Wikipedia talk:Dates in Wikipedia

Broader format dates
Hi, I saw the list of date formats that I assume was the list you used in compiling these data and I had a curiosity/suggestion regarding it--obviously, there are massively more dates for current times than far in the past. It would be interesting to see if you searched also for patterns like "the late/early/mid Nth century," "the N0's" (decade), "the Nth millenium," etc--historical articles probably have many more of these more broad dates because of the uncertainty they deal with, and although it might not make a big difference in the results I'd be curious to see if it made a significant one. Mehmuffin (talk) 14:59, 19 July 2017 (UTC)

Mehmuffin,

This is a very good question. And the answer revolves around (the law of) diminishing returns. There are any number of ways of describing a point in time in the English language and you give a couple of good examples. The problem is, as you expand your criteria for finding dates into more and more obscure constructs, you slow the whole process of finding all of the dates you can. And at some point you have to draw the line and say "Why would I want to find this obscure date format to find 10 more dates out of 40 million dates, when it may slow the whole process down by 2%?" At some point it makes no economical sense. This is where the perfect becomes the enemy of the good.

https://en.wikipedia.org/wiki/Diminishing_returns

https://en.wikipedia.org/wiki/Perfect_is_the_enemy_of_good

Jroehl (talk) 20:46, 19 July 2017 (UTC)


 * , often dates in prose are better expressed to readers in general terms, such as "in early 2015" or "in January 2015" rather than "On January 25, 2015, ..." (dateline). Are only the full dates counted? czar  07:58, 16 August 2017 (UTC)