Wikidata:Property proposal/Mount

mount type edit

Originally proposed at Wikidata:Property proposal/Natural science

Withdrawn

Description	The type of mount that an instrument (e.g., telescope) is mounted on.
Data type	Item
Template parameter	"mounting" in en:Template:Infobox telescope
Domain	bayonet lens mount (Q20181008) and telescope mount (Q1205231)
Example	South Pole Telescope (Q1513315) → altazimuth mount (Q677971)
Robot and gadget jobs	Import from Wikipedia

Motivation

This would be useful for en:Template:Infobox telescope, where this is one of the few parameters not yet automatically drawn from Wikidata. Although I'm focused on telescopes, it may also be useful in other situations, e.g. lenses. Mike Peel (talk) 14:27, 14 August 2016 (UTC)[reply]

Discussion

Support Support. Also renamed, as we have a concurrent proposal for a property called "mount", for the names of people's horses. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:34, 14 August 2016 (UTC)[reply]
Question What's wrong with has part(s) of the class (P2670) ? I'm a little unconfortable to create a specific property for this as it's seems just a specific case of this for telescopes. author TomT0m / talk page 17:34, 14 August 2016 (UTC)[reply]
- Do you mean has part(s) (P527)? I've been assuming that was for instances where we have specific entries for parts of other entries, e.g. Jodrell Bank Observatory (Q1569783) has parts that are the different telescopes at the site, rather than for descriptive articles about the type of part used in the entity. Either way, my hope would be to use this to populate an infobox field, and I'm not aware of an easy way to access specific sets of values for a property rather than just fetching all the values of that property (in this case, how would we tell 'mount' apart from 'enclosure' if they were both values for the same property)? Thanks. Mike Peel (talk) 19:36, 14 August 2016 (UTC)[reply]
  - @Mike Peel: No, I mean has part(s) of the class (P2670) (the initial name has become an alias). It's meant to link objects to kind of their parts. author TomT0m / talk page 21:33, 15 August 2016 (UTC)[reply]
    - @TomT0m: Thanks for the clarification. Isn't that more along the lines of telescope (Q4213) -> telescope mount (Q1205231) rather than South Pole Telescope (Q1513315) → altazimuth mount (Q677971) though? i.e., the general case, rather than the specific details? Thanks. Mike Peel (talk) 19:12, 16 August 2016 (UTC)[reply]
      - @Mike Peel: Nope. We took the option to decide that part of (P361) and it sister were appropriate to link items of the same (metaclass, see metaclass (Q1924819)   modelisation level : a physical object is part of another physical object (any physical object is eventually part of the universe), a class A of physical object is "part of" another class B of physical object means that an instance of A is a part of B. We created "has part of the type" to link physical object to class of physical objects, so that there is no ambiguity in the "part of" usage when we want to link physical object to both classes of physical objects and over physical objects. Hope it's clear and probably worth detailing somewhere on the documentation. author TomT0m / talk page 14:20, 19 August 2016 (UTC)[reply]
Support but as "mount type" so that it is also usable for camera lens mounts (e.g. Pentax K-01 (Q2069372) → K-mount (Q1063852)). Thryduulf (talk) 22:54, 14 August 2016 (UTC)[reply]
- That sounds good to me. I've tweaked the proposal name accordingly. Thanks. Mike Peel (talk) 06:10, 15 August 2016 (UTC)[reply]

Mixing bayonet lens mount (Q20181008) and telescope mount (Q1205231) looks like a bad idea to me. Oppose --Succu (talk) 21:03, 15 August 2016 (UTC) PS: Maybe a better generalization is photographic tripod (Q683906). --Succu (talk) 21:10, 15 August 2016 (UTC)[reply]
- @Succu: Why? Tripod is a different concept (think about why it's different from monopod (Q1306072) - and then there's the different rotation axes, although then I think I might be answering my own question...). Thanks. Mike Peel (talk) 21:21, 15 August 2016 (UTC)[reply]
  - Do you think bayonet lens mount (Q20181008) and telescope mount (Q1205231) are similar concepts as proposed, Mike? --Succu (talk) 21:28, 15 August 2016 (UTC)[reply]
    - I think they're similar enough that they could share a property, whether or not they should is up to Wikidata policies/the discussion here. Thanks. Mike Peel (talk) 19:10, 16 August 2016 (UTC)[reply]
Comment I was about to create this, but I'm not sure if it ends up getting people into translation difficulties. While the English version nicely shares the name (also with "mount" (P:P3091) which some people could use instead), isn't there a risk that other languages need to mention both ("lens mount/telescope mount") which isn't particularly useful?
--- Jura 14:18, 27 August 2016 (UTC)[reply]
- From the limited corpus available at wikt:mount#Noun_2 and the interwikis, I don't think that this is going to cause more issue than the equivalent of English "mount for a lens or telescope". Thryduulf (talk) 21:46, 27 August 2016 (UTC)[reply]
  - Hm. wikt:mounting states: 1 = Something mounted; an attachment. This is ambigoues. --Succu (talk) 22:07, 27 August 2016 (UTC)[reply]
  - Maybe mount (P3091) should be fixed to mounted by or something similar. --Succu (talk) 22:16, 27 August 2016 (UTC)[reply]
    - From the example, I'd say that the fact by a lenses is mounted by some kind of mount is redundant. I think we should stick with has part(s) (P527) as this property does not seem to really add value. author TomT0m / talk page 05:58, 28 August 2016 (UTC)[reply]
      - @TomT0m: I'm not keen on has part(s) (P527), certainly for camera lens mount types, as both the camera body and the lens need to have the same property with the same value but the mount on a lens is the reciprocal of the mount on the camera body, not the identical part (think of it like a plug and socket). Having "mount type (camera body portion)" and "mount type (lens portion)" items would be required under that ontology but that would make querying unnecessarily harder. Thryduulf (talk) 22:41, 9 September 2016 (UTC)[reply]
        @Thryduulf: Interesting. I'm not sure I understand the difference you make between a camera and the telescope however as the mount seems to be a required part of a telescope and is not really changed after the telescope is built ... It seems to be different from a w:Zoom lens for a camera for example, which is an object on his own and can be plugged into some kind of mirrorless interchangeable-lens camera (Q209918)   . I think you got something on the modelling of modular objects for which we don't have any guideline. To be clear : do you think telescope can be divided into "modular telescope" and "non modular telescope" like we can divide cameras beetween "interchangeable-lens camera" and "fixed-lense camera" ?
        
        On the "making querying harder", the comparison is between ?telescope P:mount type ?mount_kind and ?telescope P:P2670 ?mount_kind. {{Query instances|w:telescope_mount}} (actually from the example the rigth property seems to be has part(s) of the class (P2670) and not has part Search) as you link a telescope to a kind of mount and not a telescope to a mount) so I think this is ~~definitely not true~~ not really a problem as this is a pretty common pattern. If you query a specific mount kind you can even spare the "query instance" pattern, at least if it has no subclasses. Actually there is even a keyword "a" in SPARQL property path I'm implementing in Module:PropertyPath - showcase Template:Show Path Items - so it seems to be a non problem to me. But that my opinion ;) author TomT0m / talk page 07:24, 10 September 2016 (UTC)[reply]

Would restricting this to telescopes remove the objections to creating this? If so I'll happily support that and propose a new property for camera mounts. Thryduulf (talk) 14:32, 4 November 2016 (UTC)[reply]

Given that this doesn't seem to be progressing, I'll withdraw the proposal. @TomT0m's alternative approach seems to work OK over at South Pole Telescope (Q1513315) (diff), although it's not ideal (see [1]). Thanks. Mike Peel (talk) 17:26, 18 December 2016 (UTC)[reply]

Marked as withdrawn per above comment. ArthurPSmith (talk) 17:00, 6 January 2017 (UTC)[reply]