Wikidata:Property proposal/extinction date

date of extinction

Originally proposed at Wikidata:Property proposal/Natural science

Done: extinction date (P8556) (Talk and documentation)

Description	the date that a species or language is thought to have gone extinct
Data type	Point in time
Domain	item taxon (Q16521) languoid class (Q28923954)
Allowed values	values must be in the past
Allowed units	dates, ranges of dates, or estimates of dates
Example 1	quagga (Q45969) → 12 August 1883 (captivity) 1878 (wild)
Example 2	Raphus cucullatus (Q43502) → 1690
Example 3	Mesmes (Q3508617) → c.2000
Example 4	Old English (Q42365) → c.1100
Example 5	Tyrannosaurus rex (Q13098211) → 68 to 66 million years ago
See also	dissolved, abolished or demolished date (P576), date of disappearance (P746), IUCN conservation status (P141)

Motivation

The date of exitinction is an important piece of information for species and languages. It is comparable to the date of death of a person, yet there doesn't seem to be any corresponding property, yet, in Wikidata. In the case of the Raphus cucullatus (Q43502), the property "dissolved, abolished or demolished date (P576)" is used to state the date of extinction, but the description of that property states that it should only apply to an "organisation, [or] building." In fact, dissolved, abolished or demolished date (P576) has a constraint violation when used on the Raphus cucullatus (Q43502) page. We also see on Raphus cucullatus (Q43502) that IUCN conservation status (P141) is given as extinct species (Q237350) with the qualifier date of disappearance (P746): 1690. But, again, that property violates a contstraint because because date of disappearance (P746) has a type constraint of people and fictional characters. Besides, the extinction date is too important to bury in a qualifier. Therefore, I propose we create a new "date of extinction" property.

The-erinaceous-one (talk) 03:52, 6 August 2020 (UTC)[reply]

Discussion

WikiProject Taxonomy has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead.

Notified participants of WikiProject Biology

Is this date usually known? For species in the wild, the individuals may not be monitored closely and simply not seen for many years, without knowing they are extinct... ? ArthurPSmith (talk) 19:32, 6 August 2020 (UTC)[reply]

Yes, the dates are known with varying levels of percision. For example, if you look at Wikipedia's list of extinct mammels, every entry has an extinction date. A quick review of other lists show they are the generally same. The-erinaceous-one (talk) 21:36, 6 August 2020 (UTC)[reply]

@The-erinaceous-one: Ah, I guess I was thinking of the ones classified there as "Possibly extinct". How would you handle those? Also the "extinct in the wild" category? ArthurPSmith (talk) 17:20, 7 August 2020 (UTC)[reply]

@ArthurPSmith: Do you have any suggestions for what to do if extnction is only suspected? We could use nature of statement (P5102) as a qualifier with <suspected> as the object. It seems like there must be a common way of marking a statement as uncertain on Wikidata. For distinguishing "extinct in wild" vs. "extinct in captivity," we can use valid in place (P3005) as a qualifier. This would also be useful for cases where a species is extinct in some region but not everywhere. The-erinaceous-one (talk) 21:21, 7 August 2020 (UTC)[reply]

@ArthurPSmith: if extinction is only suspected, we could also use sourcing circumstances (P1480) with values unconfirmed (Q28831311), presumably (Q18122778), according to some sources (Q59783740) or disputed (Q18912752). Do you have any thoughts on whether nature of statement (P5102) or sourcing circumstances (P1480) is preferable? The-erinaceous-one (talk) 21:14, 9 August 2020 (UTC)[reply]

Why not use significant event (P793) for non fossil taxa? --Succu (talk) 18:03, 7 August 2020 (UTC)[reply]

@Succu: Thank you for the suggestion. Using significant event (P793) would work in the simple case where a species is definitely extinct everywhere, but how would it work where a species is only suspected to be extinct or only extinct in the wild? I think that we should have a distinct property that allows for qualifiers to be set. The-erinaceous-one (talk) 21:26, 7 August 2020 (UTC)[reply]

Also, I think creating a distinct property for "date of extinction" would make it much easier for Wikdata contributors to know how to add the extinction date in a standardized way. The-erinaceous-one (talk) 23:23, 7 August 2020 (UTC)[reply]

What about using an item of a new domain "extinction status" instead of "point in time"? This would allow us to use items that we can define such as "suspected extinct", "likely extinct" and "extinct". We could also add qualifiers such as applies to part (P518) -> "wild" or "captivity" (if there are still animals in Zoo). We could thus combine this with point in time (P585) to denote the time an animal went extinct in the wild and in captivity. What do you think? --Hannes Röst (talk) 14:58, 11 August 2020 (UTC)[reply]
- @Hannes Röst: We actually have IUCN conservation status (P141) for this already. So yes, a point in time (P585) qualifier could be used to model this, that's actually a pretty good solution. ArthurPSmith (talk) 18:19, 11 August 2020 (UTC)[reply]
- @Hannes Röst, ArthurPSmith: I like the idea of "extinction status" with qualifiers, but IUCN conservation status (P141) seems like it is just a little to specific. Is the IUCN the only organization that declares a species extinct? Also, Wikipedia:Conservation_status says, "for threatened species and species believed to have become extinct after 1500, the categories are based on the IUCN Red List," so how do we classify prehistoric animals? Maybe IUCN conservation status (P141) should be changed to just "conservation status" with IUCN added as a reference? The-erinaceous-one (talk) 02:52, 13 August 2020 (UTC)[reply]

Support The IUCN conservation status (P141) seems good for many use cases, but not all. I see no harm in having this property. Also, I'd change the wording for "estimated date of extinction" or similar, as it is always an estimate. TiagoLubiana (talk) 15:56, 13 August 2020 (UTC)[reply]
- @TiagoLubiana: I think that "estimated" is unnecessary because that can be specified with qualifiers and sometimes the exact date is known, such as when the last of a species dies in captivity. Right now, I'm on the fence with regards to whether "date of extinction" or "conservation status" is a better way to model this information. Using "date of extinction" will definitely make it easier for editors to add the information to Wikidata, but it is less general. On the other hand the term "conservation status" implies that there was or is an effort to conserve a species, but that doesn't make sense prior to human conservation efforts. The-erinaceous-one (talk) 00:03, 14 August 2020 (UTC)[reply]
- @The-erinaceous-one: Well, you can never be sure that the "last of a species" is really the "last of a species," in the best case scenario it is just a very good estimate. But whatever works for you. Regarding "conservation status" I believe date of extinction is better, as it is very clear, and it makes easier for editors and for people who are querying the database that use the data. Conservation status would require a way lengthier discussion. TiagoLubiana (talk) 14:55, 14 August 2020 (UTC)[reply]
  - @TiagoLubiana: In a real sense, we can never be sure of just about anything, but I get your point: there can be more uncertainty about the date of extinction than most things. However, on Wikidata we don't need to know if a claim is true, we only need a good sources that say so. So, if a reliable sources states when a particular species is (declared to be) extinct then that merits a statement, and if that source states an estimate, than we can handle it with qualifiers. The-erinaceous-one (talk) 07:45, 17 August 2020 (UTC)[reply]

Support NMaia (talk) 11:55, 18 August 2020 (UTC)[reply]

@The-erinaceous-one, ArthurPSmith, Succu, Hannes Röst, TiagoLubiana, NMaia:

Done extinction date (P8556) Pamputt (talk) 21:25, 23 August 2020 (UTC)[reply]