Support ISO 8601-2:2019 to specify uncertainty about times
Problem: Many times sources specify a date with some uncertainty. A source might say that a person died between 1341 and 1345, or that it was approximately created at a certain date.
Who would benefit: Anyone who wants to record data with uncertainty in a way similar to what the sources say.
Proposed solution: Adopt the ways of specifying Time Interval, Qualification of a date and Unspecified digit(s) from the right layed out in ISO 8601-2:2019 (described in http://www.loc.gov/standards/datetime/edtf.html). This would allow writing 1341/1345 for a date of death between 1341 and 1345.
Sounds sensible. Does the standard also cover some typical historical usages (fl., circa, before X...)? I tend to see that kind of dating are usually misunderstood for more numerical treatment.--FAR (talk) 12:18, 28 November 2020 (UTC)Reply[reply]
As far as I understand fl. stands for floruit and that's well modeled as it's own property in Wikidata and I don't see how it belongs here. The standard defines "uncertain" and "approximate". Circa can be modeled with uncertain. I didn't explicitly ask for open ended time intervals in this proposal but the standard does define them. ChristianKl ❪✉❫ 23:03, 15 December 2020 (UTC)Reply[reply]
+1 to Jarekt below: While the current practice is tedious and I don’t like it very much, a clear path to a clean future unified state would need to be defined prior to implementing the proposal. --Mormegil(cs) 14:08, 14 December 2020 (UTC)Reply[reply]
Yes, but the Community Wishlist is not the place to define standards of how things are used. Those conversation have to happen on Wikidata and having them only makes sense when there's a decent likelihood that there's work on this feature. ChristianKl ❪✉❫ 22:58, 15 December 2020 (UTC)Reply[reply]
Support Would better align Wikidata with Library of Congress best practices for archival records, for instance. Currently qualifiers are used but prevalence is uneven. Arlo Barnes (talk) 16:16, 9 December 2020 (UTC)Reply[reply]
Support This would be very handy. Right now there's no proper way to indicate something like this except for a decade (which is not very specific). Husky (talk) 16:15, 11 December 2020 (UTC)Reply[reply]
Neutral improving modeling of the dates on Wikidata is an ongoing issue and there are many long standing tickets related to it. However this proposal provides alternative modeling solution to the current practice of modeling more nuanced dates with qualifiers (see d:Help:Dates#Qualifiers), without addressing the issue how are those 2 standards going to coexist. --Jarekt (talk) 15:46, 12 December 2020 (UTC)Reply[reply]
Support IagoQnsi (talk) 18:40, 8 December 2020 (UTC)
Neutral improving modeling of the dates on Wikidata is an ongoing issue and there are many long standing tickets related to it. However this proposal provides alternative modeling solution to the current practice of modeling more nuanced dates with qualifiers (see d:Help:Dates#Qualifiers), without addressing the issue how are those 2 standards going to coexist. --Jarekt (talk) 15:46, 12 December 2020 (UTC)
• hugarheimur 08:49, 14 December 2020 (UTC)
