Misplaced Pages

This is an old revision of this page, as edited by Ohconfucius (talk | contribs) at 01:52, 6 April 2009 (add archive). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Please stay calm and civil while commenting or presenting evidence, and do not make personal attacks. Be patient when approaching solutions to any issues. If consensus is not reached, other solutions exist to draw attention and ensure that more editors mediate or comment on the dispute.

Archives

/Archive 1, /Archive 2, /Archive 3

• WT:DATEPOLL

Exit strategies

It's probably too early to say, but I think this poll is hopelessly compromised, as well.

Comment about Ryan's "What we want is for the poll to get one proposal from each section....". It can no longer be done for the linking sections. Because of the biased subtitles (link only to relevant dates), the only conclusion possible is that that statement has consensus, but not necessarily proposal 1. It's still conceivable that a clear consensus for one of the proposals could develop, but it's unlikely, as we have to consider a !vote for any of the options which says only "link only to relevant dates" as a vote not showing a preference between 1, 2, and 4. I'm not saying I think this is the only fatal flaw in the linking sections, but it seems sufficient.

As for the autoformatting, a large number of voters seem confused as to whether this refers to linking; probably enough to effect whether "oppose" gets a supermajority. I think Ryan needs to clarify that it does not refer to linking, and spam all !voters who voted before the change. (He made a change, but it doesn't seem to have helped.)

But we don't have an exit strategy, unless Ryan or ArbCom has one hidden. — Arthur Rubin (talk) 15:45, 1 April 2009 (UTC)

• “Oh dear! Nothing but confusion, confusion, confusion. What is an editor with *pinky promise* good faith to do??”

  Nothing is confusing, Aurthur. You guys have had your asses handed to you on a plate. As Ryan pointed out above (21:47, 30 March 2009):

So cease with your posturing about how the entire Wikipedian community is doomed to have this issue drag on endlessly like a herpes infection because you can reach into your wikilawyering bag of tricks and spew B.S. about how there is this or that you don’t like about how the RfC was conducted. Tough. The community has spoken: just write out the damned dates in non-linked, fixed text and be done with it. You don’t like that outcome? Fine. How about accepting that the community has spoken and accept its will with grace and dignity? Greg L (talk) 17:05, 1 April 2009 (UTC)

Define "Relevant" in a way so as to distinguish between the options. The options basically boil down to:

1. Link some dates
2. Link some more (but not all) dates
3. Link all dates
4. No guidance.

Have a look at the spread of the votes as to what "some dates" means, and tell me that you could write a MOSNUM guidance based on that. You've got the whole band between linking nothing and treating dates like other links. How on earth do you distill that down? About the only thing which is clear from this poll so far is that there is no consensus on autoformatting in either direction. The other stuff is just too non-specific.AKAF (talk) 17:19, 1 April 2009 (UTC)

Isn't the MOSNUM guidance exactly what people are voting on? Each option lists what text should be inserted into MOSNUM. Am I misunderstanding your response? I think the next RfC is supposed to further clarify how to implement the guidance. Karanacs (talk) 18:36, 1 April 2009 (UTC)

Arthur Rubin: You appear to be building a case to challenge yet another RFC result that you do not like. Perhaps it is time to accept the vote after more than three RFCs on this matter. Attempts thus far to query voters' reasoning, on their talk pages, have apparently resulted in no changes in their vote and, in a few cases, irritable responses. Tony (talk) 17:28, 1 April 2009 (UTC)

It sounds like the cases of William Penn and others. The jury found them not guilty and the judge wasn't happy with the verdicts. He said:

• "You will not be dismissed until we have a verdict--a verdict that the court will accept. And, until we do, you will be locked up, without meat, drink, fire, and tobacco; you will not think to abuse the court. By God, we will have a verdict, or you will starve for it!"

Lightmouse (talk) 17:46, 1 April 2009 (UTC)

• AKAF: Your arguments fall on the deaf ears of any rational person. As of this writing, the voting on “month-days” is 159 - 5 - 4 - 23. It doesn’t take a rocket scientist to figure out how one should proceed. What part of “Accepting what life throws at you with grace and dignity” don’t you understand? Greg L (talk) 18:17, 1 April 2009 (UTC)

For what it's worth, I would agree with you about month-days, except that some said, and I quote "link only to relevant dates" for their vote reasoning on option 1. And I did object to the subtitle before the vote, but, since I'm not on 24/7, it was after the lockdown. I also objected in the comment section, but I really don't expect most editors to read down that far. — Arthur Rubin (talk) 01:05, 2 April 2009 (UTC)

As for "any rational person": "Any rational person" would assume that any situation where where the proposer votes "no" on his proposal, is hopeless. — Arthur Rubin (talk) 01:05, 2 April 2009 (UTC)

I don't see your problem about month-day links. Option 1 is essentially option 4 plus a clarification that such links are almost never relevant. It draws attention to the fact that there is no longer a special exemption for them. This clarification has become necessary because of the past practice of making irrelevant date links for autoformatting purposes. I searched for "relevan" among voters for option 1. Most of them specifically express the sentiment that such links are almost never relevant and seem to feel (like me) that this needs saying to prevent conflicts with the minority of editors who disagree. --Hans Adler (talk) 01:46, 2 April 2009 (UTC)

No, it does more, and been used to assert much more. Septentrionalis PMAnderson 05:31, 2 April 2009 (UTC)

Care to give an example of a relevant link forbidden by this language? --Hans Adler (talk) 08:03, 2 April 2009 (UTC)

Rubin's right. Barring some unlikely surge in the polls, it's pretty obvious that the result is going to be "no consensus" on the question of autoformatting (a roughly 60-40 split is hardly definitive) so which status quo do we preserve? Each side is obviously going to argue for their own preference, and absent any clear consensus from the community, I don't see how to break the deadlock. If autoformatting is kept (and fixed) the other questions are basically irrelevant, so it's really the central issue. I wish people had taken it to heart when I pointed that out last month, and if we'd gone with a simple up/down vote "poll" on that one issue, I bet we'd have a clearer way to proceed now. I don't mind the prospect of "losing" the poll as much as I do having the cloud of uncertainty continue. That said, I'm not going to give up my argument based solely on that factor. So where do we go from here? --Sapphic (talk) 06:22, 2 April 2009 (UTC)

Well, that's a skewed argument if ever I saw one. Autoformatting has been deprecated since August on the style guides; it is totally absent from the Featured Content process, without a blink. It is whistling in the wind for a few people who don't like the results of this poll 38.5% (versus 61.5%) to claim that the clock should be turned back to the old days. Move on and get over it: the WP community has matured and is telling you yet again that it does not want dates messed around with. How many RFCs that say the same thing on this do we have to have? Tony (talk) 09:41, 2 April 2009 (UTC)

Not to throw around firecrackers but (given equal strength of argument) doesn't 61/39 give a majority? Albeit a slim majority but why are these polls regarded as "no consensus"? I know that "Misplaced Pages is not a democracy" but surely if there is a prevailing view out of two opposing views then we should go with that one? Why can't we just remove the current double brackets autoformatting system (as people have been blocked for) and then discuss a proposal for a new autoformatting system when a better (non-date-linking) system/syntax is created? If past polls show that "autoformatting through wikilinks" is deprecated then why can't we remove that old system? Sillyfolkboy (talk) 13:16, 2 April 2009 (UTC)

For some issues (such as deciding what default date format we should use, some kinds of style guidelines, etc.) a majority is fine, but when you're talking about altering core software features and editing millions of articles, there should be overwhelming support, and here there isn't. We usually won't even delete a single article (through AfD) with as slim a majority as there is here. Also, the double-bracket markup around dates isn't the autoformatting system, it's just the way of triggering the autoformatting system. We could remove (or better, just disable.. since it's one true/false setting in the config file) the existing autoformatting system, but while that would stop dates from being autoformatted, they'd still be linked and the markup would need to be removed. Just removing the markup without disabling the autoformatting software would make it too complicated when editors do want to link to a date (they'd have to use either the ] stynax or ] syntax to avoid triggering the still-active autoformatting software.) The problem with removing all the markup (and disabling the software) before a replacement system is put in place is that it would result in a lot of duplicated effort, since any replacement system would need its own markup similar to the double brackets. Removing the markup is easier than adding it back (mostly because of quotations of dates, which should never be autoformatted, and which are hard to distinguish from other unlinked-but-potentially-autoformattable dates) so it's not even a matter of just "undoing" what bots already did... unless of course we kept detailed logs of all the unlinking. --Sapphic (talk) 14:35, 2 April 2009 (UTC)

How many quotations of dates are there? What fraction of currently linked dates are going to remain linked? I'm not convinced that unlinking is going to be any simpler than re-linking, since nobody is going to argue over whether a date is inside a quotation, but lots of people might argue about whether a date should remain linked or not. As for exit strategy, I think it might be reasonable to look at all the "support" votes and see which ones actually support the existing autoformatting software and which support autoformatting "in general" and see if we can get a supermajority in favor of at least getting rid of the old autoformatting system. Then, assuming there is such a supermajority, we could disable the existing autoformatting immediately by changing whatever config setting you're talking about. That'll let every editor see the inconsistent formats, and get more people involved in fixing that problem. At the same time, we can start working on a detailed specification for the replacement software, which enough people seem to want that it's probably worth at least looking into. Yes, it might mean a lot of wasted effort in de-linking and then re-linking dates, but a lot of the effort won't be wasted, such as fixing format inconsistencies and figuring out which dates are more relevant than others, etc. —Preceding unsigned comment added by 169.229.149.174 (talk) 15:55, 2 April 2009 (UTC)

• Point of information: the community voted in August 2008 to deprecate date autoformatting, and endorsed it overwhelmingly (i.e. by a supermajority) in (1) and (2) specific questions in December 2008. Therefore, Date autoformatting as we knew it is dead, and the software should have been disabled at that point. What we are now discussing is the desirability in principle of a new system. Any eventual consensus to adopt would need to be followed by a formal consultation process and vote on detailed specifications. Looking at the stability of the 60%+ vote opposed to that principle, it is likely that a consensus will not be attained. Ohconfucius (talk) 06:28, 3 April 2009 (UTC)

Hmm.. I guess you have a point about de-linking being harder than re-linking would be, once you factor in having to deal with disputes over linking particular cases in individual articles. Figuring out which date-like-things are actual dates and which are quotations of dates (or other things that should never be autoformatted) is hard for computers, and simple for humans — but figuring out which dates are really relevant to an article is hard for both computers and humans. In other words, a date re-linker bot would probably have a low error rate and the errors would be simple for any editor to fix in a way that everybody agrees with (like reverting obvious vandalism.)

So as one of the most ardent (i.e. loudest and annoying) supporters of date autoformatting, I endorse the immediate disabling of the existing date autoformatting (set $wgDynamicDates = false) on the English Misplaced Pages, followed by the resumption of manual and/or (semi-)automated "mass" de-linking of articles with human correction (according to an as-yet-to-be-determined set of criteria) — coupled with the establishment of a community specification and review process for developing a replacement system, which will be presented in a subsequent poll for final approval or rejection (with possible abandonment of the development process, pending any changes in community opinion and/or new information and experiences gained in the intervening time.)

The key piece in getting my (and I suspect a lot of other autoformatting supporters') backing for that proposal is that the development process receive some kind of official blessing (by ArbCom?) with enforcement against anybody trying to "derail" it. If we accept that practically nobody wants the old date autoformatting, then you accept that enough people want a new autoformatting that we have to at least give it a serious go. In the meantime, you get your way 100% and don't interfere with the development process. Deal? --Sapphic (talk) 00:20, 3 April 2009 (UTC)

• Quoting you, Sapphic: (a roughly 60-40 split is hardly definitive) so which status quo do we preserve? and …coupled with the establishment of a community specification and review process for developing a replacement system . And you see a consensus that the Wikipedian community wants a new system of autoformatting… uhm… where? The old system of autoformatting with it’s attendant linking to trivia is gone. Dead as a door nail.

  So what to do next? It was conceded by Locke and UC Bill that such “specifics” as UC Bill’s “son of autoformatting” idea would be rejected out of hand by the community. So Locke insisted that the RfC be put forth just in terms of the “generalities of autoformatting”. So that’s just how we structured the RfC: on the “generalities”. Really, though, what few specifics snuck in were based on Werdna’s “specifics”. And the community’s reaction to this? There is clearly a significant majority of Wikipedians who don’t want some newfangled autformatting technology. Yet, you cite the community rejection (it wasn’t a colossal rejection, just a sound drubbing) as evidence that you should get busy, roll up your sleeves, and start working on some newfangled methods of autoformatting. Because… why???  Fine. You go ahead and work your head off. But in case you haven’t been keeping up on current affairs, the community is sick to death of this issue and doesn’t want to see it darken their doorstep for a long, long time. So if you come up with some new autoformatting idea that is the coolest thing since steam power and antibiotics, just keep it to yourself.

  You see, just because Sapphic and a handful of enthusiastic volunteer programmers really, really want something just isn’t good enough. Misplaced Pages’s Chief Technology Officer and a clear majority of Wikipedians don’t. Maybe we ought to listen to what they want, huh? Or does your right to hound the community on this issue exceed the majority’s right to be free of houding? Greg L (talk) 05:14, 3 April 2009 (UTC)

I don't see a consensus at all, at least on the issue of autoformatting. That's the point. We need to figure out a way to proceed from here, with no clear consensus in the poll results. Yes, a majority have opposed autoformatting, but that's not enough. At roughly 40% of the respondents indicating support for autoformatting, you can't dismiss the supporters as "a handful" like you have been previously.

So, as a compromise, I'm suggesting we try it your way while those who want a new software system work in peace, then we put it to another poll to see if we try that. I'd think you'd be delighted, you're getting what you want and all you have to do is not try to poison the effort to develop a new software replacement. Given how long it took to put this poll together, and the need for even more transparency and community buy-in for the new software, I imagine you'd be getting your way across the site for at least a month or two. If we agree to have the existing date autoformatting system turned off right away, it would speed up the process of getting inconsistent date formats fixed and get more people aware of the issue, because right away every editor would see the site as anonymous editors currently do, regardless of what their (now non-functional) date preferences specify. Addendum: Disabling it in the config also allows dates that should be linked (whichever those may be) to be linked using the simple ] syntax instead of some more cumbersome variety needed to defeat the autoformatting, if it were left turned on.

If after a month or two of that we don't see an increase in date format or date linking edit-warring, complaints from editors who start demanding their preferences start working again, etc. then it's entirely likely that the poll to approve the new software will show that people no longer support it at the same level as now. Maybe it really will dwindle to "a handful" and a clear consensus will emerge. Or maybe you'll see that date autoformatting (even in its current, flawed form) is really protecting us from worse headaches, and welcome the new improved replacement.

Either way, I'm willing to run a little experiment to get some real-world feedback, if you're willing to keep your nose out of the development process (unless it's to genuinely contribute to developing the specification or something, which I doubt you'd ever want to do anyway.) --Sapphic (talk) 05:16, 3 April 2009 (UTC)

• Quoting you: I don't see a consensus at all, at least on the issue of autoformatting. The trouble is, you need one to push what you’re pushing. I’m quite content to let the RfC run its course, and for the ArbCom committee and the other admins to look at how the community has spoken, and for them to instruct you handful of volunteer developers as to whether or not they think the community has asked you to keep coming back again and again, pushing your latest & greatest. I’m just not seeing this invitation from the community so far. Greg L (talk) 05:24, 3 April 2009 (UTC)

• You think they're sick to death about date-related arguments now? Just wait until they've experienced life entirely without autoformatting for a while, without a way to end the arguments by saying "just go set your preferences that way, then." If you're so convinced that any replacement system is doomed to fail, why do you care if it's developed by other people who want it and believe it's useful. Let it be developed, and let the community decide whether to use it or to continue on without autoformatting. You seem to keep missing the fact that you can have autoformatting gone as soon as Ryan (or whoever on ArbCom) convinces the Wikimedia sysadmins to turn off autoformatting for real, and not just the work-around way you've been doing so far. It could seriously be gone in a day or two from now, if we go this route. You'd still need to deal with the links, but I'd support whatever bot or scripted delinking method you wanted to use, as long as there was a clear way of dealing with disputes over whether to keep specific date links. With a large part of the dispute (autoformatting) made moot for at least a month or two, I bet those link-specific disputes would be less stressful and easier to resolve. Seriously, what have you got to lose, except the right to sabotage the development process? --Sapphic (talk) 05:27, 3 April 2009 (UTC)

• Just wait until they've experienced life entirely without autoformatting for a while. Not that! Oh the humanity! Greg L (talk) 05:33, 3 April 2009 (UTC)

• P.S. Sorry. I’m tweaking your nose and I see you are sincere. I just looked at your edit summaries and I see your mood and intentions are quite different from the last time I had the pleasure of encountering you. Now I feel bad. Please, just let the RfC run its course and accept with grace and dignity that the community isn’t asking you do do what you’re doing. Moreover, they’re really, really fatigued of this RfC issue. Let it die. At least, give it a rest for a week until this RfC concludes. Greg L (talk) 05:37, 3 April 2009 (UTC)

Wow, okay, thanks for the apology. I probably don't even deserve it, given how shitty I've been toward you in the past. Anyway, I'm concerned that the poll isn't going to resolve anything (I seriously doubt ArbCom will want to endorse one side with only a 60% majority, despite what you seem to think) and we'll remain stuck in limbo forever unless we work out some kind of compromise. If you don't come around heckling the development process, I think we really can come up with something pretty good as a replacement, and I'm convinced that when they see it (and have had a taste of the old date format wars coming back.. I don't think people have become quite so "enlightened" on the matter as you seem to believe, either) that a lot more than 60% will want it. A lot of the opposition is because of how the current software works, and if people see a working, tested, fully-specified system developed through a transparent community-driven process that doesn't suffer from the same flaws, I think they'll embrace it. Maybe I'm wrong, but wouldn't it be easier to just find out, rather than continue arguing about it forever? --Sapphic (talk) 05:47, 3 April 2009 (UTC)

• Time for bed. Goodnight. Greg L (talk) 05:53, 3 April 2009 (UTC)

• Sapphic, a > 61% versus > 39% result is not what you wanted, is it, to claim that people want any kind of date autoformatting. Now you are trying to twist the result around in contortionist ways to claim that you should still have your way, as though it were the converse result. Ah ... let me think about that ... I don't think anyone would buy that, except for you, Cole, Katz and a few other devotees. Six months of plain fixed-text dates has rapidly convinced Wikipedians that there is absolutely no problem to solve ... as though they are concerned about "realize" versus "realise"; they are not. Nor are they concerned that some people pronounce "either" with an ee, and others with an ei as in "bite". It's all too silly. We do not want dates messed with; that is what people are saying, again and again and again. Now you're talking of another RFC as though you can force people through tiring them out. Tony (talk) 07:32, 3 April 2009 (UTC)

• Yes, a very clear sub-text to the feedback is that the community is thoroughly sick of this whole lame debate. Most agree there is no problem to solve, and some have stated their annoyance at being asked their views again and again. We all know the reason for this is that the Dead Parrot is still nailed to its perch although it is "pining for the fjords". Just put the "ex-parrot" in its box, and let it rest in peace. Ohconfucius (talk) 08:44, 3 April 2009 (UTC)

It seems to me that a lot of the votes in favour of the principle of autoformatting were empty. It's a bit like being against sin - nobody could possibly argue with the principle, but the reality is a bit harder. The difficulty would come when the principle butts up against the reality of having to mark up millions of articles and dates. If the developers want to spend time trying to come up with a neat autoformatting solution, some work was done a while back on trying to develop a minimum spec. But we shouldn't be discussing it again unless and until there's a working system that meets the minimum requirements. It may indeed be that we'll welcome it with open arms in a few months as an escape from an outbreak of formatting wars, but I doubt it. And absent that, I think there would be no possibility of persuading the community to take on the massive task of building in the necessary markup, just for a 'nice to have' feature. There are plenty more productive ways to spend our time. Colonies Chris (talk) 12:51, 3 April 2009 (UTC)

I'm not asking to preserve markup anymore — that's the compromise part. Go ahead and de-link. My offer even still stands from long ago (remember the date formatting wikiproject?) to generate work lists from analysis of the wikipedia dumpfiles to help in fixing articles with the worst mismatch of date formats. All I'm asking in return is that nobody try to derail/naysay/heckle or otherwise interfere in a non-constructive way (enforced by ArbCom) with the development of some replacement software, which will be put to a final RfC whenever it's ready. In the meantime, the existing date autoformatting software is disabled by a change in the site's config file (takes effect basically instantly, across the entire site) and then de-linking and format fixing can proceed however it's decided upon. --Sapphic (talk) 04:39, 4 April 2009 (UTC)

(ec) Oh sorry, I just realized you were probably alluding to the work needed to re-link (or otherwise mark up) dates, if some new software were to be used. It has been pointed out to me that there really aren't many cases where a bot would fail, and correcting its errors would be simple for human editors ("is this inside a quotation, or not?" will people really disagree on that?) as opposed to the error rate and need for correction/disputes over which dates should be linked on their own merit. So putting markup back around dates could be done almost entirely by bot. It has also been pointed out to me that de-linking actually adds more useful metadata (in virtue of the more relevant date links that are left behind) than it destroys (by failing to distinguish between less important dates and quotations of dates) so de-linking isn't as bad as I thought, even if I hope to eventually re-mark-up the dates anyway (but at that point, preserving the new information about which ones are more and less relevant.) --Sapphic (talk) 04:58, 4 April 2009 (UTC)

• How can it be a compromise if more than half the voting public don't want it??? It is not how consensus works. What you are offering 'in exchange' for the "compromise" is the ability to delink. Well, it's mighty kind of you, but I think we have that already if not for the injunction. Remember that our objection is not about the links - it seems that you are the one confused that we are talking date autoformatting here. If you fail to achieve consensus for this principle - and I really don't see a cat in hell's chance of programmers running amok now with the share of vote of the supporters hovers just below 40%) the community is getting a raw deal if the opposers stand aside and let the techies run riot; that would be irrresponsible of us knowing it will be another big mistake. The writing is already on the wall: well over half the voting members of the community do not want it. Ohconfucius (talk) 04:54, 4 April 2009 (UTC)

I'm not sure why you think a 60-40 split is going to get you a favorable ruling from arbcom, or the right to disable the existing date autoformatting. All it's going to get us is more deliberation, more "phases" of the polling process, more drama. What I'm offering in "exchange" is to put a stop to all that right away, at least with regards to myself and anyone else I can convince. If a hard-core supporter of autoformatting like me can live with that plan, I suspect enough others will that we can get something like 80-90% of the concerned parties behind it, if not more, and actually convince the Wikimedia sysadmins to turn it off right then and there (which I have to admit I'm now getting really curious about, just as an experiment to see if it really does bring back the old date format arguments, to what degree, how quickly, what articles/topics, etc.) The dates would still be linked and there's still probably a lot to argue about there (though I'll stay out of that part; I don't care) but at least every editor would immediately see the dates the same way anonymous readers do, and be aware of the problem of inconsistent date formats that (I think) everybody agrees is a genuine problem. I think that's a better outcome for you than trying to "wait it out" and let the whole process play out, still not resolve itself, start anew, etc. that you know is pretty likely. How long has all this been going on for? --Sapphic (talk) 05:13, 4 April 2009 (UTC)

Oh I just saw the part about me be confused. (Jeez I really am tired.) No, I'm just not being clear, I guess. I'm saying to turn off autoformatting immediately. That doesn't require changing any articles, it just requires a one-line change in the server config file. If enough of the community asked for that, I'm sure they'd do it right away. So I'm saying do that like tomorrow (or however long it takes to get enough people on both sides to chime in that they're cool with it, given the other conditions about leaving the development process alone) and then proceed with de-linking according to however that part of the debate pans out. Which seems to be a lot of de-linking, which is actually fine with me. (I want the date links to be controlled by user preferences remember, so I couldn't really care less about what the defaults are, which is how I view those questions.) Now I'm really going to bed. Please think over what I'm proposing and forget that I'm the one proposing it and just consider if you really have anything to lose. --Sapphic (talk) 05:22, 4 April 2009 (UTC)

• There's nothing tempting on any channel. I'm switching off ;-) Ohconfucius (talk) 05:58, 4 April 2009 (UTC)

... could an interested onlooker get some clarification, please? does Sapphic have some sort of mandate to negotiate "deals" on this issue, and if so, on whose behalf? and who has been authorized to represent the "other side"? thanks Sssoul (talk) 06:05, 4 April 2009 (UTC)

• Even if she did, none of us individually on 'the other side' have the authority to negotiate on behalf of those who oppose, either, so it's a meaningless discussion. Ohconfucius (talk) 06:17, 4 April 2009 (UTC)

that's what i thought. thanks for confirming. Sssoul (talk) 07:14, 4 April 2009 (UTC)

• Sapphic's "What I'm offering in "exchange""—I myself would not presume to do "deals" that cut across the community's opinions. Please note that ArbCom deals only with behavioural issues. Tony (talk) 06:51, 4 April 2009 (UTC)

The only people I'm "negotiating" for are the ones I already mentioned, namely "myself and anyone else I can convince." I happen to think that if we can work out a deal acceptible to me, Locke, etc. and the people with extreme positions on the "other side" like Tony, Greg, etc. then it would be much easier to convince the more moderate folks. If something like 90% of the participants all agreed to a common plan then we could get the software setting changed (so linked dates become just linked dates, like anons see them) and the injunction lifted right away. You get the blessing of arbcom for delinking less relevant date links, ones that were linked solely for autoformatting (and nobody except the most extreme "link everything" would disagree), anyone that argues with you can have a statement from them (with enforcement) to contend with. The pro-autoformatting side isn't allowed to harass you about your date delinking. Similarly, you're not allowed to interfere with the development process. The fighting children are each sent to their own room, and not allowed to pester each other for a while. --Sapphic (talk) 22:19, 4 April 2009 (UTC)

• As I said before, this is a faux compromise. For me to agree to it would be a sellout not only to myself but also to the 200+ people who have joined the opposition to the principle of DA. BTW, I am with the majority of moderate/conservative wikipedian, and I really, truly, and wholly object to being labelled as one with "extreme positions on the 'other side'". Speaking of labels, I wonder how to label you 'guys' who are pushing this new-fangled techie agenda of a once-failed solution. Ohconfucius (talk) 03:41, 5 April 2009 (UTC)

• "Similarly, you're not allowed to interfere with the development process" There ought always to be proper checks and balances; it should always be a right and responsibility to exercise that counter-balancing role. The reason we have been dealing with this crap autoformatting which nobody wants for the last 6 years is because it was a techie project without proper community input. Techies should never be given completely free rein on Misplaced Pages. Ohconfucius (talk) 04:05, 5 April 2009 (UTC)

• Yep, while respecting the skills of "techies" and their right to have a say as individuals like all others in the community, there's a problem when programs/patches can be knocked up at a moment's notice and slapped into WikiMedia's system ... no questions asked. Ironically, to run a bot on WP, you have to go through the hoops of community input. As HWV258—himself a professional programmer—pointed out last week, there's something very fishy about this.
• Now, there's still talk here of bargains and deals. Sorry, but to repeat myself, it is inappropriate for individual editors to strike deals that cut across community opinion. Unless it's a deal like ... "I'll collaborate with you, Sapphic, Cole, and Rubin, on WikiProject working bees to fix up the lamentable state of year articles, if you agree to do so too." Then yer talkin'. Not that good year articles would change the issue of relevance and linking to them; not that it would change the fact that they are privileged in continuous main-page treatment. Worth doing, don't you think, and a damned sight better use of our time than bickering over a risky solution to a non-problem. Tony (talk) 06:21, 5 April 2009 (UTC)

• So, you're suggesting that Brion, Werdna, and the other Misplaced Pages developers just slap in any old bit of code they like into the system without thinking it through? Seems to me that "assume good faith" should be extended to the folks who maintain the system as well. --Ckatz_spy 09:21, 5 April 2009 (UTC)

I don't know what Tony would say, but I am happy to adopt your words as my own. Brion, Werdna, and the other Misplaced Pages developers just slap in any old bit of code they like into the system without thinking it through. I assume good faith, but I do not assume any ability on the part of developers to envision how their changes will affect the wider community. --Jc3s5h (talk) 15:14, 5 April 2009 (UTC)

• No, there is something which neither Katz nor Jc3 appear to understand about the need for checks and balances; Vibber may have gone into it with his eyes open, but he must've realised the DA software was a turkey when he said "My personal recommendation would be to remove all date autoformatting and let a sane manual of style recommend the fairly standard international English form …". Ohconfucius (talk) 15:36, 5 April 2009 (UTC)

I think Jc3 is generally agreeing with us on the issue. CKatz, your edit summary appeared to imply that I'm sticking the knife into programmers; I should not have to refute that implication. I assume good faith too on the part of programmers; but although they have their own particular skill-sets, they're not all professionals like Brion Vibber. Apart from his authority as CTO of WikiMedia, in the quote above he's just applying the normal, common-sense observation that simplicity is valuable in its own right. We don't have to read Edward de Bono's book on Simplicity to work that out in relation to advantages of not messing with dates.

In any case, even if members of the WP community who dabble in programming were all top professionals, the community still deserves to know ahead of time what is happening, and to be able to comment. By analogy, we have an extraordinary pool of professional talent in areas that are germane to WP (prose, research fields, copyright, to name a few), and I hear no argument that the normal checks and balances and community input on their activities is somehow an affront to their professionalism. I suspect that even Brion would not mind the establishment of a proper process for notifying programming changes/innovations/requests ahead of time, in a forum in which the community can comment. Tony (talk) 16:00, 5 April 2009 (UTC)

That "proper process" you describe is exactly what I'm calling for. I'm just asking 1) that nobody try to derail it (which is a behavioral issue that arbcom can enforce) and 2) that any end result of the process be put to a public rfc/poll/whatever-euphemism-for-vote-you-prefer. In the meantime, you get to have DA completely disabled and you get to resume delinking on a mass scale — with the blessing of arbcom and without interference from the pro-autoformatting side (the pro-linking side might be another matter, but I'm not actually one of "them" so I don't know.) I'm pretty sure I could convince most any autoformatting supporter to go along with the proposal, and if you think you could convince most autoformatting opponents, then right there we have enough people to get Ryan (and whoever) to take the proposal seriously, and act on it. --Sapphic (talk) 18:54, 5 April 2009 (UTC)

• Sapphic, I’ve learned to look at your edit summaries first, as they are a valuable resource for gaining insight into your mood and motives. The edit summary accompanying your above post was this: one last try, then we can just go back to keeping the injunction and having lots more phases of polling. In response to this, I feel I should draw your attention to something Ryan wrote (21:47, 30 March post, above) only six days ago:

Ryan’s words are important, in my opinion, because they speak not to the issue of how he feels, but of how the community feels, and how he thinks the arbitrators are inclined to satisfy the wishes of the community. This is rather like one of those classic situations where “the needs of the many outweigh the needs of the one.” The community is sick to death of this autoformatting/linking issue and I would be utterly shocked if the decision, given the past and current RfC results, was that there ought to be yet another RfC. You will note that some of the most vociferous (“textiferous”?) proponents on autoformatting seemed to have looked at the early RfC returns and accepted what fate has handed them. I encourage you to sit back and watch. And when the arbitrators have rendered their decision, I hope you accept it with peace and tranquility. Really, this entire issue isn’t at all important in the grand scheme of things. Greg L (talk) 19:49, 5 April 2009 (UTC)

Thanks for bringing that to my attention, Greg. I agree that what arbcom actually does isn't a matter of opinion, but something we can just sit back and wait for. I'd been working on the assumption that their actions following this poll would be consistent with similar actions in the past, and that a 60% majority wasn't going to be used to justify a site-wide change. But if you (and other autoformatting opponents) really believe that the result of this poll will be to sanction your position, then so be it. I still think it's more likely that we'll just have more polling and a continuation of the injunction... but I'm willing to just wait it out and see. If things don't turn out the way you like, and we find ourselves with continued deadlock on this issue, my offer will still stand. I'll pick this back up, then. --Sapphic (talk) 21:55, 5 April 2009 (UTC)

One last note on the results of this poll: I find it odd that date formatting opponents see it as a "victory" that about 60% of the respondents (actually slightly less than than now, but whatever) sided with them — when their whole argument from the get-go was that it was only a "handful" of holdouts on the other side. Over 40% of respondents want autoformatting. That's not just a handful, that's a lot. I think it would set a really really bad precedent to make a site-wide change based on those results, especially if it happens in the half-assed way of unlinking (mostly) all the dates but leaving the DA software turned on that might result. --Sapphic (talk) 22:08, 5 April 2009 (UTC)

• I’m a veteran of the IEC prefix wars (“megabyte (MB)” v.s. “mebibyte (MiB)”). They used the same argument: that it was easy to implement a dumb-ass idea with 24 hours of discussion on a remote, backwater venue with about thirty participants and only a roughly two-thirds approval. And for three solid years and fifteen “Binary” archives on WT:MOSNUM dedicated to the issue, they successfully argued that Misplaced Pages’s use of the IEC prefixes (which made Misplaced Pages the only publication on the planet that used such terminology when communicating to a general-interest audience), was somehow effectively grandfathered in because it should take an overwhelming supermajority to undo something once done. Think about it: For three years, Misplaced Pages was the only general-interest publication in the world using “KiB” and “kibibyte”; the terms aren’t even in Microsoft’s dictionary of computer terms to this day. What a stupid thing to have done.

  I’d love to see just how many developers participated in blowing the now-deprecated autformatting system out of their ass and how long it was discussed by the Wikipedian community before being implemented. I’d hazard a guess that the decision to autoformat dates was *sorta quick & easy*. Regardless, after four RfCs, the community’s mood is abundantly clear now. And now, fresh off the IEC prefix issue, I don’t have much sympathy for an argument that amounts to this:

Nope; no sympathy at all. If there is one thing I hope the arbitrators settle here, is this: A simple, clear majority of Wikipedians is sufficient to establish a consensus whenever the issue has enjoyed an enormous amount of discussion and wide community participation. Greg L (talk) 23:57, 5 April 2009 (UTC)

• I'm exhausted, and I think this thread is, too. Ommmmmm, Namaste. Ohconfucius (talk) 01:39, 6 April 2009 (UTC)

"Option 0"

I'm moving this from the main page to here, as no discussion took place before this was added to the RfC. Should this be added or not? Karanacs (talk) 19:23, 1 April 2009 (UTC)

I support Option #0 (don't link years)

1. Support - I prefer not linking year numbers at all. If you want to link the year, then do a proper link that more clearly says what it is linking. --David Göthberg (talk) 18:55, 1 April 2009 (UTC)
2. Support - I was disappointed that the date linking poll didn't have this obvious option as an option. Tempshill (talk) 19:00, 2 April 2009 (UTC)

Discussion

Although I mostly agree with the sentiment, I feel it is too late to add any more options at this point (after so many editors have already registered their opinions). I encourage people who feel this way to use comments. Karanacs (talk) 19:23, 1 April 2009 (UTC)

You first changed my vote to another option. Then when I reverted that you deleted my vote. That is you doing vote fraud and vote censoring.

Had you just moved my vote down to the comment section, then at least it wouldn't have been a clear case of vote fraud.

--David Göthberg (talk) 19:57, 1 April 2009 (UTC)

Please calm down, and note that I left a message on your talk page explaining the situation and pointing you towards this section, where I had copied and pasted your new section in full. Karanacs (talk) 20:00, 1 April 2009 (UTC)

This should be removed asap - it can go in the comments section if needs be. I'm on my iPod now so I can't deal with it.--Ryan Postlethwaite 20:05, 1 April 2009 (UTC)

• I contacted David on his talk page and suggested how he can get his point across more effectively by working within the structure of the RfC. Greg L (talk) 20:25, 1 April 2009 (UTC)

• It is customary for approval polls to add new options; when evaluating the new option, its late appearance can be taken into consideration. Since a belligerent minority supports Option 1 as a form of Option 0, and another section of opinion supports Option 1 because it is not option 0 adding it should help to clarify the real situation. Septentrionalis PMAnderson 04:05, 2 April 2009 (UTC)

• Yeah, I understand the point, PMAnderson. But can you imagine the chaos when new options that weren’t available at the start of an RfC are introduced piecemeal midway through? Throwing out new options in RfCs is better suited for the first RfC to address a new issue; it’s not a good fit at this late stage, where we are on our fourth RfC and have a well discussed understanding of the implications of all the nuances.

  If a user wants to add a comment into the comments section saying they think there should be an “Option 0” for no links at all, or an Option ΘβΔ” for some other whiz bang idea, that is still input that can be considered when trying to determine the nature of the community consensus on the matter. But the numbers of votes means a lot too in RfCs and it is probably wiser—if an editor wants to have the maximum voice in the outcome—to vote for the option that best represents their views and explain precisely what they really desire in their vote comment.

  It’s also a bit more, uhm… *humble* of an approach, since it doesn’t assume that the editor is throwing out something new that hadn’t been considered and discarded for a good reason. We had discussed this option but previous RfCs made it clear that the overwhelming number of Wikipedians felt that linking dates was appropriate in some circumstances. So there was no point in throwing out a space-filling option that we knew didn’t have a snowball’s chance in hell of winning. Greg L (talk) 04:45, 2 April 2009 (UTC)

  P.S. We could also have an option for “Negative 1: Don’t link any dates; and those editors who agitated to keep on linking the crap out of them should be given an eye‑bulging Misplaced Pages wedgie.” I bet that would have received 20+ votes. But it wouldn’t have won so there is no point siphoning votes off options that have a prayer of becoming the community consensus. Greg L (talk) 05:11, 2 April 2009 (UTC)

• Minus 1 is redundant, unless someone manages to support 0:Never link month-day, and 0:Never link years, without supporting Don't link any dates. Possible, I suppose, but not likely.
• An overwhelming number of Wikipedians felt that linking dates was appropriate in some circumstances. Thank you; I shall quote you on that.
• That the present !votes for Option 1 contain 20 votes for Option 0, (and a comparable number that would personally prefer #4 but think it bad strategy) would be very interesting. Septentrionalis PMAnderson 05:29, 2 April 2009 (UTC)

• You can also quote me on this one: “…previous RfCs made it clear that the overwhelming number of Wikipedians felt that linking dates was appropriate in some circumstances.” This RfC has added greater specificity as to what “some” means. Greg L (talk) 15:42, 2 April 2009 (UTC)

• • Not much specificity, even on the question of birthdates. As llwyrch comments, much of the support for #1 is likely to assume that birth and death dates are relevant; for such people, #1 is equivalent to #2 but with less verbiage. Septentrionalis PMAnderson 19:29, 2 April 2009 (UTC)

• Wikilawyering. There will always be editors who stand with pouted lower lip and their arms folded across their chest and say “I think the guideline leaves wiggle room to link day-month in the birthdate of the asshole who invented trivia” or some such nonsense. I’m not even thinking of trying to argue with them. Never try to teach a pig to sing; it only wastes your time and annoys the pig.

  The wording for Month-Day Option #1 says this: Month-day articles (February 24 and 10 July) should not be linked unless their content is germane and topical to the subject. Such links should share an important connection with that subject other than that the events occurred on the same date. Moreover, year linking Option #1 has similar verbiage: Year articles (1795, 1955, 2007) should not be linked unless they contain information that is germane and topical to the subject matter.

  Together, they are infinitely clear for Wikipedians in the middle of the bell curve (and your ordinary, 50th-percentile 6th-grader). And notwithstanding your protestations that you are utterly confused about what both Option #1s portend for linking birth dates, I know you really aren’t so confused. So stop with your pre-verdict posturing (accomplished via proxy by citing other editors) about how the crushing support for both Option #1s is *actually* a validation by the community allowing you to link birth dates and all the other horsecrap you’ve wanted to link all along. No, it doesn’t. And to suggest as much sounds like just the sort of argument from another editor I’m familiar with; it doesn’t impress. Greg L (talk) 21:19, 2 April 2009 (UTC)

• Remember that this is only Phase 1 of the poll, we can always address "relevance" in the next phase. Dabomb87 (talk) 12:40, 3 April 2009 (UTC)

• The difficulty is this poll doesn't really address any of the questions of importance (except perhaps which ambiguous language is to be inserted in the worthless MOSNUM; I don't care what it says, only whether it is used to harass other editors). It doesn't decide whether date links are to be treated like other links, which is the question really being disputed; it doesn't decide whether there is consent for bottery, since bots cannot decide whether a sentence is germane, topical, or relevant.Septentrionalis PMAnderson 00:53, 3 April 2009 (UTC)

• Quoting you: The difficulty is this poll doesn't really address any of the questions of importance: Well, silly us. Greg L (talk) 01:16, 3 April 2009 (UTC)

• I don't fully agree with PMAnderson's statement, although any of the options (including option #2, for which I can take primary credit/blame; remember I started drafting it less than 2 weeks before the vote), without further guidance, could be used to harass editors who do not agree with your interpretation. — Arthur Rubin (talk) 02:56, 3 April 2009 (UTC)
• However, a large number of votes for option #1 state "link only to relevant dates", or some similar verbiage, which may suggest that they didn't actually read the option. — Arthur Rubin (talk) 02:59, 3 April 2009 (UTC)

• Americans voted for Bush… twice. That would suggest they weren’t listening to the debates. But we respect the vote—even if their reasoning is “I voted for the dude who wouldn’t look funny with a six-shooter on his hip.” Greg L (talk) 05:50, 3 April 2009 (UTC)
• Arthur, I already asked PMAnderson the same question and so far he hasn't replied: Can you give an example of a relevant link that would be forbidden by option 1? People like me have voted for option 1 exactly because it is option 4 plus a clarification that linking the year of death of an obscure Albanian writer was previously considered right not because it's a relevant link (it's obviously not) but because there was an exception for such links which is no longer in effect. --Hans Adler (talk) 11:19, 3 April 2009 (UTC)

It's not a vote. The categories are to help with interpreting the results of the opinion poll, nothing more. If people place themselves in one category, but give reasons more relevant to a different category, it's perfectly valid to question what they really meant. It's more like the invalid ballots in Florida when people voted for more than one candidate — and those ended up being thrown out entirely. Since the total count isn't really the point, just determining if there's an obvious consensus view, then there's no reason to "throw out" anyone's comments, but they do need to be considered more carefully. I honestly haven't paid much attention to the linking part of the dates debate, but I'd think we'd need some kind of more flexible policy to let regular editors decide on (somewhat of) an article-by-article basis which dates were "relevant" enough to warrant links. Too strict and black-and-white a policy is just going to invite arguments. I guess I'll actually go and look over the options, so I have a more informed opinion here on some of the specifics, then maybe I'll comment further. --Sapphic (talk) 06:05, 3 April 2009 (UTC)

i'm not sure "it's perfectly valid to question what they really meant". some of the comments sound strange to me too, but being gifted at summing up all one's reasoning in succinct unambiguous statements is not a prerequisite for !voting in the poll. !voters are entitled to assume that when they've chosen "i support Option X" it's clear enough that they support Option X. Sssoul (talk) 06:56, 3 April 2009 (UTC)

That is not reasonable, if they give a justification which supports option W or Z, but not X. If they give no justification, and the subtitles are clearly misleading (as they are in this case), I'm just not sure. — Arthur Rubin (talk) 00:22, 4 April 2009 (UTC)

Bear in mind also, that people may express their opinion that seems contradictory, if that's the only way to express their opinon within the (bogus) constraints of the poll structure itself. That's what I did - and my opinion is intended to convey support for two options even in the face of any rule forbidding support for two options. Reading it otherwise, as support for only one of the two, is a misinterpretation. Likewise, ignoring part of some other voter's opinion in order to better pigeonhole it is probably also a misinterpretation. — Gavia immer (talk) 04:05, 4 April 2009 (UTC)

as noted elsewhere: if people seriously feel some of the !voters need to confirm or clarify their votes, they should get the clerk to oversee the formulation and posting of a neutral statement to be left on those !voters' talk pages. Sapphic's independent effort to get clarification from some !voters showed that most of them did indeed mean X when they said X, even if their rationale is not worded the way you or i would word it ... but asking for clarification is better than second-guessing them or "rejecting" their !votes for being worded weirdly. Sssoul (talk) 06:38, 4 April 2009 (UTC)

I just gave my opinion, and I agree that the choices are a bit confusing. I don't think it matters much though, because I think the clear (and in the first case overwhelming) majority is against almost any date linking. I like date links, but I don't think of them like regular links, and don't think they are usually "relevant" to the topic in the same way that normal topic links are. I think they're a different kind of tool, equally useful, but different nonetheless. I don't think people should be forced to see those kinds of date links, if they don't find them useful. I'd rather see us distinguish between dates that are linked by default and those that aren't, with users being able to override those defaults via preferences — but I still think the default should be pretty conservative. Then we'd have even better metadata, with the ability to distinguish between dates that are just dates, and those that are also more specifically relevant in their capacity as a date to the topic of the article (like with Christmas and 25 December, for example — or a person's birthday, in my opinion.) People that wanted more date links could have more date links, etc. You know the routine. I like the software solution. But I think on the linking issue, the outcome is pretty clearly on the side of fewer date links. So I say delink most of them using bots and/or scripts, then let people add back ones they think are important — and don't be too concerned about challenging people on that, at least right away. --Sapphic (talk) 06:29, 3 April 2009 (UTC)

Back to discussing "Option 0". Comment: why would we want to forbid a link from the article 1340s to the article 1346? The latter article clearly helps and expands an understanding of the 1340s. What would David Göthberg think was a "proper link that more clearly says what it is linking" in this case? --RexxS (talk) 22:27, 3 April 2009 (UTC)