Wikipedia talk:Manual of Style/Dates and numbers/Archive 118

Leading zeros in geographical coordinates
Are leading zeros used in geographical coordinates? I.e. should I write 9.13528°N, 4.085°W or 9.13528°N, 4.085°W? (Personally I prefer the latter, but I will follow the policy, of course.) --Apalsola t • c 19:37, 5 January 2009 (UTC)
 * I've only seen this done for minutes and seconds of time, never angle. Ideally the coords template should be programmed to strip or add zeros depending on consensus. Other templates do this, e.g. . — CharlotteWebb 20:00, 5 January 2009 (UTC)
 * There doesn't necessarily have to be consensus. If there is not, the coord template should act just as it does, leaving the choice to the editor using the template.  Gene Nygaard (talk) 06:27, 6 January 2009 (UTC)

Unifying date formats within articles
per MOSNUM

Format consistency

 * ''Dates in article body text should all have the same format.
 * ''Dates in article references should all have the same format.

''These requirements apply to dates in general prose and reference citations, but not to dates in quotations or titles.

Yesterday, I created dmy and mdy in the hopes that it could make future automated maintenance easier. There are currently AWB and monobook scripts written by my friend Lightmouse for unifying (not just de-DA) dates to a preferred format. My idea is to incorporate these tags into the relevant scripts, so that we can potentially insert these tags to the top of any given article while doing semi-automated conversions. Tags would indicate that work has been done to convert all dates (exc ISO) to a given format. Down the line, a bot can be programmed to maintain all those articles which have been so tagged, so that new edits can be made to conform to the same format.

The tags can remain invisible, as at present, or they can be made to insert markers of some description like the star at the top of each featured article. Because they are transcluded templates, categories, such as Category:Articles with American date formats or Category:Articles with International date formats could be created, if needed. However, I dare say a bot can just as easily spot the tags within the articles and go to work on them. Obviously, articles which have 'special' date formats would either escape tagging, or have another tag, for example notdate to prevent undesirable bot action. Once again note that although the tag is designed be placed on articles during or after the semi-automated process of stripping dates of the DA tags and unifying date formats (US or International), it can be put on articles by editors who wish for a bot to pass in order to accelerate transformation of an article with numerous inconsistently formatted dates to the desired format. The removal of the tag would only serve to avoid bot action on converting dates to the format so designated by the tag. As of yesterday, I have already started inserting it - see here.

For the avoidance of doubt, this discussion is not about the act of stripping dates of their baggage (sic), but about the unification of date formats. Constructive comments thereon would be welcome. Ohconfucius (talk) 02:18, 9 January 2009 (UTC)


 * Good idea. We see similar tags or notices for articles that are in a particular variety of English. It helps everyone and prevents arguments. --Pete (talk) 02:46, 9 January 2009 (UTC)


 * We already have British English and the like, but these are intended for talk pages, and they are rarely useful anyway. But it would be useful to create templates with a similar purposes for use on the articles themselves, which would produce no visible output, and, if possible, which would trigger editnotice warnings (e.g the one you get when you edit an article containing disambig). This way, if I were to write colour in an article, I wouldn't need to scan it to find any instance of a word spelt differently in different dialects in order to decide whether to include that u. -- Army1987 – Deeds, not words. 17:28, 9 January 2009 (UTC)
 * I have now moved things on. Potentially, this is a job for User:Lightbot, so there is now a discussion on the specifics of how to implement such a templating system for Lightbot on User talk:Lightmouse. Ohconfucius (talk) 06:07, 12 January 2009 (UTC)

Date linking template
As we are discussing the use of templates to tell bots what to do with dates; how about a template which would say "This link is meant to stay."? What I have in mind is a template which would take a day and month and/or a year and provide an unautoformatted link for the "rare" cases where we might want them. You'd input three parameters: day, month and year or month, day and year; two parameters: day and month or month and day; or one parameter: a plain year. The thing would add correct punctuation & nonbreaking spaces and link them up (with pipelinks to disable autoformatting which, for consistency, might be desireable). I'm just mentioning it since it could be done. Of course, you must wonder whether there are enough justified date links to warrant writing such a template ... I'm not convinced that there are. J IM ptalk·cont 12:11, 9 January 2009 (UTC)


 * Another editor insisted a year link of his worth keeping, and I showed him he could make it better without having it. He agreed it was better. I'm prepared to take on that challenge for that 1 in 10,000 date links. In the meantime, let the bots get on with the job of sweeping up the detritus. Ohconfucius (talk) 14:26, 9 January 2009 (UTC)
 * Ohconfucius, if you really believe that Hex was agreeing it was better rather than holding back to avoid an edit war, you need a reality check.Dejvid (talk) 00:52, 10 January 2009 (UTC)
 * I'm afraid Dejvid is on the mark here. — Hex    (❝  ?!  ❞)   12:59, 10 January 2009 (UTC)
 * Well, unaccustomed as I am to doubletalk, for future reference, I suggest that future postings be put in plain English. Anyway, my offer still stands. Ohconfucius (talk) 07:13, 11 January 2009 (UTC)
 * Oconfucius's offer to help is ineffective because most affected editors will not know about the offer. --Gerry Ashton (talk) 15:48, 9 January 2009 (UTC)


 * It's also ineffective because one link in 10,000 would still be thousands of links. — Arthur Rubin  (talk) 18:40, 9 January 2009 (UTC)


 * So you want others to remove 9,999 times thousands of links by hand? --RexxS (talk) 20:12, 9 January 2009 (UTC)


 * If one link in 10,000 is still thousands of links, 9,999 in 10,000 is tens of millions ... now we're talking ineffective. J IM ptalk·cont 20:19, 9 January 2009 (UTC)
 * I don't see any harm in the following process:
 * Holding off bot runs for 2 weeks
 * Creating a template that "saves" year links (presents them in a way the bot will ignore simply because they don't look like double-bracket year links -- that is, there is no need to change any bot code to support this)
 * Announce at VPP and other places, and give users two weeks to apply the template to any year they feel is necessary.
 * Let the bots sweep all remaining year links
 * Now we can use the template to see if any of those remaining year links really should be there or not, as, if as Ohconfucious suggests, its possible to change editors' minds where they are to be used, all the better.
 * We have no deadline to get rid of the links though we know we're going there. They've been around a long time, lets give editors a bit of time to make sure they're protected what they want.  The other way, letting the bots sweep any year link and then having editors add back for those they wanted, will first off seem a bit mean-spirited, but secondly, we'll have a harder time tracking when year links are put back instead of the benefit the template gives us. --M ASEM  20:36, 9 January 2009 (UTC)

Masem, I believe your inputs are in good faith, but your intermittent calls for us to pause the automated and semi-automated housecleaning in such a gargantuan project (for a week or two, here and there) are like labour-union calls not to automate telephone exchanges back in the 1950s. Sure, mistakes and false-positives occur, but we have the benefit of skilled, hard-working and sensitive operators such as Lightmouse, Dabomb, Colonies Chris and Ohconfucius, to name just a few. These users have been successfully refining the system, which I believe sets the standard for further innovations for automated assistance to WP editors. Relying on the manual correction and improvement of articles is very 19th-century. Tony  (talk)  02:21, 10 January 2009 (UTC)
 * I'm not trying to stall the removal of links - it clearly was shown, it should proceed. But if you don't set up some announcement and plan for grandfathering links that should stay (something that a bot cannot determine and has to be done by hand at some point, before or after) you are going to encourage minor edit warring.  Prep people, and they'll accept other changes more readily, instead of being resentful of "surprise" changes. --M ASEM  02:49, 10 January 2009 (UTC)
 * Masem, the last time we "notified" people about these type edits on WT:FAC, WT:FLC, Village Pump, the Signpost dispatch (after date-delinking was deprecated in August), we still had vocal complaints from a small group of people. I'm not sure how we would go about doing this. Dabomb87 (talk) 02:56, 10 January 2009 (UTC)
 * Well, the bulk of the complaints can be answered by pointing at the RFC - date delinking is a reality. But this is to help streamline removing the rest of the dates without minimal negative feedback.   If you guys don't want to do it, that's fine, but be aware you will likely get a good handful of complaints when you do it without warning. --M ASEM  03:05, 10 January 2009 (UTC)

Comment. Do bots eliminate links such as  or  ? If they don't, just mention that on those bots' userpages. (Anyway, I think that no more than a dozen articles or so should actually link to January 10.) -- Army1987 – Deeds, not words. 10:57, 10 January 2009 (UTC)

Bots could be programmed to ignore such links. What I'd had in mind was a template, which would naturally be ignored since the bot is looking for links not templates, but this might be the simpler solution (depending on how many of these links we're keeping. One advantage with the template is that we can keep track of the pages which use it. J IM ptalk·cont 11:12, 10 January 2009 (UTC)


 * How about modifying the existing template date to accept a parameter that adds a link? So writing as date as  would be converted to 11 January 2009 dynamically and show up as 11 January 2009. A bot would not touch it. &minus;Woodstone (talk) 23:33, 10 January 2009 (UTC)


 * It seems that it already does ... I guess there's our solution. J IM ptalk·cont 10:47, 11 January 2009 (UTC)


 * For the exceptional cases where date links are required to stay, neither of the following would be altered by the bot:
 * : 11 January 2009
 * : January 11, 2009
 * : 2009-01-11
 * &minus;Woodstone (talk) 13:04, 11 January 2009 (UTC)

Indiscriminate automated removal of year links
It came to my attention that Lightbot (talk · contribs) has been removing links to year articles. I read through the relevant section of the recent RfC; there isn't anything even approaching a general consensus that such a blanket removal is appropriate. Until such time as that is the case, robots should not be used to indiscriminately remove all year links, and I would ask Lightmouse to disable this feature. (I raised this issue on ver talk page originally but ve seems unwilling to do so, so I am bringing the issue here.)

Before someone misrepresents my opinion, let me point out that I am quite aware of what overlinking is and I am not advocating the indiscriminate keeping of all year links. That would be ridiculous. The issue here is that unless someone manages to invent strong AI, no robot they program will ever be able to do what is a human editor's job - reading article text and making determinations as to the appropriateness of links in their surrounding context. — Hex    (❝  ?!  ❞)   10:49, 7 January 2009 (UTC)


 * Hex is complaining about this edit, in which the link to year of foundation of the Latymer Upper School was founded, was removed. It appears Hex accepts that removal of overlinking is desirable, yet paradoxically refuses to give up the linking to this particular year and in this particular article, claiming this furthers the understanding to the subject. I cannot quite fathom what sort of 'human discretion' Hex is after. Perhaps he is after some sort of "discretionary 1 year-link per article" dispensation, even though there appears to be a strong consensus for "very limited linking only" to year articles; and the RfC to obtain dispensation for linking dates of birth and death failed due to the lack of interest - it just died and wasn't even closed properly? I say let the bots get on with removing these links, because doing it manually or semi-automatically is extremely boring and is a pain in the arse. Ohconfucius (talk) 12:54, 7 January 2009 (UTC)
 * There certainly does seem to be a consensus for limited linking only. Unfortunately it's impossible to do such limited linking because as soon as you do Lightbot or someone/thing else will come along and remove it. I am requesting that until an iron-hard consensus that all year links must be removed on sight is demonstrated and codified in policy, that robotic activity along these lines should be suspended.
 * If removing date links "manually or semi-automatically is extremely boring and is a pain in the arse", then that is probably a sign that you should move on to other, more important tasks instead, and let other people deal with overlinking on a case-by-case basis. It's really not a priority task here. — Hex    (❝  ?!  ❞)   14:23, 7 January 2009 (UTC)


 * There is a more detailed discussion on the matter at Talk:Latymer Upper School. Dabomb87 (talk) 13:46, 7 January 2009 (UTC)
 * Hex, why don't you just quit forum shopping. I'm sure you have other, more important tasks to do as well. ;-) Ohconfucius (talk) 15:09, 7 January 2009 (UTC)
 * I do not appreciate your allegation that I am "forum shopping". If the owner of a bot is not being forthcoming, then how is the talk page for the specific policy related to the issue not the correct place to raise the issue? I request that you take that back. — Hex    (❝  ?!  ❞)   15:49, 7 January 2009 (UTC)

What do you mean by "not being forthcoming" ? Lightmouse (talk) 15:57, 7 January 2009 (UTC)
 * It is well-established that the community does not want year links unless, on rare instances, they somehow increase the understanding of the reader of the topic, significantly. Please demonstrate just how that year-link you seem to be insisting on in the article above does this? It looks to me like a nuisance link. Tony   (talk)  15:58, 7 January 2009 (UTC)
 * As usual, Tony has misinterpreted the consensus here. (The word "significantly" does not appear in any form of the consensus, other than your discussion pages.)  The consensus on the detailed RfC (not yours) is that year links are sometimes appropriate, and year of founding of an organization has been given as a possible example.  I don't see a consensus that the year of founding should be linked, nor that it should not be linked.  Hence, the bot is wrong, although an honest semi-automated edit process might very well lead to the removal of that link.  (I'm trying to assume good faith, although some editors had been using semi-automated tools at a rate which makes it unlikely that any thought had gone into it.)  — Arthur Rubin  (talk) 19:07, 7 January 2009 (UTC)
 * Any unbiased editor, upon examination of the two RfCs, will conclude that Rubin's "sometimes appropriate" is not accurate, and that "link only very rarely (and only with good reason)" would be a more accurate summary.--Goodmorningworld (talk) 19:10, 7 January 2009 (UTC)
 * As I predicted in November, "If a hypothetical editor decides that ... the RfC below is 'no consensus' on ... 'When to link Year articles' (ignoring the fact that most commentators supporting 'sometimes' actually said 'very rarely'), then I guess we're right back where we started. It's a pity the committee didn't ask the question, 'Should date links be treated exactly like any other link to an article?' " How many people agreeing not to link to an irrelevant article does it take to make a consensus? --RexxS (talk) 19:57, 7 January 2009 (UTC)
 * So you did. They were in such a hurry they put out a half-baked RfC, so what do you expect? But to answer the question you posted back then, there seems to be no reason to ask that question - to do so would elevate date articles to a higher status. Ohconfucius (talk) 03:45, 8 January 2009 (UTC)
 * Indeed. Look out for a fresh new section intended to clarify these ill-planned for ambiguities in the next couple of days. Dabomb87 (talk) 04:50, 8 January 2009 (UTC)
 * The only thing ambiguous here is how to deal with people who seem dead set against accepting the results of this RFC for what they are: a clear indicator that automated tools will not work for delinking dates. —Locke Cole • t • c 05:18, 8 January 2009 (UTC)
 * "Half-baked RFC", nice... and of course the other RFC was written on a stone tablet by God and handed down to Tony directly I take it? —Locke Cole • t • c 05:18, 8 January 2009 (UTC)
 * Nope: I am God. Tony   (talk)  07:33, 8 January 2009 (UTC)
 * Locke, something which is conceived to stop a quarrel by delivering 'a satisfactory response' but -through misdesign, negligence, insufficient thought or inadequate planning - delivers an ambiguous response which only prolongs the argument can only be described as half-baked, IMHO. Any other term is just playing semantics. Ohconfucius (talk) 08:21, 8 January 2009 (UTC)
 * It seems to me that the date of foundation of an establishment which is the subject of an article is a very good example of a relevant date link. Deb (talk) 21:33, 7 January 2009 (UTC)
 * No. It is a relevant date, an important fact, but not a high-value link. Dabomb87 (talk) 22:14, 7 January 2009 (UTC)
 * On the contrary, Goodmorningworld; any unbiased editor will note the lack of consensus on questions 2 and 4 of the detailed RfC. Tony's RfC is irrelevant to the question of whether any particular year should be linked, only that years should not be linked absent a reason.  — Arthur Rubin  (talk) 22:24, 7 January 2009 (UTC)
 * Obviously automated removal of date links should cease, but none of the anti-date link crowd around here will hear of it. In their minds all date links are evil, despite clear consensus for keeping date links (be they month-day or simply year) at the recently concluded month long RFC. I'll spell it out again for Lightmouse, Tony, Greg and so forth: It is IMPOSSIBLE for an automated script/program to enact the community consensus view. Each date must be looked at individually and judged to be relevant or irrelevant on a case by case basis. You cannot simply run a script on hundreds of pages and remove date links and claim you've got the backing of the community because you don't. The only backing you have is the MOSNUM regulars who represent only a fringe view of the actual community. —Locke Cole • t • c 01:42, 8 January 2009 (UTC)
 * ...And the numerous FAC/FLC editors who allowed the date links to be removed even before the initial deprecation, as well as the numerous editors who did not protest when the articles on their watchlists were delinked. What clear consensus are you talking about? Dabomb87 (talk) 01:53, 8 January 2009 (UTC)
 *  This is old hat. We all know that it's the same old gang of five which was opposed to any form of delinking. Take a look at the article now. The OP accepts that this is a good way out, providing useful context without linking to the trivia article. Ohconfucius (talk) 02:08, 8 January 2009 (UTC)
 *  This is old hat indeed, same crowd still persisting at pushing THEIR preferred way over the top of the entire community. —Locke Cole • t • c 02:15, 8 January 2009 (UTC)
 * So FAC/FLC editors override community wide consensus? Since when? And the clear consensus at the recently concluded RFC, which you know about already, and feigning ignorance is not useful here. —Locke Cole • t • c 02:15, 8 January 2009 (UTC)
 * Seeing as those articles (FAs/FLs) are the best that Wikipedia has to offer, I think it would be suitable to consider them as model articles. Last I checked, there were two RfCs, in which one of them showed very strong opposition for linking of dates. Not only that, but most of the commenters on that RfC were against nearly all, if not all, date links (read their comments). I note that although much ado was made on the creation and development of the "detailed" RfC, there has been little to no effort to collate and analyze the results of either. I would also point to Proposal 3 the previously linked RfC, where there was clearly consensus to allow semi-automated and automated edits for these type changes. Dabomb87 (talk) 02:27, 8 January 2009 (UTC)
 * The other RFC only asked editors if they approved of inserting very specific language in to MOSNUM. The second RFC asked their opinion on various issues relating to date links and date auto formatting. It's that second RFC which controls here. —Locke Cole • t • c 02:43, 8 January 2009 (UTC)
 * Nobody asked a question about keeping date links. You asked "When should date links be made?" and the answer was "Very rarely" and "Only when relevant". The vast majority of date links are demonstrably useless in providing any relevant extra information for the article containing the link, and should be removed. The consensus is to make them when they add value and I've yet to find one that does. In fact, there's a catch-22: if a link to a factoid in a date article actually were to add value, that that factoid would almost inevitably be better incorporated in the original article. In the 1 in 1,0000,000 case where that's not desirable, then you have everyone's blessing to put it back. In the meantime, please don't hold up the removal of the other 999,999 valueless date links by insisting that they be done by hand. --RexxS (talk) 02:18, 8 January 2009 (UTC)
 * Then you can demonstrate their usefulness on a case by case basis, not in an automated fashion which runs roughshod over community consensus for allowing some (but not all) date links. I also dispute your claim of "1 in 1,000,000" being the case for keeping, obviously. —Locke Cole • t • c 02:43, 8 January 2009 (UTC)
 * Yes, because of articles like 1992 or May 19, links are going to be desirable in more cases than 1 in 1,0000,000. However, exclude articles about chronological items, certain holiday articles (April Fools' Day, New Years' Day, Christmas Day, etc.) and maybe a couple globally important articles that have important dates (Armistice Day), and there are few, if any, dates that should be linked. "Then you can demonstrate their usefulness on a case by case basis" That would be instruction creep at its worst. Dabomb87 (talk) 03:05, 8 January 2009 (UTC)
 * Then you need to get community consensus that only those select dates and years are relevant. But you haven't, so you don't have consensus. As to "instruction creep", consider your proposal: the community has said date and year links are useful under certain circumstances, but you and yours are blindly removing them from dozens and hundreds of articles each day. You're totally ignoring what the community told you in favor of your preferred method of operation. —Locke Cole • t • c 03:16, 8 January 2009 (UTC)


 * My apologies. I did engage in hyperbole. I should have said, "In the 1 in 1,0000 case where that's not desirable, then you have everyone's blessing to put it back. In the meantime, please don't hold up the removal of the other 999 valueless date links by insisting that they be done by hand." I must also admit that I was impressed by 1992. What a good looking article! Out of curiosity, I followed my own advice, but still couldn't see how George H W Bush throwing up in Japan improved my understanding of Antarctic Treaty System; or how Mike Tyson's conviction for rape helped me get a handle on Andrew S. Tanenbaum; or even how the Kentucky Supreme Court finding same-sex sodomy laws to be unconstitutional shed any sort of light on Armed Forces of Belarus (although that linked to 1992 twice). Forgive me. I just don't understand. --RexxS (talk) 03:48, 8 January 2009 (UTC)
 * That square box you're seeing this in is called a web browser. People browse the web, sometimes looking for something specific, sometimes looking for only things with only a minor relationship, but at any rate, it all comes back to browsing. Just because you do not see the value in being able to click a date and see other things that happened on that date (or year), doesn't mean everyone else feels the same way. —Locke Cole • t • c 05:11, 8 January 2009 (UTC)
 * Locke, thank you for that most patronising and mildly sarcastic response. Ohconfucius (talk) 05:20, 8 January 2009 (UTC)
 * Taking a page from your playbook. —Locke Cole • t • c 06:53, 8 January 2009 (UTC)
 * You got it wrong, amigo, I'm Ok at being sarcastic, but my ability at being patronising isn't a patch on you. Ohconfucius (talk) 13:59, 8 January 2009 (UTC)
 * Indeed, thank you, Locke. I am always grateful for education, even in my autumn years. Before you go could you explain to me the purpose of those blue words up in the top left corner of my squarebox? Yes, those - last blob down in the squarebox labelled 'navigation'. Thanks in advance. Your eager student, --RexxS (talk) 06:09, 8 January 2009 (UTC)
 * So only one method of navigation is useful? Another method isn't? For the same reason we have Special:Random, we have potentially useless (although occasionally useful) date links. It's fascinating to me all the arguments against date links: "sea of blue", "dilute the value of relevant links", etc. The only time there's a sea of blue usually is in some chronological bit of prose, and the dilution argument is just silly; anyone who browses here for more than a few minutes will understand that date links just link to articles about those specific dates (not necessarily related to the original article). WP:NOTPAPER and all that. This is all irrelevant though, the community doesn't support mass delinking and editors engaging in the act should stop. Whether you agree or not, or see the value proposition or not, is irrelevant. —Locke Cole • t • c 06:53, 8 January 2009 (UTC)


 * Thanks, Locke, I'm glad we've got to the nub of the disagreement. Would you do me the courtesy of looking at WP:Manual of Style (dates_and_numbers)/Date Linking RFC? I think we can now agree that the majority (at least) of date articles are not relevant to the articles that link to them. Correct me if I'm wrong, but you seem to be of the opinion that date article links add a kind of "browsing" value sufficient to outweigh WP:Context; others do not share that opinion. If that's the essential point, then make your case. I'm not entirely unsympathetic to that concept, and remain open to be convinced that it might be a reasonable exception to WP:Context. At present however, I just don't see the consensus for that exception. Not kidding this time, and sincerely, --RexxS (talk) 15:05, 8 January 2009 (UTC)
 * I'd love to argue this, but we're veering off here. The RFC (specifically, the second one) shows a clear lack of consensus for removing all date links in an automated fashion. And yes, I believe date links (month-day/year) should be exempt from WP:CONTEXT, and as the community shows at worst no consensus and at best a consensus to keep some of them, I think it's safe to say they are an exception. —Locke Cole • t • c 10:25, 9 January 2009 (UTC)

Summary of the RfCs
(outdent) Locke Cole, as I have said before, these are the things that we can glean from this RfC as of yet: There, I have made my case with specific citations. Feel free to respond with a similar level of detail, instead of generally saying something about the "community". Dabomb87 (talk) 13:51, 8 January 2009 (UTC)
 * Manual of Style (dates and numbers)/Three proposals for change to MOSNUM – Date fragments should not be linked unless there is a reason to do so; and at that, very rarely.
 * Manual of Style (dates and numbers)/Three proposals for change to MOSNUM – The use of automatic or semi-automatic processes to bring articles in line with the style guide does not require special consensus.
 * Manual_of_Style_(dates_and_numbers)/Date_Linking_RFC – Dates should not be linked purely for the purpose of autoformatting; again, the actual comments on the oppose section indicate that opinion against these links are stronger than that.
 * Manual_of_Style_(dates_and_numbers)/Date_Linking_RFC Month-day links should be linked very rarely.
 * Manual_of_Style_(dates_and_numbers)/Date_Linking_RFC – Year links can be made sometimes, although consensus leans toward less of these rather than more.
 * Manual_of_Style_(dates_and_numbers)/Date_Linking_RFC – Although year in field links can be made, there is no consensus of how to do this yet. Therefore, we return to the status quo: Try to avoid hidden year-in-field links in prose, but use them in tables and infoboxes as necessary.


 * Maybe I missed someone else’s summary of the RfCs somewhere else, but this is the first I’ve seen of a nice executive summary of the results of all those RfCs. Thanks Dabomb87; we need that. Greg L (talk) 06:10, 9 January 2009 (UTC)


 * Fascinating stuff, especially considering you still seem to be supporting automated removal of date links despite admitting that there are cases (even rare cases) where they should remain. How do you reconcile the inability of a script to determine if a date/year link is valuable against your read of the consensus of the RFCs? BTW, as regards the first RFC, those results are meaningless for our discussions as participants were only asked if specific language should be inserted into MOSNUM, not what their opinions in general were. —Locke Cole • t • c 10:25, 9 January 2009 (UTC)
 * If the cases are rare, then somebody can easily go back and relink that date; nobody is stopping them. It helps if they provide a reason instead of just undoing the edit. Dabomb87 (talk) 13:48, 9 January 2009 (UTC)


 * Let's assume that one percent of these links is justified. Is it easier to link one date by hand or is it easier to delink ninety-nine?  We have a mess to clean up, the job is huge—best suited for a bot.  Let it be the relatively tiny task of (re)linking any justified links (back) up that we leave up to humans.  Such links could be tagged (with a hidden comment or template) to tell bots (& humans) that the linking was intentional & any thoughtless/discussionless removal would not be appreciated. J IM ptalk·cont 11:44, 9 January 2009 (UTC)
 * JimP's logic is inescapable; however, one in 100 is rather generous. More like one in 10,000 is useful. But whatever the proportion, it's mighty small. Tony   (talk)  11:55, 9 January 2009 (UTC)
 * Find a script that can discern a valuable date from an invaluable date and we'd be golden. But since even one percent of those links may be valuable it defies belief that we would require our editors to go through after the fact and relink dates that were already linked. And again, I disagree with the low percentage of valuable links, I personally find them all to be valuable as a browsing aid (but assuming that's not the consensus view, the RFC consensus is clear that some links are appropriate). What's worse is that when people DO re-add links you have these automated scripts/bots coming through and delinking the dates again. This situation is simply untenable and those promoting it are making a mistake. —Locke Cole • t • c 12:59, 9 January 2009 (UTC)
 * Yes, but as the Date Linking RfC showed, there is much more community consensus behind the idea that date links are useless, especially for the Month-day links. Dabomb87 (talk) 13:48, 9 January 2009 (UTC)


 * On the topic of the first RfC, it was, of course, a Request for Comment, not a Request for Vote. I would guess that many commentators in that RfC would rightly be offended to hear their contributions described as 'meaningless'. The second RfC did no more to ask for general opinion, as it should have elicited opinion on whether "Date links should be treated the same as ordinary links (rather than as an exception)". I think most would agree that we could have used the answer to that. --RexxS (talk) 19:53, 9 January 2009 (UTC)

Opinions may vary as to the percentage of date links with any value but I believe consensus is putting it well below this arbitrary one percent I pulled out of my hat. One percent is generous—even the one percent of one percent Tony mentions is probably inflated—but let's go with it for now. If "it defies belief that we would require our editors to go through after the fact and relink dates that were already linked", surely it must defy belief 99 times over that we would require our editors to go through and by hand delink those dates without value. What difference makes the status quo? It's the ends not the means which are important. Let's clean this mess up in as efficient a way as possible. As for a bot-proof linking system see my suggestion below. J IM ptalk·cont 20:42, 9 January 2009 (UTC)

Purposeful automated removal of year links: there are better ways to accomplish this

 * Well, since there seems to be some editors who haven’t grasped the simple concept that “the proper response to bad speech is better speech”, I have to re-post my post. Don’t dare revert me again. I’ve removed the *s*-words since a couple of you seem so intent on playing childish games like trying to seize the moral high ground by pretending to dictate to me how I may think or express my thoughts. I’d rather get my below point across than play such stupid games. Hex, it is not “indiscriminate” removal; it is “purposeful” removal. Get your facts straight. The RfCs seemed clear enough to me on two points: 1) that bots may be relied upon to semi-automate the tasks that would be well beyond the capability of a mere mortal, and 2) that routine linking of dates to Wikipedia’s utterly mindless trivia articles is to no longer be done. If you think linking to trivia is such a fantabulous idea, let’s see you read through just four of them (here at Sewer cover in front of Greg L’s house) and earn yourself a Sewer Cover Barnstar in the process for your unbridled bravery. Doing so I will at least help us to accept that you actually like the puke you expect our articles should be linking to. But just because you can prove you can stomach through reading that crap will only prove that you like reading mindless crap; it will come up short of convincing proof that those trivia articles are “compelling reading” that most readers appreciate. Greg L (talk) 23:08, 7 January 2009 (UTC) P.S. If you think readers of an article on an upper school in England will be fascinated to learn such breathtaking trivia as [In 1624:] The Japanese Shogun expels the Spanish from the land and severs trade with the Philippines, then add a link in the See also section that says • Other notable events of 1624. That way, readers will know precisely what they will be clicking on, you can better advise of the availability of the topic, and one accomplishes that end without turning yet more main body text into a giant blue turd. Greg L (talk) 23:37, 7 January 2009 (UTC)

I see Greg L has restored his abuse. My comment, which he removed, follows. I intend to raise the issue of his conduct on this page in the appropriate forum shortly. — Hex    (❝  ?!  ❞)   00:00, 8 January 2009 (UTC)

Abuse removed for a second time: there are better ways to get people to listen.

"Don’t dare revert me again"? Who does Greg L think that he is that he can threaten other editors? As I'm involved in this discussion I won't be taking administrative action by myself, but will refer this to others should such behavior continue. — Hex    (❝  ?!  ❞)   23:51, 7 January 2009 (UTC)


 * As you can see, I’ve already started an ANI for your outrageous conduct. Your protestations of “abuse” are absurd. Greg L (talk) 00:05, 8 January 2009 (UTC)


 * Your potty mouth and outraged comments are more absurd still. Bring it on.


 * Now you've got that out of your system, I look forward to continuing the discussion with editors who are able to express themselves cogently. — Hex    (❝  ?!  ❞)   00:19, 8 January 2009 (UTC)


 * “Potty mouth”? LOL! Are we in 2nd grade now? What part of my above post is “abusive” den mother? That’s what you cited when you censored me. Methinks thou doth protest too much. Don’t try to hide behind the apron strings of “potty mouth” or “abusive” or “personal attacks” because you are just flat wrong. Just address the issues and stop trying to delete posts you find to be *inconvenient*. Damn, there will be hell to pay if you keep that up. And stop trying to throw your admin weight around. Answer me this please. Just how old are you? Seriously, I’m trying figure out how all that power went to your head here that you can censor my above post. Greg L (talk) 00:30, 8 January 2009 (UTC)


 * Further, as I see you have dropped a hint, above, that you are an admin who might go solicit some friend to do some dirty work, I’ve also asked a crat to keep an eye on you. Based on your lack of judgement here, I wonder how it is you got by Wikipedia’s vetting process. Finally, in my first edit summary, I wrote Earth calling Hex: The proper response to bad speech is better speech. Don’t presume you can censor others. Yet you chose to overlook that little gem of good advise, and instead chose to focus on “Don’t dare revert me again” and claim that is some sort of “threat”. I think you are damaging to Wikipedia and should be stripped of your administrative privileges. I suggest you go cool off and get out of this discussion if you can’t control yourself any better. Greg L (talk) 00:21, 8 January 2009 (UTC)
 * Have cup of coffee, a Pall Mall (or whatever your vices are) and relax. Come back in a day or two, put the guns on the table, shake hands and try starting the discussion over with a clean slate and clear heads.  &mdash;  MJC detroit  (yak) 02:29, 8 January 2009 (UTC)
 * Mmmm… Coooooffeeeee. That just might be able to make my censored (completely deleted) posts seem like a non-issue. “Might.” As for getting on track: the general consensus on RfCs is abundantly clear and it is not at all sporting for some here to posture about how it’s not fair to let bots loose to fix the dates. The RfC on bot activity is clear too. The vast majority of our date links gotta go because they are not germane and topical to the articles in which they are linked. But… Two notable exceptions are “February 5, 1965” and “September 13, 1965” in our Trivia article: those, curiously, were not linked. Now fixed (and with piping so they don’t autoformat). Greg L (talk) 03:56, 8 January 2009 (UTC)
 * Sound advice indeed. Ommmm, Ommmm, Ommmm. Ohconfucius (talk) 03:58, 8 January 2009 (UTC)


 * Oh, RexxS pointed out that I missed “March 1” in our Trivia article. That is now fixed too. Some editors here want dates linked. I can not possibly think of a more suitable article within which to link dates to articles on historical trivia. It is a perfect way to provide readers interested in trivia with more understanding of the topic. Very topical. Very germane. Greg L (talk) 04:26, 8 January 2009 (UTC)
 * Ahh, Tony didn't see the connenction :( Dabomb87 (talk) 13:33, 8 January 2009 (UTC)
 * Gee, for an article on trivia of all things, linking three dates there seemed imminently germane and topical to me; a great way to better learn about the subject. Perhaps not. Greg L (talk) 15:08, 8 January 2009 (UTC)

Let's bring them before the ArbCom
Since there's no consensus for indiscriminated automated removal of all year links, let's bring all owners of robots (User:Lightmouse for his User:Lightbot and User:Dabomb87) before the ArbCom. Forcing of their POV is totally unwikipedian. If they wish to orphan articles 499 BC to 2059, the only proper way for doing it is WP:AfD —Guy Peters Talk • Contributions • Edit counter 20:36, 9 January 2009 (UTC)
 * Arbitration is only a last resort for dispute resolution. Dabomb87 (talk) 23:42, 9 January 2009 (UTC)
 * And ArbCom will just say it's a content dispute, anyway. I'm not sure I agree, but that's what they'll say.  — Arthur Rubin  (talk) 00:33, 10 January 2009 (UTC)
 * The way to go is to organize a call for votes on the proposition that bots should never be used to delink dates. I don't believe they would go against such a vote.Dejvid (talk) 00:47, 10 January 2009 (UTC)
 * Perhaps, but their behavior here (ignoring consensus, performing mass disputed edits, ignoring calls to stop or do it in a way that acknowledges the community consensus, etc) can be dealt with by ArbCom. Besides, there's a new slate of arbitrators, perhaps they'll see this differently. At any rate, the non-stop endless debate is disruptive in itself. —Locke Cole • t • c 01:18, 10 January 2009 (UTC)
 * Maybe you missed it, but you've been talked to on your talk page, there's been an RFC over the issue (because many of you didn't believe the community wouldn't support you), and we've had to two talk page threads here trying to get those operating scripts and bots to see reason and compromise on the issues. But instead the steamroller just keeps on running... you're not leaving us many options. —Locke Cole • t • c 01:18, 10 January 2009 (UTC)
 * Locke Cole, as long as you continue to claim that dates are being unlinked "against consensus" without specifying detailed diffs or links (as I did above), your claims will be ignored, as they often have been. The last time you tried to "take action" (in the form of repeated ANI threads), you were not very successful. I must say that in all your haste to set up the RfC to get "real consensus", you and the others involved with it never set up a way to analyze the information to get a definite read on the consensus. This is the product of that haste: we are now back to where we started. Dabomb87 (talk) 01:44, 10 January 2009 (UTC)
 * You need some perspective: I was actually very successful at getting many of the people delinking dates against consensus blocked for disruption (quick note: I'm not particularly proud of that, I think blocks are a last resort, but I wasn't left with many options). Only once was a request turned down, and I stopped trying after that. We are not, in fact, where we started. Here's your link: WP:MOSNUM/RFC. The questions about when to link month-day and years are strictly relevant here. The community has decided that some month-day and year links are valuable, and implied with that is the inability for scripts/bots to be used in delinking dates (a script/bot is incapable of "reason" and seeing when such a link might be appropriate). Those of you wishing to continue delinking dates are doing so in a disruptive manner. You've been warned about this repeatedly throughout these discussions. I've urged everyone to try and reach compromise on this, but despite my best efforts you all seem unwilling to discuss the matter (at least in a rational and calm manner, instead intent on forcing your ways through). —Locke Cole • t • c 02:13, 10 January 2009 (UTC)


 * Thanks, Dabomb. I agree, except that I don't think we're "back to where we started"; I believe the community has continued to move on, now that it sees the sky hasn't fallen in and that a "smart-linking" approach is a definite improvement. What is telling is that:
 * apart from the same noisy "Gang of Five" in-house users, there's absolutely no stream of complaints about "where have our bright-blue dates gone?", or indeed "our bright-blue years/centuries/decades" ... just silence;
 * WP:FAC nominators promptly adopted the new guideline without a blink, and our best articles are being promoted using the modern approach as though it always existed.

It seems like a non-issue, and I'm sure an ArbCom complaint would lead straight down the drainpipe. Now, rather than doing this circular dance ad infinitum like little kids in the playground, we need to focus on more important things, like getting year-articles onto a better footing. Currently, they're way under our standards. Who's gonna help? Locke, where are you? Tony  (talk)  02:12, 10 January 2009 (UTC)
 * I'm still dealing with this mess because most of you are intent on forcing your way through "like little kids in the playground". —Locke Cole • t • c 02:14, 10 January 2009 (UTC)
 * What "mess"? Speaking of more important things, see Manual of Style (links)/merged. Instead of continuing these stream of generalizations that claim the same things without substance, your input on this overdue merge of three redundant and contradictory guidelines would in fact help to resolve some of the ambiguities on date linking. Dabomb87 (talk) 02:19, 10 January 2009 (UTC)
 * The mess caused by mass unlinking of dates against consensus. Maybe you missed it, but people are still complaining here. —Locke Cole • t • c 02:28, 10 January 2009 (UTC)
 * Mostly the same people though. When Tony refers to the "Gang of Five", he may be only half-joking. Dabomb87 (talk) 02:33, 10 January 2009 (UTC)
 * No, not really the same people all the time. But that's how you guys seem to convince yourselves that it's okay. Further, even if it WAS the same group, it's not just one editor (me), it's numerous editors. Ignoring us instead of gaining consensus accomplishes nothing for you. And I take the "gang of five" remarks he's made as personal attacks. Disparaging remarks have no place in a community driven by peer respect, communication and consensus development. —Locke Cole • t • c 08:11, 10 January 2009 (UTC)
 * I concur; labeling other users in such a fashion is incivil. And now that I've turned up, must I expect to be similarly labeled for committing the crime of not throwing up my hands and abandoning the issue? — Hex    (❝  ?!  ❞)   13:11, 10 January 2009 (UTC)


 * Tony1: "apart from the same noisy "Gang of Five" in-house users, there's absolutely no stream of complaints... just silence" - given the reaction that anyone with a differing opinion seems to receive here, I am absolutely not suprised. It took a positive amount of force of will for me to get involved after reading the tone of many people commenting. — Hex    (❝  ?!  ❞)   13:11, 10 January 2009 (UTC)

Locke Cole, are you, seriously, making an admission that you abused admin privileges as an INVOLVED admin? That is a significant breach of WP:ADMIN. Whom did you block?

"'I was actually very successful at getting many of the people delinking dates against consensus blocked for disruption (quick note: I'm not particularly proud of that, I think blocks are a last resort, but I wasn't left with many options).'"

Tony  (talk)  02:26, 10 January 2009 (UTC)
 * I'm not an admin, and if I was, I would never use the tools in a dispute I was involved in (conflict of interest). I was referring to the reports I made to WP:AN/EW regarding the edit warring over date links which resulted in a half dozen blocks of those refusing to stop. —Locke Cole • t • c 02:28, 10 January 2009 (UTC)
 * Seems perilously close to gaming the system, that you now feel free to boast of your success in "getting" people blocked for what you construct as "disruption", particularly when the blocks appear to be framed according to your personal claim of what "consensus" was. Did this involve convincing an admin of your particular view of these two concepts, and of disregarding what the style guides said? It's got the makings of a very serious issue. Tony   (talk)  07:33, 10 January 2009 (UTC)
 * If you say so Tony, but I wasn't boasting. It was a statement of fact: edits like those done by you, Lightbot, Dabomb87 and on are disruptive when you know you don't have consensus for the automated removal of all date links. It also hinders serious discussion on reaching agreement when the threat of inevitability hangs over your head. —Locke Cole • t • c 08:09, 10 January 2009 (UTC)

YEAR ARTICLES ARE NOT 'TRIVIA' ARTICLES
For those who failed history, year articles are NOT trivia articles. They are meant to give a global perspective that intersects with a time perspective. If anyone checks out, for example, the World Almanac, they give lists of historical events in chronological order. This far predates the existence of Wikipedia. However, it seems that some people thinkr that the only thing that is important is what exists NOW, while forgetting that those that analyzing the past can help us follow trends, threads, and figure out where we may be headed in the future.

Just because someone has the biggest trash-mouth doesn't make them right. Ryoung 122 23:45, 7 January 2009 (UTC)
 * Oooh, shouting now. Your words are like putting lipstick on a pig and trying to pass it off as a prom date: it’s not working for me here. The facts speak for themselves for anyone who has tried to read through those trivia articles. The only people who’ve managed to read through them have earned themselves a damn barnstar for their heroic efforts (and only six editors have so-far managed to throw their bodies onto the barbed wire to earn their medals). The RfCs are clear; you don’t have to believe the consensus view is the right thing, you just need to abide by it and stop wasting our time with declarations that the consensus is other than what it clearly is. Greg L (talk) 23:50, 7 January 2009 (UTC)
 * Yes, the RfCs are clear. The consensus is that year links may be appropriate in some cases, so a bot to remove them is inappropriate.  (As I agree that most year links are inappropriate, as indicated by the consensus, it might be suitable for a bot to remove them if a year link could be marked as non-removable — not only by bots, but without appropriate specific discussion.  I don't want to see edit wars about year links, either.)  — Arthur Rubin  (talk) 00:29, 8 January 2009 (UTC)
 * No Arthur. There are far, far too many linked dates for any mere mortal to delete them all. They are used all over Wikipedia and must be fixed with a bot. The few cases where links to trivia makes sense can be hand-restored easily enough. Greg L (talk) 00:58, 8 January 2009 (UTC)


 * Your use of circular 'reasoning' by calling date links "trivia" does not make your case..."it's trivia because I call it trivia" is not a good argument. Ryoung 122 01:02, 8 January 2009 (UTC)
 * Because that it what it clearly is. Radio talk show hosts can—and probably do—use our trivia articles on their morning broadcast during their “on this day throughout history” segments. You can call our trivia articles “cogent, well-organized history” if you want. But the articles still virtually never have jack to do with the articles from which they are linked. Links have to be germane and topical to the subject matter. If not, they gotta go. If you can figure out a better way to handle the millions of date links that doesn’t require a bot, I’m all ears. But we both know a bot is the only way address such huge numbers. Greg L (talk) 01:31, 8 January 2009 (UTC)
 * And ironically enough, the same criticism of 'circular reasoning' applies to the original: Just because you say (read: shout), "YEAR ARTICLES ARE NOT 'TRIVIA' ARTICLES" doesn't make that true either. We need to look at the evidence, not just a set of assertions. Pick a year at random; go to that article; look at "What links here"; pick one of those articles linked to the year article; then, as dispassionately as you are able, examine the linking article and see if anything in the year article adds value or understanding to the linking article. Now ask yourself the question: "If that kind of link was made to a non-year article, would I even think twice about removing it per WP:Context?" Go on, be honest. Now repeat that until you can find a decent example of a relevant link to a date article (and let us know). Or give up now that you know what the odds are of finding a good link. --RexxS (talk) 02:32, 8 January 2009 (UTC)


 * A list of unrelated events (except for the fact that they happened on the same year) will hardly provide "global perspective that intersects with a time perspective". On the other hand, systemic bias and recentism will ensure that this is will not be the case. Now, if all year articles were organized like 1345, then I might concede. Dabomb87 (talk) 00:46, 8 January 2009 (UTC)
 * This is refuted for me by a recent LightBot edit I saw (provenance unknown, it's made too many edits to track through). The article was along the lines of a monarchic succession or some such in 1943, in a nation then involved in, umm, World War II. I did check the 1943 article and yeah, it had a bunch of "trivia" about WWII. For someone who was born after the beginning of time in 1990, that may be trivial, but it does a great disservice to our general readership. 1943 does indeed provide "global perspective that intersects with a time perspective" - I suppose maybe you had to be there to understand (thankfully, I wasn't).
 * The LightBot/automated removal discussion may be moot anyway, since last I checked it's finished "Z..." articles and moving on to template-space. Mission accomplished - stall the objections and keep operating the bot until it's done. A successful implementation of "facts on the ground".
 * Now I'll accept the RFC that indiscriminate year linking should not be done, but I also observe that limited and appropriate year-linking is acceptable. This will be impossible if the bot (or other scripts) are going to again apply an indiscriminate rule, i.e. if I judiciously restore a year link, is it going to be blindly removed, or will it be subject of editorial discussion i.e. the talk page?
 * Will Lightmouse stop the bot runs after the comprehensive removal is complete? Franamax (talk) 02:14, 8 January 2009 (UTC)
 * Did the articles link to World war II? If they did not, they probably should, as that is definitely a good link. Most of the relevant facts from 1943 are probably in the WWII article already. Dabomb87 (talk) 02:19, 8 January 2009 (UTC)


 * (ec) First off, I'm going to challenge the fundamental premise of this thread. The OP said "For those who failed history, year articles are NOT trivia articles. They are meant to give a global perspective that intersects with a time perspective." This is completely false, as there is no such intention at all. A visit to WP:YEARS and Timeline standards yields no mission statement to that effect. The stated objective is to standardise the format, thereby creating an almanac. To my mind, that intention has contributed to this pisspoor state of trivia in years articles. In fact, these projects have been so 'successful' in imposing that format that, out of the thousands of articles which come under their banner, there is not one single GA or FA among them. Ohconfucius (talk) 02:25, 8 January 2009 (UTC)
 * Yes, exactly. The upshot is that better ways to organize timelines and year articles do exist. See Timeline of the 1994 Atlantic hurricane season and 1345. Obviously, even 1345 needs tweaking; it failed its GA nomination. Dabomb87 (talk) 02:32, 8 January 2009 (UTC)


 * Newsflash: It's not the year articles themselves that are trivial (although most are not in good shape) – it's the indiscriminate linking to them in the assumption that they are relevant to a particular context. There's almost always a stony silence after I ask indiscriminate linkers for evidence that the link will enable our readers to increase their understanding of the topic. This is suspicious. Tony   (talk)  07:44, 8 January 2009 (UTC)
 * Well said Tony; a cogent nutshell of what the issue is about. Please, let’s all get back to expanding articles and ridding them of errors. There has been far, far too much bickering over formatting and linking of dates. It’s really OK if dates look and behave like regular old text; they’re not all that special. More than that, it’s better if they look like regular old text unless there is something very topical and germane to the subject matter of the article. In the mean time, a great way to let editors know about relevant material in a historical-list (trivia) article is to do as has been done here on Kilogram—a GA article—where two, unambiguous links are first in line in its See also section. Those two linked articles now probably get more visits by virtue of this method than they ever would have with the old method. Greg L (talk) 00:37, 9 January 2009 (UTC)
 * If this indiscriminate delinking of dates would stop then we would be able to stop this "bickering over formatting and linking of dates". And just because you do not find linking of dates as useful do not assume that others do not.  Given that history covers such a large span there is a danger of learning things as isolated events. Checking what was happening at the same time as a specific event helps tie things together.Dejvid (talk) 19:07, 9 January 2009 (UTC)
 * It doesn't help encourage responses, Tony, when your requests are combined with trash talking like "dislexia?" or "mental illness?". Tennis expert (talk) 09:59, 9 January 2009 (UTC)

You all might be interested in the discussion I've started at Wikipedia talk:WikiProject Years. -- Army1987 – Deeds, not words. 14:02, 8 January 2009 (UTC)

Given the RFCs, Army's suggestion above, and the like, it seems clear that what is being argued towards on both sides is to aim towards using more "year in field" type links, with the current year pages being reorganized to provide sections/links to such "year in field" articles. This is not all years, of course, but when pointing to a year in field will help to provide chronological context for the linked event. This will help to remove the trivial-ness of the year articles (and if organized right, will allow users to explore the events of other parts of the year), and make any linking more relevant. The only issue is how these year in field links are to be made, as there wasn't strong consensus of any way. The only time a bare year should be linked directly is the discussion of that year from a calendar/timekeeping format, as well as for organizing of the "year in field" pages. --M ASEM 15:17, 8 January 2009 (UTC)
 * I'm the editor of 1345, 1346 (which is completely cited and now a GA nominee), and 1347. I think that those articles are just beginning to show how incredibly useful year articles can be. Yes, I hate the listy trivia our year articles are now, but the answer is not to split them up and delete them, as Army suggests. All you would have is similarly low-quality split-offs of the old year article. Masem, you say that "This will help to remove the trivial-ness of the year articles," but it wouldn't. It would just be a bunch of useless trivia split off into even more useless articles than we have now. What improvement would that be? The answer is not to create a new problem in thousands of useless split-offs, the answer is to improve the year articles themselves, the HARD WAY, with work, research, and negotiation. Anything else is an artificial solution. Otherwise, someone will come along later saying, "Why are we linking to these useless, trivial, Year in X articles?" Splitting it off like army suggests will just make more of a problem. I'm in favor of ending autoformatting and removing most links to year articles, but deleting year articles is taking it way too far.
 * I'm interested in, as Masem says, figuring out how to divide year articles into relevant sections, so that someone could link to 1346 or something. Some articles would be harder than others to sort out, but it could be done. Just having a list of events is less helpful when it comes to context. Perhaps some of the editors here would be interested in joining the years project. As of now, we have representatives from both sides of the spectrum: eg. Tony1 and Arthur Rubin. There is no doubt we need to fix the dismal state of year-related articles on wikipedia. Perhaps figuring out how to section off and organize year articles would be a good place to start. Wrad (talk) 18:04, 10 January 2009 (UTC)
 * For "deep" history articles, I see the approach you used to be a good approach as to write an encyclopedia article about a year, though I'm having trouble based on the RFC results of still when we link to those - this is potentially an issue though arguably these could stand alone with minimal linking from article space. When you get more contemporary (say 1969), I'm not sure if "per region" works, but there's bound to be a better breakdown.  However, we should still encourage on a separate mapping (through lists or categories) the "year-in-field" type articles/categories which would be adjunct to this and one would not have to worry too much about figuring out breakdowns. --M ASEM  21:39, 10 January 2009 (UTC)
 * Newer articles couldn't be broken down as well by region it's true, but we should also avoid making them too broken down. An article should be organized as an article, it shouldn't be organized purely out of concern for linking policy. More recent year articles are already split into dozens of different category articles, and splitting the article similarly into sections that match all of those subarticles would make it schizophrenic and indecipherable. We need to avoid that extreme. Wrad (talk) 21:46, 10 January 2009 (UTC)
 * Also, I don't know when 1346 should be linked too, either, and I concede that it is probably linked to too much, but I would prefer not having a bot decide, if you know what I mean. Wrad (talk) 21:49, 10 January 2009 (UTC)

If I understand User:Wrad correctly, he/she is concerned about a bot delinking articles that are themselves about dates. Yes, Lightbot does attempt to avoid delinking articles that are themselves about dates. One of the major problems that reduces the effectiveness of Lightbot is that it is not currently permitted to delink valid autoformatted dates in *any* article. I think delinking autoformatted dates is something it should do. Lightmouse (talk) 11:37, 11 January 2009 (UTC)
 * I would also like to register my opposition opposition to year de-linking. When a Wikipedia reader is reading an article for example a biography I think it is entirely conceivable that this user may click on the year of birth of the subject of the article in order to gain some insight into the state of the world at the time of the birth. From that article he/she can then zoom out to the decade or the century to gain a broad overview of the given time. He/she could also examine the events in any nearby year by clicking appropriately. It's called hypertext guys which was invented in the 1960s! The same applies to an article about any significant event. The global or regional context is relevant in many cases. Year de-linking will give the impression that historical events occur in isolation to each other rather than as part of an organic whole. Furthermore year de-linking will stifle the identification and observation of historical trends which human history is replete with. Consequently I would like to urge the reversion of all year de-linking carried out by bots so far and that de-linking only be done on a case by case basis. Davidzuccaro (talk) 06:04, 11 January 2009 (UTC)
 * I think you've missed the huge volume of discussion about this that has occurred over the past six months. Perhaps you could consult the archives to read the arguments the put this into perspective. WP has been slowly moving towards smart linking practices, away from the scattergun approach we had at the beginning, when we were like kids with a new toy. Every link comes at a slight cost (read up on it?). Links to seas of almost entirely irrelevant information are now discouraged. If you're keen to type in the four digits of the year of someone's birth: go right ahead. No one is stopping you. But we do not want every bio article to start with blued-out years that hardly anyone will click. Avoid the dilution of high-value links in the vicinity (especially at the opening) is a key way of making wikilinking work better. Tony   (talk)  07:34, 11 January 2009 (UTC)
 * David, it is a trivial business to go to the "search" text box, type the four digits of the relevant year and click the "Search" button—delivering the same result you are seeking without the whole heap of issues (including formatting) that accompany date-linking.  HWV 258  23:55, 11 January 2009 (UTC)
 * Well, most articles about dates don't use auto-formatting, anyway, so I guess we're ahead of the curve. :) Wrad (talk) 20:20, 11 January 2009 (UTC)


 * Indeed. Would you (and anyone else reading this) support Lightbot delinking autoformatted dates? Lightmouse (talk) 23:20, 11 January 2009 (UTC)
 * My only concern was with links in articles about dates, so I guess I would support. Wrad (talk) 00:02, 12 January 2009 (UTC)

Year, decade and century articles are sometimes useful - especially if what you are looking for is of great significance. I have sometimes used them to try to find out information - e.g. what was the important battle between the Tang Chinese and the Turks/Arabs...--Toddy1 (talk) 21:01, 11 January 2009 (UTC)

Italicizing g (acceleration)
Here on Talk:G-force, is something new get our attentions off dates. Our G-force article asserts that the unit symbol g for acceleration (a unit symbol now) is italicized. Whereas it might be a great idea if this were the case, it doesn’t appear to me that it is the case in the real world. Besides, such an assertion violates the basic principle of mathematics that unit symbols are always roman and variables are always italic. When we get over there to scratch each others eyes out (no bitting now), how about we agree in advance that just because italicizing g (for acceleration) might be a peachy-keen idea, it is not our job here on Wikipedia to advocate change by stating as fact something which is not fact. Can someone please come up with an authoritative citation that g (acceleration) is supposed to be italicized? All I can come up with is gobs of NASA usage with a roman g. Greg L (talk) 05:53, 13 January 2009 (UTC)


 * ''All discussion moved to Talk:G-force

Temporary injunction in Requests for arbitration/Date delinking
The following temporary injunction has been passed in Requests for arbitration/Date delinking;

''Until this case is decided or otherwise directed by the Arbitration Committee, all editors are instructed not to engage in any program of mass linking or delinking of dates in existing articles, including but not limited to through the use of bots, scripts, tools, or otherwise. This injunction is entered as an interim measure and does not reflect any prejudgment of any aspect of the case.''

For the Arbitration Committee,  Ryan Postlethwaite See the mess I've created or let's have banter 11:59, 13 January 2009 (UTC)

Date delinking arbitration
I've started a request at Requests for arbitration which those reading here may wish to comment on. —Locke Cole • t • c 06:41, 10 January 2009 (UTC)


 * Yeah, we were laughing about that this morning. Greg L (talk) 19:32, 10 January 2009 (UTC)
 * What do you mean "we", Kemosabe? — Hex    (❝  ?!  ❞)   20:38, 10 January 2009 (UTC)
 * Some of us besides you, Tonto. Greg L (talk) 00:05, 11 January 2009 (UTC)

(outdent) Depends on what type of timeline. See Timeline of the 1994 Atlantic hurricane season. You should probably ask User:Lightmouse about it. Dabomb87 (talk) 22:14, 10 January 2009 (UTC)
 * (More appropriate here, then there.) The "I refuse to comment" comments seem to be good grounds for an injunction against those people adding or removing any date links, either by bot, AWB, even if there were a prior consesnus.  They show a lack of respect for Wikipedia process.  — Arthur Rubin  (talk) 20:00, 10 January 2009 (UTC)
 * I have ample respect of admiration for the process (such as the extensive and wise work that went into this decision to reign in an editor who was POV-pushing on Cold fusion). I simply feel that Locke’s complaint is abuse of the system by a whiner who wasn’t getting his way. I fully expect ArbCom to refuse to get involved—in part—for that very reason (and because this is clearly a dispute over article content and how rapidly to fix that content). As for my refusing to be involved, it is a matter of principle. I’m not about to let one Mr. Locke Cole (who resides who-knows-where on this pale blue dot), dictate how I might enjoy my hobby here on Wikipedia or control my life in anyway whatsoever—particularly over such a whiny-ass, sore-looser stunt. WT:MOS and WT:MOSNUM are marketplaces where ideas are exchanged. Locke isn’t impressed with the direction things are going here right now (apparently you aren’t either). Well, so sad—too bad. In the grand scheme of things on a 1–10 scale of importance, this issue over which you two have your panties in a bunch is a nice solid 1.2 (or maybe 1.3). Concern over which restaurant Brad Pitt and Angelina Jolie ate at yesterday is about the only damn thing I can think of at the moment that might possibly be less important. I have zero interest whatsoever in getting swept up in your überdrama and wikilawyering. As is typical with your other arguments here, I take great pride and pleasure in pronouncing that I agree with nothing whatsoever in your above post, Arthur. Greg L (talk) 21:15, 10 January 2009 (UTC)
 * Is this bot seriously set up to remove all links to years and dates everywhere on wikipedia? If the answer is no, which dates is it not removing? Wrad (talk) 21:21, 10 January 2009 (UTC)
 * Wrad, with articles on en.Wikipedia, there are probably over a million linked dates. That is far too many for humans to manually correct. Bots are automated tools—like a power saw—operated by humans. The only practical way to fix our article content is do our best—within the limits of available technology—to ensure that bot activity is within the scope of the latest consensus view on date linking, and let bot-operating editors do their work unimpeded by obstructionists. Those dates that prove to have been improperly swept up in bot activity can easily be re&#x2011;linked by hand. Note that this will also give us an opportunity to let editors put some of these dates in articles’ See also sections. For instance, check this link out: Giovanni Fabbroni after several years of research chose to redefine the standard in 1799 to water’s most stable density point. Click on that link please. Where do you expect to go? When dates such as these are de&#x2011;linked, and we want to re&#x2011;link them, the existence of articles such as these need to be conveyed in a better way to readers, such as seen here in Kilogram. Greg L (talk) 21:51, 10 January 2009 (UTC)
 * I'm assuming that the answer, then, is yes, the bot is removing all date link everywhere on wikipedia? It would be a massive and necessary task to relink everything that needs relinking. That task could be avoided by identifying categories which are more likely to have appropriate links and have the bot avoid them. For example, have the bot avoid all articles in categories specifically discussing dates or which are timelines themselves. I shudder at the thought of the enormous task it would be for us in the Years project to relink everything when all we have to do is tell the bot to stay away from certain categories. Wrad (talk) 21:57, 10 January 2009 (UTC)
 * To answer your question Wrad, Lightbot, the only fully automated date delinker, removes year and decade links only (perhaps also century links). The AWB script and monobook.js script, which is used by humans, requires human oversight. Those scripts delink all chronological items and have the option of making the delinked dates the same format, either international and American. Additionally, these scripts have the function of delinking common terms, such as well-known geographical locations. The scripts allow for their users to view and modify their changes before saving. As an example, see what I did here to Armistice Day. I used the script to delink dates and common terms, then re-linked two dates, setting them so that autoformatting is disabled. As an aside, I have never seen a useful date link on an article, with the exception of holiday articles and date articles themselvs. Dabomb87 (talk) 22:06, 10 January 2009 (UTC)
 * I want to make sure I'm being understood. Date links always link to timelines. If you click on a date link within a timeline article, "Where do you expect to go?" (to quote Greg). If I was in a timeline article, I expect to go to another timeline article. That is completely in context with what I was reading. Therefore, there is at least one place in which date links are appropriate, and that is within timeline/date articles themselves. I am asking that the Lightbot, then, not be allowed to delink date links within such articles at all, ever. That should be a human issue. Wrad (talk) 22:11, 10 January 2009 (UTC)
 * This is a very good suggestion - if only everyone else in this discussion were as constructive and unbiased. Deb (talk) 18:59, 12 January 2009 (UTC)
 * Thank you Dabomb87. Wrad: I’m saying that there are clearly dates currently being de-linked that editors feel should be linked. I’m further saying that isn’t a big deal. No, I’ll go beyond that and say it is a very small deal. Further, it is actually a good thing because there are better ways to let readers know about our specialty year links like “this year in music” and “this year in science”, etc. If you want to know the details of how these bots work, stop by on the talk pages of those who operate them, such as Lightmouse. Finally, with regard to your …massive and necessary task to relink everything that needs relinking… and …shudder at the thought of the enormous task…, that too is A) your opinion, and B) is überdrama. Any reasonable reading of the RfCs shows that the current consensus is that it is a rare date indeed that ought to be linked; ergo, bot delink, and hand re&#x2011;link only those dates that a reasonable interpretation of the current consensus says ought to be linked. Greg L (talk) 22:16, 10 January 2009 (UTC)
 * Greg, I agree that almost all date links are useless. I believe I've said that several times. However, as a member of the Years wikiproject I find that very callous of you to say this really doesn't matter and is just uberdrama. I worked hard on 1345, 1346, and 1347, hoping to inspire people, and so far it is working, slowly but surely. We at the Years project work hard to improve the years portion of wikipedia. Far from being unimportant, year articles are among the most edited on the site, consistently. It would be a massive and unnecessary task to relink everything, and oh-so-simple a thing to avoid. I'm watching out for this project. We have enough things to do, and would rather not be stuck with tasks that could been avoided in the fist place. I would appreciate it if you would show at least a little empathy and help us out here. I have one other question: How is a link from one timeline to another "out of context"? Isn't it what you would expect? Why, then, should we have a bot delete link from timelines when it would be so easy to tell it to ignore them? Wrad (talk) 22:23, 10 January 2009 (UTC)
 * Very well, Wrad. I deeply respect your contributions to Wikipedia. I just looked at the 1345 article. Although I think something could be done with those first four, unwieldily paragraphs (they look quite a bit like Sewer cover in front of Greg L’s house), the rest is quite attractive and impressive work to pull together wide-ranging topics. There is too little of that (building and error correction) going on. And, though we don’t see eye-to-eye on the techniques being used to link to those articles, I completely agree that we need to find a good way to let readers know of the availability of these articles you are working so diligently to improve. The tradeoff here is that the old way of letting readers know of these ‘years’ articles was cluttering up our main body text with blue links. I strongly feel (and many others here too) that body-text links should be solidly germane and topical to the subject of the article in which they are linked. If I am reading an article on the years leading up to the Great Depression, I will be interested to know that Black Tuesday is something I can click upon for further reading (or perhaps Social Security). But if I read that The Bank of New York, in 1925 started loaning to individuals in a way…, I will find an enormous amount of information that is totally (colossally) irrelevant to the topic of Great Depression. It is irrelevant that the 1925 article includes such information as October 16 - Angela Lansbury [is born]. Similarly, if I am reading up on Louis VI the Roman, I am likely interested in Roman-related and nobility-related links. He married Cunigunde in 1345. I propose that since the 1345 article contains so precious little information relevant to to Louis V, but since Louis VI the Roman is clearly a historical article and since 1345 is all-things-historical, a much better way to let readers know of the availability of this information is to add a See also section to Louis VI the Roman and to put a link there that reads • Other notable events of 1345. I this is a much superior method and truly think more readers will be inclined to click on that and explore the article that are currently tempted to click on yet another one of our millions of blue dates. This technique is expanded upon in WP:Why dates should not be linked. Greg L (talk) 23:03, 10 January 2009 (UTC)
 * I completely agree with that entire post. I'm just concerned about date links within timeline articles. I don't think that it would be "out of context" to link one timeline article to another through a date link. I don't think that the bot should, say, remove links to 1904 from the 20th century page, see? Wrad (talk) 23:23, 10 January 2009 (UTC)
 * I agree, Wrad. You are saying that when the topic of the article is intrinsically historical in nature, it is entirely appropriate to link to another historical article. This is nuance I’ve long spoken of. I suggest we try to pull Lightmouse, Tony, and others into this discussion. The details of how to implement the general principles of the RfCs are quite important. They are important because we A) want to improve Wikipedia for our readership, and B) because there should never be activities that discourage hard-working editors from improving Wikipedia by taking all the fun out of it for them. The en.Wikipedia project has a wide variety of subject matter. Moreover, there is more than one tool in our arsenal of available techniques for making readers aware of the existence of other articles for further reading; the art is matching how germane and topical the link is to the technique employed. Greg L (talk) 23:39, 10 January 2009 (UTC)
 * If the bot were to avoid certain categories (e.g. any category that is simply digits, like 1966; or the same followed by 's'; or 'centuries') then we could look at those articles manually. I just looked at 20th century, as Wrad mentioned it. The date links there go to specific articles (like 1948 Arab-Israeli War, piped to 1948) or to a decade. Nobody would suggest removing those, and a bot wouldn't touch them anyway. Now, if you look at 1960s, the links at the top to nearby years are generated by an infobox, so no bot problem there, but the body contains about a dozen 'pure' date links, several of them seeming to be just for autoformatting (and should go). Of the rest, I'm once more skeptical about any value or relevance. Examine 1960s - the first link is to the article itself (resulting in an odd-looking bolding); the next is In 1963, with Betty Friedan's revolutionary book, The Feminine Mystique, the role of women in society, and in public and private life was questioned. So I click on 1963 only to find that among the 100+ events listed (and twice as many birth & deaths), just February 27 - Female suffrage is enacted in Iran is even remotely connectible, although I'm still no wiser about whether one influenced the other, or it was just coincidence again. And so on for the others. So, are all of those links adding anything? If not, even in this date-related article, I would argue that any uninvolved editor would believe it right to remove them. YMMV --RexxS (talk) 03:51, 11 January 2009 (UTC)


 * Wrad, there is a large navbox at the top of the 20th-century article, yes? Why link individual years in the list or running prose when it's staring in your face at the top? What I find difficult is the formatting of years and month–day items in these articles when they're constructed as lists. Bluing a lead item runs it into the very next item, which itself breaks a MOSNLINK rule. Why not bold them instead? Tony   (talk)  01:49, 11 January 2009 (UTC)
 * I really don't care as long as people can fairly and easily navigate through timelines. Wrad (talk) 02:52, 11 January 2009 (UTC)

I see that the WikiProject Years may be indirectly canvassing for people to make comments at the ArbCom event, by linking to this section. While at WikiProject Years (of which I am a member), I made this comment about merging year-articles into decade articles from many centuries ago. No one seems to engage with the idea, yet I believe it is compelling. Tony  (talk)  02:32, 11 January 2009 (UTC)
 * I don't believe I was canvassing. I believe I'm actually agreeing with most of the people here(?) Wrad (talk) 02:39, 11 January 2009 (UTC)


 * I'm not too concerned about this topic of removing date links, though I can see it has become a contentious issue. I am more concerned about Lighbot after seeing it changed a link on one of my watched pages and I went to its talk page to determine why.  Above, Greg L suggested If you want to know the details of how these bots work, stop by on the talk pages of those who operate them, such as Lightmouse, but Lightmouse's page doesn't describe the Lightbot's operation.  It mentions removing date links, but it doesn't describe the scope or parameters for removal. Many users are asking similar questions without adequate response that I can find.  Perhaps much of this debate could be resolved with transparency as to what the bot is doing.  Are there date links in certain areas or formats that will be skipped?  Will this bot constantly run, removing date links that editors restore?  I think many editors are concerned that the bot may the few legitimate links and prevent them from being restored quickly, easily, or permanently.  With users understanding the answers to these and other common questions, it may be unnecessary to defend or alter the bot behavior.  I find it difficult to support either side without understanding the bot my fully. —Ost (talk) 16:46, 12 January 2009 (UTC)

I will try to answer: It has been running since May 2008 and has made multiple runs for different purposes. On each occasion it runs, it may be running with a different code and with different purposes. Sometimes it is run in manual mode where the operator (Lightmouse) supervises each edit. The code is always being improved in response to new ideas and to feedback. It generally (but not always) stays away from articles that are themselves specifically about dates. The code is quite complicated and is very long. For example, just one piece of the code would print out to 15 pages of A4 paper.
 * Lightbot scope: The scope of Lightbot includes (but is not limited to) units of measurement and dates.
 * A 'date' is any sequence of characters that relates to time, chronology, or calendars. This includes but is not limited to seconds, minutes, hours, days, weeks, fortnights, months, years, decades, centuries, eras, and can be in any sequence or format.
 * Edits may add, remove or modify the sequence or format of dates.
 * Edits may add, remove or modify templates that involve dates.
 * Edits may add or modify autoformatting. Edits may remove autoformatting where it is invalid, broken or itself breaks a date for readers.
 * Edits may add, remove or modify links to dates.
 * A 'unit of measure' is any sequence of characters that relates to measurement of things. This includes but is not limited to units defined by the BIPM SI, the US NIST or any other weights and measures organisation or none at all. This includes but is not limited to time, length, area, volume, mass, speed, power.
 * Edits may add or modify metric or non-metric units.
 * Edits may modify the format.
 * Edits may add, remove or modify templates that involve units.
 * Edits may add, remove or modify links to units.

A lot of people appear to be saying that links for autoformatting should be removed. Ironically, that is the one thing that Lightbot is not permitted to do. It could easily be done by Lightbot and all we need is for people to support a change to the bot authority. Would you (User:Ost316) or any other user support removal of autoformatted links? Lightmouse (talk) 18:35, 12 January 2009 (UTC)


 * Thanks for the response. I appreciate this summary of the bot.  I tend to think if the bot should be running to cleanup date links, then it should act on autoformatted links as well due to their deprecation.  However, it may be nice if  could be designed to retain date autoformatting without links on the date to retain readability throughout regions. —Ost (talk) 21:29, 12 January 2009 (UTC)
 * Ost, to return to a basic issue: why is it that the order of month-day or day-month is such a concern? Part of the reason that date-autoformatting has been dispensed with is the wide acceptance that it's a solution in search of a problem. Why does it matter, I'd like to know? If one makes some new-fangled toy that autoformats for logged in, preferenced WPians alone, it's kind of elitist and stops us from housekeeping the mess of different formats within articles (and the wrong global choices). I've yet to see how a mechanism could format for every visitor, unless they explicitly choose which date preference they want before being allowed to access articles. That would be silly. This whole thing is not worth the time and effort. Only computer programmers and associated editors seem to be upset about the demise of a toy. Tony   (talk)  02:22, 13 January 2009 (UTC)


 * Tony please don't misstate things. WP:MOSNUM/RFC shows a good amount of support for "some form of auto formatting". And I've stated, many times, that I would only support a new system that addressed your primary concern (that it doesn't work for unregistered readers/anon editors). That you don't understand how it could work or why it's "worth the time and effort" aren't relevant. Every dev on MediaWiki is a volunteer except for I believe Brion and perhaps one other paid dev. But we've been over this... why do you keep repeating the same mantra? "Waste of time", "pointless", "no support", when all those items have been addressed previously? It doesn't help move us forward to have the same arguments repeated ad nauseam.  —Locke Cole • t • c 02:51, 13 January 2009 (UTC)


 * Excuse my profession, Tony; I didn't realize that knowing the power of coding was a bad thing on Wikipedia. Coming someone who seems to support scripts and bots for date delinking, I would think that you would appreciate computer programmers.  I'm not trying to be elitist, but I don't see the need to disenfranchise users if there is and adequate solution to maintain functionality.  Why should an editor care if users with date autoformatting settings see a page differently than the editor intended?  Does it matter if the user doesn't see the editors' date format if the user is comfortable with their chosen date format and everything is formatted the same?  In the course of reading some of months your arguments about date delinking, I've seen other users expressing concern for the date formatting.  I was unaware that autoformatting was also such a controversial issue.  I don't understand how implementing a form of autoformatting negates your goal of cleaning up pages.  The  template could be designed to autoformat only if the user has autoformat preferences set and otherwise use the editor's formatting.  Or it could be an optional parameter that can enable autoformatting.  I also think that you could generally base it off IP to guess users preference based on region to help unregistered users if desired.  There are likely other solutions, but the point is that it seems to me autoformatting could be implemented without date linking while still giving editors control of the page's date format.  I'm no expert on Wikipedia nor this topic as I just recently learned of it.  I may be over simplifying the changes that would have to be made to  or a similar template and I welcome comments that explain why my approach is flawed. —Ost (talk) 16:00, 13 January 2009 (UTC)
 * I have nothing against your profession: I have a patent on a computer application myself, although I'm not a programmer. And some of my friends are ... I'm curious as to why month/day and day/month order matters so much to you. And why you're seemingly unconcerned about color/our, travelling/traveling and the like. Your concern doesn't make sense to me. The problem I have with many programmers is that they appear to have a heightened concern for the products of programming, irrespective of their functionality/utility/consequences in real-life. A cost-benefit analysis—even a back-of-the-envelope one, would be a useful exercise here. Tony   (talk)  16:29, 13 January 2009 (UTC)
 * Your insinuation that I haven't considered a cost-benefit analysis is unfounded. While you see hardly any benefit (despite those given by others and me), I see hardly any cost (perhaps due to your lack of details on negatives of autoformatting).  My problem with this date autoformatting over the language is that the feature already exists and you're removing it because of its implementation.  It's presumptuous to assume it isn't appreciated by users and I think it's lazy not to keep the feature if there is a simple solution.  You choose not to entertain the idea of a simple solution even if it can be built into an existing template.  This problem doesn't bother me personally, but you seem to ignore users that it may impact; that does bother me.  Just because users don't visit discussion pages and weigh in on issues doesn't mean that they don't use features.  You assume that silence is acceptance when often in Wikipedia it is a reflection of users being oblivious to debates occurring.  You've evidently been arguing this topic for months and yet I only recently learned of it, quite by accident.  I apologize if this means I don't know the intricacies of the debate.  You may only be concerned with yourself, but I can think of broader user impact and convenience.  Instead of giving broad generalizations about how programmers act, feel free to give me the costs that autoformatting create; I've already provided the benefit of usability and not disenfranchising users of the feature.  I know changing the template takes time, but in all your months of your arguing, it could have been created, tested, and perfected several times by now.  Is there a major performance hit from autoformatting that I'm unaware of?  To me it seems autoformatting is just going away because of date linking was used for it and no one wants to modify the behavior, so I'm still waiting to be compelled otherwise. —Ost (talk) 18:20, 13 January 2009 (UTC)
 * "...in all your months of...". Are you aware that the "debate" at bugzilla?id=4582 has been going on for almost three years? And that's the re-opened discussion! Please understand that the current debacle is a direct result of the lack of action for so long that the only way forward is to take drastic steps to remove square brackets around dates (in order to rid WP of the bizzare linking of dates—something that is overwhelmingly supported by the community).
 * Thanks for taking an interest though. If you wish to "know the intricacies of the debate", have a careful read through this (closed) RfC and this (closed) RfC.  HWV 258  21:54, 13 January 2009 (UTC)
 * Again, I apologize for being new to the subject and not having your years of experience on it. I really only have the information I find traversing through links such as your 2008 RfCs and Bug report.  I was unaware that this was considered a bug nor that debates occurred at Bugzilla.  I appreciate you providing me with this information as I now have a better understanding of the issue.  In the RfCs, I still see mixed support for autoformatting by user preference and opposition to a new date autoformatting because of date linking (current implementation) or assuming that autoformatting would be forced on all dates typed.  But the Bugzilla report shows many users concern with the unwieldiness of using a template instead of markup along with server issues and guessing issues with if markup were used. With the debate dead at bugzilla, no markup or template seems likely to be agreed upon if autoformatting isn't directly requested and defined.  I still contend that autoformatting by user preference increases the user experience, and I acknowledge an agreed upon solution to provide this has yet to be agreed upon.  I am satisfied of the complexities of the issue, though the comments on this topic still reflect scope confusion that doesn't signify resolution. — Ost (talk) 15:03, 14 January 2009 (UTC)
 * I haven't thought of the bugzilla debate as being "dead", but now that that word has been used, I'm tempted to concur. In terms of "user preference" (and other related issues), have a read of this. Cheers.  HWV 258  21:59, 14 January 2009 (UTC)

Rather than autoformatting being foisted on the community by programmers and developers as a solution in search of a problem, I believe it more likely that a few too-accommodating techies who were trying to be helpful offered to resolve a WP:ENGVAR dispute by making both sides happy. (Someone made this point before but I can't remember who and where…) A cautionary tale with a double moral: to the programmers, it is "no good deed goes unpunished"; to the editors it is "careful what you wish for, you might get it!"--Goodmorningworld (talk) 04:57, 14 January 2009 (UTC)

Trying to stack outcomes through misinformation

 * Looking at the thread here on WT:WikiProject Years, all I see is Wrad trying to understand the issue after being mislead and spun up by Arthur Rubin. There, Arthur wrote [Looking at the RfCs,] I also see a consensus not to unlink month-day links from year articles. Arthur then did edits like this, restoring autoformatting and linking of dates to trivia articles, which I don’t believe is supported by any realistic and honest reading of the RfCs (kindly summarized here by Dabomb87). Stubbornly editing against consensus like this isn’t helpful. When are editors going to stop acting like fourth-graders? Greg L (talk) 03:07, 11 January 2009 (UTC)
 * I think the situation is fixed now. We changed the newsletter a bit. Wrad (talk) 03:11, 11 January 2009 (UTC)
 * It’s a pleasure working with you, Wrad. Goodnight. Greg L (talk) 03:40, 11 January 2009 (UTC)
 * I decided from the start not to do anything with dates on those types of articles. I could have done, and the yield per edit would be considerable just by targeting those. To me, these articles are an extended walled garden of the day-month articles. Ohconfucius (talk) 03:51, 11 January 2009 (UTC)

Coordinates format
Hello, I have traveled from WikiProject Geographical coordinates, where we seek wider opinions on whether coord should offer a N/S/E/W labeled format for decimal coordinates (example: [43.12_N_79.34_W_ 43.12° N 79.34° W ] ) either as an option or by default, or if the existing unlabeled format (example: 43.12°N, -79.34°W) is sufficient. Please comment there if you have an opinion on this. Thanks! --GregU (talk) 18:15, 17 January 2009 (UTC)

Microns and micrometres
A request for clarification here: I have no particular preference either way, but which out of "microns" or "micrometres" is preferred, and would this depend on the general type of article (e.g. "microns" seems to be used quite frequently in astronomy, even if "micrometres" is preferred elsewhere)? Icalanise (talk) 23:26, 14 January 2009 (UTC)


 * We used to have a principle written into MOSNUM that was slowly gutted during the “kilobyte v.s. kibibyte” holy wars until it appears to have suffered a fate worse than the Dead Sea Scrolls. It was FCL, for Follow Current Literature. The point of an encyclopedia is to educate and prepare readers so they can be conversant in the art. Then if they go get some books on astronomy, they will quickly understand the material with minimal confusion. I would say that if there is a rough mix of micron and micrometer (or micrometer) in the real world (as evidenced by current, most-reliable literature on that subject that is geared for a general-interest audience), it would be preferable to go with the formal SI name for the unit of measure. One might write as follows: the brown dwarf was detected at 2 micrometers (2 microns) before later writing but its companion was detected at 3 micrometers. If current literature on that particular astronomical topic predominately uses microns, then I would alias the link and write follows: the brown dwarf star was detected at 2 microns (with the link to micrometre). BTW, I’ll eventually be pushing for getting FCL back into MOSNUM; much of our most common-sense guidelines, such as write 22% and not the SI-compliant 22 % flows from this fundamental principle that underpins the proper practices for authoring content in any encyclopedia. Also, just because cc is not part of the SI never stopped motorcycle manufacturers from rating their engine displacements in ccs, nor Wikipedia from following current literature on that subject. A lot of bickering will be quickly resolved when editors are encouraged to follow modern, most-reliable literature. It is far superior to coming here and having someone say “since User:FireInHisHairTheSI-Nazi is active today, let’s see what his opinion is”. There is ample white space below for other editors to weigh in here so we can kick & gouge each other over microns v.s. micrometers now. Greg L (talk) 01:06, 15 January 2009 (UTC)


 * Waiving my Greg-given right to kick & gouge, I'll just point out that this reply was spot-on and that indeed the FCL principle needs to be readded. -- Jao (talk) 13:06, 15 January 2009 (UTC)


 * Micron is the older, "traditional" name and still sometimes used, but nowadays the SI-based name micrometre is much more common. -- Army1987 – Deeds, not words. 13:15, 15 January 2009 (UTC)
 * No doubt, Army. But that isn’t the case in all disciplines and that what we’re talking about here. We don’t start writing about a 750 cm3 Kawasaki motorcycle engine because the rest of the world, for the most part, has gone that way (or to mL); for articles on specific subjects, we follow current literature. Greg L (talk) 18:02, 15 January 2009 (UTC)
 * It doesn't matter, Greg. There is no discipline which never follows the current rules.  We have a perfectly good right to establish our own Wikipedia style in any of these cases.  "Current literature" is a standard impossible to address accurately, especially when it is not monolithic and it almost never is.  I was involved in one case here where an editor insisted that I shouldn't convert to standard units because it wasn't used in his field.  When I showed that it was, he then complained about the country in which the specific papers I had found were written.  After showing more examples, he had to admit that not only are the SI units used, they are in fact the units recommended by his own professional organization.  But most people don't have the werewithal to address these wild-ass claims that an editor is "following current literature".  Far better that we establish our own rules here on Wikipedia, our own standards.  We can use the standards of various standards organizations as our guidelines; in this case the General Conference on Weights and Measures (CGPM), the primary standard-setters of the whole world,  specifically abolished "microns" as an acceptable name for this unit back in 1964, more than 44 years ago.  We ought to do the same in our guidelines.  Gene Nygaard (talk) 14:56, 19 January 2009 (UTC)

There might be good reasons to permit a non-SI format. But permitting a non-SI format does not mean you have to ban SI formats. We don't want Wikipedia to be more hostile to SI than the outside world. Kawasaki itself does use SI formats "cm&sup3;" It is common enough to use that format in Kawasaki references on yahoo. It is common enough in other motor publications. Allowing FormatA is not the same as banning FormatB. Lightmouse (talk) 18:33, 15 January 2009 (UTC)


 * I see you haven't figured out the rules yet, Lightmouse. When GregL talks about "follow current literature", the usage in current literature is whatever GregL says it is.  So he says we cannot use "cm³", even though it is in fact often used.  Gene Nygaard (talk) 15:04, 19 January 2009 (UTC)


 * Like Jao I see no need for kicking and gouging, but I wish to point out that there was never a consensus for "Follow Current Literature". On the contrary, its unilateral addition to MOSNUM sparked an edit war that lasted more than a month. Thunderbird2 (talk) 18:54, 15 January 2009 (UTC)
 * No, be honest, T-bird. I used Follow Current Literature as the common-sense principle to illustrate how Wikipedia’s continued use of the IEC prefixes was terribly unwise. FCL got trampled into dust during all the vitriol because T-bird et al. at the time—and still do—wanted Wikipedia to use terminology like “kibibytes” (KiB) instead of the “kilobytes” (KB) used by the rest of the world. Smooth move: you try to purge FCL from MOSNUM because it stood in the way of your singular purpose in life. Give it up. We will never be using kibibytes and mebibytes and the rest of the IEC prefixes until after they are common and well recognized in the real world. It is irrational to keep agitating on this subject; we’re all tired of your incessant drumbeat on this issue. You’re flogging a dead horse. We’re moving on. Greg L (talk) 22:44, 15 January 2009 (UTC)
 * Don't tell me what I do or don't want. Instead I will tell you.  What I want is for MOSNUM to reflect consensus, and if the principle of FCL had been accepted its inclusion would not have sparked an edit war.  You are free to argue again for its inclusion now to see if consensus has changed.  That's how WP works. Thunderbird2 (talk) 16:48, 16 January 2009 (UTC)
 * Correction, the consensus is that you are being disruptive on this issue about what you think consensus is. This is because it has become apparent that what you think is consensus is actually "what Thunderbird2 wants regardless of what everyone else says". You kept on beating the same dead horse with your claims there was no consensus despite there being obvious consensus and despite that consensus being demonstrated to you numerous times. The question is, when are you going to comply with the RfC against you and stop repeating your obviously false claims about lack of consensus? Fnagaton 03:40, 19 January 2009 (UTC)
 * The technical articles I've seen, mostly in the area of electronics, generally use the symbol "μm" and don't spell the word out. --Gerry Ashton (talk) 19:43, 15 January 2009 (UTC)
 * Concur with Gerry's last. Older publications often used "μ" for "micron" but I can't recall seeing that in a decade or so, likely because the instrument makers (HP, Tektronix, etc) switched when "nm" came into common usage and authors followed. LeadSongDog (talk) 21:19, 15 January 2009 (UTC)

A further related point - would converting "microns" to "micrometres" (or "micrometers" if the article was in US English, say) in an article be against guidelines, in a similar way that converting between UK and US English would be? Icalanise (talk) 23:11, 15 January 2009 (UTC)


 * Oooh, gouge-eyes time. The unit is *officially* spelled “meter” in the United States (and micrometer, etc.). If the article properly uses American-dialect English, then be consistent with all spelling. And, to preempt those who say the name is universally and officially spelled “metre”, not it’s not. The French invented the metric system and the SI system that it became. In French it is spelled “mètre”. When the BIPM translates to English, they translate to British/International English and it becomes “metre”. That is the spelling for that dialect. For U.S.-dialect English, it is officially and properly spelled “meter”. Greg L (talk) 23:31, 15 January 2009 (UTC)


 * Yeah that was why I tried to imply either "micrometres" or "micrometers" - I'm not particularly interested in micrometres↔micrometers because the rules about dialects seem to be fairly clear, and that's not a fight I want to pick :-) I'm more interested in the conversions micrometres↔microns or micrometers↔microns. Icalanise (talk) 23:36, 15 January 2009 (UTC)


 * Micron (n) A unit of length equal to one micrometre. No longer in technical use. That says it all. If you're using microns, you're technologically obsolete. Accept that fact, take early retirement, and let the young whippersnappers take over.RockyMtnGuy (talk) 23:43, 15 January 2009 (UTC)


 * Chill out, no need to bite people's heads off. Regardless of whether "micron" is officially deprecated or not, the term does seem to still be in use in a small number of fields. As an editor primarily interested in astronomy topics, I have seen it in use quite a lot, e.g. the abbreviation 2MASS stands for "Two Micron All-Sky Survey" not "Two Micrometre All-Sky Survey" (and in no way am I suggesting we change what the article says it is). Icalanise (talk) 00:01, 16 January 2009 (UTC)


 * "Micron" is an obsolete synonym for "micrometre". Sure, some people still use it, but I think the issue is whether readers of the article will understand it. There are all kinds of oddball units still in use, but most people don't understand most of them. Micrometre is the current standard unit, and fairly generally known to the educated readership, so it should be the preferred unit.RockyMtnGuy (talk) 03:32, 16 January 2009 (UTC)


 * Seems like Greg's spoiling for another spelling fight. How about we get going with the "wikipaedia" name change?LeadSongDog (talk) 23:52, 15 January 2009 (UTC)


 * LeadSongDog: stop personalizing everything and grow up. RockyMtnGuy, you are ignoring why this thread got started as well as exactly what I wrote, which was very specific and conditional. Icalanise started this with "microns" seems to be used quite frequently in astronomy. I responded with a conditional:




 * I added the underlining. And the advise is good advise. It doesn’t matter if micron is obsolete if, for instance, modern infrared spectroscopy articles related to the Hubble Space Telescope to this day usually use microns. The operative word here is “if.” I don’t know what the facts are in this particular issue that Icalanise is speaking about, I don’t care what the facts are, I don’t need to know in order to address the principle. The world-wide oil industry prices oil in cost per barrel and Wikipedia follows this practice with Price of petroleum: barrels, not cubic meters. It doesn’t matter if a 44-gallon (US liquid) barrel of oil is obsolete and poopy and unscientific. If any discipline generally uses a unit of measure or spelling for that measure or whatever, Wikipedia follows current literature. We’ve done it all along, we just haven’t memorialized that in a guideline because of there are a few User:SI Nazis who editwar here. Greg L (talk) 00:09, 16 January 2009 (UTC)


 * It's a 42 (US) gallon barrel, and it's not universally used. Europeans use tonnes, Canadians use cubic metres. The US press converts everything to US dollars per US barrel, but that's an oversimplification for the benefit of the US readership. The world is much more complicated than you read in the papers.RockyMtnGuy (talk) 03:32, 16 January 2009 (UTC)


 * Somebody's sarcasm detector needs a tuneup! Or perhaps "Oooh, gouge-eyes time" is the way folks say "Hi, how ya doing?" in your part of the world? Really, anyone who thinks they need to stockpile "dispute resources" may just be taking this stuff too seriously. Have a nice cuppa Joe.
 * There's no need (or even possibility) for uniformity on this at WP. Try as you might, "anyone can edit" and will. A guidline on this is just needless WP:CREEP that will mostly be observed in the breach. Let it go.LeadSongDog (talk) 01:01, 16 January 2009 (UTC)


 * In Italian TV news programmes, the price of petroleum is usually stated in dollars per gallon, too. (Last year, when the price of oil was quickly rising, but the value of a dollar was falling more quickly, someone pointed out that doing that gives the wrong impression to the audience: the price of oil in euros was actually falling, but petrol sellers could use "the price of oil is rising" as an excuse to increase the price in euros of petrol, as the general audience who is only told about the price of oil in dollars wouldn't see how they are being teased.) -- Army1987 – Deeds, not words. 12:17, 16 January 2009 (UTC)


 * Icalanise, regarding your 23:36, 15 January 2009 (UTC) post, I don’t get it. Micron and micrometer are different names for the same unit. What is to convert? Or are you wondering about the atomic-level details of the best way to show the *conversion*(?), in which case, your instinct is as good as mine. Greg L (talk) 00:14, 16 January 2009 (UTC)
 * As in conversion between the words. So to be specific: is an edit which replaces occurrences of "micron" with "micrometer" or "micrometre" in an article in a field in which the term "micron" is used frequently in line with guidelines/policy? Icalanise (talk) 00:20, 16 January 2009 (UTC)
 * Changing to a formal SI name is probably better supported by current MOSNUM guidelines precisely because of its current silence on the basic principle of FCL. As I mentioned above, Wikipedia has a de-facto practice of FCL, MOS and MOSNUM just haven’t memorialized this reality (see Yamaha YZR-M1) with an expressed guideline. It doesn’t sound wise to me for an editor to do what you are asking about, and I don’t know why anyone would want to do such a thing if current literature on the subject predominately has a specific practice. It’s not like one needs to employ Jimbo’s own personal contribution to Wikipedia: Ignore all rules (in order to improve Wikipedia) in order to simply exercise a common sense principle of technical writing. Why are you asking? I do have e-mail. Greg L (talk) 00:30, 16 January 2009 (UTC)
 * I'd come across an edit where micron was converted to micrometres and was curious as to whether doing that was encouraged or not. Icalanise (talk) 00:38, 16 January 2009 (UTC)
 * I see. Greg L (talk) 00:51, 16 January 2009 (UTC)
 * It is acceptable to change it. Gene Nygaard (talk) 15:09, 19 January 2009 (UTC)

Let me just throw a layman's spanner (U.S.: monkey-wrench) in the works from Left field. When I was not-very-successfully taking a BSCS sophomore high-school biology course in the sixties, I seem to remember using "micro-millimeter" (Br: micro-millimetre), abbreviation μ or μm, as the name for a thousandth of a millimeter/re, or ten Ångstrom units.

And believe it or not, there's actually a plausible (if hardly decisive) practical reason to avoid micrometer, because it's the name of a very-commonly-used physical measuring device (emphasis usually on the second syllable). The important thing, as suggested above, is to let someone who has heard of, has seen, is familiar with, or has previously used measurements in microns (or for that matter micro-mm or tens of ångstroms) know how something measured in micrometres (micrometers) compares.

—— Shakescene (talk) 10:53, 18 January 2009 (UTC)


 * Compound prefix units such as "micro-millimetre" are now forbidden by the SI standard. And, outside of the US, "micrometer" is used for the measuring device and "micrometre" used for the unit of measure - for the exact purpose of eliminating any confusion between the two. For instance, in Canada the official standard is that "theater" and "theatre" are variant spellings for the same thing and either can be used, but "micrometer" always means a measuring device and "micrometre" always means a unit of measure.RockyMtnGuy (talk) 21:15, 18 January 2009 (UTC)
 * It doesn't matter if something "is forbidden" by some organsiation if that commandment is mostly ignored by the secondary sources we cite for articles, what matters is what is used in the literature covering a subject.Fnagaton 03:43, 19 January 2009 (UTC)


 * The "millimicron" or "micromillimeter" also wasn't equivalent to an angstrom, but rather to 10 angstroms. It is, in both forms and various hyphenations, an obsolete equivalent to nanometers, just as microns are an obsolete equivalent to micrometers.  We shouldn't allow micromillimeters or millimicrons or microns in general use.  Gene Nygaard (talk) 15:12, 19 January 2009 (UTC)

It is funny that when some fields switched from Non-SI units that used ten raised to a number other than an integral multiple of three, like myriametres, they came to use other units that became obsolete. This is the case for visible wavelengths that once were measured in Ångströms in early spectroscopy. Microns (µ) would have psychologically felt closer in written, spoken, and symbol length to Angstroms (Å) than micrometres (µm) but recent student books tend to actively discourage micron use as do some IUPAC pages.

I prefer micrometres because I have done calculations that multiply micrometres by units that broke down to metres and, should I have written µ m instead of µm m, I might have mistaken µ m for µm. In addition, when one has poor English skills such as dyslexics or non-native speakers, micrometre is a clear, expressive term but micron is not. I do not see any reason to continue using an old psychological crutch like microns when software that forced filenames and such into a limited number of characters (often 8) was superseded over a decade ago and for most people by now has gone out the windows ^_^. In answer to the 'should I convert' question, I think we should sometimes write microns [micrometres] in citations for those with poor English skills because the micron was not such a big hit in other languages.

By the way, thanks for raising the micrometer/micrometre issue. I come from a mixed Commonwealth/US background, so I use that issue to decide that it is better to always use the "-re" spelling despite examples of both throughout Australian texts. Should it ever be unclear which you are writing about and the units column is the last significant figure, just put a decimal point after it to emphasize that the number of micrometres could have been fractional which is highly unlikely of instruments like micrometers. :)--Thecurran (talk) 13:59, 19 January 2009 (UTC)

Sq km vs km²
I don't want to revive something that must have been discussed exhaustively many times before; but on the other hand, I don't really have the patience to wade through a hundred MoS Talk archives, or the technical skill to search those archives in one batch. To me as someone who lived in Britain before metrication and then moved to the United States, something like km² looks like something out of my high school algebra and physics classes, which suggests that it's not intuitive or readily apparent to a significant number of English-language-Wikipedia users in the U.S. and other places that still use "customary" measures, for example the many schoolchildren and other readers who haven't yet reached (or may never take) those algebra and physics classes—whereas "square km" (avoiding the Br./US spelling difference), "sq. km" or "sq km" makes sense by analogy with square mile, sq. mile or sq. mi. I just don't know what the ordinary non-mathematical person in metric places, especially English-speaking ones, would be familiar with. I can see that, for example, a German would not recognize [kilo]mètre carré, so perhaps km² is used more on the Continent and in other non-Anglophone countries. (I see that my 2004 Petit Larousse Illustré, a French desk cyclopedia, uses km². But what's obvious to scientists and Francophones might not be so obvious to all of the Anglophones who use en.Wikipedia.) That doesn't mean by any means that I want to abandon, deprecate or discourage km², but that I feel that in non-scientific articles, editors should have the option of using "sq km" or "sq. km" according to their best judgement, particularly in places such as table headers where there's no first spelling-out of "square kilometer[s]" or "square kilometre[s]". At the moment, "sq km", etc., is reverted to km² by browsers such AWB (WP:AutoWikiBrowser). The Manual of Style—as is so often forgotten—is in principle mainly for guidance, so I'm sure I could argue something like WP:Ignore All Rules, but I'd prefer not to get into such wrangles (let alone suggest a change to the Manual) without understanding the reasons given on all sides about this question.

—— Shakescene (talk) 06:48, 18 January 2009 (UTC)


 * I think you’ve raised a very good point and your instinct is spot on. Wikipedia tends to look like a mathematics handbook rather than an encyclopedia. IMO, we should write “240 square kilometers” more often in articles, and only resort “km2” (after a proper parenthetical introduction) if the unit of measure appears frequently enough in the article that it is tending to be ponderous. Greg L (talk) 09:11, 18 January 2009 (UTC)


 * As Greg says, it's usually best to spell units out unless they are used very frequently (don't worry about spelling variation, we have ways of handling that). However, from the perspective of someone who grew up with the metric system, "sq km" does not seem more humane than "km2", it seems downright odd.  We just don't use "sq km" in metricland and, with all due respect, it is for us metriclanders rather than you footpounders that the encyclopædia has such units. So what I'm putting to you is that for those of us to whom "km2" is not intuitive or readily apparent we can provide conversions to square miles (and abbreviate those to "sq mi").  I'm suggesting that not only need we not mention the option of using "sq km" but we should maintain the guideline against it. Let's not model our metric abbreviations on what's done with customary ones, it's not how it's generally done in the real world. J IM ptalk·cont 11:08, 18 January 2009 (UTC)


 * I agree with Jimp. Everyone will understand at least one of the measures in 54 sq mi (21 km2) or in 21 km2 (54 sq mi). Using an abbreviation such as sq km is no advantage over km2, as everyone who knows how long a kilometre is will also know what the 2 means, and someone who doesn't know what the 2 means will probably be very unfamiliar with kilometres. The form sq km is also potentially more confusing, because it's not used in the "real world" and thus readers will be more unfamiliar with it. In cases such as measures without a conversion to square miles, where km2 can be confusing to some readers, the solution is spelling out square kilometre in full, not making up an abbreviation which can only be confusing to many more readers than km2 is. -- Army1987 – Deeds, not words. 16:05, 18 January 2009 (UTC)


 * I come from a country with two official languages, and km2 is understandable in both of them, whereas sq km is not - and is used in neither official language. That is the whole point of SI notation, to make symbols internationally understandable. The abbreviation sq km is not internationally understandable, and therefore should not be used. The fact that km2 looks like something from your high school physics textbook is a good thing, not a bad thing.RockyMtnGuy (talk) 19:02, 18 January 2009 (UTC)


 * "sq km" is nonstandard, ugly, unfamiliar, and unlikely to be understood by people who are familiar with the metric system. Either "square kilometres" or "km²" are fine, and the appropriate one should be chosen according to context. —AlanBarrett (talk) 09:27, 19 January 2009 (UTC)


 * I agree with AlanBarrett and RockyMtnGuy using "sq km" in articles would be a retrograde step.Flaviusvulso (talk) 09:42, 19 January 2009 (UTC)

Poor old ArbCom
See this section. Tony  (talk)  15:05, 15 January 2009 (UTC)

I suggest you leave them to do the arbitrating and leave the rest of us to get on with contributing to the encyclopedia. Deb (talk) 17:46, 15 January 2009 (UTC)
 * I'll do nothing of the sort. This is an interactive project in which cross-feedback is encouraged. Did you say you were tired? How about logging out? Tony   (talk)  18:03, 15 January 2009 (UTC)


 * Sorry. Too busy creating articles. Deb (talk) 18:07, 15 January 2009 (UTC)
 * Suggest disengagement. Deb, remember that Tony is a preeminent FAC and FARC reviewer. A healthy bit of calm discussion is always nice. NuclearWarfare  ( Talk ) 22:00, 15 January 2009 (UTC)


 * Well… I was going to ignore this thread until I saw Deb rattling Tony’s cage with this “I’ve got so much more important things to do in life than fret about trivial issues”-stuff. Unfortunately, I think you are both right. Tony should leave that petty name-calling to the arbitrators; they’ll see through it, recognize the fourth-graders responsible for it, and focus on the few salient issues. Tony’s reacting to being personally attacked is perfectly human and understandable; he, at least, gives a crap about his reputation (unlike some, who obviously don’t care what anyone thinks). After Tony has been on the receiving end of so much slander, I’m sure he’d like to cordially invite some of the editors there to do something to themselves that isn’t generally considered to be physically possible. But, Deb, your dismissive, flippant *I’m a big-picture kinda guy who seems to be uniquely alone in improving Wikipedia*-tone, is uhmm… unimpressive (even though I enjoyed your 18:07 post). And finally, Tony is right; the latest ArbCom looks like a bunch of third-graders have run amuck and taken over the place. What Tony needs to understand is this: That’s GREAT! Greg L (talk) 23:05, 15 January 2009 (UTC)
 * Time and again my block log is held up in the air like it's the Heisman Trophy. Of course my block log is largely irrelevant. Anyone objectively looking at it will note that most blocks are undone and some of the entries are of a technical nature (autoblock was broken, and one was undone because the admin didn't understand the edit I made, he even noted it shouldn't be held against me). —Locke Cole • t • c 03:41, 16 January 2009 (UTC)
 * No, “most” of your blocks (here) were not unblocked. And the portion that were unblocked weren’t because the blocks was in error (as was the sole block for Tony), they were unblocked because you and/or the other editor showed contrition—which I’ve unfortunately never seen from you here. Your block log betrays the whole reason so much bickering has transpired over a practice which has clear community consensus. Your record paints an unambiguous picture of an editor with a chronic propensity of being bullheaded and of being uncivil to others. In short, you don’t “play well with others.” Tell me I’m wrong. Or are you just a “misunderstood genius” and you suffer only from knee-jerk reactions from admins? Since the community consensus was so clear with regard to how it should be a (very) rare date that is linked in body text, you decided your only available recourse was to resort to wikilawyering in an attempt to squelch the most effective tool in our arsenal (prolific editors like Lightmouse and his exotic bots). Earth calling Locke Cole: It doesn’t matter what happens this week or next; you are fighting a battle in a loosing war. The community consensus on date linking is abundantly clear; your arguments here are unpersuasive and you won’t be able to reverse that. The community consensus that bot operators don’t need permission from you and others to do their thing is also clear; it just might need to be more clearly proven in yet another RfC; ergo, the part about you loosing the war in the final analysis.  Greg L (talk) 18:37, 16 January 2009 (UTC)
 * All the time, without combing through the entries in Cole's block log, I had been wondering about how he appeared to know so much about the ArbCom process. Then, it became clear that Cole had been there before, and had received a one month block for stalking. About his block record, sure, there were "some of the entries are of a technical nature", but the most illuminating entries were not - and there were a few of these. Most entries were for edit warring, and user harassment. Your record could not be more relevant, for I and a few others had received a good dose of same under your hands. Now he risibly moves on ArbCom as if he was the victim, portraying Tony (predominantly) evil. Can you still defend your block record now, Cole? Ohconfucius (talk) 04:27, 16 January 2009 (UTC)
 * My past misbehavior does not excuse Tony (or you, or Greg L) from misbehaving now. —Locke Cole • t • c 04:42, 16 January 2009 (UTC)
 * Locke is correct here. — Hex    (❝  ?!  ❞)   11:52, 16 January 2009 (UTC)
 * Aw, geez—why don't you give him a paper cut and put lemon juice on it while you're at it. And just when I thought some degree of calmness was sneaking back into things. :-(  HWV258  04:46, 16 January 2009 (UTC)
 * Because his block log underlies the whole reason why we are engaged is a pitched battle over a common-sense change. See my above post. Greg L (talk) 18:40, 16 January 2009 (UTC)
 * How, pray, am I "misbehaving" now? That is a term usually used reserved for children. In what way have I been childish? Ohconfucius (talk) 06:10, 16 January 2009 (UTC)
 * My fault, I started this (kind of). Let's try and forget that there's an arb-thingy on, shall we? Deb (talk) 13:01, 16 January 2009 (UTC)


 * You mean the "Get Tony Page"? Tony   (talk)  14:34, 16 January 2009 (UTC)
 * No, try to see if Tony (and some others) can behave like civilized human beings: a pledge to that effect should still, since ArbCom is not yet coherent after their internal troubles, defeat the proposed  sanction against him. Septentrionalis PMAnderson 20:58, 18 January 2009 (UTC)
 * That sounds like a personal attack. Tony   (talk)  09:50, 20 January 2009 (UTC)

Date ranges
I see no decision on date ranges, so I will WP:BB and enter one. Here it goes.
 * In dmy, if a range of dates occur in one month, 8 – 17 January 2009, 8 January – 17 January 2009, and 8 January 2009 – 17 January 2009 are all acceptable, though the shorter the better. 8 – 7 January 2009, on the other hand by shortening a datum from two significant figures to one, is unacceptable. Similarly for mdy, January 8 – 17, 2009, January 8 – January 17, 2009, and January 8, 2009 – January 17, 2009 are acceptable but January 8 – 7, 2009 is not; the second is most preferable because it leaves at least one full date for searching. When spanning months, dmy users may apply 28 January – 3 February 2009 or 28 January 2009 – 3 February 2009 but not 28 – 3 February 2009; the shorter the better. mdy users may apply January 28 – February 3, 2009 or January 28, 2009 – February 3, 2009 but not January 28 – 3, 2009; the shorter the better. When spanning years, one format alone is used for each. dmy is applied as 8 December 2008 – 17 January 2009 and mdy is applied as December 8, 2008 – January 17, 2009.

How was that? Feel free to edit it to make it 'flow'. :)--Thecurran (talk) 03:43, 17 January 2009 (UTC)

Okay, I cannot WP:BB here because I need admin privileges. Please help, admins! :)--Thecurran (talk) 03:52, 17 January 2009 (UTC)


 * You wouldn't need a space if the elements separated by the dash have no space (i. e. just the day): e.g. 8 January 2009 – 17 January 2009 but 8–17 January 2009; January 8, 2009 – January 17, 2009 and January 8 – January 17, 2009 but January 8–17, 2009. Also, the suggestion not to write 8 – 7 January 2009 or January 28 – 3, 2009 to mean 8–17 January 2009 and January 28 – March 3, 2009 sounds like WP:Don't stuff beans up your nose: nobody would do that anyway, so, better not to mention that at all. -- Army1987 – Deeds, not words. 09:35, 17 January 2009 (UTC)
 * i concur with the WP:Don't stuff beans up your nose assessment, which leaves us with something like:
 * In articles using the dmy format, if a range of dates occurs in one month, 8–17 January 2009, 8 January – 17 January 2009, and 8 January 2009–17 January 2009 are all acceptable; for articles using mdy format, January 8–17, 2009, January 8 – January 17, 2009, and January 8, 2009 – January 17, 2009 are all acceptable; in most cases the shorter forms are better.
 * When a date spans months, the dmy format is either 28 January – 3 February 2009 or 28 January 2009 – 3 February 2009; in mdy format, either January 28 – February 3, 2009 or January 28, 2009 – February 3, 2009 are acceptable; again, the shorter forms are generally preferred.
 * When a date spans years, dmy format is 8 December 2008 – 17 January 2009 and mdy is December 8, 2008 – January 17, 2009.
 * something like that, anyway - i've no doubt messed up the spacing but i trust Army1987 will adjust it as necessary - thanks Sssoul (talk) 10:47, 17 January 2009 (UTC)

OK. Let's give a try: Date ranges

For date ranges, the following formats can be used: The shorter formats are usually preferred ; the exception being the shortest mdy same month format, "January 17–24, 2009", because it obstructs text searches for both start- and end-dates. -- Army1987 – Deeds, not words. 11:22, 17 January 2009 (UTC)


 * None of this is necessary. All that are required are two simple rules, and they're there already.
 * Space the en dash when there's an internal space in one or both elements; otherwise, don't space it.
 * Consider avoiding repetition ("January 3–17" rather than "January 3 – January 17"). Tony   (talk)  11:26, 17 January 2009 (UTC)


 * Where are those rules Tony? I can't see them in the date ranges section, at least not explicitly. Woody (talk) 12:29, 17 January 2009 (UTC)


 * Take a look at WP:Manual of Style. Hope that helps. --RexxS (talk) 13:55, 17 January 2009 (UTC)


 * In auditing the date ranges in military-history pages, I was aware of how such ranges—which are very important in that field—are poorly formatted, right at the top of articles. I wrote a brief section into the style guide at the MilHist WikiProject, which was well-received. Tony   (talk)  15:28, 17 January 2009 (UTC)


 * With Wikipedia, I have come across the "If I tried to do it, someone else will, too", line of logic many times and it always seems to ring true. Now, if you folks agree that it is explicitly clear on the WP:ENDASH section how all date ranges should be formatted, I think it would be sensible to have a link to that section in MOS:NUM, simply because I came looking for how but could not find it, so others must be doing the same (but probably not WP:BBing and asking why) and will continue to do so.


 * Personally, I think WP:ENDASH does not have enough information to construct the rules you normally abide to. I wrote the beanish phrases above because only a few subsubsections below, I saw WP:YEAR asking us not to abbreviate a year into three digits or one, which at first glance is beanish, but I have seen it in adult classrooms and community newspapers. Beyond which there is always room for dyslexia and typographical errors, so I tried to think of which "mistakes" would make a well-read person cringe but are likely to be made by someone poorly-versed in English (but with valuable subject knowledge) and tried to prepare for those contingencies. :)--Thecurran (talk) 20:20, 18 January 2009 (UTC)


 * BTW, sorry for misusing the endash spacing. between numbers. I find the lack of spaces, visually disturbing but I do not wish to override convention. I seriously have seen it both ways many times (even here) and thought it was up to editor preference. Now I see we have had this rule for a while.


 * I like User:Army1987's graphical approach. My only qualm is that I think the "January 17–24, 2009" one should be discouraged simply beacuse it cannot be parsed for the start-date, "January 17, 2009", nor the end-date, "January 24, 2009", like each of the other entries can and we are text-searched a lot. What do you guys  think? :)--Thecurran (talk) 20:35, 18 January 2009 (UTC)


 * I strongly dislike the addition. We've gone through that before, and it clearly fails both common usage and common sense.  — Arthur Rubin  (talk) 01:02, 19 January 2009 (UTC)


 * I used "ins" and "small" in User:Army1987's graphical approach to address my qualm. The small text has the lowest importance but may solve FAQ's. :)--Thecurran (talk) 20:48, 18 January 2009 (UTC)
 * There is no need for a huge blow-out of advisory text: just the two principles I entered above, with a few brief examples. Tony   (talk)  08:24, 20 January 2009 (UTC)
 * Sure I am up for that addition. I certainly think something should be entered. Does anyone else think that January 17–24, 2009 should be avoided because it causes problems for parsing? :)--Thecurran (talk) 10:32, 20 January 2009 (UTC)

Just what did the RfC on month-day date-links show?
I've examined the third of the "detailed" RfCs. It has been trumpeted as indicating consensus for "at least some form of month-day linking". Let's look at why this is very misleading, and again why we need to be very careful in the way we frame RfC—indeed surveys of any type. This one is by far the most problematic of the three I've looked at.



Summary

 * Allowing only three possible choices significantly swelled the numbers in the middle "certain cases" category.
 * The language at the opening may have skewed the results towards the favouring of linking.
 * More than 90% of respondents appear to have a very conservative attitude towards the linking of month-day dates.
 * There may be consensus to allow the piping of a few annual events to month-day pages, but this RfC demonstrates little public favour for more than a very small amount of month-day linking as exceptional cases.

The available choices distorted decision-making
In their rush to prepare the RfC, the authors did what I might have first thought of doing—giving everyone just three choices: that month-day links should:
 * 1) always be made;
 * 2) be made in certain cases; or
 * 3) never be made.

While on the face of it this seems reasonable, the tripartite choice placed many participants in a difficult position: not wanting to support an inflexible, total ban on the linking of any class of item (Choice 3), they were forced to put themselves in the middle category, that month-day date links be made "in certain cases". There is clear evidence of this conundrum; one of the more explicit pleas was from Pagrashtaki— "I cannot support the phrase 'Month-Day links should never be made'. I would agree that, in general, Month-Day should not be linked. I cannot quickly think of a high-value Month-Day link (25 December on the Christmas article is not high-value in my opinion) [Pagrashtak]" In good faith, I think, a stated aim of the authors was to encourage feedback and discussion within these three categories; but they may not have been aware of how this aim worked against the drawing of definite conclusions from the survey—witness the difficult task of sorting out within that middle category users' actual attitudes to the question at hand. I've attempted to do this as best I can for the more than 50% of participants who placed themselves in the "in certain cases" category, :
 * Very rarely: some 22 participants. Many used these actual words (one said "Very very rarely", and one "Very very very rarely"); the rest I've included in this subcategory because they used equivalent language. Two mentioned exceptions such as "April Fools Day" and "All Saints Day", and two might well have been appropriately classed as "Never" (one of them User:It is we here), although I didn't shift them.
 * Rarely: 12 users.
 * Specific events: 17 users did specify that annual events such as "All Saints Day" might be linked, and four mentioned dates of birth (one of them death-date as well); three people felt that month-day dates might be linked in infoboxes (one of them in the lead, too). Total of ~ 23.Note 1
 * Unclear, leaning towards mostly: Four, but including the remark, "only where of value" [BarkerJr].
 * Mostly: two participants, who (conversely) couldn't bring themselves to declare "Always" (Choice 1).


 * Note 1: Critically, however, the distortions imposed by being funnelled into one of three choices were all too evident in numerous responses – for example, "Otherwise very rarely" [RainbowOfLight] and "But other than that, it's slightly pointless. For example, I sometimes read articles on fictional characters and soaps that link a date. It makes the whole thing seem redundant." [londonsista].

The basic breakdown was presented a while ago: However, the psychological funnelling of large numbers of users into the middle (Choice 2, "in certain cases") has made such a count highly misleading. Given that the style guides do not speak in terms of total proscription of any links (words such as "normally" and "generally" are used), four or even five categories—more finely graded—would have yielded less distorted results, and these subcategories I've identified would have been much less of a structural feature. It's hard to say exactly, but my reckoning is more like this:
 * 1) 5 Always [4.2%]
 * 2) 63 In certain cases [52.5%]
 * 3) 52 Never [43.3%].
 * 1) 7 – 5 Always, plus 2 probably more appropriate here than in the middle category ([5.8%])
 * 2) 4 – Leaning towards the previous category (3.3%)
 * 3) 23 – Only "All Saints Day", "April Fool's Day", etc ("annual events") [19.2%]
 * 4) 86 – Don't link, with rare or very rare exceptions (52 Never, plus 34 rare/very rare, who were put off by the extreme framing (i.e., "no exceptions") [71.7%]

Conclusions
More than 90% of respondents appear to have a very conservative attitude towards the linking of month-day dates. Typical comments were ""We don't need a world view of what was going on that day in order to better understand the subject of the article." and "These articles are almost trivia collections." It appears that there may be consensus to allow the piping of a few annual events, at editors' behest, but the conclusion that was drawn by more than one editor that this RfC demonstrates a public favouring of the linking of month-day items "in some cases" (vague) needs to be re-examined. I propose that piped links to annual events such as "All Saints Day" be explicitly permitted in the style guides; however, the overwhelming evidence favours no more than this.

Distortions in the opening text
Two other factors may have skewed the results to favour linking.

POV wording: The opening "Background" contained a major error (as far as I can see), and the title and formal question contained a hidden POV assumption. Rather than using NPOV wording, such as:
 * "Should month–day links be made in articles; if you believe they should be made, comment on when this should occur."
 * the formal question encouraged the false assumption that these links are or should be made, by asking only when they should be made:


 * "When should Month-day links be made in articles?"

The wording implied when, not if, so it was almost a fait accompli that month-day dates should be linked some of the time.

Apparent error of fact: first-time link rule in guideline? This statement appeared:
 * "It should also be noted that per the current style guidelines for linking, normally an article should link to a date – if at all – only the first time the date appears in the article."
 * The link was to the top of the MOSLINK page in general; the complete absence of the claimed guideline therefore required the participant to hunt around to try to find the statement referred to. More likely, most participants accepted this apparnetly false claim at face value. This had the distorting effect of encouraging users to believe that the linking of month–day items is already mandated at the top of articles. This was not and is still not mandated in the style guides.

Final comment: This is not a blaming exercise; it is all too easy to unwittingly introduce skew into a survey. Tony  (talk)  14:28, 18 January 2009 (UTC)


 * FWIW, as that question was not edited much beyond my initial draft, my point of the "should, then if" instead of "when" was that for the purposes of a first-run RFC to determine several date linking issues, it did not make sense to try to work out the exact details of when date fragments should be linked in light of trying to strip autoformatting dates, only if the general attitude was always, sometimes, or never; if consensus pointed to either absolute choice, then Lightbot et al would know what to do with those links. I fully expected that further discussion would be needed to establish that if it was "sometimes" there would need to be determination of when. Thus, there was no intended distortion, it was only trying to get clarity per the aim of what Lightbot and others should be getting ready to do. --M ASEM  15:25, 18 January 2009 (UTC)
 * Whatever the claims of "distortion", most of the "sometimes" and some of the "never" said something like rarely, which is the important point. Tony's effort to construe a majority for rarely as consensus for never is curious; it should be obvious that there is no consensus. Masem's further discussion may produce one; but in the meantime, why don't the minority that would like to routinely link, and the minority that is blinded by the sea of blue in linking May 1 in May Day go back to the showers, and decide whether it is worth coming up with better arguments to win so small a prize? Septentrionalis PMAnderson 21:05, 18 January 2009 (UTC)
 * That's gobbledygook. Did you read what I said? I made no claim that there was consensus for "never"? Do not twist my words for your own purposes. Heavens, Anderson, no one will take any notice of what you say, soon. Tony   (talk)  01:28, 19 January 2009 (UTC)
 * If the result is no consensus, as I believe, then what do the details matter? Tony's last post at least half concedes this.


 * Without consensus, the bots and scripts flying about have no justification for their bulldozing, which is the sole issue actually in question. Septentrionalis PMAnderson 17:47, 19 January 2009 (UTC)
 * Don't twist my words, Manderson. Read what I said in the analysis. Tony   (talk)  08:23, 20 January 2009 (UTC)
 * This acknowledges that if there is a consensus, it is to link rarely, and a few of the specific categories that several of us think may be worth linking. Other than that, it follows the model of a Bushie report: it comes to the conclusion the reporter wishes the world to believe, even though it is unsupported by the evidence. In particular,  Typical comments were ""We don't need a world view of what was going on that day in order to better understand the subject of the article." and "These articles are almost trivia collections." is unsupported, even by Tony's own count; that's one of three views that a significant number of commenters expressed. Septentrionalis PMAnderson 16:12, 20 January 2009 (UTC)

I endorse Tony's summary
And just by the way, I can't wait for mass delinking to be put back on track very soon. I just gave up half way through removing stupid datelinks and other distracting blues from Ernst Nolte because it was turning me into a vegetable. That is work for a machine, not a human.--Goodmorningworld (talk) 00:10, 20 January 2009 (UTC)
 * Agreed. (I completed the date formatting and linking edits for the article.)  HWV258  00:46, 20 January 2009 (UTC)
 * Thank you!--Goodmorningworld (talk) 01:19, 20 January 2009 (UTC)
 * I hope I won't be dragged before ARBCOM for asking this, but isn't there a bot that scans for punctuation marks before and puts the comma or period after the where it belongs? If there isn't yet then there should be.--Goodmorningworld (talk) 01:27, 20 January 2009 (UTC)
 * Yes, I've always found it "funny" to see the after the punctuation, but I've never bothered to investigate the accepted practice. Not sure about a bot that does it either. Sorry. Perhaps I should start ARBCOM procedings on myself for wasting Wiki-ink with this less-than-helpful post?  HWV258  01:56, 20 January 2009 (UTC)
 * Aaaargh, I meant it the other way around, and what is worse, I just checked and discovered that according to WP:REFPUN (just down the street from Minitrue, and remember: Ignorance is Strength) I did it wrong, either way is okay but now that I abandoned changing around the REFPUN order midway through, Ernst Nolte is stuck in limbo, neither dead nor alive, just as the horrible fate suffered by Monsieur Valdemar! though on 2nd thought, it couldn't have happened to a nicer guy. --Goodmorningworld (talk) 02:15, 20 January 2009 (UTC) Update: request made at WP:BRFA for a bot to clean up after me.
 * Gruesome! I wonder if "Valdemar" is where Rowling got her charcter name "Voldemort"?
 * Perhaps the date-delinking evidence page should be handed over to Miniluv?
 * The Ernst Nolte article is a pretty full-on page. (Personally, I'd prefer to put effort into other topics.)  HWV258  02:49, 20 January 2009 (UTC)


 * Tony’s summary seems like an accurate portrayal of the facts to me. Loose the links. Let bot operators prolifically nuke dates as they see fit. Greg L (talk) 04:21, 20 January 2009 (UTC)
 * Fixed that article. Dabomb87 (talk) 13:19, 20 January 2009 (UTC)

Exposing the distortions in the "detailed" RfCs on dates
One or two editors seem to be using the responses to a set of RfCs to support claims that various kinds of consensus exist against the community's moves to "smart linking" and the removal of date-autoformatting. When light is shone on these RfCs, I believe claims that they deliver any useful information that amounts to consensus is highly questionable. As advertisers and those who write surveys and questionnaires know well—sometimes to their advantage—the information they present to participants has significant potential to contaminate the data.

RfC1
The first RfC is a good example of this; I regret that this much space is required to get to the nub of why credibility is lacking.



Contamination. The first RfC explicitly sought to reveal the following: "Do you support or oppose retaining the following statement? Dates should not be linked purely for the purpose of autoformatting (even though in the past this was considered desirable)." However, the otherwise NPOV background statement ended with a prominent "Note" claiming that MediWiki has created a new, improved autoformatting system. "[Background statement:] Note: Mediawiki's software developers have created a patch to correct problems with the current method of date autoformatting. It has not been established when or if the patch will be implemented. (For more information on this, please see the discussion on Bugzilla.)"

Independent technical opinion. In the opinion of a skilled and experienced programmer (s/he has taken Cole to task about this claim already—and Cole probably knows whom I'm referring to): "That is disingenuous. After three years [of discussion at Bugzilla], I would say that not only isn’t there a patch ready to go, but there is no consensus over what should actually be in that patch.... you could poke the space shuttle through [the claim].... Trust me, I’ve been in the computer industry long enough to see that the recent debate on the page at the link you sent indicates there is no specification for development (let alone implementation). That doesn’t mean a developer hasn't created “something”, but in my experience there is no way known that that “something” is going into production WP any time soon."

Confusion reinforced. The "Note" was followed by no fewer than four rejoinders and a statement during the RfC, mostly in prominent positions towards the top. Under the pretext of stimulating discussion, this appears to have had the unfortunate and predictable effect of significantly blurring the stated issue in the minds of many participants. "[Rejoinder:] You understand that the bugzilla patch would remove the links and provide formatting to all users (logged in or not), yes? Also you need not say 'support' as your comment is in a section of the same title. =) —Locke Cole" "[Rejoinder:] Removing date links invalidates the work [of developers] done so far and is strictly punitive in nature.'—Locke Cole" "[Rejoinder:] ... a developer has already created a patch that can address the concerns raised against autoformatting. The developer has indicated that retaining the existing links simplifies the process, as it is more complicated to identify unlinked dates. Furthermore, the new patch - if enabled - would not require a modification to the existing markup for most dates; instead, the autoformatting would remain and be improved, while the links (the most contentious issue) would simply vanish. (Ckatz)" "[Rejoinder:] Or maybe Ephebi read the RFC entirely and realizes there's a patch that addresses this concern. —Locke Cole" "[Lead to the 'Oppose' section:] Note also that the current formatting system (using wikilinks around date fragments) is the method being addressed by the MediaWiki patch at Bugzilla noted in the background above. Removing these links would undermine and harm the work being done by developers to fix these issues. —Locke Cole"

A second element of confusion. In addition, the language in the lead inadvertently invited confusion between date-autoformatting and the linking of chronological items to actual pages; this is a great pity.

Conclusion: As as result, a sizeable proportion of the some 48 users who entered "Oppose" showed evidence of either (i) influence by the claim that "son of autoformatting" is at hand, almost ready to go ... or (ii) confusion between two quite different functions. By my reckoning,
 * only 25 entered a straight "Oppose" without such contaminating influence or confusion;
 * 17 showed that they were influenced by the spectre of some new "son of autoformatting" invention;
 * seven confused date-autoformatting with linking to chronological pages;
 * one was a confused Oppose.
 * one, "Edmund Patrick" appears to have entered his Support in the wrong section.

The confusion is also seen in the seven participants who explicitly made "Neutral" entries.

To be generous—since a few of those in the confused categories probably would have opposed in a pure vote—the 48 claimed Opposes are more likely to boil down to about 30–32 (10–11%) of more than 300 participants, when the contamination of the leads and the rejoinders is taken into account. There is no clear way of determining what the opinion of the others would have been in the absence of confusion. As Anomie said in the discussion section: "'there seem to be at least 6 that assume anons would still see unformatted dates, at least 5 who oppose because it would be a waste of developer time (it's up to the developers, particularly the unpaid volunteers, what they want to spend their time on), at least 7 who assume it would apply to every date rather than just dates marked up with whatever new syntax, at least 2 who seem to think dates would still be linked, and too many to count who think the only reason for date formatting is MDY versus DMY (only 1 comment considered the template issue)."

This goes much of the way towards explaining why this RfC yielded a more favorable result for the opposers of deprecation than the other RfC, which asked a straight, binary question without exposing participants to contaminating/confusing text before they made their entries.

The lesson is that RfCs, like all surveys and questionnaires, need to be framed very carefully if their results are to be considered reliable. Unfortunately, the framing of this first RfC set the scene for serious issues that damage the credibility of the subsequent RfCs in this set. I will deal with these in a day or two. Tony  (talk)  15:38, 13 January 2009 (UTC)


 * That is an excellent posting, Tony. I did see that the RFC claimed "Note: Mediawiki's software developers have created a patch..." and the quotes show that some people voted under the influence of that note. Lightmouse (talk) 15:49, 13 January 2009 (UTC)
 * Interesting indeed. I had been blinded by all the hubbub. Had I looked more closely, perhaps I would have objected to the attempt to surround the questions by 'loading' it with information of questionable validity (ie vapourware). I also took the imminent availability of the patch at face value. Looking at it again like Tony has done, I can see the questions were still the same flawed ones worked on by Masem, but they were given a gift-wrapp in shiny, glossy paper with ribbons and bows by Locke. Ohconfucius (talk) 06:42, 14 January 2009 (UTC)


 * Eh, I've seen patches applied to MediaWiki in a matter of hours, so this is provably false. Since the rest of your claims rely on that, I won't respond to the rest. —Locke Cole • t • c 15:16, 14 January 2009 (UTC)
 * There's no evidence of specs or a roadmap. Forgive my caution; it comes from experience. The matter is one of multilayered complexity. Tony   (talk)  16:01, 14 January 2009 (UTC)
 * On at least two occasions I've tried to discuss what type of syntax editors here at MOSNUM would prefer, what concerns they might have with a new system, and in both situations discussion died without reaching a consensus (and in some cases you helpfully responded it was a "waste of time" to discuss it). —Locke Cole • t • c 16:05, 14 January 2009 (UTC)
 * I'm sorry, but it's not just me: many people here are well aware of what technical, administrative and logistic problems you would be up against. I believe that is why discussion died, not because I said at some stage that it's a waste of time. It is because of the larger, messy issues that you can't achieve worthwhile discussion on specs and a roadmap. After three years at Bugzilla, going around and around in circles, I do not believe it will ever happen. Tony   (talk)  16:24, 14 January 2009 (UTC)
 * I've already explained why it hasn't happened: we haven't come up with a description of what we want here on the wiki for the devs to act on. And I already went over part of the problem (namely your behavior in those discussions, the "waste of time" comments, etc). I'm all for having a discussion on it again if you feel you can discuss it calmly. —Locke Cole • t • c 22:07, 14 January 2009 (UTC)
 * Interesting to contrast "we haven't come up with a description of what we want here on the wiki for the devs to act on" and "a developer has already created a patch that can address the concerns raised against autoformatting". You are right though—we should try to remain "calm".  HWV 258  22:23, 14 January 2009 (UTC)
 * Nothing "interesting" about that at all: regulars here seemed to shoot down that patch when it was discussed because they didn't like the way it worked. Again, the issue isn't the devs, the issue is actually deciding what we want. —Locke Cole • t • c 22:31, 14 January 2009 (UTC)
 * You have "contrasted" exactly one half of my quotes. The "interest" sparks from trying to reconcile both quotes (in the light of the section heading under which they were posted: "RfC1"—specifically "contamination").  HWV 258  22:40, 14 January 2009 (UTC)
 * I did not add that passage, you'd have to ask the person who did. For my part I think that patch sounds fine, but again, those here turned it down. —Locke Cole • t • c 02:07, 15 January 2009 (UTC)
 * For a software implementation to be successful, programmers must work to a rigidly-defined specification. Locke, please demonstrate to the community the specifications for the "patch" that could be "applied to MediaWiki in a matter of hours".  HWV 258  21:37, 14 January 2009 (UTC)
 * When ParserFunctions were first authored by Tim Starling it only took a matter of hours, maybe a day tops, for it to get pushed out to the Wikimedia servers IIRC. ParserFunctions are, comparatively, far more complicated than any date formatting code we're likely going to ask to have added... —Locke Cole • t • c 22:05, 14 January 2009 (UTC)
 * So that would be a "no" then to producing specifications? Software development and release (even involving complexity) can happen very quickly, but only based on clear and detailed specifications. Please take time considering the issues here—no one is denigrating WP's programmers as being skilled developers and implementers, however no programmer can produce effective code in the absence of specification.  HWV 258  22:14, 14 January 2009 (UTC)
 * PF wasn't created from any community issued specification, that was Tim going off on his own I believe (but being aware of the demands needed to address the communities desires). And I have no idea exactly how long he worked on it prior to it being submitted, but I don't believe it was long. I misunderstood your question as being how long from the time finished code was completed until it was pushed out to us for use. —Locke Cole • t • c 22:21, 14 January 2009 (UTC)

Thank you HWV. I love the way Cole is constructing much of his argument around "my behaviour"; this is being done both here and at the ArbCom hearing, which strangely allows users to freely post trumped-up and irrelevant accusations, and tittle-tattle, without space limit. If Arbitrators do not see through this smoke screen, we will be witnessing the destruction of the peak judicial process. When you look at the huge tranches of text devoted to smearing me (dear me, I look like a dragon from the headings and amount of text under them), you find that almost all of it is based on diffs to the normal cut-and-thrust of the discourse about date links, yet is loudly drummed out as incivility.

I can't bring myself to play that game. Cole has still not answered HWV's queries. I'm sorry, but we're not stupid around here. Tony  (talk)  01:33, 15 January 2009 (UTC)
 * Don't forget the "258"—people are always forgetting the 258 (there is another user named "HWV").  HWV258  02:00, 15 January 2009 (UTC)

HWV258: "For a software implementation to be successful, programmers must work to a rigidly-defined specification." Er... no. I've spent the last decade largely working agile commercially, and I can tell you that's one of the least accurate statements I've heard in a long time. To then try and claim that Locke has to produce some kind of "specification" to justify himself is absolutely ludicrous. And yes, I can confirm what he said about Tim's implementation of ParserFunctions. I'm guessing that you don't actually have any experience of working on FLOSS to make such a statement. — Hex    (❝  ?!  ❞)   11:51, 18 January 2009 (UTC)
 * FLOSS doesn't suggest that software development will work better without a well-defined specification. Without a specification, how does the programmer know what to develop? Think about why the (reopened) debate on exactly what should be in the "patch" has stalled after three years. If it was easy to know what to program, the whole thing would have been concluded long before the first date-delinking took place. On WP it is even more important that (at least) functional requirements are produced as there is a very large community that must adopt the results of the development.
 * In terms of the quoted statements, I wanted to include in the debate the clear discrepancy between "we haven't come up with a description of what we want here on the wiki for the devs to act on" and "a developer has already created a patch that can address the concerns raised against autoformatting". Please note that the first statement clearly indicates the desire to produce a specification before development (and quite right too!). Nothing "ludicrous" at all ("absolutely" or otherwise).  HWV258  21:57, 18 January 2009 (UTC)

RfC2
As promised, I have examined the second of the set of RfCs at issue, with disappointing conclusions in a technical sense.



The danger that language will funnel users into one option. I believe the language of the RfC has the effect of manipulating users into declaring a "Support". Let's look closely at the opening.

First, users are warmed up for a "Support" declaration by two key positive words: "desirable" in the title ("Is some form of date autoformatting desirable?") and a few seconds later—if they'd forgotten this word—combined with the use of "encouraged" in the official question (my italics):

"Do you agree with the following statement: The ability for the Mediawiki to convert dates into a form either appropriate for the page, or to user-defined preferences, is desirable, and the MediaWiki developers should be encouraged to find a solution that works without the problems of the current date autoformatting system."

The subsequent "Background" is a minefield:

"Background: ... Currently the MediaWiki developers are discussing methods of improving autoformatting to address these points, including possibly correcting the problems in the current system. To make sure their time is being used effectively, it is necessary to understand if a date autoformatting approach that works correctly is desired on Wikipedia. If not, the developers should be informed of this so they may focus on other aspects of the software that need improving."

Here, the claims in the first RfC above are extended in the assumption that WikiMedia is investing time in the development of a date-autoformatting system that works. The now-moribund three-year-old debate at Bugzilla is dragged out and dressed up as an ongoing, dynamic process. The same unevidenced claim is made that the MediaWiki developers (read User:UC_Bill (aka Bill Clarke)?) are close to delivering a spanking new generation of autoformatting. Just what "correcting" means is not explained, and there is no indication of the significant technical, logistic and administrative issues that would lie in the path of this option—a roadmap and specifications are nowhere to be found. Please see the independent programmer's opinion above on this claim.

Caught between two choices. Users were given a stark choice. By entering a "Support", they were somehow ensuring that developer time will be used effectively ("To make sure [MediaWiki developers'] time is being used effectively")—it seemed like the easy, positive thing to do. By contrast, declaring Oppose was framed as turning down the opportunity to have developers deliver a date-autoformatting approach that works—and worse, as interrupting professional work towards this goal, something that many folk would think twice about doing ("If not, the developers should be informed of this so they may focus on other aspects of the software that need improving"). I'd feel a heel myself at spoiling their ongoing project.

Support responses. A look through them clearly shows that many supporters were influenced by the blue-sky prospect of a new generation of technology, just as in the first RfC; again, many showed a confusion of the issues and technicalities, as would be expected when non-specialists are faced with a complex feature.

Conclusion. Again, writing an RfC is an exercise in trying to avoid bias and contamination, a difficult task indeed; the data are only as good as the NPOV of the stimulus. I submit that the language and choices presented to users rendered the result significantly unreliable, and explains why it generated much higher "Support" numbers than the first RfC above or the more straightforward RfC. Regrettably, this RfC does not deliver useful data.

Tony  (talk)  16:24, 14 January 2009 (UTC)


 * You had no problems !voting that no form of autoformatting is desirable; neither did I. This is a red herring; there is no hope of the sort of consensus against it we would like, for separate reasons, because too many editors actually like it. In any case, this does not matter unless a developer is actually moved to provide a new style of autoformatting, and you should know better than I do how likely a developer is to do anything on this issue.  Septentrionalis PMAnderson 19:35, 14 January 2009 (UTC)
 * No, Cole and a hard-core unrelenting band of four or five others like it. See Colonies Chris's stats as to the vanishingly small number of complaints he has had after thousands upon thousands of DA removals. Don't be swayed by the loud noise of the few. (Another diff here for my "incivility", Cole?). Tony   (talk)  01:38, 15 January 2009 (UTC)
 * Wrong again, Anderson. You are not qualified to say all those who voted "had no problems". The statement certainly had an influence, although we will never know to what degree, unless we questioned the respondents all over again. Now that you have openly said so, I will happily share with you what I actually felt: of course, instinctively and emotionally, I would feel very guilty about turning down "the hard work of those poor guys and gals and for no pay so that we could all have some pretty autoformatted dates, and it's nearly ready." That's how it was put. It's just like you arrive home one evening to find that your mother/girlfriend/ significant other had prepared a scrumptious meal - how inclined would you feel to suddenly announce to her/him that you don't like paté de foie gras and pressed duck. I can tell you I was nearly swayed. What made the difference for me is logical reasoning which took place. Looking at the bugzilla thread, I figured it was no where near ready to fly. It was only then, I felt I owed nothing to the little elves, who had not in fact started working all that hard - in fact, they hadn't quite worked out what needed doing: no vision, no mission, no strategy (roadmap). Just talk. Ohconfucius (talk) 02:14, 15 January 2009 (UTC)

Not to derail this, but...
I don't think this exercise is necessary the best of use of time (though WP is volunteering you're free to do so), for a couple reasons:
 * First, after creating the core structure, I did open the pre-RFC to discussion and improvements from editors here. I know you looked at it, along with others.  I realize some of the language that is in there now was after I last edited it (the part in the first question about the promised fix in the current system), and that this occurred after you initiated the second RFC and I myself decided not to pursue the other one further, until others decided to put that forward (at which point I tried to make sure both RFCs were acknowledged in watchlist notices).  However, still, prior to making it a true RFC the questions were free to be edited by anyone, including yourself.  So stating the questions are misleading after the fact seems counterproductive as they could have been fixed at the start.
 * Second, we're not marketing - we trying to achieve consensus, so good faith should be assumed that editors responding to the RFCs have read the questions and understand the implications, not responding to a survey question with the first answer that pops into their heads. As long as the core question is clear and understood, anything else around it should have been evaluated personally by every editor, and ignored if they felt it was biased.
 * Third, what really should be happening now is seeking a neutral third-party on both RFCs to evaluate and provide conclusions based on the results, which will take in account any apparent biasing the questions may have had. There are some points that are very clear - date autoformatting is dead, for one - but the points on bare day/month and year links are not clear from either.  Let that neutral party determine if the questions were truly rigged to get a specific response and declare that consensus is not clear on the answer due to the biasing.

The only reason I state this is that if you want to be pedantic, there are similar problems with your RFC (getting users to bring support on a negative statement, and starting from the assumption that the language of MOSNUM was accepted to begin with), though I'm not going to try to invalid those because the results between the two RFCs pretty much agree when the questions overlap: DA is deprecated, and only select links to day/month and year pages should be made. We shouldn't be trying to argue those points anymore, but instead making sure specifically for the day/month and year links when and where these are acceptable prior to having Lightbot et al run at full bore. Again, you're free to continue this analysis, but it does smack of wikilawyering and being a sore winner/loser, and is not going to be productive to resolving the conflict. --M ASEM 16:59, 14 January 2009 (UTC)
 * I don't see myself as a sore anything here; no, it's clear that the amount of steam being generated from these RfCs at the ArbCom hearing, where they're used as the foundation for all sorts of claims to consensus, required a proper scientific analysis. The lesson to come out of this is that we must be very careful in constructing RfCs.
 * No matter who contributed to the wording, I can only analyse the wording that ended up being used. Is that not fair enough?
 * Marketing and soliciting information from people (consumers, WPians, etc) bear a lot of similarities; this is not sinister, but simply a technical observation. Tread carefully, I say, or some bastard will come along later and debunk the results. Tony   (talk)  17:21, 14 January 2009 (UTC)
 * You can analyze these as much as you want, but really either we should get a third, neutral party to review the RFC, or you can raise your concerns that the RFCs were tainted by bad questions at the ArbCom case and let ArbCom evaluate that evidence. You're not neutral in this, nor am I (even if I'm just trying to resolve this, not push for an acceptable answer either way) --M ASEM  17:33, 14 January 2009 (UTC)
 * My methodology is quite open—no secrets. You're free to criticise it. Tony   (talk)  17:41, 14 January 2009 (UTC)
 * Tony you're not neutral in this, at all, and you've shown a continued unwillingness to compromise on issues surrounding this dispute. Why would anyone trust your judgment here? For similar reasons, while I appreciate Dabomb87's efforts, they will still not be as acceptable as getting someone outside the dispute (someone uninvolved) to give a fair opinion of the results. —Locke Cole • t • c 17:46, 14 January 2009 (UTC)
 * Note that Carcharoth recused himself from the arbitration, and he is far less involved than anyone in this conversation. Septentrionalis PMAnderson 19:58, 14 January 2009 (UTC)
 * True, we could ask him, but I'd personally prefer three uninvolved (or at least, as involved as Carcaroth) make this decision as a group. This way the decision isn't entirely on one editors shoulders (not a fair burden to ask anyone to take on) and so it receives reasonable consideration. —Locke Cole • t • c 21:45, 14 January 2009 (UTC)
 * What I did was a general summary of the RFCs; I made that before the Arbcom case had started. I intend to post that here soon. Dabomb87 (talk) 22:21, 14 January 2009 (UTC)


 * Masem, I much appreciate the parenthetical in your initial post: I don't think this exercise is necessary the best of use of time (though WP is volunteering [so] you're free to do so). Indeed, Tony is volunteering his efforts to keep working this issue. WT:MOSNUM is a venue devoted exclusively to the exchange of ideas. If Tony has an idea and he wants to rally others to his way of thinking and create an RfC that 1) provides a mechanism for everyone to have a free exchange of ideas, and 2) quantify and qualify the varying opinions, he is certainly free to do so. Date linking and formatting has been a battleground where we’ve been scratching each others eyes out and questioning the parentage of others for over six months now. It must end and Tony thinks he can help. I hope to be second in line on any RfC he develops to register my views on this subject. You are free to do the same. May the editor(s) with the best and most appealing ideas prevail. Wikipedia will be the prime beneficiary when we finally have a clear community consensus on all of this. Greg L (talk) 21:30, 14 January 2009 (UTC)
 * Scientists and scholars are rarely "neutral" in relation to the data they produce: what counts is that their methodology is open and discussed. That is what I have done here, yet zilch, niente, null has been offered in substantive criticism of my analysis or conclusions. Thus, they appear to stand. Sorry people, you have to actually deal with the substance, not trumped-up complaints about "Tony1's behaviour".  Tony   (talk)  01:41, 15 January 2009 (UTC)
 * Your claims have no basis in fact and are reaching. Further, they're biased, and this isn't science, it's an attempt at consensus. —Locke Cole • t • c 02:08, 15 January 2009 (UTC)
 * Masem, I don't think Tony was focussed on your RfC questions. Sure, they were problematic, and I did not hesitate to tell you so. The problem Tony seems to have is the emotionally appealing background to the "son of DA" (based on some unproven premises) which Locke et al had inserted into it in an attempt to influence. Ohconfucius (talk) 02:27, 15 January 2009 (UTC)


 * The ArbCom will, hopefully, formally declare precisely what has been decided by the RfCs. To the extent that ArbCom doesn’t settle some disputed points, I have every confidence that Tony is perfectly capable of producing a new RfC that is clear as glass to all but the most intransigent of editors (which don’t have to be factored into a consensus in the final analysis). What is clear to me is Tony’s last RfC made it extremely clear that the community consensus was that it was a rare date indeed that should be linked in normal body text. I find it probative and amusing to note which editors here dispute that simple fact. Greg L (talk) 05:21, 15 January 2009 (UTC)
 * Stupid question: What would be the purpose of this RFC? Deprecation of linking per DA was affirmed in both RFCs, and both pointed to highly limited use of date links otherwise.  The only thing left on the table is what those limited uses are.  (The issue at ArbCom is the race to start the bots before establishing this aspect, which hasn't been discussed in any detail). --M ASEM  05:33, 15 January 2009 (UTC)


 * So, let me guess, Cole: it's not the methodology of an analysis that counts, but who's doing the analysis. Only people who pass Cole's power of veto may analyse the data from his RfCs. Again, there has been no substantive argument with my analyses and conclusions. That kind of figures. Tony   (talk)  12:23, 15 January 2009 (UTC)


 * Maybe everyone is tired and no one can be bothered to argue with you any more? Deb (talk) 12:58, 15 January 2009 (UTC)
 * If you're tired, don't log in. Those who are not too tired to write reams of discourse that use the results of these RfCs as supporting evidence have nothing substantive to say to a proper analysis of their language and structure. You don't? Tony   (talk)  14:33, 15 January 2009 (UTC)


 * Oh, and Cole, if you're going to accuse me at ArbCom of incivility, at least have the courtesy to use the correct word. UNcivil; INcivility. Got it? Tony   (talk)  09:53, 20 January 2009 (UTC)


 * Well… that reminded me of something. I wonder if G.H.W. Bush (Bush #41) got all hot & bothered when Saddam Hussein accused him of being an infidel and a liar? Now, no one should get all hot and bothered by suspecting that I think for a New York second that anyone here is a really, really bad person like Saddam; he used poison gas on his own people and was responsible for many other very bad things. No indeed, that would be taking the analogy much further than intended; everyone here are clearly splendid, law-abiding citizens and all that. But, what you just wrote, Tony, reminded me of that for some reason. I don’t think I’d let accusations of incivility get under my skin. In fact, since I seem to almost never agree with anything Locke writes here, and since I don’t admire his record of interacting with others and of chronically breaking rules of conduct here on Wikipedia, I think I’d be terribly proud of being on the receiving end of his criticisms. Congratulations. Greg L (talk) 16:32, 20 January 2009 (UTC)


 * incivil. —Locke Cole • t • c 08:55, 21 January 2009 (UTC)

The structure of the dates and numbers article concerning BC vs BCE etc..
I sometimes forget the BC vs BCE convention at Wikipedia and frequently have to re-look it up. This is a question of upmost importance regarding style. Unfortunately, the current structure of the document makes the reader spend more time than they should to find the answer and the current outline is logically incorrect. First the BC vs BCE (and of course AD vs CE etc.) issue should be handled under the Dates section and not the "year numbering systems" section. BC, CE, and so on all are used with dates. The section heading should also make it easy to spot that this is the section that deals with this question... something like "BC vs BCE etc." would be a great heading title. The "year numbering systems" should be called something like "Scientific units of time". After time for feedback, I might just make the changes if nobody objects. Jason Quinn (talk) 00:05, 19 January 2009 (UTC)


 * Yes, BC/BCE/AD/CE are used with dates, but they qualify the year alone. They belong in "year numbering systems" as much as BP, bce and mya do. It is probably a minor matter whether that section is listed as a fourth level heading under "Dates" or as a third level heading as it is now. Optionally, if editors have problems finding the section, perhaps another bullet point in the Dates section could read:
 * Guidance on the use of BC/BCE/AD/CE is given at WP:ERA.
 * Personally, I've never had a problem finding the guidance, but if others have problems, then suggestions for a fix would be welcome. --RexxS (talk) 00:31, 19 January 2009 (UTC)
 * And since the guidance consists of "we use all four of BC/BCE/AD/CE, in their usual format", why is it hard to remember? Perhaps we should rephrase. Septentrionalis PMAnderson 18:04, 19 January 2009 (UTC)
 * It's not hard to remember but policies change, so before an editor goes and makes a big change to several articles it is often useful to re-check with the MoS. I recheck the manual of style constantly even for things I'm pretty sure I still remember. Plus sometimes if you haven't fixed any dates in articles for 6, 8, or 12 months, you do forget. This is absolutely one of the most common formatting questions that arises it deserves to be as easy as pie to find. Currently a user trying to find it from the MoS main pages, goes straight to the MoS date and number section, and then goes to the Dates subsection only to spend a minute skimming it only to discover the information they want isn't there. It then takes another moment to realize there's some goofy semi-ambiguously named "Year numbering systems" section outside of the dates section where the info they want is located. All of those prefixes BC/BCE/AD/CE/BP are always part of the formatting for dates and should be under that section, not outside of it. The WP:ERA shortcut should point to a subheading under Dates perhaps still named "Year numbering systems". It is clearly more logically incorrect than correct to have "year numbering systems" outside of the "dates" section. Lastly, I never said I wanted to rephrase the policy language itself, but to reorder the outline and perhaps rename the section titles. "Year numbering system" is a confusing phrase and I promise you that most users that click on it aren't sure what it contains; but if the section title had "BC vs BCE and so on" in it, they would know instantly what the section was about before even clicking. Jason Quinn (talk) 19:46, 19 January 2009 (UTC)
 * I recheck the manual of style constantly even for things I'm pretty sure I still remember. Dear Heavens, why? Ignore all rules you find on Wikipedia (Wikipedia is not a reliable source) and write English. Septentrionalis PMAnderson 20:36, 19 January 2009 (UTC)
 * You don't feel inclined to help make it a "reliable source"?
 * MoS is a guide (not a rulebook).
 * "Ignore all rules you find on Wikipedia"—could that be the final nail in the coffin for the "civility" evidence?
 * Just out of interest, will you ignore this one?  HWV258  02:21, 20 January 2009 (UTC)
 * Tony (and others) appear to believe it's a rulebook which must be followed. —Locke Cole • t • c 02:27, 20 January 2009 (UTC)
 * I wouldn't have said that, but they are good at implementing its guide.  HWV258  02:48, 20 January 2009 (UTC)
 * I'm inclined to move article space closer to being a reliable source; we will not succeed before WP:DEADLINE. This page is an impediment to that goal insofar as it takes up editor time or attention. Septentrionalis PMAnderson 03:53, 20 January 2009 (UTC)
 * In my humble, that's a bit like saying the road-rules handbook is an impediment in getting from A to B. Surely once you've read the "rules" you've become empowered to safely and confidently participate? (Which is not to say that people can't behave like a maniac from time to time.)  HWV258  04:43, 20 January 2009 (UTC)
 * No. There are two differences
 * I said this has nothing to do with reliability (or neutrality, or verifiability); nothing on this page affects any of them. Nor would this page, even if it got things right.
 * But it doesn't. This page has the greatest possible flaw in a road-map: It doesn't match the territory. This page does not describe what Wikipedia does, or what good English should do; it describes the prejudices of a handful of underqualified editors.Septentrionalis PMAnderson 06:19, 20 January 2009 (UTC)
 * I didn't saw "road-map", I said "road-rules". I don't believe the page is designed to describe what "Wikipedia does". I still believe you could play a valuable part in improving the page to help explain to others what "good English should do".  HWV258  21:15, 20 January 2009 (UTC)
 * That would mean cutting about 90% of it out, where it pointlessly adjudicates between two things, both of which good English does; or, yet worse, requires a rule which good English does not follow. Septentrionalis PMAnderson 21:32, 20 January 2009 (UTC)
 * For any particular example, if both are "good" English, then what troubles you with picking one?  HWV258  21:43, 20 January 2009 (UTC)
 * Because there is no reason to pick one. Both are useful, and we will get editors who reasonably prefer both; quite often each is more useful in some situation than the other. We should allow both editors to edit, because anyone can edit, and not make rules for rules' sake. Septentrionalis PMAnderson 21:55, 20 January 2009 (UTC)
 * But there is—consistency across articles—hence improving the casual reader's experience on WP. I really don't know how else to say it, but I'll have another go: both editors are allowed to edit, and therefore make their contribution to WP. The fact that a "worker ant" mops up behind them without changing the semantics of the original edit shouldn't bother anyone (and, in general, it doesn't). MOS is a guideline, so if there is a localised situation where one format should be adopted over another, it can be discussed and adopted by consensus. You are failing to convince me.  HWV258  22:07, 20 January 2009 (UTC)
 * Consistency across articles on 17 vs. seventeen is a Golden Calf. We shall never obtain it, and I have never seen an actual reader complain of its absence; the efforts to obtain it suppress the real occasions where one or the other has some point. We would do better to leave it alone, as we leave honor/honour alone; readers don't boggle at that inconsistency either. Septentrionalis PMAnderson 22:17, 20 January 2009 (UTC)
 * I disagree with everything you have written above. WP is an evolving process, and as you have pointed out, there is no deadline. Please remember that the vast majority of readers can't complain as they don't log in. I have pointed out that localised consensus can over-ride MOS in the case where "the other has some point". I don't believe "honor"/"honour" is an inconsistency as there are guidelines as to regional spelling.  HWV258  22:26, 20 January 2009 (UTC)
 * Please remember that the vast majority of readers can't complain as they don't log in. It's news to me that anons can't post; they do regularly, usually to complain. But this seems of a piece with the rest of this "argument". Septentrionalis PMAnderson 23:54, 20 January 2009 (UTC)
 * Disingenuous! I would confidently state that the vast majority of readers at WP don't even realise they can edit. I've personally talked to many many people in order to advocate the use of WP, and almost universally I hear a surprised response when I point out that they are free to edit. Your point of "I have never seen an actual reader complain of its absence" is moot.  HWV258  00:17, 21 January 2009 (UTC)
 * De gustibus non disputand' est. Just please confine your taste to the edit wars on this page, and don't inflict it on the rest of Wikipedia. Septentrionalis PMAnderson 22:45, 20 January 2009 (UTC)

Anderson, you can huff and puff, but you won't blow the house down. Too many people here recognise your agenda to considerably reduce the role of the style guidelines, especially their contents that you have decided to take personal objection to. I see that you've spread your campaign to the current ArbCom hate-page, now almost an encyclopedia itself, due in no small way to your input. I'm surprised that you haven't evolved to take on a more positive role here over the past few years; that option is still open to you. I note that your prose on the clause level has improved significantly, probably because of your contact with the guidelines themselves and the experts who contribute to the talk pages. Tony  (talk)  13:28, 20 January 2009 (UTC)
 * (outdent) Well… I go away for a while and check in here before getting myself a cup of coffee. And what do I find? On a thread that doesn’t even have Tony weighing in, I see you two feeding off each other and busy nipping away at Tony’s heels for seemingly no good reason other than to not miss out on yet another opportunity to get some digs. For one thing, this crap is getting is tedious. So it would be just splendid if you guys did a better job of acting like the grownups I presume your drivers licenses say you are. And, FWIW, what MOSNUM eventually ends up doing will ultimately be decided by the community consensus. And that can be influenced only through sound, reasonable, logical arguments that persuades others to a way of thinking. If you guys think that trying to drag down the reputation of one of our most respected editors, then, by all means, keep at it. For I take great pleasure in pronouncing that I rarely agree with anything you write here and am perfectly content to sit back and watch you act like fourth-graders in a vain attempt to influence the community consensus. Greg L (talk) 02:54, 20 January 2009 (UTC)
 * There should be no "MOSNUM community", there should be the community of Wikipedia, which deliberates here or elsewhere as may be most convenient. Underneath the petty squabbles, that's the real point at issue. Septentrionalis PMAnderson 03:50, 20 January 2009 (UTC)
 * Don't you feel that guidelines influencing consistency across articles aid readability? Otherwise you'd end up with (to take some sort of weird hypothetical case) the situation where tennis-based articles have linked dates, but music-based articles don't have linked dates—which could be confusing to the average reader. (And I hope that everyone is welcome at the MOSNUM "community".)  HWV258  04:13, 20 January 2009 (UTC)
 * No, I don't. This page (with all its dicta) has even less influence on readability than the differences in style and vocabulary which are inevitable in a jointly written project, and on which we have long since decided to live and let live. If switching between color and colour doesn't phaze me, none of this will. Septentrionalis PMAnderson 04:20, 20 January 2009 (UTC)
 * When you say "jointly written", are you refering to academic-type articles with less than a handful of authors? If so, do you see a difference when there are thousands of (non-academic-type) authors editing a WP article (over time)? Also if so, do you see a difference based on the target audience being very different for an academic journal versus a WP article?  HWV258  04:34, 20 January 2009 (UTC)
 * No, I mean Wikipedia, which is jointly written by nature. I have no idea, although there must be a study, how many academic papers have one of the joint authors as a drafting committee. Septentrionalis PMAnderson 04:38, 20 January 2009 (UTC)
 * I might have to get someone else to help me here as I can't follow the introduction of "nature" or "drafting committee" into the discussion. Is your previous point something you could reword to help me follow your meaning?  HWV258  04:47, 20 January 2009 (UTC)
 * Done, immediately below. Septentrionalis PMAnderson 06:22, 20 January 2009 (UTC)
 * This page is no service to communication with our readers whatsoever. Being inconsistent from page to page on the issues this page decides (such as date format, or whether to use seventeen or 17) is no impediment to our readers; varying between honor and honour would be a greater impediment, and we have decided to live with that. Septentrionalis PMAnderson 06:19, 20 January 2009 (UTC)
 * I was under the impression that things like honor and honour are consistent—in context with the underlying article's region. I have no problem with regional spellings as they make sense. On the other hand, it would be better to be consistent with "17" versus "seventeen". Not a biggy, but again, I feel you have failed to explain the over-riding problem. (And truth be told, I'm still not overly sure about how "nature" got into the discussion.)  HWV258  21:43, 20 January 2009 (UTC)
 * Many articles don't have an "underlying region"; they happen to be in British or American or Australian English. This is intentional, to permit anyone to edit, without prefering any of the national varieties of English.
 * Similarly, there are people who prefer 17 in spme given context; there are others who prefer seventeen. We should let them both edit, without putting them through the dozen bullet-points on this page, which are both arbitrary and incomplete. Septentrionalis PMAnderson 21:51, 20 January 2009 (UTC)
 * The point is that any one particular editor doesn't have worry about the "dozen bullet-points on this page" as someone else will fix syntax without injuring semantics. And as you've admitted that "17" and "seventeen" are both "good", the original editor shouldn't be dismayed to see the occasional change. Overwhelmingly, it is a system that works well.  HWV258  21:58, 20 January 2009 (UTC)
 * In which case the editors will find their prose tweaked by illiterates who know this page, but can't write English. Articles will be turned down at FA on the grounds that they don't comply with the rules here, even when they are wrong. This page is a source of power-gaming for the lazy, the semi-literate, the bigoted, and the editor who Knows the True System of Metric Prefixes (there are two True Systems; for the slambang between them, see any of the archives labelled B).
 * The alternative is to memorize the incomplete, inaccurate, and ill-phrased guidelines here, instead of trying to write English and be accurate, neutral, and verifiable.
 * In short, this page is a waste of electrons. Septentrionalis PMAnderson 22:10, 20 January 2009 (UTC)
 * You fail to see the bigger picture. The work of the "illiterates" will be addressed in due course. Plenty of articles get FA status, so MOS is not the issue. I keep stating it, but you just can't address the point: there is no need to even know about the existence of MOS to contribute to WP, as someone else will tidy up the syntax without damaging the semantics. The original editor should still be able to log-off tired, but happy after being edited. "In short, this page is a waste of electrons" is dismissive and disappointing, but I guess you are free to unwatch and simply walk away.  HWV258  23:30, 20 January 2009 (UTC)
 * At last, something I can agree with:  there is no need to even know about the existence of MOS to contribute to WP. Indeed, WP would be better off without it, and without the incompetent, self-appointed, tidiers, following rules invented by prejudiced ignoramuses. If MOS would keep itself to its own fantasyland, and not impinge on the useful WP, I would indeed unwatch it; but it breaks out, every so often, and does lasting harm.  Septentrionalis PMAnderson 23:51, 20 January 2009 (UTC)
 * I'm no longer sure if you miss the point intentionally or accidentally (but I'll be patient and give you the benefit of the doubt). There is a point to MOS and it is possible to edit without referring to it (the fact you linked your two sentences with "Indeed" clearly indicates that you have not followed the argument). Perhaps at another time you'll be able to rationally address the points I've raised elsewhere on this page today. I hope so.  HWV258  00:10, 21 January 2009 (UTC)
 * Could there be a useful MoS? Of course there could.
 * Could some of this page be part of such a MoS? Yes; the admonition not to date war.
 * Is all of it? By no means. Much of it is not consensus; much of it is not English. Some that is both is nonetheless WP:CREEP. All this should be done away with, or at least not used to bother productive editors.
 * Is working on MOS productive? Only in the negative sense of deterring harm. Septentrionalis PMAnderson 22:14, 21 January 2009 (UTC)
 * Well, have a go at improving it. Could you pick a small section of it and write the "useful MoS" version (on a temp page somewhere)? I'm not being facetious—I really would like to see what you have in mind. Cheers.  HWV258  22:33, 21 January 2009 (UTC)
 * No, Tony, only you claim to be able to read my mind. Then again, only you claim this page has authority . Why do you need it so much? Septentrionalis PMAnderson 19:58, 20 January 2009 (UTC)

Why have MOSNUM? Examine the alternative. Somebody adds "53 BC" to an article. Somebody else changes it to "53 BCE", claiming that "BC" is unduly religious and therefore unacceptable. The first editor disagrees arguing that his history book uses "BC" and therefore "BCE" is unacceptable. After the warring, wiser heads prevail and all agree that either is acceptable as long as the use is consistent throughout the article. Eventually that consensus is codified into MOSNUM. Now, those who wish to scrap MOSNUM are asking for that war/debate to be repeated on every article that contains a date before the year dot. Or asking everybody to read through the archives of talk pages to see if a local consensus was made. The same applies to every piece of collected consensus in every guideline. It would seem to me to be a very heavy price to pay, simply to allow a few editors to have the freedom to write "just as they see fit". Wikipedia is not an anarchy. --RexxS (talk) 20:16, 20 January 2009 (UTC)
 * That's the really useful part of MOS; the don't edit war over trivialities conduct guidelines. But that's one paragraph, at most; it doesn't need to be here; it could be boiled down and stuck next to WP:ENGVAR where it belongs. Septentrionalis PMAnderson 21:28, 20 January 2009 (UTC)
 * Yes (to RexxS). Hence the idea that the page provides a skeleton of consistency, upon which articles can be fleshed. Of course editors are free to ignore MOS, but they shouldn't be surprised to see their words subsequently made consistent with other pages (of course without changing the meaning of their writing). I fail to see why that disturbs you?  HWV258  21:43, 20 January 2009 (UTC)
 * A foolish consistency is the hobgoblin of little minds. Observe the contending SI and IEC rule-mongers above. Septentrionalis PMAnderson 21:58, 20 January 2009 (UTC)
 * You say it's "foolish", others say it isn't—I guess that's the end of the debating points then. Perhaps you could provide examples of the foolishness of "contending SI and IEC rule-mongers" (I'm keen to learn). "Rule-mongers" is such a pejorative perspective; is it impossible for you to take a step back and see the work of Tony (and a whole heap of others) as aiming towards improving WP? You surely can't rule that out; can you?  HWV258  22:18, 20 January 2009 (UTC)
 * No, I quoted Emerson; for instances of that foolishness, see above, and all the pages in the archive marked with a B, for Binary. Septentrionalis PMAnderson 22:24, 20 January 2009 (UTC)
 * I have no problem with people working out which of "microns" or "micrometres" is preferred. Doesn't change a single point I've raised above.  HWV258  22:49, 20 January 2009 (UTC)
 * Outside this page, however, we acknowledge that both words exist because they both have uses, and do not call on external style guides to decide for us. Septentrionalis PMAnderson 22:54, 20 January 2009 (UTC)
 * I cannot rule out that Tony and his three or four friends intend to improve Wikipedia; that's what WP:AGF means. In practice, however, their efforts result either  in a campaign for a New and Improved, politically correct, English, or in preferences for Tony's native dialect over others. When they fail to accomplish that, they resort to obscenity, abuse, and  try to blackball editors out of this page, as here.  Septentrionalis PMAnderson 22:29, 20 January 2009 (UTC)
 * I fear you have taken a worse-case view. I'll let the four or five mentioned take over if they wish.  HWV258  22:49, 20 January 2009 (UTC)


 * "Resort to obscenity"—Anderson, a good start would be if you could stop obsessing about my pooh at ArbCom. You appear to have led the writing of paragraphs about it. I wish you would desist now; it is offensive. Tony   (talk)  10:30, 21 January 2009 (UTC)
 * You chose to write what I have quoted. If you find a certain blowback from that tactic, you may wish to consider others. You do courtesy rather well when you try; you can appeal successfully to reason and evidence. Do carry on. Septentrionalis PMAnderson 22:08, 21 January 2009 (UTC)

Yards and metres
Can a value in yards be supported by a value in metres? I tried on two occasions to add a value in metres to the phrase "within 200 yards of the Dudley factory" but it was reverted twice. Please can somebody take a look at Bean Cars and see if I am missing something? Lightmouse (talk) 14:27, 16 January 2009 (UTC)
 * Should definitely be converted to metres. Tony   (talk)  14:32, 16 January 2009 (UTC)


 * Perhaps you should consider raising the issue with the editor in question at Talk:Bean Cars, rather than fleeing here for backup. — Hex    (❝  ?!  ❞)   14:39, 16 January 2009 (UTC)


 * What has happened to everybody lately? Can't we get along anymore? Lightmouse (talk) 14:52, 16 January 2009 (UTC)


 * It looks to me like the kind of article where the vast majority of readers can be expected to know that within the level of precision used here there is no difference between yards and metres. To such readers it does indeed look silly. Now if this was the article Tower Bridge, the readership would be quite different, and I would argue to include the indication in metres or perhaps to drop the yards altogether. --Hans Adler (talk) 15:17, 16 January 2009 (UTC)

I agree with Tony, put it in. If it was ...200 metres of the Dudley factory, I'd convert to yards. The MOS does back you up on this. However, in my opinion, this type a of discussion probably should have taken place on the Bean Cars article's talk page first before coming here. &mdash;  MJC detroit  (yak) 18:07, 16 January 2009 (UTC)


 * Some of the international readership may not know what a yard is, since metre is the only unit used in their country. People in the US and UK need to realize that some people in other countries (particularly those who speak English as a second language) may never have used it.RockyMtnGuy (talk) 20:00, 16 January 2009 (UTC)


 * Hehe - as an amazing example of "small world", the Coseley factory is just a few hundred yards (metres :p) from where I live! I've added a little more info to the article, plus conversions, and mentioned on the talk page that they are a courtesy to readers who are not familiar with yards. Hopefully we'll reach a consensus. --RexxS (talk) 04:30, 19 January 2009 (UTC)
 * Some of the "international readership" doesn't know what of means (and edit anyway). The solution to their problem is the Simple English WP. Septentrionalis PMAnderson 04:29, 20 January 2009 (UTC)
 * Some of the "international readership" has a better command of English than most Englishmen and Americans (at least in my opinion). However, there are several factors which are relevant:
 * The United States is the only country which still recognizes the yard as an official unit of measurement.
 * In English-speaking countries outside of the US, traditional English units such as the yard are no longer being taught in schools.
 * English is becoming a global lingua franca used by educated people world-wide.
 * As a result, there are increasing numbers of people who have a good command of English as a first or second language, but who may not be familiar with traditional English units of measure such as the yard.RockyMtnGuy (talk) 05:45, 20 January 2009 (UTC)


 * Agreed with Rocky, I have many European friends who speak excellent English, but would not have good grasp of the distances involved if they were given only in yards. On the other hand, I spent 25 years teaching maths in England and had to recognise the fact that, for general use, miles, yards, feet and inches were what most folk understood. I certainly taught metric units, but also had to teach imperial if the students were to be equipped to deal with the realities outside of the classroom. A school-leaver who was comfortable with both sets of units was far more useful than one who could only deal with one set. The guidance to provide a conversion, wherever possible, seems like eminent good sense to me. --RexxS (talk) 20:35, 20 January 2009 (UTC)


 * This makes absolutely perfect sense to me; in fact my father told me of a foreign-born colleague many years ago who asked the proctor at a fairly-advanced science or maths exam how many inches were in a foot. It's only a question of what's the most natural, least intrusive, least pompous way of letting metric Anglophones know what "200 yards" means. —— Shakescene (talk) 21:02, 20 January 2009 (UTC)


 * In most of the former British colonies, the schools no longer teach miles, yards, feet and inches. The students have heard these words from their elders, but can't really visualize them unless you convert them to kilometres, metres, and centimetres. If you say "200 yd" it becomes clearer. Also, the schools in some English-speaking countries don't seem to teach grammar any more, which means students couldn't put together a coherent sentence to save their lives. On the other hand, people learning it as a second language often get a lot of grammar. I had a French teacher who met her husband at the Sorbonne in Paris. Being Anglo-Canadian, he thought it would be an easy credit to take English. He failed the course. Just because English is your mother tongue doesn't mean you know how to speak it properly.RockyMtnGuy (talk) 00:51, 21 January 2009 (UTC)
 * I think that 200 yards has only a trace over one significant digit, so 180 metres might be probably misleadingly over-precise. I go crazy whenever I read something like most blades of some grass or most book margins are 2.54 cm, or that some creature is about 25.4 centimetres long, when the writer obviously intended to say about an inch or about ten inches. ¶ As for English, I'm not that surprised because most of the original grammatical names and categories try to impose Latin or French rules on a fundamentally-Germanic grammar. (French, for example, uses different forms for the conditional and subjunctive where English makes little or no distinction, while most moods and tenses are formed from finely-differentiated auxiliary verbs that fit Latin rules rather badly.) Norwegians learn Norwegian and the Greeks are taught their Greek. —— Shakescene (talk) 02:00, 21 January 2009 (UTC)
 * You don't know really how many significant digits 200 yards has unless someone tells you, e.g. 200 yards ± 10 yards - the convert utility does its best, but it has to make assumptions. As for the quote from Pygmalion, I've always found it amusing because I understand a little Norwegian and know that the Norwegian language has two different official forms, Bokmål and Nynorsk, and there are hundreds of different Norwegian dialects, with considerable controversy over their grammar and spelling. In other words, it's a linguistic in-joke. However, most Norwegians I have met speak English extremely well - they sound like they come from Minnesota rather than Norway. But when you tell them distances in miles, they think you mean Norwegian miles (10 km).RockyMtnGuy (talk) 06:32, 21 January 2009 (UTC)
 * That line would have been a great example of devious Shavian wit, because the Greeks were another post-Napoleonic nation who similarly battled endlessly over which Greek to teach: Katharevousa or Demotic (see Greek language question). One example my friends who opposed the Greek military junta of 1967-1974 gave of its repressive backwardness was making the formal and uncolloquial Katharevousa compulsory in schools, which is why the Greeks had to be taught their Greek. (Katharevousa was banned from Greek schools in 1976, after the junta fell.)
 * It would have thus been a doubly great Shavian line, had it been written by Shaw. But I can't find the line in Pygmalion, only in the lyrics to My Fair Lady by Alan Jay Lerner (of Lerner & Loewe). So render unto Lerner the things which (deliberately or not) are Lerner's, and render unto Shaw the things that are Shaw's. [And while I'm not sure if I'd made the connection to Henry Higgins' outburst, I did know there were two Norwegians, though I'd known them as Landsmål and Riksmål]. The debates engendered by diglossia (cf. classical vs modern Hebrew & Arabic) make Wikipedia's squabbles over (e.g.) MOSNUM look friendly and tame.
 * And if you look at the Minnesota Star Tribune's recent files, you can see the local puzzlement that Governor Sarah Palin of Alaska, born in Idaho, sounded so Minnesotan. The indirect answer seems to be that FDR's New Deal settled thousands of desperate (and later gratefully Democratic) Minnesota farmers on virgin lands in the Alaska Territory during the Depression, producing the local accent picked up by newer migrants like Sarah Palin. —— Shakescene (talk) 08:44, 21 January 2009 (UTC)
 * RockyMtnGuy has this spot on. I know many English speakers who are unfamiliar with yards and inches. They should be converted.  The appropriate number of significant figures depends on the situation. Thunderbird2 (talk) 19:24, 22 January 2009 (UTC)

Bits - IEEE 1541 defines b as symbol not bit
TechControl (talk) 15:36, 21 January 2009 (UTC)
 * Manual of Style (dates and numbers): " By extension, the symbols for the units of data rate kilobit per second, megabit per second and so on are kbit/s (not kbps or Kbps), Mbit/s (not Mbps or mbps), etc. Similarly, kilobyte per second and megabyte per second are kB/s (not kBps or KBps) and MB/s (not Mbps or MBps). "
 * This contradicts somehow Bit: "IEEE 1541 a commonly-quoted relevant standard, specifies "b" to be the unit symbol for bit and "B" to be that for byte. This convention is also widely used in computing."
 * Maybe somebody needs to tell Microsoft and the computer manufacturors then. If I click on my internet connection it tells me I am connected at 54.0 Mbps and if I download a file it reports the download speeds in the form of Kbps. All of the literatures that came with my PC and laptop also talk in these terms. So where does kbit/s and Mbit/s come from exactly? 21stCenturyGreenstuff (talk) 16:04, 21 January 2009 (UTC)


 * TechControl does not seem to understand the role of this guideline. It is not necessarily a compilation of how other publications handle style issues; it is a statement of the style choices made by the Wikipedia community. Since some standards use b as a symbol for bit, and other standards always spell out bit, this guideline has chosen to recommend that it always be spelled out. This is not incorrect; it is a choice. --Gerry Ashton (talk) 18:09, 21 January 2009 (UTC)
 * I am talking about relevant standards, not any arbitrary behind the moon hidden standard. If you have anything more relevant than IEEE 1541 I am happy to see you updating Bit. TechControl (talk) 01:36, 22 January 2009 (UTC)


 * Thing is there are many standards out there. IEEE uses b, IEC uses bit. Confusion can arise when b is used, so the community decided to use bit all accross. So basically what Gerry Ashton said above. Same goes for Mbps vs Mbit/s. When the entire computer industry follows one standard instead of 20 at once, then the community will be more than happy to rank and file. In the meantime, it's a choice. Hennce B for byte, bit for bit. Headbomb {{{sup|ταλκ}}κοντριβς – WP Physics} 05:42, 22 January 2009 (UTC)
 * I suspect that no active IEC standard defines bit. Which IEC standard do you refer to? TechControl (talk) 22:28, 22 January 2009 (UTC)

Broken reference
Someone with edit privs: MOS:UNLINKDATES refers to this page via WP:CONTEXT; the sub-reference has apparently has been renamed over there, to WP:CONTEXT. Studerby (talk) 21:53, 22 January 2009 (UTC)
 * Done, thanks for noticing. Karanacs (talk) 22:04, 22 January 2009 (UTC)

Currencies
There are two independently maintained sections regarding Currencies - Manual of Style and Manual of Style (dates and numbers). The former should be merged into the latter, noting there are some content differences between these two Currencies MoS sections.

For this proposal, a mergesection notice should be added to the WP:MOSNUM section. This editprotected request is not to perform the actual merge yet to allow for discussion first, especially to confirm which statements should remain in the merged section. Dl2000 (talk) 00:36, 11 January 2009 (UTC)
 * ✅ Ruslik (talk) 10:06, 11 January 2009 (UTC)

Version comparison
A comparison between the versions would be useful for reference during the discussion:

There seem to be no major contradictions, mainly some wording and formatting differences. Ideally, the priority should be to merge what we have first, then do any debate on these statements later. The Serial number column can be used as a shorthand to refer to each statement. Dl2000 (talk) 03:52, 12 January 2009 (UTC)

Analysis and recommendation from the two versions
Having seen no comments so far, it's time for a proposed merged wording. Some comments on the items, noting that column A represents existing WP:MOS wording, column B for WP:MOSNUM. In the following, the convention "B6" means column B, serial row 6.


 * Serial 1: Coding in column A is more concise, so suggest using A1.
 * Serial 2: Identical, no evident issues, merge as is.
 * Serial 3: Column B text builds on column A without wording contradiction. A has the "xt" green formatting for US$123. Therefore suggest B3, with US$123 example highlighted as done in A3.
 * Serial 4: B4 is mutually exclusive with A8 - ISO 4217 is more a type of formatting than choice of currency, therefore leave this out of the Which one to use section. i.e. go with the empty "A4".
 * Serial 5: B5 generally builds on A5, no contradictions seen. However, some extra xt formatting is seen in A5. Also note the last words of B5 about avoiding EU with the € sign are similar to wording in A7. Therefore B5, but xt-format all currency value examples.
 * Serial 6: A6 seems the better version to move forward, with xt-formatting and more straightforward wording.
 * Serial 7: Since B5 is recommended, this suggests B7 is the better wording, although should do xt-formatting as done in A7.
 * Serial 8: Use A8 for reasons indicated in Serial 4.
 * Serial 9: Go with "signifiers" for now - therefore use A9, with its xt-formatting. The wording could be revisited after the merge.
 * Serial 10: B10 expands on A10, includes the rounding of conversion. Therefore, B10 but with xt-formatted values.
 * Serial 11: B11 guideline not found anywhere in A, so use that.
 * Serial 12: B12 generally expands on A12, with more explanatory guidance. Go with B12, with xt-formatting.
 * Serial 13: A13 does not appear anywhere in B, therefore use A13.
 * Serial 14: B14 does not appear anywhere in A, therefore B14.

Proposed wording of merge
Based on the above, the merged wording becomes:



Which one to use

 * In country-specific articles, such as Economy of Australia, use the currency of the country.
 * In non-country-specific articles such as Wealth, use US dollars (US$123), the dominant reserve currency of the world.  Some editors also like to provide euro and/or pound sterling equivalents, formatted as described in the next section.

Formatting

 * Fully identify a currency on its first appearance (AU$52); subsequent occurrences are normally given without the country identification or currency article link (just $88), unless this would be unclear. The exception to this is in articles related entirely to US- or UK-related topics, in which the first occurrence may also be shortened and not linked ($34 and £22, respectively), unless this would be unclear. Avoid over-identifying currencies that cannot be ambiguous; e.g., do not place EU or a similar prefix before the € sign.
 * Do not place a currency symbol after the value (123$, 123£, 123€), unless the symbol is normally written as such. Do not write $US123 or $123 (US).
 * Currency abbreviations that come before the number are unspaced if they consist of or end in a symbol (£123, €123), and spaced if alphabetic (R 75).
 * If there is no common English abbreviation or symbol, use the ISO 4217 standard.
 * Ranges are preferably formatted with one rather than two currency signifiers ($250–300, not $250–$300).
 * Conversions of less familiar currencies may be provided in terms of more familiar currencies, such as the US dollar, euro or pound sterling. Conversions should be in parentheses after the original currency, rounding to the nearest whole unit, with at least the year given as a rough point of conversion rate reference; for example, 1,000 Swiss francs (US$763 in 2005).
 * For obsolete currencies, provide if possible an equivalent, formatted as a conversion, in the modern replacement currency (e.g. decimal pounds for historical pre-decimal pounds-and-shillings figures), or at least a US-dollar equivalent as a default in cases where there is no modern equivalent.
 * When possible, always link the first occurrence of a symbol for lesser-known currencies (₮146); some editors consider it unnecessary to link the symbols of well-known currencies, but doing so can often be helpful to readers, as many countries use "dollars" or "pounds" as their base currency, and not all readers are familiar with the euro.
 * The names of currencies, currency subdivisions, coins and banknotes should not be capitalised except where normal capitalisation rules require this (for example, at the start of a sentence).
 * The pound sterling is represented by the £ symbol, with one horizontal bar. The double-barred ₤ symbol is ambiguous, as it has been used for Italian lire and other currencies as well as that of the British. For non-British currencies that use pounds or a pound symbol (e.g. the Irish pound, IR£) use the symbol conventionally preferred for that currency.

Comments/poll on the proposed wording of merge
Support/Oppose/further Comment? Dl2000 (talk) 00:20, 17 January 2009 (UTC)
 * support - Looks good to me. It's clear a merge is needed, and the proposal seems to have preserved all the semantics. Nice work. -Kieran (talk) 04:12, 20 January 2009 (UTC)
 * It's more than a week and there has been support and no objections for the proposed merge text. It is therefore proposed that the merge occur on or soon after 25 January 2009 00:01 (UTC) unless any objections or concerns are raised by then. Dl2000 (talk) 00:09, 24 January 2009 (UTC)

Conclusion
editprotected Based on the consensus of the discussion above, it is requested that the merge proceed, namely replacing the existing WP:MOSNUM with the Proposed wording of merge text as colour-boxed above. Once an administrator approves and proceeds with this, the redundant material in WP:MOS should be deleted. Dl2000 (talk) 03:46, 25 January 2009 (UTC)
 * Yes check.svg Done--Aervanath (talk) 10:23, 25 January 2009 (UTC)

Fix/replace date autoformatting
Well, now that there's an injunction in place to put a hold on mass date delinking (or relinking), perhaps it's a good time to discuss how to fix Date Autoformatting (DA). Although I disagree, the majority (not consensus) view is that dates should not, in general, be linked. So, I suggest that we make that into a preference option that logged-in users can specify, and have the default (applied to both anons and logged-in users who haven't otherwise selected a preference) be for autoformatted dates to not be linked. The next most common complaint seems to be the inconsistency in date formats seen by anons (and editors who have selected the "No preference" preference). The simplest way to fix that would be to just use DMY as the default format, while still allowing logged-in users to set a preference if they choose. That way anons would see a consistent format (DMY) and anyone who prefers a different format can create an account, login, and set their own preference. These changes can be implemented very quickly if there's community agreement (and I'll gladly write the code and shepherd it through the approval process on bugzilla) and can certainly be done before ArbCom lifts their injunction. Once in place, all dates on the entire site will instantly be made consistent and de-linked (except dates that have already been delinked, which will need to be edited manually to fix any inconsistencies in format.. or just put back in square brackets to re-enable autoformatting sans-linking).

I'm sure there are a ton of additional features we'd like to see, but for the purpose of building consensus &mdash; which should always be thought of as the minimal set of criteria to which (almost) everybody can agree &mdash; I thought we should start simple. Most of the people here (and who have spoken up elsewhere) don't like dates being linked &mdash; and those that do (including me!) will probably be satisfied with a user-specific preference to turn linking back on. Similarly, most everyone agrees that everybody &mdash; including anons &mdash; should see a consistent date format. While it would be nice to custom-tailor that format to each target audience, we don't have any good way of doing that. Using the IP address or browser "locale" settings won't work because of the way Wikipedia's servers work (the squid cache issue) and trying to guess at the majority readership of an article based on the "strong national ties" exhibited by the subject of the article is both inefficient and prone to causing arguments (and in many cases simply doesn't apply).

Please keep this conversation on-topic and impersonal; There's no need for "so-and-so is just going to derail this anyway" or "this is pointless" as you can simply choose not to participate in the discussion if you feel that way. Since none of us are supposed to be linking or delinking dates anyway, there's no reason to feel pressured by either side, or to look at this call for discussion as an attempt at delaying tactics. --UC_Bill (talk) 00:29, 22 January 2009 (UTC)
 * This always did seem like the most reasonable way forward to me (it addresses the concerns of those who dislike the "sea of blue", as well as the objections of those who think date links devalue other links). Especially with regards to making date linking a preference in Special:Preferences. —Locke Cole • t • c 00:40, 22 January 2009 (UTC)


 * I oppose any form of date autoformatting. However, if it is implemented despite my opposition, I have two requirements and one wish:
 * Requirement: The window in which users select the format either will not include the YYYY-MM-DD format, or will clearly state the format is NOT governed by ISO-8601
 * Requirement: No style manual and no instructions for any template will state that any date visible to article readers is in the ISO 8601 format (exception: template fields that clearly are only useful for dates with years greater than 1928 [when Nationalist China adopted the Gregorian calendar])
 * Wish: separate choices be available in the user preferences for dates appearing in articles and dates appearing in system messages, logs, watchlists, etc. --Gerry Ashton (talk) 00:49, 22 January 2009 (UTC)


 * There would need to be the ability to form a link, in specific cases, where a link to a specific date or year article is required. Keith D (talk) 01:13, 22 January 2009 (UTC)


 * This feature seems reasonable but I fear that scope creep may start to occur. Let's be on the look out for scope creep.
 * Flaviusvulso (talk) 02:03, 22 January 2009 (UTC)


 * Can you give an example of what you're asking for? —Locke Cole • t • c 02:07, 22 January 2009 (UTC)
 * Autoformatting never worked with links like January 1 . Auto-unlinking would probably not work with such links, either, without more rewrite to Mediawiki than it's worth. If any form of auto-format or auto-unlink exists for wikilinked dates, piped dates will be likely to exist as a way to wikicode an exception. Gimmetrow 02:19, 22 January 2009 (UTC)


 * I endorse the suggestion of UC Bill. I was thinking about exactly the same approach/solution but did not voice but felt uncomfortable voicing this view due to the controversial nature of this debate. I also endorse the reasoning of Locke Cole. This approach is a win-win scenario for both sides of this debate.
 * Flaviusvulso (talk) 01:25, 22 January 2009 (UTC)


 * I would strongly endorse having such an option. It resolves the issues expressed by both sides, while allowing personal choice to be the deciding factor. --Ckatz chat spy  01:48, 22 January 2009 (UTC)

Comment: Any documentation or examples should explain what to do about BC or BCE. Also, if the YYYY-MM-DD format is offered, dates such as AD 0033-03-21 or 0005-01-13 BC are going to look rather strange (but I suppose people who choose that format can change their preference if they don't like it). --Gerry Ashton (talk) 02:44, 22 January 2009 (UTC)
 * Is there anyone that seriously prefers this date format? This format is typically only used in CISAM databases and the like to facilitate date sorting.
 * Flaviusvulso (talk) 03:11, 22 January 2009 (UTC)
 * I use that format, at least. I don't see that ISO 8601 is really a crucial issue here – exactly the same issue of calendar confusion appears in any date format (autoformatting is simply a rearrangement of what was typed, after all). &mdash; Carl (CBM · talk) 03:16, 22 January 2009 (UTC)
 * I agree. The date format options which are offered are low level requirements and can be determined later.
 * Flaviusvulso (talk) 03:24, 22 January 2009 (UTC)


 * CBM wrote "exactly the same issue of calendar confusion appears in any date format" which is false. Every ISO 8601 date is either in the Gregorian calendar or wrong. Of course, a Julian date could be written in the YYYY-MM-DD format, provided it is made clear that the ISO 8601 standard does not apply to the publication in which the date is written. --Gerry Ashton (talk) 03:33, 22 January 2009 (UTC)
 * Again this is a detail that can be hammered out later. First we need to determine if the consensus is to use autoformatting or not.
 * Flaviusvulso (talk) 03:51, 22 January 2009 (UTC)
 * People who knowingly claim to conform to a standard and do not do so are liars. Telling lies is not a minor detail that can be hammered out later. Of course, date autoformatting is currently depricated, and the question of whether ISO 8601 applies to dates in the YYYY-MM-DD format is in a state of confusion, so the English Wikipedia is not currently telling lies on this issue, but the danger of becomming a pack of liars is real. --Gerry Ashton (talk) 04:10, 22 January 2009 (UTC)
 * It's not really clear to me why a discussion about "Whether or not Wikipedia claims to follow ISO8601" is relevent here.Flaviusvulso (talk) 04:25, 22 January 2009 (UTC)
 * The ISO 8601 standard requires every date written in accordance with that standard to be in the Gregorian calendar. If the English Wikipedia were to indicate that dates autoformatted into a YYYY-MM-DD format conform to ISO 8601, then the English Wikipedia would also be claiming that all dates that are marked up for autoformatting are Gregorian calendar dates. If that were not in fact true, then we would be reckless, a pack of liars, or both. --Gerry Ashton (talk) 04:32, 22 January 2009 (UTC)
 * OK, so lets not claim iso8601 conformance. Would that be ok?Flaviusvulso (talk) 04:41, 22 January 2009 (UTC)
 * In the design process (such as it was) leading up to the existing form of date autoformatting (which I summarize at User:Gerry Ashton/History of ISO 8601 and date autoformatting in Wikipedia) it is clear that date autoformatting was intended to conform to ISO 8601 whenever dates were in the YYYY-MM-DD format. Also, the text used in the existing preference window to choose YYYY-MM-DD format ("2001-01-15T16:12:34") is unmistakably ISO 8601. Given this unfortunate history, it isn't enough to just not mention ISO 8601; it is necessary to actively disclaim conformance with it. --Gerry Ashton (talk) 04:55, 22 January 2009 (UTC)
 * Gerry, your concerns are again, duly noted. I dont believe anyone disagrees with you. dm (talk) 04:50, 22 January 2009 (UTC)


 * DMY (if this means numeric) is ambigous. I have frequently seen extension writers on mediawiki.org writing things like 8/7/2005. Is this 8th July or 7th June? From edit history I often determined that it was US style. I know US people are often ignoring international standards, but maybe we can start making a CHANGE (yes we can) and do not allow this ambigous date format.
 * In tables or in other places where space is scarce I prefer YYYY-MM-DD, but in texts this maybe is not so good. If exact days get linked, the WP software should redirect /wiki/1999-08-07 to things like /wiki/June_7,_1999 or vice versa. Or a bot should create redirects. Really YYYY-MM-DD is very much more compact, it should be allowed in tables.
 * If linking is not generally allowed I support the idea of allowing this via preferences. Maybe it would be good to additonally allow some change in display for autolinked dates, so also users who have this preference=ON can see what the default is, without going to edit mode. TechControl (talk) 04:05, 22 January 2009 (UTC)
 * Everyone involved wants to use spelled out months, no one disagrees with you.  When creating dates with yyyy-mm-dd in tables, you should probably use dts or dts2 to sort columns properly, but still create unlinked text strings.   Check it out. dm (talk) 04:50, 22 January 2009 (UTC)

I'm opposed to anything that requires me to put dates in brackets. For a prolific writer, that is actually a pain. Even when date linking was encouraged, many new users didn't know that they were supposed to do so, leaving a lot of articles bare. Other new users (and many experienced ones) didn't understand how the date-linking was supposed to work, so we got a lot of January 30, 2007 and just plain 2005 because that is how those users thought it was supposed to work. Why confuse people unnecessarily? Just leave off the brackets so it is simpler for everyone. Karanacs (talk) 20:43, 22 January 2009 (UTC)
 * I think we tend towards using brackets because they're a simple syntax. Anything else we come up with will likely be some HTML/XML-like syntax like  which is even harder to use when writing prose. Plus, in this proposal at least, there will be an option to link dates, so using the bracket syntax makes sense with that goal in mind. —Locke Cole • t • c 20:57, 22 January 2009 (UTC)


 * Karanacs, if using brackets is too bothersome or complicated for a user they don't need worry about it. They can write the dates in plain text. This is a wiki so either a human or a bot can correct the date at another time. Flaviusvulso (talk) 22:11, 22 January 2009 (UTC)


 * Bots can't fix dates because they can't tell if they are within a direct quotation, which should not be autoformatted. Until a human fixes it, the date will appear inconsistent with the rest of the article for some readers. --Gerry Ashton (talk) 22:18, 22 January 2009 (UTC)
 * Semi-automated scripts can fix dates though. And nothing on here will ever be perfect, but we can strive towards perfection.. —Locke Cole • t • c 23:52, 22 January 2009 (UTC)


 * I don't see why it is so important that dates in quotations not be autoformated. If it is really important then the date format can be overridden by using the pipe symbol. Flaviusvulso (talk) 23:52, 22 January 2009 (UTC)
 * It is always important that quotations be character for character exact; in a digital age, that's how they will be searched for. Septentrionalis PMAnderson 06:40, 23 January 2009 (UTC)

One problem with the existing date autoformatting syntax is that it provides no way to describe a date range. The proposed solution would lead to an anon/no-preference user (they're the same thing, under the proposal) seeing a sentence like "The election was 3 November 2008 and inauguration festivities occurred January 17-22, 2009." --Gerry Ashton (talk) 01:36, 23 January 2009 (UTC)
 * This is a very very good point - either the date autoformatting functions have to supply the ability to format a date range themselves, or (the better solution) provide as a template magic word/variable/whatever an indicator of what date format to use so that templates can then be used to figure out the right date format. Neither are easy. --M ASEM  01:50, 23 January 2009 (UTC)


 * Date ranges are outside of the scope of this thread. We are only discussing how single dates are rendered. Flaviusvulso (talk) 02:00, 23 January 2009 (UTC)
 * I agree that single dates are the focus, but he makes a good point about the inconsistency in an article. —Locke Cole • t • c 02:07, 23 January 2009 (UTC)


 * I don't think this is an insurmountable problem, but I do agree it's likely this should be addressed first. See discussion below where I've mentioned your concerns to UC Bill. —Locke Cole • t • c 02:07, 23 January 2009 (UTC)


 * Date ranges should not be left out of the scope of the thread—this is one of the blunders made when the current system was put in place. Either propose something that will work or forget it. J IM ptalk·cont 10:24, 23 January 2009 (UTC)
 * Yes indeed, Jimp. You've emphasised just one of a number of complicated problems that would have to be sorted out if this solution chasing a non-problem were ever to see the light of day. Just one problem of the many that hung around Bugzilla for an extraordinary three years of circular discourse. No one has answered my basic question of why anyone would care about month-day or day-month order, when we're all perfectly used to both. Just a stony silence whenever I take the elephant in the room for a walk. Just on date ranges, en dashes are required, and they must be spaced or unspaced according to whether there are internal spaces within either one or both items. Minimal repetition is important, especially in infoboxes and tables, so we don't want "September 5 – September 8, 1991", or worse, the year twice as well. See the MilHist style guide for the info; it's so important for battles and the like, and I've cleaned up a lot of redundant date expressions in their articles.
 * So it gets back to: what is wrong with complete editor control over this by manual keying in? It's less work, and highly correctable in both edit and display modes. Why this insistence on a redundant toy that carries risk in so many respects, and is ... pardon me ... likely to make editing WP just a little more exclusive? Tony   (talk)  11:08, 23 January 2009 (UTC)

I'll reiterate my opposition to date autoformatting. Furthermore, I don't think the wishes of those who do want it have been fully captured. Some people don't want to see any date links. Some want to see occasional, or rare, links to especially relevant dates. Some seem to want every date to be linked (it isn't clear if this group would want dates that are not to be autoformatted to always be linked, even if the date has no relevance to the article). So it is necessary to modify the user preferences; rather than having a yes/no choice, a choice of never/only identified dates/always is needed. And it is necessary to have four date syntaxes, which I will illustrate by example: --Gerry Ashton (talk) 15:05, 23 January 2009 (UTC) corrected 17:50 23 January 2009 UT.
 * 1) January 23, 2009: don't autoformat and don't link
 * 2) January 23, 2009 : autoformat and only link if the user chooses "always"
 * January 23, 2009 : do not autoformat, and only link if the user chooses "only identified links" or "always"
 * New, undefined syntax: autoformat, and only link if the user chooses "only identified links" or "always".


 * I'm not sure about the third option (do not autoformat, link if "only identified links" selected) because there's a question as to whether user preference or editor intention should take priority. I usually favor user (i.e. reader) preferences in which case we'd need to consider whether to ever allow editors to force a link.  A similar situation applies to the "undefined syntax" example, for which I'd suggest something like what Sapphic describes below, using link, link to indicate that both the month-day and year are "identified" links that should be autoformatted, or link, 2009 to "identify" just the month-day (and similarly for year).  I'm certainly not opposed to having three different date linking options in user preferences, but I'm wondering if it's actually necessary.  --UC_Bill (talk) 17:06, 23 January 2009 (UTC)