[00:02:14] Creating random 'orphan' items for people that don't fill a 'structural need' (like, authors of works cited on some wikipedia) get bad admin attention. [00:02:57] But you can also ofc end up with seeming 'totally random' people because they are cited somewhere. [00:05:38] What I can't see (what would matter a lot) is what kind of connections to other items he had. [00:07:00] (FWIW, Wikipedia USES imdb a lot, but being on imdb isn't considered to contribute to notability... though 'fact noted there' can make it obvious that someone is notable [00:10:11] as far as what mahir said, WD does include '3rd party databases' to be in scope, for mass injestion in fact, partly because doing so intelligently points out lots of missing items beeded to describe them. [00:10:15] er, needed [00:11:13] Like, you could create records for people listed in Library of Congress Authorities all day long, and nobody would complain, they would probably thank you. [00:13:04] The 'bio' part of imdb is (usually) written by that person...what is useful as a 'source' is really the list. [00:16:36] ..which I believe is ibdb's own (aggregate) database, apart from the stuff people can write [00:40:08] WD takes the 'relational' part of relational database to the extreme, lol... when it can take creating half a dozen items to meaning state one thing. [00:40:18] *meaningfully [00:40:45] Yup it's a bit extreme in some places [00:43:12] Plus imdb (given that it's business model isn't as a data provider) probably includes trap data [00:43:51] Yeah getting reliable biographical data is a HUGE pain in the butt [00:43:55] It's why I usually don't [01:02:54] I've got to say, figuring out that I can...I dunno, 'compose' data in OO and mangle it into QuickStatements is going to make it a lot easier to contribute here. [01:03:56] OO? [01:49:43] OpenOffice [01:50:11] Calc, specifically [01:54:36] ahhh [02:25:00] thanks for your input [02:50:59] Q49041745, lol, nice when some concept already exist for you. :) [03:10:23] perryprog, https://pasteboard.co/Kinc60X.png https://pasteboard.co/KincfIa.png <- much betterfullness, tho more to mung before I dump this into QuickStatements. Mainly being able to figure out and ten reliably stick to a schema without dump mistakes [03:11:29] Neat! [03:22:23] Years ago, I worked (as a temp, actually) in the parts warehouse at a big factory, when they had 'gone computerized' they had hired a bunch of random temps to input data....people who had no idea of the difference between a gear and a sprocket, or how to usefully describe either. I learned then how useful a 'spreadsheet type' view can be. [03:22:48] Tho trying to fit the idea around a RELATIONAL database is kinda broken. [03:24:16] ^ That two week temp job became a summer of hunting a thousand random misspellings [04:40:21] Yeah, lol, creating lots of 'academic departments', 'faculties', 'research groups', 'research institutes', to add affiliations for these researchers. One advantage to doing something like this with a bot, I guess, I can actually read. :) [04:41:51] I should probably set up another spreadsheet, do the whole university at a shot, lol. The way tasks here can expand exponentially. [06:58:25] > WD takes the 'relational' part of relational database to the extreme [06:58:25] WD is not a relational database... [07:20:26] haansn08, I didn't meant literally, hence the air quotes...just that it's more about 'relationships' (this is a property of that) that just being a flat db [07:20:41] *than [07:22:00] I'm sure there is a term... [14:52:36] are senses for lexemes always in the original language? [14:53:48] oh, nevermind, finally found one that's got more languages in the sense (Earth( [15:38:18] Jarnsax: maybe a graph db? [15:40:07] * graph db is the term you are looking for? [17:07:30] jimman2003, After some reading, yeah. " In contrast, the graph database would walk from Gone With the Wind to Clark Gable, gather the links to the movies he has been in, gather the links out of those movies to other actors, and then follow the links out of those actors back to the list of movies. The resulting list of movies can then be searched for "submarine"." <- sounds about right [17:09:04] the enwiki article calls it a 'knowledge graph' [17:09:38] https://en.wikipedia.org/wiki/Knowledge_graph [19:13:15] Is there a better way to describe something like "Pauline Jewett Institute of Women’s and Gender Studies [19:13:15] " than "field of work [19:13:15] " = "undergraduate education [19:13:15] ", "postgraduate education [19:13:15] ", "women's studies [19:13:16] ", "gender studies [19:13:18] ", "sexuality studies [19:13:22] ", "disability studies [19:13:25] ", and "critical race studies [19:13:26] "? [19:13:28] Ouch, dammit spam [19:13:37] invisible linebreaks [19:14:55] i.e. is there a property for 'academic fields in which an academic program offers degrees'? [19:28:41] found it... https://www.wikidata.org/wiki/Property:P5460 (grants)