Talk:Q18593264

Merge with collectable (Q2342494)?

@Oursana, Andreasmperu, Nurni, Spinster, Avatar6: Should we merge this item with collectible (Q2342494)? --Marsupium (talk) 08:22, 26 March 2018 (UTC)

the interwikilinks are only about collecting, so I would not merge. What about collectible (Q2342494) subclass of (P279) > item of collection or exhibition (Q18593264)--Oursana (talk) 11:40, 26 March 2018 (UTC)

work (Q386724)

I added subclass of (P279)work (Q386724) to this, but Nurni reverted it without comment. To me, it seems fairly evident that not all items of a collections are physical objects. They can be digital art, archived web pages, electronic records, audiovisual, so it would make sense to me materials, datasets and computer code, and so on. What is the issue here? Dominic (talk) 20:11, 31 May 2019 (UTC)

@Dominic: Collections may contain objects that are not made by human (e.g. fossils, minerals etc.). Nurni (talk) 20:23, 31 May 2019 (UTC)

@Nurni: Yes, I agree. What does that have to do with my edit? Dominic (talk) 20:46, 31 May 2019 (UTC)

@Dominic: work (Q386724) means made by humans. fossil (Q40614) for example, can be items of a collections but were not made by humans. So, you can't say that all item of collection or exhibition (Q18593264) are work (Q386724) Nurni (talk) 21:37, 31 May 2019 (UTC)

@Nurni: Agreed, but by the same token, physical object (Q223557) refers to a physical object, which not all items of a collection are, as I stated above. Rather than simply removing either claim for being too narrow, do you have an idea of how we can resolve the problem and accurately describe this item? I have noticed some issues that are caused by the way this is described, such as an items which are classified as instance of (P31)item of collection or exhibition (Q18593264) will run into statement constraints when trying to add reasonable fields that belong to collections objects, such as title and production date (which wants to be applied to works and not physical objects). Dominic (talk) 23:04, 31 May 2019 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Dominic, Nurni: I see two different problems here. Collections do include both physical objects and digital objects, so perhaps we can change the parent of “Item of collection” to a different class that would include both physical object (Q223557) and digital object (Q59138870) as its subclasses. (One of our items called “object” perhaps, though I’m not sure which one - they don’t make a lot of sense to me as they are currently used).

For the issue that some items in collections are also works, I like the simple solution of making those items <instance of> both “item of collection” and the appropriate subclass of “work” which should resolve the constraint violations (and is also correct in my mind). - PKM (talk) 03:25, 1 June 2019 (UTC)

@PKM: Sounds good to me. Nurni (talk) 04:46, 1 June 2019 (UTC)

@Dominic, Nurni: digital object (Q59138870) is sublass of concrete object (Q17553950). Are we okay to make concrete object (Q17553950) the parent of "item of collections"? - PKM (talk) 20:33, 1 June 2019 (UTC)

I think that makes sense. Though, having studied them more, I admit to being confused over the technical distinctions between concrete object (Q17553950), concrete object (Q4406616), and physical object (Q223557). (concrete object (Q17553950) is described as "tangible", which is different from "physical". Are digital objects "tangible"?) But it's likely an improvement in any case. Dominic (talk) 15:04, 5 June 2019 (UTC)

@Fuzheado: ^^ Multichill (talk) 16:45, 28 January 2023 (UTC)

Thanks, I don't feel strongly about this, so I'm happy to follow the lead above. - Fuzheado (talk) 19:37, 28 January 2023 (UTC)

instance of = item of collection or exhibition

@Dominic, Zolo, Multichill:, in c:Template_talk:Artwork#Object_type_:_item_of_collection_or_exhibition There is a proposal to rewrite c:Template:Artwork to ignore instance of (P31) = item of collection or exhibition (Q18593264) in items like Still Life with Sweets (Q2349662). I kind of agree with it since, likely 99% of artworks out there are part of some collection or were part of some exhibition. A better way to model that would be to add collection (P195) and exhibition history (P608). So for items with those properties I would like to propose to remove instance of (P31) = item of collection or exhibition (Q18593264) as redundant. --Jarekt (talk) 14:05, 22 March 2020 (UTC)

Agree. Calliopejen1 (talk) 23:02, 23 March 2020 (UTC)

@Jarekt: bit late reply. Agree. I ran into this item again on paintings. This item should generally not be used. For me this is the item you would use when you import a diverse collection and you have some left overs you're not sure about how to classify (others/misc/etc.).

Instead of hiding it in the code, I would prefer we clean up the data. So if it already contains a suitable instance of (P31) like on Ravana’s sister Shurpanakha complains that Lakshmana cut off her nose and ears, from Chapter 31 of the Aranya Kanda (Book of the Forest) of a “Shangri” Ramayana (Rama’s Journey) (Q60483058) (permalink), we just remove this item. Agree? Multichill (talk) 17:43, 7 July 2020 (UTC)

@Dominic: based on https://w.wiki/WLJ , you're probably the top user of this item. Any comments?

On https://w.wiki/WLL it's easy to see the amount of items we can remove this property from which have another valid instance of (P31). Multichill (talk) 17:56, 7 July 2020 (UTC)

Agree "item of collection" does not add anything clear when another P31 is provided. -Zolo (talk) 16:36, 15 July 2020 (UTC)

https://w.wiki/X$e with labels. Fired up the removal for the paintings. Multichill (talk) 08:35, 27 July 2020 (UTC)

Sorry I missed my ping earlier. I added these in large numbers, but only because I was dealing with large amounts of data, not because I feel particularly strongly on this. So I am happy to defer to others if there is agreement about it. Note, while I understand what people are saying about most artworks (described in Wikidata) being museum objects, most of the uses of this in P31 are not for artworks at all. The NARA ones are mostly items like Q66319120, a record in an archives which could have been any format or media, art or not, where there was no other machine-readable (or mapped) possible value for P31. If you remove it from ones with an alternative P31 value, most of those NARA ones will probably remain, which is fair. I certainly agree with the idea that this value makes no sense to display in Template:Artwork.  – The preceding unsigned comment was added by Dominic (talk • contribs) at 19:08, 27 July 2020‎ (UTC).

@Multichill@Jarekt@Calliopejen1 I agree with P31 restriction for artwork or similar. However, now I'm working in camera (Q15328) and its subclasses and I found items like camera inv 389 (Q30250682), Kodak F3 DCS-CnAM 43706 (Q55221073) or Hasselblad 500 EL, 1968 - Musée Nicéphore Niépce (Q65719563) that are one specific camera among the others like it, so, I consider that it is not a camera but a item of collection or exhibition (Q18593264).

In these cases, what should be the P31 in your opinion, and which would be the correct property to indicate that is a item of collection or exhibition (Q18593264). Thank ! Amadalvarez (talk) 12:07, 13 January 2023 (UTC)

This distinction is also important for artistic printmaking. It's similar to the FRBR hierarchy Work–Expression–Manifestation–Item. Getty Art and Architecture Thesaurus has “impression” [1] versus “print”. Paintings on the other hand tend to be unique manifestions (though the artist may paint multiple versions of the same composition), so that they are instances of both an artistic expression and a physical object. (We tend to treat, say, an oil sketch and the final studio painting as separate works, rather than different versions of the same artistic intent.)

@Multichill @Dominic @Zolo @Amadalvarez

I ran into the constraint whilst trying to model Victoria, Princess Royal, with her nurse (Q124002522) (the work), Victoria, Princess Royal, with her nurse; VAM E.2175-1932 (Q124038028) (a printed object in one museum), Victoria, Princess Royal, with her nurse; BM 1926,0109.33 (Q124038029) (printed object in another museum). If the original printing plates had survived, they could also be museum objects; but like Amadalvarez's cameras, not all printing plates are necessarily also museum objects.

Agree that it's not needed to show in a painting template, but I do think this class should be allowed as a P31 where the wikidata item is about a specific physical instance that has a museum catalogue record.

There are also some dependent constraints. For archival items, level of description (P6224) requires the subject to be instance of (Q21503252) item of collection or exhibition (Q18593264) (instance of item of collection or exhibition).

Ideally, properties like is-member-of-the-collection would only apply to things that are actually items of a collection. Donning my OOP hat: the ability to hold such properties should inherit from that class/interface. (I think this should be explicit, rather than inferring the class from the existence of the property, as @Jarekt suggested.)

If the assumption is that all paintings are also items of some collection (possibly an unknown private collection), then should we not have painting (Q3305213)subclass of (P279)item of collection or exhibition (Q18593264)? ⁓ Pelagic ( messages ) 23:00, 28 December 2023 (UTC)