This page is an archive. Please do not modify it. Use the current page, even to continue an old discussion.
Spidean Mialach (Q24677975) different from Spidean Mialach (Q13131534)?
Hi, I noticed you marked these items as different from one another, but I believe they should be merged because they share the same location and height on all linked wikipedia articles. I'm also only aware of one Munro with the name "Spidean Mialach" --SilentSpike (talk) 18:23, 3 January 2019 (UTC)
Dear Laddo I don't understand what's wrong in the model ?
then, why not
- Hi @MKar:, is also wrong. (I will fix that) ;) Check Help:Basic membership properties to clarify what is an instance and what is a type/subclass. Generally speaking: an individual object or entity is an instance (my dog "Rex" is instance of (P31) of dog (Q144)). A type/class/kind of something is a subclass of that thing ( ) (Subclasses may additionally be instance of (P31) of class-like entities, like dog breed (Q39367) or ship type (Q2235308) or class of award (Q38033430)). LaddΩ chat ;) 03:04, 29 January 2019 (UTC)
Empty category items
Hey Laddo, there are items Q61237312, Q61237365, Q61237397, Q61238762, and Q61238833 which are apparently missing sitelinks. Can you please add them? Thanks, MisterSynergy (talk) 20:50, 5 February 2019 (UTC)
Alves da Silva
"Alves da Silva" in and by itself, is an unproductive combination of two disparate surnames (Alves and Silva), and it is of no use to treat it as a unit. Only spcific “dinasties” of related people could gain from being bagged together; people with random, volatile instances of this or any other combination of surnames do not. Tuvalkin (talk) 00:27, 31 May 2019 (UTC)
- @Tuvalkin: I don't have specific opinion on that case, but guess it depends; I know some compound names that constitute a unity despite that they are made up by two family names, like the French case Miville-Dechêne (Q59196647), see this search. I would expect such cases to be written with a dash ("-") in between the two surnames, generally. Lαδδo chat ;) 00:35, 31 May 2019 (UTC)
- Not a matter of opinion, a matter of facts and knowledge thereof — up which decisions concerning how onomastics are to be curated can be made and opinions about those decisions can be developed. You don’t know the first thing about Portuguese onomastics (or else you would not be refering the unrelated matter of hyphenated surnames), yet you think «it depends». Well, it kinda does: It depends on Wikidata’s rules on how to treat Portuguese onomastics — which has its challenges and doesn’t fit the simple "given name + surname" system, but which is relatively simple campared with Spanish, Icelandic, Slavic, Hungarian, Arabic, or Far Eastern onomastics. Sadly, Wikidata doesn’t seem to have developed a meaningful way to deal with onomastics. Is has been merely vampirizing human-entered data from Wikimedia Commons and serving it back via its toxic, non-wiki interface. I wish Commons would take back control over its own data, given the huge disappointment Wikidata has become. Tuvalkin (talk) 01:02, 31 May 2019 (UTC)
- @Tuvalkin: You are correct I know nothing about Portuguese surnames, though I now have an idea of some differences. Wikidata attempts to provide machine-readable that reflect reality, and it is not without flaws. If you know for sure that the concept of compound family name does not exist in Portuguese, feel free to fix those but understand that other languages may require that. Lαδδo chat ;) 01:30, 31 May 2019 (UTC)
- Wrong again: In Portuguese onomastics yes there are compound surnames (we have been plagued for centuries with endless strings of Pacheco Pereira, Morais Sarmento, Câmara Pereira and many others), but that’s almost never the case of multuple-name surname sections of the typical Brazilian or Portuguese person. A clever system would have tools to curate that and a bunch of other issues of Portuguese onomastics such as, say, particles ("e" and "d*") and two-word surnames — necessarily harmonized with analogous issues of Spanish and other onomastics and with special consideration to things like the differential treatment these issues get when integrated with onomastics of yet other cultures in the case of immigrant’s names (in France, the US, etc). You in Wikidata don’t have such a system, as said, and you don’t seem to have either the will or the resources for it. And you personally chose to engage in petty edit wars. Very well — at least things are clear now. Tuvalkin (talk) 13:48, 31 May 2019 (UTC)
Both of these are archaeological sites and have the same GeoNames ID, which mentions both names.(http://www.geonames.org/search.html?q=buthrotum&country=) They look like duplicated items; is there a difference between them? Peter James (talk) 21:22, 28 August 2019 (UTC)
- Hi @Peter James:, I believe that some time ago, Butrint (Q188694) referred to the old settlement site, and Butrint (Q22002796) was the modern place, since the second item indicates "named after" the first item. Now it all seems mixed up. Only Swedish WP has articles for both: sv:Butrint and sv:Butrint (fornlämning i Albanien). Good luck if you wish to clean it up ;) Lαδδo chat ;) 22:42, 28 August 2019 (UTC)
Item change request
Barłóg to Gawra - Change request
Hello! In the Q11699447 in most of items except Polish one it leads to the winter resting place of bears. The word "barłóg" in Polish means daytime resting place of bears. Winter resting place is Gawra in Polish. I am not expert in makin such item changes. Could you chenge it or show to me how to do it? Rembecki (talk) 08:56, 7 October 2019 (UTC).
- @Rembecki: It's not easy for me since there are no equivalent terms, as far as I know, in French or English. I gather that lodge (Q2472317) ("Barłóg" in pl) is the normal resting place while gaura (Q11699447) ("Gawra" in pl) is the winter/maternity place. Please check their interwiki links and confirm to me whether or not Polish items need to be exchanged between the two Wikidata items. Thanks - Lαδδo chat ;) 01:41, 9 October 2019 (UTC)
Roll back #quickstatements; #temporary_batch_1575680929394
country (P17) is a statement for Wikidata property to indicate a location (Q18615777) and should not be applied to navy ships or anything else that is not a geographical feature (Q618123). You should use country of origin (P495) instead. /ℇsquilo 10:05, 7 December 2019 (UTC)
Please review this page and collect valuable data
Human name disambiguation pages
Please do not mark items as "Human name disambiguation pages" unless the pages list only names of humans. You have been tagging pages that also include names of animal species, asteroids, and book titles. --EncycloPetey (talk) 01:39, 13 December 2019 (UTC)
Not sure what you meant here... Not only do they share same birth and death date as well as occupation, but sitelinks also cite the same publications... Where did you get that they are not the same person? --Nono314 (talk) 21:29, 15 December 2019 (UTC)
- @Nono314: My bad - a birth year without a date ends up as January 1st when downloaded in batch, and it messed up my comparison scheme. I removed that "different from" statement and merged these items. Good catch! Lαδδo chat ;) 22:23, 15 December 2019 (UTC)
Hi! I think you should pay more attention to the latter part of item description "item that is different from another item, with which it is often confused". You apparently add this property merely on basis of identical item labels, while there is little reason to believe that anyone ever actually confuses these items (e.g. in case of Q3417000 and Q3462299). In such cases "different from" statements in my opinion contribute to unnecessary clutter up rather than serve a useful purpose. That is, compared to cases where item descriptions are similar or overlap to a degree, and labels may be identical in some languages (e.g. Q42948 and Q2547819), or labels may not be identical (e.g. Q12089641 and Q25496059), and for this reason items may be actually confused. Surely on Wikidata people occasionally link to wrong items that normally wouldn't be confused, due to misclicking or due to being just careless, but adding "different from" everywhere probably doesn't prevent that kind of mistakes from happening. 2001:7D0:81F7:B580:C0E7:9E8B:9BD8:A9F0 16:27, 28 December 2019 (UTC)
- Hi 2001:7D0:81F7:B580:C0E7:9E8B:9BD8:A9F0, thanks for your comment. Indeed we have different perspectives on property different from (P1889). You use it to highlight differences between slightly different concepts, thus it represents meaningful information for the reader (wall (Q42948) vs. wall (Q2547819)).
- My use of P1889 is more technical: prevent careless merges and highlight some key difference (through qualifier criterion used (P1013), such as in Carmen Quesada (Q40686080)). I try my best to carefully select where I add the property, only where items overlap by more than their labels - river and village in same country, some ethnic group and their language, people with identical names, villages in different countries like the pair Undi (Q3417000) and Undi (Q3462299) that you pointed out. Much more often, similarly named items get added to whitelists such as Wikidata:Do_not_merge/14 without using P1889.
- The level of item overlap and resulting "confusion" needed for P1889 is not clear to me. Possibly a different property should be used for my purpose. I'll stop using it for now. Lαδδo chat ;) 20:46, 28 December 2019 (UTC)