an article's text is "directed at the initial editor".  The entire concept simply does not happen here.  As for what's wrong with "citation needed", it is overly long and interruptive.  See elsewhere on this talk page and on the talk pages of various other templates in this class.  There is a general putsch to make them all shorter and concise, and move any verbiage that they need to hover-over tooltips. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 23:07, 11 February 2007 (UTC)
 * Let me try to explain the problem that Geni has with "source" (or, at least, the problem I think that Geni has). The "source" of a statement can be taken to refer to whatever prompted the original contributor to write the statement. If I read in book X that there are 3 million people in Melbourne, and I add this to Melbourne, then book X is the source. If book Y also has this information. then book Y is not strictly speaking the source for the statement that I added.
 * You might think that's nitpicking, but I tend to agree that there is a possibility to misunderstand "source". This is only a small possibility, but on the other hand, "source" is only two letters shorter than "citation". So the question becomes: is saving two letters more important than running a small risk that the text is misunderstood? (I've no opinion either way, just trying to explain the issue).
 * By the way, I think "citation" is also open to misinterpretation. I thought that the main meaning of citation is quotation (i.e., reproducing a piece of text word by word), and I'd use "reference" for what's meant here. But of course, "reference" also has multiple meanings (and I'm not a native speaker). -- Jitse Niesen (talk) 02:40, 12 February 2007 (UTC)
 * I as well see nothing wrong with Citation needed, and it should be directed at the initial editor, or any editor attempting to ad the material, WP:V states under burden of evidence the burden lies with them to provide the source if they want it kept in the article, no one else is under any obligation to hunt it down for them, so really the message should in a way read like its directed at them.--Crossmr 18:34, 10 February 2007 (UTC)
 * How about "no ref"? --Random832(tc) 06:43, 11 February 2007 (UTC)
 * It is very short, but using abbreviations is probably an iffy idea; not all users here are native English speakers, and even natives aren't necessarily going to intuit that "ref" means "reference" rather than "referee" or some other "ref" word. Was why I proposed "cite source" (or of course "source?"). &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 23:07, 11 February 2007 (UTC)


 * I think that we should go with something less passive. "Citation needed" sounds like we're asking someone to help improve the article by editing. "Unreferenced" sounds like we're warning the reader that we don't know where this information came from.  We should go with the one that encourages participation in improving the article.  Jkelly 23:30, 11 February 2007 (UTC)


 * I think one of the ideas of the tag is to ask whoever inserted the text or whoever is reading it to cite the source if possible. "Unreferenced" doesn't exactly say that. Citation needed expresses that well and also tells anyone who is reading the article that the text is not cited and not entirely reliable (maybe hovering over should say something like "this text is not cited and may not be reliable" and clicking it too should make that clear instead of simply redirecting to WP:CITE. Specially since someone who doesn't know what it means may well go ahead and click it and get confused by WP:CITE.
 * It's not ideal, and I agree its too long, but I think its the best option we have cos I don't think any of the above proposed variations make the two objectives of this template clear (i.e.
 * (i) Ask someone to cite it
 * (ii) Tell anyone who is reading that it may not be reliable --<b style="color:#9696A0; font-family:Tahoma;">snowolf</b><b style="color:#0A0096; font-family:tahoma;">D4</b>( <b style="color:#339966; font-family:Constantia;">talk</b> / <b style="color:#CC0099; font-family:Constantia;">@</b> ) 03:24, 12 February 2007 (UTC)
 * Thats the point right there. We're not a primary source and too many people treat wikipedia as it is. If we have information we can't verify, it needs to be tagged in a visible way to indicate to readers that that information hasn't been verified. Shortening it to be "less intrusive" is a waste of time and opposite its purpose.--Crossmr 14:11, 12 February 2007 (UTC)
 * How about "Please cite a suitable references which meets Wikipedia's criteria as a reliable source of this information"? Oh wait, it's not April yet. Rich Farmbrough, 16:02 14  February 2007 (GMT).

Addition to template (ctd..)
See. Some editors think the date is unhelpful. I have invited them to discuss here, rather than just talking with me about it. Rich Farmbrough, 09:49 16 February 2007 (GMT).

vs. vs.
I don't understand the difference between items 1 and 2 on the template page, namely:
 * 1. if it is likely true, but needs specificity, you may use
 * 2. if it is not doubtful, you may use or  tag to ask for better citation in order to make the article complete.

I'd appreciate a clarification: when should i use, when should i use and when ? Itayb 15:20, 16 February 2007 (UTC)
 * Specify is used if the statement you've made is a bit too general. The tag is also outdated. Basically its used to prevent people from using weasel words. i.e. "Many people believe this".
 * fact is used when there is doubt about information, but usually not if the information is just ridiculous. Then you'd remove it. Even if you don't doubt it, if you think people could doubt it, you should tag it.
 * citequote is for when you put a quote in article and forget to include a source.--Crossmr 15:23, 16 February 2007 (UTC)
 * Thanks, Crossmr. Then may i suggest removing item 1, since it's outdated, or replacing it with a mention of the weasel-inline tag? Itayb 15:36, 16 February 2007 (UTC)
 * Italyb there is already a template who which I believe may replace the specify template. It already links to the WEASEL words article so that may be what you mean. --<b style="color:#9696A0; font-family:Tahoma;">snowolf</b><b style="color:#0A0096; font-family:tahoma;">D4</b>( <b style="color:#339966; font-family:Constantia;">talk</b> /  <b style="color:#CC0099; font-family:Constantia;">@</b> ) 16:39, 18 February 2007 (UTC)

Incorrect behavior
Something is wrong with this template. Take sentences that are supposed to be separated by two line breaks, add the template and suddenly the line breaks disappear. This is remedied by adding a single space after the template, but that is incorrect behavior for the template in the first place and many people will not think to try adding a single space. — coe l acan t a lk  — 20:21, 17 February 2007 (UTC)
 * Yeah, up at, someone removed the space. I think they may've done this to other similar templates too. Sounds like it needs to go back in. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 09:47, 18 February 2007 (UTC)
 * I've readded the space. Picaroon 21:39, 18 February 2007 (UTC)


 * The space inside the template didn't change this. In fact, removing the space after the template in the article causes the odd behavior right now, with the space in the template. On the other hand, having the space creates odd spacing in normal text. Gimmetrow 16:05, 12 March 2007 (UTC)


 * I don't understand the behavior. When I remove the space in the original article, the paragraph breaks disappear. However, when I remove the space in my personal copy, the correct behavior results. --Kevinkor2 20:39, 12 March 2007 (UTC)
 * The problem is related to the if statement, which is why the behaviour is restricted to namespace 0. Try copying the article (or the Example below) to a blank page in article space and viewing it in preview. Note that the problem happens with the next line begins with a wikilink; if you remove one of the wikilinks, the fact tag before it works fine without a space. Gimmetrow 21:04, 12 March 2007 (UTC)

Example 1

Line 1.

Line 2.

Line 3.


 * Aha! The code is eating following newlines if followed by a wikilink! In Example 2, the two lines should be separate paragraphs: --Kevinkor2 21:44, 12 March 2007 (UTC)

Example 2

Line 1.

Line 2.


 * Seems the problem is having the categories after the superscript. Putting them before fixes this problem, no spaces required. Gimmetrow 21:36, 12 March 2007 (UTC)

This seems related to : " Don't remove whitespace before ".--Kevinkor2 21:55, 12 March 2007 (UTC)


 * Yes, it seems related. So is there any problem with putting the categories first? A fact tag should, after all, follow a statement without a newline before it, so it shouldn't run into display issues like trailing newlines. Gimmetrow 04:47, 13 March 2007 (UTC)

This is a request to examine moving the includeonly block before the initial sup, and removing any space between the trailing sup tag and the noinclude that will follow. Putting the categories first avoids a bug with categories eating newlines, as described above, because the category is then placed between article text and the superscripted fact text, and there should not be any newlines in that context. The trailing space was supposedly added to fix the newlines disappearing, but it doesn't. It forces a trailing space, which should be under editor control. As far as I can tell, moving the categories neither improves nor hurts the bracket-wrap issue described below. If this edit doesn't cause any other unknown problems, I would ask the same be done to Verify credibility and verify source. Gimmetrow 03:45, 15 March 2007 (UTC)


 * Done. Let's wait and see whether there are any complaints. -- Jitse Niesen (talk) 06:05, 15 March 2007 (UTC)


 * (Note: Putting the categories first was previously done by Omegatron here, and was later changed per this request which doesn't seem to specifically mention having categories first.) After this change I saw a citation needed tag wrap. I'm assuming other people will notice this too, and since it's appearing on pages it didn't before, is going to be a complaint. However, at least on Mozilla, was able to fix it by putting the brackets inside the span tag. I think this edit tried to address this by moving the no-wrap to the sup tag, but that doesn't work. Going through the history, I didn't see a version which tried the following for the superscript:


 * There is probably some popular browser that will still break the span block between the bracket and text. Could someone verify how this version of the span tag works? In any event I agree with SMcCandlish here that all the related templates should be coded similarly. Gimmetrow 23:38, 17 March 2007 (UTC)

The existence of a category
The discussion of whether Category:Articles with unsourced statements, used primarily by fact, ought to exist at all has been restarted as:

Deletion_review/Log/2007_February_20.

Dragons flight 16:45, 20 February 2007 (UTC)

SmackBot Kevinkor2
To: Rich Farmbrough, Kenosis

Cc: Template talk:Fact

From: Kevinkor2

Currently, Rich, your bot is expanding  by adding the current month, resulting in.

Kenosis, whenever you see this on the Truth, Pragmatism, and a few other articles, you revert it. As you noted at User talk:Rich Farmbrough, it would be useful to have fact dating/nondating under control of an article's editors.

I suggest we adopt one of three possible compromises:
 * 1) Manually change  to   for facts where we do not know an accurate date.
 * 2) Add  to the top of the article.
 * 3) Research the page history for the first appearance of the  tag to give it the correct date.

I recommmend the first alternative. --Kevinkor2 17:43, 23 February 2007 (UTC)


 * Hi Kevin, Good set of ideas, but actually all the articles are done now (with a vanishing small set of exceptions, such as those where people have subst'ed the template, or reverts of SmackBot). If anyone wants to move the dates further back, they can of course do so, Kenosis and others claim that it is easy to find the dates from history - I think "easy" is a relative term here.  The names of the dated categories are supposed to reflect that the tags are at least that old (since month boundaries, for example, will never be neat), however the rate of addition of articles suggests that a typical "current month" will be around 20,000 at the end, so that the February block is not as oversized as it appears. Rich Farmbrough, 00:23 24  February 2007 (GMT). 00:23, 24 February 2007 (UTC)
 * I did not make the claim that it is easy to find the older dates using the history, I believe that was another commentator on the issue in the current DRV for Category:Articles with unsourced statements. The difficulty of correcting these errors (that is, correcting the dating of fact-tags within roughly 50,000 articles containing one or more fact-tags as February 2007 irrespective of when they were actually tagged, the vast majority of of which are now incorrectly dated) is part of the reason I've chosen to look into how this situation developed.  Obviously there are many interwoven issues, some of which plainly will need further comment from the larger community as the wiki moves into the future. I believe I already responded to Kevinkor2 on his talk page and Rich's talk page. Good regards everybody (except for the bot, which i'm peeved at at the moment). ... Kenosis 01:03, 24 February 2007 (UTC)
 * Apols for misattributing that. In terms of the vast majority, yes it probably is the majority, but by less than you might think. Something like 700 fact tags are added every day so the February total will already be over 15,000.  And my concern was that the longer it was left the worse it would be, which is why there are 1600 articles in the January category.  The situation is not ideal, but the dates on tags are not "content", they are housekeeping, so not that critical. What is more, as the old tags are removed and new tags dated, both the absolute number and percentage with correct dates will increase. Rich Farmbrough, 10:46 24  February 2007 (GMT).
 * Hold on a second. I have a lot of respect for the detail work Rich Farmbrough has done around the wiki. But there are some genuine issues here.  Firstly, the category is "articles with unsourced statments."  There are some 50,000 such articles involved at present.  I don't at the moment have the method to calculate the total number of fact templates involved in those 50,000 articles which were dated "February 2007", but it appears likely that in excess of 65% that are now incorrectly labeled.  And so the earlier argument was that they needed to be dated and categorized, and now the argument is that it's "not that critical" that they be dated properly? ... Kenosis 21:03, 25 February 2007 (UTC)
 * For what its worth, it sounds great to me. The fact that some of the dates will be wrong for a while is really of no consequence at all.  At least they have a date that is an upward bound - that is we know the latest that they might have been added - and that is more helpful than harmful, by far. &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; Éc 15:36, 24 February 2007 (UTC)
 * So in other words it's now a new wiki-wide policy, starting February 2007, which will improve its accuracy over time. ... Kenosis 21:03, 25 February 2007 (UTC)
 * Exactly, except it's not "policy", it's more of a project. Rich Farmbrough, 21:56 25 February 2007 (GMT).
 * I've yet to see a valid argument for the inclusion of the category or putting dates on fact tags. The category "page" (for want of a better term) is a neverending nightmare list of 200 articles per page.  Just out of curiosity, does anyone here really think people are going to wade through that pile of detritus in order to bring enlightenment to the dim and dark great unsourced masses?  Furthermore, the argument that it is beneficial to show how long a fact tag has existed was poorly thought out from the PR angle.  Six months from now, when someone looks up Heebie-Jeeebie-Tiddly-Poo and a fact tage is dated to February it'll be pretty damned obvious that we are pisspoor at maintaining articles.  &#0149;Jim 62 sch&#0149;  21:21, 25 February 2007 (UTC)
 * <Reset indent> It is already "pretty damned obvious that we are pisspoor at maintaining articles". This is part of an attempt to address that, not to hide it.  You will already see that some tags are extremely old, these, in my opinion need, need attention.  Ideally they are source or removed, but as a reader and editor I would like to know if something has not been sourced, and for how long.  If nothing is done, the "fact" tags will simply accumulate 'til kingdom come.  The more difficult problem is how we prevent unsourced statements that are eventually removed reappearing. Rich Farmbrough, 21:53 25  February 2007 (GMT).


 * Further to Kevin's original proposal, there's no reason people can't change the template to "date=unknown" if it floats their boat. Rich Farmbrough, 21:57 25 February 2007 (GMT).


 * I have to agree with Rich and SMcCandlish. Dating the tags will let us know how long the text has remained uncited and will make it easier to remove text that has remained disputed for too long. So fact tag or fact tag with date? Definitely with date, cos SMcCandlish puts it, it is more helpful than harmful. Not adding the date to simply hide how long the thing has been uncited is totally against the principles of Wikipedia, in my opinion.
 * And in the end, if it comes to that, a few years from now dedicated editors could start going through uncited articles month by month (starting from the earliest) and either help find citations for the text or remove it completely. Huge task, yes, but who'll have thought something like WP:PROXY will have managed to ban so many open proxies from editing and disrupting Wikipedia? --<b style="color:#9696A0; font-family:Tahoma;">snowolf</b><b style="color:#0A0096; font-family:tahoma;">D4</b>( <b style="color:#339966; font-family:Constantia;">talk</b> / <b style="color:#CC0099; font-family:Constantia;">@</b> ) 22:11, 25 February 2007 (UTC)
 * Partially agreed, but substantially disagreed, as follows. These templates don't tell us anything except that someone placed the template. Some indicate that a fact is contested; others indicate that it's simply questioned; others indicate that an editor has made a "note-to-self" about finding a citation; yet others indicate an broader POV dispute, etc. Each depends on the topic, context, how many editors are involved, controversial or non-controversial nature of the article or statement(s), and a host of other factors.  Some don't deserve to stay one minute, and others ought be left alone as long as it takes, even if it's ten years. Overall, the existence of the template tells us very little, if anything at all, about actual fulfillment of WP:VER, what with the million and a half articles with actual unsourced statements, and tens of millions of actual unsourced statements that arguably ought be sourced in writing.
 * What instead has happened is that in essence, with little or no real discussion of the issues involved, a pronouncement has been made: "OK, Wikipedians, here are your new marching orders. Starting February 2007 y'all are going to start keeping track of these "citation-needed" templates. Date-tagging is now mandatory, or at least automated. All "citation-needed" tags that were lodged prior to February 2007 are hereby granted amnesty under our new project to more strictly enforce WP:VER (the ones in those approximately 40,000 articles containing fact templates that were lodged in 2005, 2006 and January 2007, which fell through the cracks of our new project). We don't mind if you fail to put a "citation-needed" for those tens of millions of statements that should ideally be cited. But by golly, if you're going to use that template, we're going to keep track of those dates (starting February 2007 of course)."
 * All this was done with no feedback from the larger community, not from WP:BOT, not from WP:VER, not from anybody in the community really, except for several on this page that regarded automatic fact-tag dating as a good idea. Recall that Category:Articles with unsourced statements, after being deleted in a CfD with an sound consensus, was unilaterally undeleted by a single admin without DRV on the basis that a "related" category had been subject to undeletion in a DRV, and then a new CfD was cut short asserting it to be "obviously" needed (with no explanation why). Then the bot was argued to be needed to clean up the category, and then after incorrectly dating the majority of fact tags on the wiki as February 2007, the new policy of robotically dating fact tags is argued to be OK because in time the correctly dated tags will outweigh the incorrectly dated ones. And it's now argued on the category talk page that the category is needed to implement the automatic tag-dating.  And now it's further being argued that accuracy of the dating is "not that important", because in time the accuracy rate will increase.  ... Kenosis 23:28, 25 February 2007 (UTC)
 * Para 1 "These templates don't tell us anything except that someone placed the template." If that were literally true, they should all be removed. But there is information,  if it needs to be finer grained, change the process "Cite needed now!" "cite would be quite nice here" etc..
 * Para 2 segues from "y'all are going to start keeping track" to "we're going to keep track" - the latter is more accurate - this imposes no "load" on other editors, unless they liked to go through the all-in-one category in alphabetical order. How this "amnesty" clause came into being is a mystery.  Older unsourced cites still require sources just as much (or more) than current cites. Perhaps if the first run had used the category "... since Feb 2007" rather than "...  from Feb 2007" this wouldn't be a problem. Oh, hang on... it did.
 * Para 3 is about process. DragonsFlight brought the undeletion up at ANI where it has a large audience than DRV, and has since DRV'd it as well (there is no URV).  If anyone wants to nominate it for CfD, they can.
 * This really should not be a problem, it is a simple change that imposes no workload on anyone, and if it proves not to be useful, can be backed out in about five minutes.  Rich Farmbrough, 15:25 1  March 2007 (GMT).


 * Rich: I still fail to see a compelling argument as to how the dates will help us manage squat. Do you really believe that adding a date will compel someone to do the research to fix it?  We still have tens if not hundreds of thousands of article in the "stub" categories, some of them quite old.  We have thousands of articles in the "cleanup" category, dating as far back as May 2006.  Of our much-ballyhooed editor base, only about 1,000 have 10K edits and another 1,400 have 5K or more, and of those a number are either inactive, have left, or have been perma-banned.  In addition, most editors mostly edit articles in which they have an interest.  Nope, can't see the dating having any value.
 * Cleanup used to go back as far as at least Aug Jun 2004, I believe. Also as has been pointed out, depending on the nature of the statement, consideration should be given to removing old unsourced ones, or escalating the hunt for sources.  Yes it's a judgement call, but so is any editing. It is also clear as Kenosis says that there is neither a perfect correspondence between stmts that most need citing and those with the tag, nor an ordering of need that will put the tagged items at the top.  However someone has thought it worth putting the tag on, so it is a reasonable assumption that the majority of editors would agree that it is reasonable to request a cite for the majority of items.  Suppose something is tagged "cite needed", at what point do we decide that there is no reasonable chance of a cite being forthcoming?  A difficult question, but one which we cannot even address without some form of dating.  Rich Farmbrough, 15:02 1  March 2007 (GMT).
 * Kenosis: Your points are excellent, although I fear the autarchs who imposed the policy will never get your points. I think I know how Galileo felt.   &#0149;Jim 62 sch&#0149;  22:31, 26 February 2007 (UTC)

Minor formatting error
Breaks when used in table by overlapping with text in an adjacent cell, at least in my browser (Firefox). See Comparison of operating systems.

Thanks, ArmedBlowfish (talk|mail) 21:44, 27 February 2007 (UTC)