Wikipedia talk:Manual of Style/Dates and numbers/Archive 46

'Note'': This extended discussion on the linking of dates, during March and April 2006, is 171 kilobytes long. It therefore occupies archives 42 through 46.'''

The second-level header section "Examples of when to link and not link solitary links" continues from archive45.

The third-level header section "P8c" continues from archive45.

P8c (continued)
I see no further objections, so I'm moving the P8c version to the guideline page. --Francis Schonken 10:58, 11 April 2006 (UTC)


 * I'm very sorry to revert you, but I still object and don't feel that this represents the consensus view at all. As I said above, there may be a possibility of a compromise along these lines, but this version is fatally flawed. Did I make a mistake by not writing "object" in bold type?
 * I also note that several of the other proposals above got much more support than this one. By the time this one came along, I think everyone had got bored of the discussion, and tuned out. I'm afraid I don't feel it commands widespread support.
 * Stephen Turner (Talk) 11:21, 11 April 2006 (UTC)
 * I agree with Stephen Turner. There were several other options which received much more "agree" votes amongst the original 8.  I think even if you add all of 8+8a+8b+8c, you will still not have a concensus. Neier 11:39, 11 April 2006 (UTC)

@Neier, please make content remarks, like for instance Stephen Turner did. "add all of 8+8a+8b+8c" is nonsense, P8c was distinctly created to cut out some of the recommendations that had been proposed before, while, for instance, there was no consensus on "it is good practice to link all dates that respond to your readers' auto-formatting preferences", as established above, not only in this section, but for instance also by the Bach cantatas, KV and gnossienne examples.

@Stephen Turner, you said you objected because you "think there is widespread agreement that WP:CONTEXT applies to year links as well as to any other type of link." – I gave no further attention to that remark, because there was enough evidence on this page that, for instance, full dates would easily be linked in excess of "Only make links that are relevant to the context"'s general principles, while there is no other way to trigger date preferences ("trigger date preferences to work" is not the same as "the general principles of "Only make links that are relevant to the context" as far as I understand English). Further, this page sufficiently made clear that there is no consensus on the point of linking separate years; on the point of repeat links for dates in the same article;... Sorry, your statement of what your preferences are regarding when to link dates, only confirmed "There is no consensus whether the general principles of Only make links that are relevant to the context apply for dates." So, please establish "consensus" on this point if you think there is. I fail to see any consensus on the point that there would be "widespread agreement that WP:CONTEXT applies to year links as well as to any other type of link"; so the expression "There is no consensus whether the general principles of Only make links that are relevant to the context apply for dates." is perfectly applicable, and reflects the present "consensus" that there is "no consensus". --Francis Schonken 12:03, 11 April 2006 (UTC)


 * Full dates are irrelevant. Everyone agrees those should always be linked even if WP:CONTEXT doesn't apply.
 * As for the year links, the disagreement is not about whether year links are somehow "special" so that WP:CONTEXT doesn't apply to them. The disagreement is about whether most year links are relevant to the context (and so should be linked), or whether most year links are not relevant to the context (and so shouldn't be linked).
 * Stephen Turner (Talk) 12:26, 11 April 2006 (UTC)


 * No, re. year links the disagreement *also* regards year links being "somehow special so that WP:CONTEXT doesn't apply to them": this was argued above by, among others, Jdforrester/James F. ("[...] an overall consensus and generally-held concept that dates and year references generally would always be linked"), as being the original-and-still-effective consensus. First I hadn't understood that either (even made a remark in that sense, addressed at a comment by SlimVirgin), but no, effectively some still defend to link (almost) every year, whether relevant to context or not, as a general exception to WP:CONTEXT, so there's no consensus to say the contrary. Your statement:
 * "there is widespread agreement that WP:CONTEXT applies to year links as well as to any other type of link"
 * and Jdf's:
 * "I distinctly remember discussion leading to consensus some 3 years ago or so that the don't-link-to-the-same-article-twice is waived for years, as part of an overall consensus and generally-held concept that dates and year references generally would always be linked."
 * are so diametrically opposed, while none of the two are views held by an extreme minority that could easily be convinced in its opposite, that I'd see no way for the time being to formulate as "consensus" what obviously is not. Of course, you're free to try to convince your fellow-wikipedians of your view (and several other wikipedians would support you). But as long as a consensus in that sense is not established, it is better to have a statement about the "consensus to disagree" in the guideline.
 * Please appreciate I'd have preferred a less open-ended consensus too. But currently there is none in that sense. I'd not agree to your particular preferences either, so if any finite "consensus" type of formulation would end up in the guideline, it would be the consensus established several years ago (but I don't think that a good one either, but anyway better than your proposal). --Francis Schonken 13:24, 11 April 2006 (UTC)
 * If it's a less open-ended consensus that you're after, P1, P3, P4, and P5 all seemed to draw positive comments before everyone got tired of the discussion as Stephen suggested. If lack of a protest is sufficient for modifying the MoS, then all four of them seem to be worthy as the only one who made a negative vote towards any of them was me (P5 and P3).  I also notice that they all four have the same structure which reads cleaner than the list you made.  And, all four of them address the impetus for this discussion in the first place; which was whether or not editors should be able to go through the database and change date links on the basis of not being a full date or not.  I think that's what your third bullet point is trying to get at; but, it does not come across as clear as the other four proposals I mentioned. Neier 14:13, 11 April 2006 (UTC)
 * P1, P3, P4, and P5 were drafted - and commented upon - before (1) we discovered there's no consensus on linking all full dates, before (2) we discovered there's no consensus to go by general WP:CONTEXT standards, and before (3) we discovered dates on dab pages are different matter, about which consensus was established elsewhere.
 * Re. non-consensus "(1)" you can turn a blind eye (as you apparently try to), but that will still not make a contrary "consensus" emerge.
 * I objected to P1, P3, P4, and P5, explaining why above. So, no consensus.
 * Concensus does not have to be unanimous. Otherwise, I doubt that there would ever be a concensus to do anything on WP.  But, concensus does need for everyone to understand what it is that they are supporting or opposing — even if the support/opposition is passive (see below). Neier 23:16, 11 April 2006 (UTC)
 * Don't know what you're trying to say here... I only see you trying to put yourself outside the consensus on P8c. Even Stephen Turner, though in appearance opposing the text, implicitly confirms the "non-consensus" option by proclaiming yet another approach: no, there's no "consensus" on the application of WP:CONTEXT for dates and Stephen Turner only confirmed that with yet another stance; even Singkong2005's comments appear not to oppose P8c, as Singkong2005 appears only to comment negatively on the proposals that "seem to be implying that linking dates is good for allowing date preferences to work"... I'm quite confident that if Quiditty and Donald, who earlier supported P1/P3/P4/P5 type of solutions, now support P8c that that means there's a broad enough consensus on this proposal... which, as far as I can assess, also includes Talrias (who approved an earlier "P8" version, don't see why Talrias wouldn't agree to P8c - there's nothing fundamentally new, just a correct representation of the breadth of the consensus); James F.; and all others (apart from Neier) that supported earlier P1, P3, P4 or P5 solutions. Of course, you're free to try to prove me wrong on my current assessment that P8c has broad enough consensus. --Francis Schonken 08:34, 12 April 2006 (UTC)
 * "before everyone got tired of the discussion", nonsense. Only the "voters" (those that weren't interested in the discusion in the first place) didn't return.
 * Actually, I didn't return because it was never made clear that P1-P7 were apparently being trashed (despite the voting) so we could focus on modifying P8. The manipulations of P8 seemed quaint, but since I was never in favor of it in the first place I didn't see how my commenting on it would be beneficial, or a productive use of my time. Neier 23:16, 11 April 2006 (UTC)
 * Well, you did return to the discussion, didn't you? All what's still missing is content arguments: I still see none - sorry, you only rehash what has been discussed and answered to long ago above.
 * Re. "The manipulations of P8 seemed quaint", I don't know what you mean by that. Probably better that I don't, while on first sight it indicates you having trouble with WP:AGF... Anyway, trying to fight something that apparently has a relatively broad acceptance in the community, might not be your most "productive use of time"... don't say I didn't warn you... but of course, you're free to do with your time what you like. But don't try to shift the responsibility of what you do with your time to someone else. I tried to stop the apparent disruption/wheelwarring/etc resulting from (semi-)robotised delinking of dates. Thus afaik my efforts in this sense have been worth it. I'm not complaining. --Francis Schonken 08:34, 12 April 2006 (UTC)
 * "lack of a protest is sufficient for modifying the MoS", nonsense – but note that Bobblewik better never had tried to apply that as a rule, which still is the first cause of all this trouble. Confirmations of the validity of the P8c version include those by Quiditty, Ambi, Donald and myself.
 * Maybe I misinterpreted I see no further objections, so I'm moving the P8c version to the guideline page. --Francis Schonken 10:58, 11 April 2006 (UTC)
 * Maybe I should have added that I saw broad enough consensus over my latest proposal, even by opponents of my prior proposals: maybe this would have made clearer what I meant. Also Donald's "I'll go back to my editing" was quite convincing to me; maybe time to put P8c in the guideline, and let it rest for some time while going back to editing, and I think "editing" in this context has some other meaning than (semi-)robotised "transformations" steering to bring articles in line with a style about which there is no consensus. --Francis Schonken 08:34, 12 April 2006 (UTC)
 * Just as clarification of my stance, if any is needed, while I would prefer a guideline that strongly discourages solitary year links, and encourages their removal, I believe that there is no consensus on the issue, and unlikely to be one any time soon. So I think the best we can say now is, there is no consensus, so behave yourselves and don't try to make a point. --  Donald Albury ( Talk )  11:00, 12 April 2006 (UTC)
 * "P1, P3, P4, and P5 structure reads cleaner", nonsense. As I explained before they're more verbose and contain contradictions.
 * Please point out a contradiction within any of P1, P3, P4, P5. Of course, there are contradictions between them – and that's why I presume we were holding the straw poll.Neier 23:16, 11 April 2006 (UTC)
 * I did, didn't I - didn't I already give you the link to above; here's another link to part of the discussion above:, where others (e.g. Cyde Weys: "What sense is it to say, We think X is good, but you can only do X in a roundabout way, and if you do X with the sole intention of doing X you will get in trouble for it.") spotted the contradiction in P1.
 * Well, and you clearly didn't follow the "impetus" of the discussion, both because you say it lacks contributors, and because you apparently missed non-consensus "(1)" and "(2)".
 * Further, no, it's primarily the initial statement of P8c ("There is no consensus whether the general principles of Only make links that are relevant to the context apply for dates.") that adresses going through the database and change date links on the basis of not being a full date or not. If there's no general consensus, you can't do that in a (semi-)robotized operation. But, e.g., you can do that for secondary date links on dab pages, per the second bullet of P8c. --Francis Schonken 15:00, 11 April 2006 (UTC)
 * MoS:DAB already has a big bold don't wikilink any other words in the line. So, I think that delinking dates on those pages doesn't need explicit permission here anyway. Neier 23:16, 11 April 2006 (UTC)
 * ??? P8c does not give "explicit permission to delinking dates on dab pages"... What are you trying to say? P8c only avoids possible contradiction with MoS:DAB, by referring to it... --Francis Schonken 08:34, 12 April 2006 (UTC)

Re Jdforrester's comments that Francis refers to above. I didn't say that there's unanimity that WP:CONTEXT applies to date links. I still maintain that there's a very clear consensus in favour of this, and that the disagreement is over what that means in practice. P8c seems one-sided to me: it says that there are good times to link years, but it doesn't say that there are ever bad times to link years, except in one specialist situation. If we're going to change the guideline, I would like to see a bit more symmetry. Stephen Turner (Talk) 11:19, 12 April 2006 (UTC)
 * I will follow Francis's advice, and move to commenting on the content only. I think that P8c adds nothing to the original wording, and as Stephen said above, it detracts strength from the "if it is not a full date, don't link it".  Most of the other proposals left that intact, while also giving various degrees of free reign to go through and correct the damage which is currently done on most articles with standalone year, month, day links.  P8c does a good job at describing what to do for new articles, but so did the old wording.  What has been at issue (afaict) is how to deal with correcting what most of us see as a problem (too many year links which aren't full dates), while not causing a new problem (removing "relevant" year links that can add historical context, etc).  This is at the very top of the page, with Bobblewik's first proposal.  I think that this issue should be solved without expanding the debate to cover full dates, because it will just cloud the issue (as it seems it already has). Neier 12:55, 12 April 2006 (UTC)
 * I voted when P6 was on the table. i did not comment on P8 and I 'm not sure I like it that much.  Neir makes good points here. It seems to me that this whole process has been good to narrow the options down but to say that P8 is favoured among all those that voted is not true. i had assumed this was an exploritory discussion and had been monitoring it when i had time.  i had not realised that this would be binding for the next three years. If we really want to hammer this down, comments need to be made by everyone on the finalised versions.  Preferably no more than three options that represent the spectrum of opinion. David D. (Talk) 18:02, 12 April 2006 (UTC)

P8d
This is a proposal that retakes an earlier suggestion by me, and is primarily intended as a response to Stephens request to give the formulation more balance. It's a bit longer – while generally I prefer guideline recommendations to be short I think this one is pretty much OK for me, and it doesn't need to give additional descriptions in WP:CONTEXT (that was my prior suggestion):


 * P8d - Francis' new proposal (more balance according to Stephen Turner's suggestion) : The general principle of Only make links that are relevant to the context doesn't always apply for dates. These are some recommendations for when to link dates:
 * it is good practice to link dates when relevant to context;
 * generally, it is good practice to link all dates that respond to your readers' auto-formatting preferences, but not necessarily in:
 * lists, e.g. List of Bach cantatas by liturgical function;
 * elaborate tables, e.g. Köchel-Verzeichnis;
 * other pages that contain many dates and where linking every full date may become encumbering, e.g. Gnossienne.
 * it is good practice not to link auxiliary dates on disambiguation pages, per Manual of Style (disambiguation pages).
 * it is good practice that editors look for consensus on over/underlinking issues on a by article basis.

Maybe this proposal doesn't cover all the concerns by Neier and David yet, but please, feel free to comment.

I'm a bit puzzled by David's "binding for the next three years."? Bobblewik's "problematic" formulation has been in the guideline for over a year now, without *any* prior discussion, so anyway I'll remove that formulation if we're going to live by principles that guidelines "can't be changed for the next three years" (who ever said that?), and if "everyone" has to comment before a guideline can be updated to current standards (Bobblewik is still under a sort of informal probation-supported-by-Jimbo that he gets blocked when bringing the current guideline formulation in practice). So better get rid of that problematic and inadequate guideline formulation immediately, put P8d instead, and if anyone has something better, with a broader consensus, change it the next day. Please stop that "next three years" nonsense, it leads to WP:LAME situations and inability to cope with issues when they present themselves. No, guidelines are not *frozen* for three years. They should adjust to current situations (like e.g. the situations created by AWB's raising popularity, which also may decline in the next months, who knows; or software updates, for example changes to the "user preferences" mechanism; or whatever evolution in wikipedia that is directly or indirectly relevant to this issue). --Francis Schonken 09:27, 13 April 2006 (UTC)


 * I think this formulation confuses things by getting into the issue of full dates again. Full dates should have their own section, and the issue of whether it's appropriate to link bare years and months should be discussed separately.


 * I do admire Francis's persistence in trying to solve this, and it's given me the impetus to try and come up with my own proposal. I've made no secret of my own views on this subject, but I have some ideas which I hope will be balanced enough to be agreeable to everyone. It will involve quite a lot of rewriting and may take a day or two to formulate it fully, so I hope everyone can wait that long. And maybe it will get shot down as soon as it appears, but I'd like to try.


 * Stephen Turner (Talk) 12:03, 13 April 2006 (UTC)


 * My three years comment is reference to James Field who said that the consensus for the last three years has been to link all dates. This, despite the fact the MoS had been edited by bobblewick and many people were fine with his addition. To imply that the MoS can be easily changed "change it the next day" is not correct since there are a significant number of people here saying that Bobblewicks changes were invalid. MoS rules from three years ago were invoked.  I assume this is because not everyone agreed with bobblewick. Given this rhetoric, it IS important that everyone has a say so that there is a significant consensus. Throwing in wording after everyone has voted subverts the system and will create more confusion (and probably arguments) in the long term. Again the votes seemed like a fact finding, consensus building exercise. It was a starting point not an end point. David D. (Talk) 16:39, 13 April 2006 (UTC)


 * Sorry, my proposal is not ready yet. I'd hoped to have it done by now, but I haven't had time. I haven't forgotten, and I'll get round to it as soon as possible. Stephen Turner (Talk) 18:01, 16 April 2006 (UTC)


 * I've managed to finish this at last. Apologies for the delay. I've started a new section for the discussion below. Stephen Turner (Talk) 20:08, 20 April 2006 (UTC)

Linking Dates
This is probably a very stupid question, but I can't make head or tails of this guide line: some people delinked years I linked. Okay ... but why? What is the policy: "generally do not link". But when do I link at all? For instance, from Social Democratic Workers' Party (Netherlands). "The party was founded in Zwolle in 1894." Do I link this year or shouldn't I and why would I and why not? Thanks! C mon 21:44, 30 March 2006 (UTC)


 * Link it if you feel it is useful. Ambi 21:45, 30 March 2006 (UTC)


 * I would put it this way: if someone followed the link to 1894, would it help them understand the article about the Social Democratic Workers' Party (Netherlands)? Stephen Turner (Talk) 09:44, 31 March 2006 (UTC)


 * Make a judgement yourself. Is the link useful? Ambi 12:50, 31 March 2006 (UTC)


 * Thanks for the answers, I'll rely on my own intuition then, but I hope that a manual of style will in the future provide some guidelines for this. C mon 4:22, 31 March 2006 (UTC)


 * Yes, do make the judgement yours, but please balance (1) the disadvantage of making the text slightly harder to read, the possibility of diverting the reader to an unfocused body of information, and the risk of diluting the high-value links in the text, with (2) the advantage of providing a one-click link to a useful, focused article that will help the reader to understand the orginal article. Often, piped links are in this category. Chronological links to ancient times may be useful. Tony 10:30, 3 April 2006 (UTC)
 * I agree with Tony here. Especially the comment "the possibility of diverting the reader to an unfocused body of information, and the risk of diluting the high-value links in the text".  The piping of dates to relevent information should be encouraged.  For me, the 1894 collection of information is next to useless. Since the article is related to politics a much more useful link would be to pipe the date to  1890s or 1890s. I believe that if users discover that the date links are actually piping them to relevent information then users will be much more likely to click on them in the future.  At present the information gained from the date pages is so minimal that we are training people not to use the links. We need to make these links useful, that is the priority. David D. (Talk) 16:35, 3 April 2006 (UTC)
 * Personally the only thing I find unusual is that if you go to 1894 - why are there ANY red link articles? I would have thought that if it is worth mentioning that something happened in a certain year (on the page for the year) then it should be worthy of it's own full article. Garrie 06:12, 5 April 2006 (UTC)

See Jack Tatum, currently an FAC, for excellent use of piped year links. If less focussed, please don't link. Tony 07:51, 5 April 2006 (UTC)
 * Completely agree. and Piped links might just be the perfect solution to much of this. --Quiddity 08:45, 5 April 2006 (UTC)
 * This is exactly what would make me use date links. David D. (Talk) 09:01, 5 April 2006 (UTC)
 * Yeah, but often not available. Use 'em when appropriate, sure, without adopting a blanket practice of bluing years, decades and centuries for no particular reason. Tony 09:06, 5 April 2006 (UTC)
 * Agreed. David D. (Talk) 09:10, 5 April 2006 (UTC)
 * "For me, the 1894 collection of information is next to useless." ... IMHO not a problem of the Social Democratic Workers' Party article. Whether or not the 1894 article is a "next to useless collection of information" should not affect the writing of the Social Democratic Workers' Party article. If the 1894 article does not live up to standards, then that article should be improved... this should, IMHO, be unrelated to whether or not to link 1894 in the Social Democratic Workers' Party article. Also, in Wikipedia one doesn't know what will be in the 1894 article tomorrow. If today it fails to give a good overview of social developments in that year, then I hope it will give such overview soon. Not linking 1894 in the Social Democratic Workers' Party article does not help to overcome the flaws of the 1894 article. Neither does probably the linking of that year... linking the year in one article and the quality of the linked article are IMHO unrelated. But, anyhow, the 1894 article should picture some useful social context. And the person writing the Social Democratic Workers' Party article should not be hampered if it doesn't do that yet.
 * Further, the piped link theory for years is of limited importance IMHO. If I think it useful to make a link to the 1893 article in the gnossienne article because these compositions were published for the first time in the same year and the same city as the first version of The Scream was painted, I think 1893 as a link does the job perfectly, while there is the navigation that allows to explore the 1893 year. Neither of the alternative "year" articles is really appropriate or useful in the gnossienne article IMHO:
 * 1893 in art, mentions The Scream, but has no "music" topics (so why should it be linked from gnossienne?) - neither do I want to push the idea that for this composition developments in other art forms were as important as those in music in 1893 Paris;
 * 1893 in music, although mentioning the premiere of Debussy's string quartet this article does not really show anything specific on fin de siècle Paris.
 * 1890s mentions some fin de siècle topics... but not what other artists were doing in Paris at the time. And also: not all Gnossiennes were composed in the 1890s (some were composed in the late 1880s...), so not too appropriate as a link from the gnossienne article.
 * 1893 in Paris... does not exist yet... but if it would ever exist it would be linked from 1893, so 1893 as a non-piped link does perfectly fine for me in the gnossienne article.
 * Also, don't be a nanny, wikipedia readers don't need the steering apparently preferred by some (which is also a POV, if you want to convey to a reader what that reader should appreciate as "relevant" or not about a topic). I mean, what was said above: "the possibility of diverting the reader to an unfocused body of information, and the risk of diluting the high-value links in the text"... It is no judgement call to decide for others what dilutes and what doesn't. For example, people that want to know more about the time when the gnossiennes were composed should not need three clicks to go to the article of 1893 in literature. Links play a role in easy navigation, which is somewhat opposed to navigation by someone's POV preferences.
 * So, I'm not a proponent of the universal application of piped links for years (although this may be useful in some instances!) - the risks of steering for a POV structure of knowledge should be kept in mind too IMHO. I'd rather say: let facts speak for themselves as much as possible. --Francis Schonken 09:40, 8 April 2006 (UTC)
 * Have you considered that when every fact is included in the year links then to navigate and find the relevant information will be practically impossible? The piped links help focus in the most relevant area. Sure the most relevant area could be POV but once a user finds its not what they want is is quite easy to continue the search for better context. Certainoly better than throwing the user into a sea of facts. David D. (Talk) 18:08, 12 April 2006 (UTC)