[11:28:51] I'm having trouble understanding how some Wikispecies data fits into Wikidata. I wonder if someone can explain it to me, and #wikispecies doesn't seem to have many people in it. [11:28:53] I'm looking at an item called "rabbit" https://www.wikidata.org/wiki/Q9394 [11:29:03] It is an instance of 'taxon' but it has no 'taxon name' property and doesn't seem to exist on Wikispecies. Is this a real taxon? If not, how could I filter it out when I'm processing data items? [11:30:55] * phschafft listens in as he is also interested in the answer. [13:26:08] mtorpey: there's a good chance the answer is it's incomplete [13:27:17] I'm not even sure if "rabbit" is usually considered a taxon [13:27:41] like, wouldn't that be leporidae? [13:28:33] for how could you filter it out... not having a taxon name seems like a decent filter [13:28:49] But you could always fix the entries you come across yourself ;) [13:38:18] It occurred to me to fix problems I find, but I'm not a biologist and don't want to do anything stupid! [13:39:57] I've been making progress the last couple of hours, and it seems like "rabbit" is not a clade, being used for some branches of Leporidae and not others. So I'm guessing that's why this stands out. [13:41:09] Strictly I think "taxon" can basically mean any set of organisms that you give a name to, but we're usually interested in the ones that are clades, which seem to be the vast majority of taxa on Wikidata. [13:41:37] I've been filtering based on whether they possess a "taxon name" property and that's avoided similar problems. [13:43:07] see also 'folk taxonomy' [13:43:36] indeed! [13:44:00] which rabbit is an instance of such a taxon. [13:44:12] but it is not a clade. [13:44:42] and it can be frustrating if you need to handle those cases and when you're not sure which of those it is. [13:45:35] extra points for those that exist in both worlds: like fish. by folk taxon it means funny water animal, by clade it includes humans! [13:46:48] lol, indeed. I love being a spiny fish. [13:47:12] It'd be nice to have an explicit property for taxa called something like "paraphyletic" for non-clades, but I'd have no idea how to start. [13:48:34] mtorpey: yeah, that all sounds right to me re: clade. (Also not a biologist) [13:48:53] In the meantime I've been struggling a bit with "taxon name" for new undescribed species. Some have no taxon name, some have an empty array, some have an array with one entry that's explicitly marked as having no value. Fortunately I'm not interested in any of them! [13:49:11] so what's the struggle :P [13:52:55] okay here's one that I could surely fix! [13:53:25] I've found a handful of items that have the same taxon name listed twice, I guess one could be deleted: https://www.wikidata.org/wiki/Q21308564#P225 [13:53:43] for sure! [13:55:29] What do you think of this one? Same name, same author, different years: https://www.wikidata.org/wiki/Q718688#P225 [13:55:49] there you just want to use the preferred one [13:56:11] Sure! But should the other one be deleted or combined with it somehow? [13:56:39] or is it valuable data to know that the same author used that name in two different years? [13:56:53] maybe the same author made two publications, which are not referrenced, but the years might point to that. [13:57:15] in that case maybe adding the publications would be the better option. [13:57:48] it's kinda a weird area—https://www.wikidata.org/wiki/Help:Ranking#Deprecated_rank has some helpful stuff [13:57:51] maybe also have a look at the history. [13:58:53] makes sense [13:59:06] There seem to be a few of that form [14:00:42] I mean, it can be an error or an outdated way to do things. but I would when in doubt consider that whoever entered the data had some reason for what they were doing. [14:00:51] ^ [14:01:14] though anyone who doesn't use reason for deprecated rank is smelly [14:02:33] I combined two entries on this beetle, which I think must be the right thing to do: https://www.wikidata.org/w/index.php?title=Q30232589&diff=2530898509&oldid=2459271708