Wikipedia talk:Manual of Style (dates and numbers)/Archive 43

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 40 Archive 41 Archive 42 Archive 43 Archive 44 Archive 45 Archive 50

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

previous page: /archive42 < (long discussion)
archive43
> next page: /archive44

The second-level header section "linking of dates" continues from archive42.


linking of dates (continued)

Other approach

I'm about to simplify Wikipedia:Manual of Style (dates and numbers)#Avoid overlinking dates, in order to make it less contentious. Also I'm proposing to move it under the section that explains in which cases the "date preferences" don't work (Wikipedia:Manual of Style (dates and numbers)#Usage of links for date preferences) while IMHO that's a more logical sequence: first make people acquainted with the issues, and let the recommendation based on these issues follow.

Here's the present text:

If the date does not contain a day and a month, date preferences will not work, and square brackets will not respond to your readers' auto-formatting preferences. So unless there is a special relevance of the date link, there is no need to link it. This is an important point: simple months, years, decades and centuries should only be linked if there is a strong reason for doing so. See Wikipedia:Make only links relevant to the context for the reasons that it's usually undesirable to insert low-value chronological links; see also Wikipedia:Manual of Style (links)#Internal links.

Here's the alternative I propose (reworked 10:22, 22 March 2006 (UTC)):

See also: Wikipedia:Only make links that are relevant to the context and WP:MOS-L#Internal links
If the date does not contain a day and a month, date preferences will not work, and square brackets will not respond to your readers' auto-formatting preferences. It is discouraged that separate months, years, decades and centuries are linked, unless when such link is relevant.

The related Edit Summary would read: "(1) not an "important point": rationale is based on Date Preferences which is only an optional accessory; (2) "strong reason" contentious: relevance is enough; (3) Wikipedia:Only make links that are relevant to the context does not give any more specific reasons re. low-value chronological links" --Francis Schonken 00:23, 22 March 2006 (UTC)


  • IMV, it's a good move. Here's a suggested modification, which preserves the substantive meaning.
If the date does not contain both a day and a month, date preferences will not work, and square brackets will not respond to your readers' auto-formatting preferences. Therefore, unless the date link has special relevance to the topic, there is no need to link it. Editors are encouraged to link separate chronological items, such as months, years, decades and centuries, only if the information they would lead to is relevant to the topic.
See also Wikipedia:Make only links relevant to the context and Wikipedia:Manual of Style (links)#Internal links.

The proposed edit summary is too long to fit. Tony 00:56, 22 March 2006 (UTC)

summary too long – yeah, I know, but it's here now, so I can make "see Wikipedia talk:Manual of Style (dates and numbers)#Other approach"
Dunno about the "special relevance": why "special" relevance (as if there were something like "ordinary" relevance, which would not suffise)? Something is relevant or it's not, and if something is "relevant" (which btw always relates to some "context"), then it can be included in wikipedia. That always has been thus, regardless of whether it fires an optional user preference feature. Changing this would be changing "official policy" (and we don't want to change "official policy" by a guideline paragraph do we?). And changing this order of importance would be too inviting for users (and their semi-bots) going bazurk over something that really isn't of first plane importance. --Francis Schonken 08:55, 22 March 2006 (UTC)
An amplifier (perhaps 'particular' is better than 'special') emphasises that to be useful as a link, a chronological term should be focused on the topic, and not be just a year or decade in which the topic is based. But I don't feel strongly about it. Tony 11:14, 22 March 2006 (UTC)

Example

Edit of "Fibonacci" article (edit summary: "formatting, fixed deadlinks, unlinked lone years per Wikipedia: Manual of Style")

The only useful part of that edit was IMHO the checking of external links, and adding some new ones instead of one old one that had gone dead.

Update: after a closer look, I removed both new links:
  • "The Fibonacci Association" contains no biographical information, is about Fibonacci numbers, and already listed in the external links section of that page.
  • "The Fibonacci Numbers" also exclusively about the *Fibonacci numbers*. Despite this being a rather old webpage (1996), among 15-odd more recent websites linked at the Fibonacci numbers page, I added it to the list there, for completeness, but afaik it may be removed there per WP:NOT web directory.

Not useful formatting (really, I object to Wikipedia's resources being used for such edits):

  • Leonardo Pisano (Pisa, c. 1170 - Pisa, 1250), also known as Fibonacci,... → Leonardo Pisano (1170 - 1250), also known as Fibonacci,...
  • In 1202, at age 32, he published what he had learned in Liber Abaci,... → In 1202, at age 32, he published what he had learned in Liber Abaci,...
  • In 1240 the Republic of Pisa honoured Leonardo,... → In 1240 the Republic of Pisa honoured Leonardo,...

Especially the first two of these are deteriorating the article IMHO. If the Liber Abaci is not mentioned in the 1202 article, then that's no fault of the Fibonacci article, but a failure of the 1202 article. One doesn't solve faults by denying their existence.

The first alteration is also obnoxious while removing place of birth and death contrary to the MoS (biographies). It *might* seem obvious that he was born and died in Pisa, while named "of Pisa"/"Pisano"/"da Pisa". It becomes less so when acquainted with the fact (as mentioned in the article) that in Pisa he was rather known under the nickname "bigollo" ("traveller", actually a better translation would be: "vagrant" or "tramp").

Further pedantic and useless changes from the same edit:

  • changing subtitles from == References == to ==References== (is there a rule about leaving out the spaces? The section title button doesn't: here's what you get when you click it:
== Headline text ==
  • Adding a redundant category sort key to all categories: (e.g. [[Category:1170 births]] → [[Category:1170 births|Fibonacci]])

I don't know whether this was "bot", "semi-bot", or simply a wikipedian thinking (s)he should outwit all others, but I reverted some of it (not the pedantry) --Francis Schonken 09:19, 23 March 2006 (UTC)

See updates above: in sum the only benificial effect of an edit that changed the Fibonacci article in 20-odd places, was the removal of a deadlink... We should definitely make the MoS pages thus that they discourage changes that appear to serve no other goal than their own. --Francis Schonken 14:01, 23 March 2006 (UTC)

Specific subissues

Dates in tables

While repetitive linking is normally a bad thing in a table the only way people can view the whole table by their date preference is if all the dates are wiki-linked. Maybe this should be mentioned here. Discordance 21:12, 2 March 2006 (UTC)

AFAIK, nobody is proposing to delink full dates, as that falls within the guidance of the existing MoS already. So, this feature shouldn't be affected, right? Neier 09:20, 14 March 2006 (UTC)
Aboslutely. Rich Farmbrough 19:05 14 March 2006 (UTC).
Correct. In any case, this is not a proposal to change the guidance. bobblewik 20:04, 14 March 2006 (UTC)
I would want May linked from Maypole, but not from 1992 United Kingdom General Election. This seems eminently reasonable to me. The grey area is really over reasonably ancient years, and those who think they should be unlinked would by and large be happy to leave them if the tens of thousands of bare-year links to twentieth century years, decades, to months and days could be addressed. Rich Farmbrough 11:32 16 March 2006 (UTC).

For full dates in tables (and in lists starting with the date), it often makes sense to either not link dates, or to disable preferences by using piping (e.g. [[14 November|14 Nov]] [[1968]]), to make the length more consistent for easier comparison and to keep the column from becoming unnecessarily wide by using consistent three letter abbreviations of the month, no matter what order is used (YYYY mon DD, DD mon YYYY, mon DD, YYY). (Personally, I also like fixed width fonts and padding of DD less than 10 with zero or nonbreaking spaces.) Gene Nygaard 15:00, 24 April 2006 (UTC)

As of 2006

Hey - I just wanted to make a note of something in case you didn't use it already. I often use the as of 2006 syntax in articles where it's useful. These links redirect to 2006, but the use is that if you backtrack to the "as of 2006" redirect page and click "What links here", then there's a big list of articles containing statistics that were valid in 2006 but will need to be updated later. In 2016 some editor team may choose to rip through and update all those statistics as of 2016. And then in 2026, and 2036, etc.

You probably already knew about this, but I wanted to make sure your laudable effort to get rid of the "useless sea of blue" didn't extend to the "as of" date links. Thanks - Tempshill 02:39, 17 March 2006 (UTC)

I don't see any difference here: readers are quite able to type "as of 2006" into the search box on the left on the rare occasions when that article is required; takes five seconds. Same for all other trivial links. Tony 00:43, 22 March 2006 (UTC)
The point is that the 'as of 2006' link implies there is information that will need to be updated. With the link in place it is easy to find the data that needs to be updated with the 'what links here' option. Givben the huge task of updating this type of informtion I assume this type of data is used sparingly. 19:50, 22 March 2006 (UTC)

Date delinking: a proposed compromise

Another editor was blocked today for delinking dates. There seem to be very strong feelings about this issue, and it's a great pity to see editors and admins being blocked for adhering to the MoS, so I'd like to propose that this compromise text be inserted:

While editing articles, editors are encouraged to remove links that do not conform with the guidance on this page. However, as with all style issues, going through the encyclopedia with the sole purpose of changing from one style to another is discouraged (whether it involves date delinking, changing British to American English, or BCE to BC), because those who oppose the style change will see your actions as provocative. Remember that the MoS is a guideline, not policy, although editors are strongly encouraged to adhere to the advice in the MoS in the interests of harmonious editing and consistency across pages.

Any thoughts? SlimVirgin (talk) 20:18, 24 March 2006 (UTC)

Sounds like a very reasonable compromise, and it is line with previous Arbitration Committee rulings. Jayjg (talk) 20:19, 24 March 2006 (UTC)
I don't like it. It puts the encyclopedia second and "peace, love, and harmony" first. What sense is it to say, "We think X is good, but you can only do X in a roundabout way, and if you do X with the sole intention of doing X you will get in trouble for it." --Cyde Weys 20:25, 24 March 2006 (UTC)
First of all thanks, Sarah, for suggesting this. Which part of the MoS page would this replace, or if it's not replacing anything, where abouts would it be placed? I'll think more about this in a bit. Talrias (t | e | c) 20:25, 24 March 2006 (UTC)
I'm thinking it could be the second paragraph in the section called "Avoid overlinking dates" here. SlimVirgin (talk) 20:29, 24 March 2006 (UTC)


There is an important difference here. British vs. American English are both styles that are accepted by the Wikipedia community. BC and BCE are both considered valid, according to the MoS. But overlinking years is not a valid "style", according to the MoS. Neither is linking every word. Both of those are discouraged, whereas using BC or BCE is not discouraged.

Consider the standard AutoWikiBrowser changes. In those cases, AWB-users going through the encyclopedia with the sole purpose of enforcing MoS guidelines (e.g. not capitalizing all words in headers). This is encouraged. How is what I'm doing any different? – Quadell (talk) (bounties) 20:31, 24 March 2006 (UTC)

You're right, Quadell, and perhaps the paragraph needs to be tweaked a little to make clear that it's not entirely comparable to BCE/BC etc. You're also right about the headers, and in principle date delinking is no different. The only reason I'm proposing a compromise is that it seems to have triggered strong feelings, and so it would be good if both sides could be accommodated, as the dispute has been going on for weeks, and two people have been blocked because of it. SlimVirgin (talk) 20:35, 24 March 2006 (UTC)
It takes two people to have a block; one to be blocked and one to do the blocking. You're laying all of the blame on the former. --Cyde Weys 20:39, 24 March 2006 (UTC)
No, not at all. Sorry if I gave that impression, CydeWeys. SlimVirgin (talk) 20:54, 24 March 2006 (UTC)

Then it seems to me that this compromise would have the following result:

  1. There are some style choices which are up to the individual user. These should be done in articles you're working on, but should not be done wholesale. The choice between BC and BCE is an example.
  2. There are other style decisions which have been made by the community and are our official guidelines. These should be enforced in articles, and making these changes should be encouraged. An example would be changing titles to not capitalize every word.
  3. But then there are style choices which are actually in category 2, since they are official guidelines, but should be treated as if they were in category 1, solely because a few people will get upset. And they block people for it. Therefore we make an exception.

This doesn't seem like a good idea to me. If overlinking years is acceptable, then the MoS should say so. If it's not acceptable, then I should be able to make changes like this without fear of retribution. (That edit is in accordance with the MoS, and is, in my opinion, badly needed.) – Quadell (talk) (bounties) 20:44, 24 March 2006 (UTC)

I completely agree with Quadell's analysis. It needs stronger wording, a more formal decision. Good example too, we could use more examples to prove/support the issue maybe? --Quiddity 21:40, 24 March 2006 (UTC)
Examples are a good idea. Gflores Talk 23:09, 24 March 2006 (UTC)

Well, I'm glad that everyone who has previously made their support for the manual of style section in question has come along and said "no" to this compromise, now all we need is for everyone who disagrees with the manual of style suggestion to come along and also say "no" to the compromise. Then we're all back at square one! Compromise and consensus requires some give and take people. Talrias (t | e | c) 20:59, 24 March 2006 (UTC)

Ok, the MoS currently says this about linking years:

"Unless there is a special relevance of the date link, there is no need to link it. This is an important point: simple months, years, decades and centuries should only be linked if there is a strong reason for doing so."
"Generally, do not link unless they will clearly help the reader to understand the topic."

From this guideline, it clearly advises users not to link dates and thus, I think it is perfectly acceptable to remove them manually, as Quadell has done. As SlimVirgin has said, there should be something said about bots not being allowed to traverse the encyclopedia unlinking years. I don't quite understand the opposition of unlinking dates manually... Gflores Talk 21:34, 24 March 2006 (UTC)

I think Gflores' formulation very much OK. Even better than what I proposed above in #Other approach.
The part of the guideline that now says "This is an important point:" should go. While, as long as that overestimation of the importance of the issue remains in the guideline the problem as scetched by Quiddity remains. In fact linking or not linking dates should be option "1" in Quiddity's analysis. An overestimation of the importance of the issue leads to people being allowed to automatically work away this type of overlinking. Then other wikipedians don't accept that, and people who do it get blocked.
The rationale of the de-linking of dates being a more important issue than general "overlinking" issues is in the guideline *exclusively* argued on the basis of a minor issue regarding an optional user preferences setting, that is unrelated to the use of separate years (months, days) in article text. Redundant to say I think that can only be qualified as a bogus rationale: this should be not different from other "overlinking" issues.
Quiddity is right that if it is said that this is "important", then afterwards you can't make other rules to say when to follow the "important" rule (and when not), and that's indeed something not yet fully solved in SlimVirgin's alternative proposal (using words as "strongly encouraged"). So I like Gflores' proposal better, while it avoids that trap. --Francis Schonken 22:58, 24 March 2006 (UTC)
I did some digging around to find the basis for that section of the Manual of Style, to try and find some basis for the differing viewpoints we find ourselves presented with today. I found that it was originally added by, of all people, Bobblewik, back in mid-April, 2005, where he made an edit saying he was eliminating an inconsistency. He also added the text "Other date forms such as year only (e.g. 1981) should be treated like any other words and linked only if there is some particular relevance.", without mentioning this in the edit summary, nor discussing it on the talk page. A related issue was raised on the talk page, which can be seen in archive 17, however this mentions only the duplicate linking of years (i.e., where a year is linked more than once in an article). I believe this change should have been reverted and discussed on the talk page, especially since the edit summary does not mention this change.
Over time this gradually was changed by various people into the text we have today, as far as I can tell without much discussion, into requiring "special relevance" and a "strong reason" before linking a year. This is a good example of "creeping guidelines", where an important page has been modified slowly but surely over time into what its advocates can now declare are "uncontroversial" guidelines. Now, when people (such as myself) are questioning the guideline, I am being told "this is a part of the guideline, shut up and do something else". Well, I think that is appalling. There are only a few policies on Wikipedia which are not subject to change. I can count them on one hand, and this is definitely not one of them. I've tried to have a debate on this issue a number of times before, but it has always been the case that "the mos says this, end of discussion". Raul's 7th law, "[a]s time goes on, the rules and informal policies on Wikipedia tend to become less and less plastic and harder and harder to change", applies well here. This is the reason I suggested task forces - to make sure policies and guidelines are discussed openly, are always available for debate and are not subject to this gradual and incremental modification we've seen on this dates linking fiasco. Talrias (t | e | c) 23:23, 24 March 2006 (UTC)
On the other hand, there's strong support for delinking. The poll above (on whether Bobblewik's suggestion that "Editors are encouraged to remove links that do not conform with the guidance on this page" should be added, I count 26 support, 4 support with the rider than editors shouldn't go through Wikipedia with the sole purpose of delinking, and 9 oppose. SlimVirgin (talk) 23:33, 24 March 2006 (UTC)
(oops, edit-clashing too...) Hi Slimvirgin, did it maybe occur to you (as it did to me), when reading the vote justifications of the 26 pro-voters, that o so many of them are circular to the point it was already in the MoS? So I second Talrias: I think I resume his research correctly in saying that he pointed out that thus far this only has been a self-fulfilling prophecy, initiated about a year ago by Bobblewik. So, I think Talrias and Gflores are correct in pointing out that it's about time to break out of this spiral (at least, that's what I understand). --Francis Schonken 00:14, 25 March 2006 (UTC)
PS, in that context, I don't think I'm revealing any particular secret that someone put on my user talk page:

Thank you for taking this matter forward. I am very pleased that you have taken the time to do this (even though you voted "oppose" and I voted "support"!). Whilst not feeling very stongly on the date issue as such, I have been horrified by the retaliation being taken against Bobblewik. It seems to me these people have utterly failed to make any move at all to try and bring the guidelines towards what they feel ought to be said.

...Which also confirms not to take too many importance in the vote results thus far, but rather "try and bring the guidelines towards what is felt ought to be said.", avoiding a blocking behaviour that is not justified by what the actual text says. --Francis Schonken 00:42, 25 March 2006 (UTC)
A simple binary straw poll with 40 participants hardly shows strong support, especially on an issue which affects almost every user of Wikipedia, editors and readers alike. Doing a straw poll and suggesting the results should be adopted as policy is absolutely a terrible way to run things. Talrias (t | e | c) 00:09, 25 March 2006 (UTC)
Maybe we should have a two or three-week straw poll with a wider audience (posting this on Community bulletin board, etc) to settle this once and for all? Gflores Talk 23:51, 24 March 2006 (UTC)
No, settling something "once and for all" is very very bad (as I explained in the two paragraphs above). In any case, voting on something before we've decided what to vote on is premature. We need to discuss this, and come up with a collaborative and mutually acceptable phrasing. We should aim to meet in the middle, not to ostracise and annoy the people who feel the opposite way by voting for two polar alternatives. Talrias (t | e | c) 00:09, 25 March 2006 (UTC)
Err, that was bad wording. What I meant is that so we don't have this debate every month... I'm not sure how you think the present views are polar, though. Gflores Talk 00:42, 25 March 2006 (UTC)
(edit clash with SV) Did you ever consider that the MoS is written this way because most people agree with it? I also agree that overlinking dates is not an alternative option similar to BC and BCE and I like the the wording in the MoS that warns against linking dates for no reason other than one can. Every link should be meaningful, adding useful context. Linking every dates does not do this.
I see no reason why manual unlinking should be frowned upon. The only valid complaints from those who block users who systematically unlink dates is if they are running a bot to do it. I am sympathetic to their frustration of linked dates with context being removed, nevertheless, it appears that there is a consensus that overlinking dates is not good for wikipedia. In my opinion, those who choose to use their admin powers to block for manual delinking are abusing their administrator tools. Under the current guidelines if bobblewick or others wish to manually delink overlinked dates there should be no complaint. What is worse, instruction creep that represents a concensus or admins who enforce their own POV regardless of what others think? Is the latter any different to POV pushing? David D. (Talk) 23:39, 24 March 2006 (UTC)
Yes, I did of course consider this. However, looking at the results of the straw poll which Sarah highlighted above as strong support shows that a number of people simply said, "yes, the guidelines suggest this, so make it a bit clearer". That absolutely cannot be read as support for delinking (or for that matter, linking) dates. That should be read as a simple statement that the current text is unclear. Anything else is overinterpreting. Also, I don't think anyone is arguing that every date should be linked. I am arguing that some dates should be linked so people can click on them, to get historical relevance. As an example, I for one like reading history and politics articles, which obviously have a lot of dates included. Being able to click on those dates allows me to see easily other noteworthy events which happened during that year, which of course affect the political landscape. I don't think this issue should be presented as an "all dates linked" vs. "no dates linked" issue, which is what some are seeking to present it as. There is clearly a midpoint between the two we can reach. I'm against scripts or bots determining where to draw it (or even recommending where to draw it!) for us. Talrias (t | e | c) 00:09, 25 March 2006 (UTC)
Well from what you've written above I can see we agree more than we disagree. I have never been of the opinion that ALL dates should be unlinked, clearly links are more relevant in some articles than others i.e. you political example above. That is a good start. I agree with your bot opinion that it's better to do manual delinking. At present I'm not quite sure why you disapprove of bobblewick since he does not seem to be randomly delinking (or using a bot). Certainly there are still dates that are linked after he has passed through. I'm happy to see what the community consensus is on this issue although i think that most supports above are citing MoS since they are supportive of those quidlines nt just becuase they are there. A poll could be enlightening. David D. (Talk) 00:28, 25 March 2006 (UTC)

Imagine the following hypothetical vote: "Should the MoS recommend, as it currently does, that years be linked only when they are parts of full dates (or else when there is a particular reason for there to be an exception)? Or should the MoS be changed to allow years should be routinely linked?" If such a vote were held and were widely advertised, I think a consensus (80%+), or else a strong majority (70%+), would support the former approach, with Talrias and Ambi strongly objecting. I don't advocate such a vote at this time because it would take up time and harden feelings, and I don't think it would tell us anything we don't already know: that a consensus exists, but that there are strong feelings of opposition from a few well-respected admins. If this consensus exists, and I think it does, then I would like to be able to make these changes without fear of repercussions. We could hold a vote just to prove that this is the case, but it would be unfortunate; the very fact that such a guideline has existed for a year says a great deal in its favor. – Quadell (talk) (bounties) 00:04, 25 March 2006 (UTC)

I don't know what your intentions are, but as far as I can see that would be absolutely the wrong question. Nobody supported or even suggested that years should be routinely linked. So the question as you formulate it comes down to: either you choose something absolutely dumb, or you support the de-linking camp requiring "particular" reasons for exceptions — thus far the word "particular" in that context hasn't been cleared out: in fact the whole discussion is about the meaning we should give to "particular" in this context: is it the same "particular relevance" required for any other internal link, or does "particular" aim at something different than what is said in general against overlinking. So, no, such way of putting the question is no more than a booby trap to keep the discussion going on ad infinitum. --Francis Schonken 00:24, 25 March 2006 (UTC)
Quadell makes a good point. This has been in the MoS (a much-watched page) for a year, and on the few occasions someone has tried to remove it, the deletion has been quickly reverted, so delinking clearly does have consensus. Talrias, you say you're not arguing that every date should be linked, but only relevant ones. Can you try to come up with compromise wording that reflects your concerns, but that also bears in mind there is a consensus against overlinking? SlimVirgin (talk) 00:53, 25 March 2006 (UTC)

Sorry, SlimVirgin, delinking does not clearly have a consensus, at least not as practiced by Bobblewik and some others: such users have been blocked (and then, sometimes followed by a wheel-war). So this consensus *existing* is nothing more than a myth. I have put this MoS page on my watchlist because of the disruption following from delinking and relinking spreading to so many pages. I suppose I'm not the only one, so I think your suggestion that this page was "much-watched" at the moment that it got more specific about overlinking might be unjustified conjecture.

So, I'm open to any suggestion by Talrias – but as far as I'm concerned there is no prerequisite to state a "consensus against overlinking". Stating that there is no consensus that this should be interpreted any stricter than the general recommendations against overlinking would be as viable to me. --Francis Schonken 01:26, 25 March 2006 (UTC)

I was under the impression that there is consensus and that the only reasons why Bobblewiki was blocked was because he was using a bot and issues relating to it (quick successive edits, bot errors, etc). I think SlimVirgin is correct in saying there is consensus. I would actually like to hear from those who opposed Quadell's and other users' delinking, when they think years should be linked. Because judging from the reverting that went on, it seems like almost every year should be linked, which I believe is a minority view. Maybe someone can clarify here? Gflores Talk 01:54, 25 March 2006 (UTC)
Bobblewik didn't use a bot: I understand Bobblewik used AWB ("is not a bot") and some script for some time published on user talk:Bobblewik; The delinking edits were high speed, requiring "bot permission", which was denied in two consecutive votes. Then in the end I think it was decided (but I don't know where, WP:AN/I?) Bobblewik could be blocked for de-linking a single date. Which occured. Got blocked. Wheel war followed ([1]).
There were no "bot errors", just that Bobblewik delinked all separate years, in all processed articles (there was no pattern of topics Bobblewik was interested in; no sign of awareness of context where a linked date could actually be useful; and what enervated most, to me at least, Bobblewik first delinked, and thought it good behaviour to think that if such delinking was not questioned, then it was by definition an "improvement" – in a "shoot first, ask later" style,...)
So, no there's no consensus that "delinking" applied to separate years should be treated different from general recommendations against overlinking, that is: according to what you quoted: "Generally, do not link unless they will clearly help the reader to understand the topic." (sorry, above I thought that sentence was of your making, but it's simply the general recommendation against overlinking); it is my contention that no "separate" rule should be made against overlinking of single years. Which means:
  • This can never be done by bots (while needing interpretation of the text where the year is inserted); the use of semi-bots is questionable (see also the elaborated #Example above);
  • This is not a fixed-rule issue, e.g. in most cases when I write an article about a book or a composition I would link the year when it was published or written, I did one without any such link recently (gnossienne) as a try-out: no, this doesn't work, and I'll be adding some year links soon. Which means it is not very different from the BC/BCE style issue, or maybe even more comparable "style of references conversion", things with ArbCom rulings against them. --Francis Schonken 02:47, 25 March 2006 (UTC)

I think the issue that Talrias is wanting discussed, is what determines when a date Does get linked? How is importance determined (allowing for subjectivity)?

  • Example of when de-linking should NOT occur (ie if a bot were run, this would create a mistake edit)
  • Example of when de-linking SHOULD occur

I would support the issue going to some sort of vote.
I would insist that we wait until the bug 4582 is fixed.
Until then we can discuss it peacefully :) --Quiddity 01:37, 25 March 2006 (UTC)

We're not discussing years in dates, but the links in solitary years (e.g. In 2004, Jimbo did this...). Gflores Talk 01:54, 25 March 2006 (UTC)
But years in dates are a subsection of the overall discussion. The decision about one will end up being applied to the other (and then redebated...). We must consider the larger issue if a "Guideline" is to be established, yes? as discussed in the preamble i added. --Quiddity 03:28, 25 March 2006 (UTC)

discussion continued from /archive42 < (long discussion)
archive43
> discussion continues in /archive44