Wikipedia talk:Date linking request for comment

Arthur's view

User:Arthur Rubin said [21:00, 19 February 2009 (UTC)] "and if not, should an error-prone tool (such as User:Tony1 for delinking or User:Tennis Expert for linking) be allowed." I feel that this comment is far too targeted and personal, and should be refactored. Ohconfucius (talk) 04:14, 25 February 2009 (UTC)[reply]

Have you suggestions? I want to imply, although stating it outright might be better, that certain editors have been acting as if they were unintelligent bots. The names are not that important. — Arthur Rubin (talk) 15:27, 25 February 2009 (UTC)[reply]

OK, fixed. Names removed, but specifically noting that I include as a an "error-prone tool" editors who link or delink without thought. — Arthur Rubin (talk) 15:32, 25 February 2009 (UTC)[reply]

Response to Deb

What I called 'relevancy' is based on the statement: "Internal links add to the cohesion and utility of Wikipedia by allowing readers to deepen their understanding of a topic by conveniently accessing other articles" as found in WP:MOSLINK. I was hoping that the general principle of only making relevant links would be accepted, but I read your view as indicating that you prefer the position where each individual editor decides unilaterally what links (of any sort) they may make. I would reject that as it runs counter to the principles of collaborative editing, consensus and consistency between articles - the three 'C's, if you wish. You could always make a proposal to test that stance in an RfC. If I'm mistaken and you are suggesting only that date-links should be treated differently from any other links, then please make a suggestion for testing that in an RfC. In either case, it would be instructive to see the balance of opinion on such questions. --RexxS (talk) 23:38, 25 February 2009 (UTC)[reply]

The problem with this "new understanding" -- which is, after all, only a few months old -- is that it would preclude most of our traditional links. For example, in the article on Ben Franklin, we would no longer link to Philadelphia, or electricity, or kite, or United States one hundred-dollar bill because none of those links are likely to "deepen" a reader's understanding of the topic of Ben Franklin. -- Kendrick7^talk 01:19, 26 February 2009 (UTC)[reply]

See my reply here. Dabomb87 (talk) 01:35, 26 February 2009 (UTC)[reply]

I certainly agree that links to 'electricity', or 'kite' do nothing to further one's understanding of Ben Franklin -everyone knows what these are. However, the banknote is germane (and therefore relevant) because this former President's likeness features on one of them; likewise, Philadelphia is where much of the revolution stuff happened. Ohconfucius (talk) 01:45, 26 February 2009 (UTC)[reply]

I suppose I do question the 'traditional' links, but I'm willing to be open-minded about their value. I really think that the link in Ben Franklin to Philadelphia is a wasted opportunity. The Philadelphia article is very large and much of its modern content is of as much value to the Ben Franklin article as the link to Paris is (i.e. not much). There is an excellent article History of Philadelphia that is begging to get a (piped?) link from Ben Franklin. Perhaps by questioning tradition, we will find better links that do much more to enhance our encyclopedia? --RexxS (talk) 02:53, 26 February 2009 (UTC)[reply]

That is certainly a lot smarter way of linking. Ohconfucius (talk) 03:38, 26 February 2009 (UTC)[reply]

Sorry, how does History of Philadelphia enhance a reader's understanding of Benjamin Franklin? That link would be forbidden, and presumably enforced by bots, under the version of Wikipedia you all seem to advocate. -- Kendrick7^talk 09:18, 26 February 2009 (UTC)[reply]

I would have thought that the sections History of Philadelphia#Early growth and History of Philadelphia#Revolution provides high quality and very relevant background to the article Benjamin Franklin, deepening the understanding of the topic, especially expanding the line "Franklin became a newspaper editor, printer, and merchant in Philadelphia ..." which is where the link would be. Surely you can see that sort of link is exactly what WP:MOSLINK asks us to make? Such a link is clearly not 'forbidden ... under the version of Wikipedia [ I ] seek to advocate' and I can't see how you could possibly draw that conclusion if you read the two articles. Of course, that's just my own personal opinion, there is no "you all" as far as I'm concerned. Should the consensus change to move away from having relevant links, I will (sadly) accept that. Can you make the same assurance? --RexxS (talk) 14:43, 26 February 2009 (UTC)[reply]

No, I'm pretty much going to link whatever I want whenever I want. MOS is just a guideline, remember? -- Kendrick7^talk 20:22, 26 February 2009 (UTC)[reply]

Conversely, when editors see links that they don't want, they are free to remove them. Dabomb87 (talk) 22:56, 26 February 2009 (UTC)[reply]

As you can see, we have guidelines, which are defined as collections of consensuses, to avoid such pointless edit-warring. Of course, guidelines are rules which are expected to have exceptions - when those exceptions can be justified. So, go ahead and link whatever and whenever you want. After all WP:POINT is 'just a guideline', isn't it? I'll settle for going along with consensus and collaborating to build an encyclopedia. --RexxS (talk) 01:43, 27 February 2009 (UTC)[reply]

The main thing to take away from the ArbCom should be that running around the project at superhuman speeds creating WP:LAME edit wars isn't constructive. Guidelines should reflect the good faith practices of our editors as carried out on a human scale and not try to unduly change them. Because the reality is that's not possible. At the end of the day, I simply don't see overlinking as a huge threat and I've yet to see anyone actually advocate for a pro-overlinking position. -- Kendrick7^talk 05:10, 27 February 2009 (UTC)[reply]

I certainly would agree that running around the project at superhuman speeds creating WP:LAME edit wars isn't constructive. But that's the symptom, not the cause. The dispute arises for several reasons and you've expressed one of them above. There is a disagreement about the nature, purpose and status of guidelines. You honestly believe that guidelines should summarise all the good faith editing practices of editors. But that turns out to be untenable. Editors' good faith practices may conflict with each other. Guidelines therefore document the consensuses that have emerged when those conflicts have been resolved. They carry the weight of that consensus, but since consensus may change, they are descriptive, not prescriptive. However that is not the same as saying editors are free to ignore them at will. This much common ground, at least, will need to be established before any future RfC stands a chance of clarifying the consensus on particular issues such as date-linking. --RexxS (talk) 13:21, 27 February 2009 (UTC)[reply]

←(Although possibly thread drift.) That's (part of) the problem here. This guideline clearly does not reflect consensus. (Ummm, well, it does not reflect a clear consensus, anyway.) I agree that, once consensus is established, that edits contrary to that consensus may be reverted unless a good reason is given indicating a local consensus. The fact that people are "running around the project at superhuman speeds creating WP:LAME edit wars," some enforcing this "guideline", is just a symptom. Regardless of the overall purpose of the RfAr, the purpose of this RfC is (or should be, IMHO), to establish consensus. — Arthur Rubin (talk) 15:03, 27 February 2009 (UTC)[reply]

Response to User JonH

If we're going to keep some sort of markup around dates, we should just fix the built-in date autoformatting system and get the full benefit of a preference-based system, rather than just using templates. Unless I'm mistaken, templates can't format dates (or optionally create links) according to a user's preferences, and so they have fewer features than the (flawed) date autoformatting system we have currently (and also fewer features than the proposed replacement.)

It all really boils down to the markup around dates. If we have it, it's only a minor step (the changes to the software really aren't that complex, just a few extra regular expressions and case statements) to get a fully functional preference-based autoformatting system that replaces the one that's already built in to the mediawiki software. If we don't have the markup, we're pretty much stuck with manual editing and manual enforcement of style guidelines. --UC_Bill (talk) 18:49, 8 March 2009 (UTC)[reply]

The proposed new date formatting system knows nothing about the preferences of readers who are not logged in. Therefore the preceeding passage is more than 90% false. --Gerry Ashton (talk) 19:04, 8 March 2009 (UTC)[reply]

Your point only applies to a part of the last sentence, so that's a strange 90% you're seeing. If we have dates in some format like {{dt|xxx}}} then we might as well just process them using the built-in date autoformatting system, appropriately modified. It can already do everything templates do, and it can also do things that templates can't. HOWEVER you're correct that my last sentence was a bit misleading. Templates can automate much of the enforcement of style guidelines, as anonymous users would see things. That's a huge improvement over "plain text" dates, and I by no means want to diminish the value of those date templates. My point is that we already have code in the existing Wikipedia installation that does date autoformatting, and that we should just fix it, rather than replace it with something that has fewer features. --UC_Bill (talk) 19:18, 8 March 2009 (UTC)[reply]

The reason I wrote "View by User JonH" is that the draft RfC seems to confuse the concepts of linking and markup, so people who just do not like links to May 6, etc, will say that they oppose any markup. I am more interested in the overall result than the details, and in particular:

I have no strong views about whether the dates should be processed by the MediaWiki platform or a template, but I was under the impression the MediaWiki developers are not keen on changing their software.
I think it must be possible for users to customise the output of a template by setting up CSS and JavaScript files. However, I do not know how to do it, and I guess most users do not know, so this would not be much help. Also it would be good to follow the date format preferences of existing users, and that might be impossible or require some programming trick.
While I am happy for users to be able to set preferences, I am more concerned about the default behaviour. I think eventually dates should be linked to a GeoHack-style system which would provide links such as "May 6 in History" and "Events of 1972". Thus everyone could click through to these pages, and no-one would need a preference for direct links.
A system that allows editors to write the default output, like the present system, is more likely to be accepted than one which tries to force a particular default format, like the "proposed default date formatting system".

JonH (talk) 22:39, 8 March 2009 (UTC)[reply]

I strongly disagree with that last point you made there. The fact that the current system displays an inconsistent mix of formats is probably the biggest motivation (on both sides) for this entire debate. If editors were okay with "fixing" date inconsistencies by simply changing the format inside the markup, then there wouldn't be a crusade to eliminate the current autoformatting system in the first place. "Plain text" proponents want to fix (e.g.) "[[11 September]] [[2001]]" by making it "September 11, 2001" while "Software fix" proponents want to leave the wikitext unchanged but change the system so it applies a site-wide or per-page default for anons. To my knowledge, nobody is saying to fix it by making it "[[September 11]], [[2001]]" — except maybe you? I'm of the opinion that the current system needs replacing anyway (mostly because it intermingles autoformatting and autolinking in ways that are unnecessary and problematic) but I'm not totally opposed to keeping the old system, or something like it. --UC_Bill (talk) 22:54, 8 March 2009 (UTC)[reply]

I might well be wrong with my last point, but I thought the debate started with people who did not like the linking of dates. Editors who are happy that the default format follows the source are probably quiet at the moment; it is not unlike the WP:ENGVAR compromise. However, they might start complaining if a site-wide default was introduced. Still this is a distraction from my main point: markup, linking, and formatting are distinct matters which should considered separately. JonH (talk) 23:32, 8 March 2009 (UTC)[reply]

You raise two interesting points: One, that there might be a "silent majority" (or at least "silent constituency") of people who are more concerned with the date linking aspect of the current automated system than with the mess of formats anons see as a result of the formatting features. Two, that the issues surrounding dates should be broken into the three categories you mention: markup, linking, and formatting. I definitely agree with the second point, and you've piqued my curiosity with the first. I'll attempt yet another rewrite of some of the proposal text, based on your suggestion(s), when I get some time (hopefully today or tomorrow.) --UC_Bill (talk) 16:21, 9 March 2009 (UTC)[reply]