Talk:Q102165

Autodescription — nucleon (Q102165)

description: particle that makes up the atomic nucleus (proton or neutron)

Useful links:

Pretty display of data  (with Wikidata:Reasonator)
Pretty display of data (with Wikidata:Sqid)
View it! – Images depicting the item on Commons
Scholia
Report on constraint conformation of “nucleon” claims and statements. Constraints report for items data
Number of statements about nucleon by property (query)
All claims about nucleon (query)
List of items with same label (query)
List of lexemes with sense nucleon (query)

Classification of the class nucleon (Q102165)  

For help about classification, see Wikidata:Classification.

Parent classes (classes of items which contain this one item)

Subclasses (classes which contain special kinds of items of this class)

⟨nucleon⟩ on wikidata tree visualisation (external tool)(depth=1)

Generic queries for classes

Union and disjoint queries

Instances of nucleon (Q102165)   that are instances of two (or more) of the classes: [1]
Instances of nucleon (Q102165)   that are instances of none of the classes neutron (Q2348)   and proton (Q2294)   [2]

See also

This documentation is generated using {{Item documentation}}.

has part

@Succu: Per Help:Basic Membership properties, part of (P361) and has part (P527) are for either physical objects and their parts, or classes of part of physical objects like wheel and car (as classes). So this is not a coorect use of has part (P527), it is not an inverse property of subclass of (P279). TomT0m ^(talk) 18:44, 6 May 2015 (UTC)[reply]

Any references supporting your opinion? --Succu (talk) 18:49, 6 May 2015 (UTC)[reply]

@Succu: I'll just quote Help:BMP, an instance of class X is part of an instance of class Y. Part of can't be both an inverse of has part and of instance of, or they would be kind of equivalent properties ... this would be a mess :) For example see http://www.w3.org/2001/sw/BestPractices/OEP/SimplePartWhole/simple-part-whole-relations-v1.3.html also, and Wikidata:Requests_for_comment/Refining_"part_of" ... TomT0m ^(talk) 18:57, 6 May 2015 (UTC)[reply]

Succu, you are pretty clearly incorrect by definition. A subclass relationship carries inheritance of some or all of the parent class's attributes, in addition to being a "part of" the set associated with the parent class. It is however simply incorrect to say that the parent class has part subclass because of this. --Izno (talk) 19:27, 6 May 2015 (UTC)[reply]

A biological example: taxon has part species is a clearly incorrect statement, while species subclass of taxon is not. --Izno (talk) 19:29, 6 May 2015 (UTC)[reply]

(1) So define the term nucleon with a natural language statement. (2) Depends on the constraints of your model. --Succu (talk) 19:38, 6 May 2015 (UTC)[reply]

@Succu: : Enlight me, is this a sick way to say to me that you don't want that I touch taxonomy items ? TomT0m ^(talk) 19:50, 6 May 2015 (UTC) And ... you answered to Izno[reply]

I replied to Izno. If you want to help? - No objections, TomT0m. --Succu (talk) 20:01, 6 May 2015 (UTC)[reply]

I don't see a need to. You're simply wrong according to how we talk about ontological terms. Bene* below probably does a better job than I would do. --Izno (talk) 13:53, 7 May 2015 (UTC)[reply]

Just from my knowledge about Physics, I'd say that a proton and a neutron are both subclasses of a nucleon. Wikipedia also says: "There are two known kinds of nucleon: the neutron and the proton.". Using the has part (P527) property implies that nuclein actually is an instance which consists of a neutron and a proton. This however would be the atomic nucleus (Q37147), which obviously is a different thing. So I agree that using has part (P527) is not correct in this place. -- Bene* ^t_a^l_k 20:12, 6 May 2015 (UTC)[reply]

Bene*, both parts are a subclass of baryon (Q159731). Both are a part of atomic nucleus (Q37147) and are called together nucleon (Q102165). BTW: You cited enwiki, but you omitted the details. :( --Succu (talk) 20:57, 6 May 2015 (UTC)[reply]

@Succu: They are subsets, complementary subsets of each other in the set of all Nucleons, we agree on that. But part/whole relationship are reserved for relations on physical objects. It relates part-token to whole-token of classes or part-token to classes of whole-token. Nucleons and baryons are not tokens, they are classes. As no neutron-token is a part of a nucleon-token, has part does not fit. This is to clarifies things like that that I created Help:Classification. TomT0m ^(talk) 06:35, 7 May 2015 (UTC)[reply]

Seems we need a new property owl:unionOf to express this correct. --Succu (talk) 09:44, 7 May 2015 (UTC)[reply]

The question is as always: Is there an actual need for such a reverse property or can we handly this using queries? -- Bene* ^t_a^l_k 10:07, 7 May 2015 (UTC)[reply]

@Bene*: That depends. It's not always enough to say that there are subclasses :

If nucleons is exactly the union of neutrons and protons, this is not enough to say they are both subclasses of nucleons. An instance can be an instance of nucleon and none of the subclasses as is.
If a neutron is nether a proton and converse, the subclasses are disjoint and we currently don't express that. An instance can always be an instance of both the subclasses as is.

Those too things together can express that a subclass hierarchy is a real tree : an instance is classified in one and only one leaf of a tree. These properties (alldisjoint(sub)classes and unionof) are not reverse properties. TomT0m ^(talk) 16:46, 7 May 2015 (UTC)[reply]

I understand that point. So do you agree with Succu that we need a new property like "owl:unionOf"? -- Bene* ^t_a^l_k 17:36, 7 May 2015 (UTC)[reply]

Yep, I would support owl:unionOf , and even more a DisjointUnionOf is perhaps even more relevant in practice (this would combine alldisjointwith and unionof). TomT0m ^(talk) 17:56, 7 May 2015 (UTC)[reply]