[15:49:49] nikki: awesome, thanks so much! [16:30:44] Hi! Our page was blocked and disabled. I am looking to reactivate it please. [18:32:52] Hi, I found a problem with a bot [18:32:55] Hi, I found a problem with a bot [18:41:06] Hi faqr, tell us :) [18:49:21] some bots create backwards relationships [18:49:33] eg [18:49:45] A -> parent astronomical body -> B [18:49:51] so the bot assumes [18:50:12] B -> child astronomical body -> A [18:50:28] that makes sense except if A is a fictional object [18:50:41] Hmmmm [18:51:10] Fictional objects are inconsistently represented in general, from my point of view :S [18:51:24] I suspect it's Matěj Suchánek's bot? [18:51:27] I'm going to expose that one of these days [18:51:31] A = Q1088708; B = Q1052822 [18:51:59] here is the wrong edit [18:52:03] https://www.wikidata.org/w/index.php?title=Q1052822&diff=next&oldid=621472984 [18:52:08] Pasleim eh [18:52:17] Feel free to leave a message on https://www.wikidata.org/wiki/User_talk:DeltaBot [18:52:51] ok :) [18:54:03] The bad thing is that the behaviour is correct, but not the practice of using a property to mix items from different universes [18:58:33] ^ [19:02:35] (Excepting properties to intentionally, and always, link a subject of a universe with an object of another, such as P1074) [19:03:52] Dates should also be expressed separately, in different calendars, if the entities are from different universes, but I see that more complex :S [19:04:15] https://twitter.com/WikidataFacts/status/961636126468423681 [19:04:58] These aren't dates in our calendar/universe, but in the calendar/universe of the Harry Potter series [19:05:49] But they are treated in the same way right now [19:06:47] Because only calendars of our universe are allowed (Gregorian/Julian) [19:11:36] WikidataFacts, hey :) [19:11:43] hi :) [19:11:47] I've just mentioned your tweet... [19:11:55] 19:02:52 < abian> Dates should also be expressed separately, in different calendars, if the entities are from different universes, [19:11:59] but I see that more complex :S [19:12:00] 19:03:15 < abian> https://twitter.com/WikidataFacts/status/961636126468423681 [19:12:01] * WikidataFacts checks logs [19:12:03] 19:03:59 < abian> These aren't dates in our calendar/universe, but in the calendar/universe of the Harry Potter series [19:12:06] 19:04:50 < abian> But they are treated in the same way right now [19:12:08] 19:05:47 < abian> Because only calendars of our universe are allowed (Gregorian/Julian) [19:12:12] * abian floods the channel [19:12:38] ah, I see [19:12:40] hrm [19:13:09] so I would like to have separate calendars for, e. g. First / Second / Third Age and Shire Reckoning [19:13:22] but I’m not sure I agree that dates from Harry Potter should refer to a different calendar [19:14:43] In the case of Harry Potter, if its calendar is inspired by the real Gregorian one, we could use a fictional Gregorian calendar of the Harry Potter series [19:15:29] And that fictional calendar could be P1074 of our Gregorian calendar [19:15:47] Or something like that :S [19:15:53] https://scifi.stackexchange.com/questions/55020/do-wizards-use-the-same-calendar-as-muggles [19:16:28] Yeah, but muggles are also in the fictional universe of Harry Potter :/ [19:16:53] it’s the second part of that answer that worries me more tbh [19:16:58] because the calendar just isn’t consistent [19:19:51] We should tolerate inconsistencies in fictional calendars, I think [19:20:21] The Legend of Zelda's one is directly a nonsense xD [19:22:09] Is it a good idea to consider an arbitrary set of calendars (which should be items), fictional or not, apart from predefined Gregorian and Julian ones? [19:23:52] tbh I feel a bit uncomfortable discussing the addition of fictional calendars while we don’t have any more non-fictional ones [19:24:24] if we have a Harry Potter calendar before properly supporting, say, the Chinese or the Islamic calendar, that doesn’t look too good… [19:25:34] Perhaps we're mixing different ideas of "calendar" and, then, we should say that a date is only valid for a certain universe [19:26:48] Right now, we're mixing dates of design of a fictional object by its artist in the real world with the date when that object was "created" in the fictional world [19:27:16] Both are, simply, "start date", and there's no way to distinguish them in an automated way [19:31:13] The contemporary constraint will detect some of these inconsistencies, dates have sometimes no sense because they belong to different universes [19:35:47] We could also put all the fictional dates in the same bag with something like "fictional calendar" instead of Gregorian or Julian (if we try not to use an item for more than a fictional universe) [19:37:18] ^ What do you think about this option? [19:57:13] No opinions? You can tell me if you think that the issues I say aren't such, that's possible :) [20:13:12] sjoerddebruin? WikidataFacts? [20:13:24] Hmmm idk [20:13:47] Okay [20:14:02] I don’t think a generic “fictional calendar” is possible [20:15:34] What about a "novalue calendar"? :/ [20:16:09] has the same problem imo [20:16:14] how do you input data for this calendar? [20:16:23] how do you format it? (in the user’s language!) [20:16:28] how do you export it to RDF? [20:16:41] Good answer, you're right :) [20:16:46] any generic calendar, whether fictional or novalue, could only be a free-text input [20:16:52] not so structured data :) [20:17:59] we already have problems just with the existing calendars… see https://phabricator.wikimedia.org/T95553 for example [20:26:50] Yeah, WikidataFacts :( [20:27:07] Honestly, I've been thinking on this for some days and haven't come up with any perfect solution for the current representational ambiguities and other issues with fictional data [20:28:31] We need a kind of bubble that isolates fictional universes, the other issues around fictional data seem less important to me [20:28:59] Seperate Wikibase instances :P [20:29:14] :D [20:29:49] Yeah, but still a "start date" could be a date in the real world (which created the fictional one) or in the fictional world [20:31:06] When was Springfield (The Simpsons) created? In the real world, when the series were created; in the fictional one... no idea, xD but a different date [20:31:19] I feel like that shouldn’t be impossible to solve though [20:31:39] start time (or inception, no?) of the series belongs only on the series, not on the fictional items [20:31:50] but I have no idea how much existing data there is like you describe [20:32:02] We can use qualifiers such as "valid for the fictional universe" for fictional dates, too [20:32:28] I thought “from fictional universe” was only used as qualifier when it’s ambiguous, when the item is in several fictional universes [20:32:41] I feel like putting that on every statement would be a lot of redundant data [20:34:41] They would be, yes [20:42:25] sjoerddebruin: I think your idea is much better than any of mine :) [20:44:43] It's indeed possible to separate the real dimension from the fictional dimension of an entity as different instances, the "start date" of each would be the real and the fictional ones, respectively :D [20:45:25] Did you saw the case of dinosaur versus fictional dinosaur? [20:46:02] Hmm... I think I didn't, but sounds great :) [20:51:10] Not found, sjoerddebruin :/ [20:52:22] I can find the similar dragon case though. https://twitter.com/LucasWerkmeistr/status/947460047331840000 [20:53:21] Ahhh, yeah, that sounds familiar to me [20:53:35] Thanks for the link! [21:02:56] rip chanserv [21:04:36] It's now a fictional entity [21:06:58] One of my favorites is still https://www.wikidata.org/wiki/Q18246091 [22:37:31] 291 items are described in English as just “Middle-earth”, nothing more… -.-