Wikidata talk:WikiProject Mathematics

From Wikidata
Jump to navigation Jump to search

algebraic structures and the transitivity of the subclass property[edit]

(i hope this is a somewhat appropriate place for this..)
the items for algebraic structure (Q205464)s, for example monoid (Q208237) often have seemingly redundant subclass of (P279) properties. it's like saying penguins are birds but also animals. i think at, least in mathematics, this property should be seen as strictly transitive and redundant properties should be removed accordingly. --opensofias (talk) 09:37, 16 May 2014 (UTC)[reply]

@Opensofias: - Currently there are a lot of mistakes in the subclass of (P279)-tree. A lot of people don't yet have a clear idea how this works. An easy way to view the tree and clean it up is to use this tool:
I already saw some mistakes in the first one - People are using "mathematical conjecture" (conjecture (Q319141)) as a general term, instead of the strict meaning we need to have in a semantic database. -Tobias1984 (talk) 09:58, 16 May 2014 (UTC)[reply]

pre-Hilbert space[edit]

Two Russian wiki-pages translate to pre-Hilbert space: inner product space (Q214159), Hermitian space (Q15629785). Might need some site-link sorting. -Tobias1984 (talk) 23:34, 17 May 2014 (UTC)[reply]

Almost fixed, only the German label in Hermitian space (Q15629785) needs to be changed. Danneks (talk) 14:43, 20 August 2014 (UTC)[reply]
@Danneks: - Thanks. I translated from the English label. -Tobias1984 (talk) 14:56, 20 August 2014 (UTC)[reply]

Simple mathematical operations[edit]

I'm thinking of which properties we need to express addition (Q32043) or subtraction (Q40754). For instance "intrinsic property:associativity (Q177251)", "ruled by:order of operations (Q845118)", "calculation algorithm:method of complements (Q4741052). And then binary operation (Q164307) would need "arity:2". What do you think?--Micru (talk) 12:53, 22 May 2014 (UTC)[reply]

for associativity (Q177251), we should use "Rule_of_replacement". Cuvwb (talk) 15:32, 21 September 2014 (UTC)[reply]
Well, we can say that associative operations are models of a definite formal theory (but I'm not sure whether it is possible to define formal theories and deductive systems in a database, I'm not a computer scientist). Danneks (talk) 20:07, 22 September 2014 (UTC)[reply]
@Danneks: Well, you will be happy to hear the existence of Web Ontology Language (Q826165) family language of description, who are based on a family of logics called description logic (Q387196). Wikidata can not do inferences as of now (it can't even be queried yet ;) ) but it gives an overview that databases and inferences are not an oxymoron. For an application on math expressions, see this discussion on how to model priority of operators on math expressions. TomT0m (talk) 20:52, 23 September 2014 (UTC)[reply]
@TomT0m: I've read something on representing mathematical knowledge in RDF-based languages. Sorry, but I won't help with that, it is such a pain... I can only help with informal descriptions, e.g. that some concept/theorem was introduced by A in NNNN, published in B, and used in some proof by C. Or if some concept was s generalisation of another one. Danneks (talk) 20:04, 24 September 2014 (UTC)[reply]
@Danneks: It's already great ! as you said, this is a complex task, we probably can learn from these on how to represent those basic knowledge in a hopefully consistent way with those more advanced ontologies. But in a project like Wikidata I don't think it's reasonable to try to fully represent maths in full details. TomT0m (talk) 10:25, 25 September 2014 (UTC)[reply]

Launch of WikiProject Wikidata for research[edit]

Hi, this is to let you know that we've launched WikiProject Wikidata for research in order to stimulate a closer interaction between Wikidata and research, both on a technical and a community level. As a first activity, we are drafting a research proposal on the matter (cf. blog post). It would be great if you would see room for interaction! Thanks, --Daniel Mietchen (talk) 01:35, 9 December 2014 (UTC)[reply]

Mathematicals constructions, algorithms, functions[edit]

How to model Mathematical constructions ? Do you think they can be modelled just like algorithms ?

Then, how to model algorithms :) author  TomT0m / talk page 09:42, 19 August 2015 (UTC)[reply]

  • @TomT0m: If they are regarded as definitions of corresponding objects, then maybe the property facet of (P1269) will be enough, e.g. ? Danneks (talk) 05:17, 20 August 2015 (UTC)[reply]
    • @Danneks: : it's a constructive definition, meaning we know more than a mean declarative definition ... Relative to maths fundations, for example, first come set theory build on top of some logic, then comes the natural numbers built from sets, then relative numbers as pairs of natural numbers, ... constructible universe (Q2777107)  View with Reasonator View with SQID represents everything than can be built that way, so I think we can do best than just this :) author  TomT0m / talk page 07:45, 20 August 2015 (UTC)[reply]
      • @TomT0m: I agree that we know more, but probably there are several ways to express this, so I need more motivation :) For example, why do we need a link between natural numbers and integer construction? Maybe all we know can be expressed as Kantian schemas, maybe inductive types, who knows. Danneks (talk) 08:22, 20 August 2015 (UTC)[reply]
        • @Danneks: How and what for maths are useful is a philosophical matter, and in some sense dependant of what you want to achieve or what you are interested in :) So I'd say in Wikidata what's important is what we can express about maths in a reasonable way (that means without too much efforts :) What I know is that fundaments of maths lead to important results in maths such as Godel's theorems, the Turing machine in computer scince, and later important tools such as automated reasoners and automated theorem proving that can actually be used in modern practices. So I would say it's a not so hard to model and important and productive part of maths :) author  TomT0m / talk page 08:56, 20 August 2015 (UTC)[reply]

Wikimania 2016[edit]

Only this week left for comments: Wikidata:Wikimania 2016 (Thank you for translating this message). --Tobias1984 (talk) 11:57, 25 November 2015 (UTC)[reply]

Formulas[edit]

I found "general formula", "chemical formula", but no property for mathematical formulas. Is there one? --- Jura 13:21, 29 November 2015 (UTC)[reply]

What use cases do you have in mind? For example, 2 (Q200) would have a formula 1+1? :) Danneks (talk) 18:13, 29 November 2015 (UTC) There is a property TeX string (P1993), but,as far as I understand, it is not intended to give any mathematical information about an object, only a TeX representation. Danneks (talk) 18:22, 29 November 2015 (UTC)[reply]
Yeah, fairly simple stuff, but with variables, otherwise it wouldn't be fun. Things that can also be found in Wikipedia. --- Jura 06:32, 30 November 2015 (UTC)[reply]
A generalization of my proposal in
   Under discussion
Data typeMISSING
Example 1MISSING
Example 2MISSING
Example 3MISSING
about units and mathematical expressions could be used for that. There was a proposition "tex string" which was proposed in
   Under discussion
Data typeMISSING
Example 1MISSING
Example 2MISSING
Example 3MISSING
but someone opposed because it's not standard Snipre? if I recall well because it's not standard. For the use of my proposal I'll account that the semantics of the formulas in terms of "sum/product/ ..." is more explicat than using a "formula" with say the string property, with no semantics at all except "the usual meanings of the symbol are used. On the contrary with the "sum" and the like properties we can fairly easily generate a tex formula. author  TomT0m / talk page 10:11, 30 November 2015 (UTC)[reply]
Could you try to link your proposal without a template? URL is fine: sample: https://www.wikidata.org/w/index.php?title=Wikidata_talk:WikiProject_Mathematics&oldid=277521950#Formulas --- Jura 18:30, 30 November 2015 (UTC)[reply]
Please see
   Under discussion
Data typeMISSING
Example 1MISSING
Example 2MISSING
Example 3MISSING
, sorry for the disturbance /o\ author  TomT0m / talk page 07:28, 1 December 2015 (UTC)[reply]
Ok. Thanks. In the meantime, I formulated another approach below. It seems to be more generic, but slightly less formatted than the 3rd proposal. --- Jura 07:45, 1 December 2015 (UTC)[reply]
@TomT0m: maybe we could have instead something like <metre per second (Q182429): mathematical expression: fraction> with qualifiers "numerator" and "denominator" (like \frac{numerator}{denominator} in LaTeX) and so on for sums and powers? For one thing, the word "of" can not be translated precisely into Russian, we use declension of nouns for that. Danneks (talk) 17:49, 1 December 2015 (UTC)[reply]
@Danneks: I would not use "fraction" as it's a division, not a fractional number. For "of", could you get another formulation such as "composed of" or "made of" or similar ? We already have such a qualifier anyway. author  TomT0m / talk page 18:15, 1 December 2015 (UTC)[reply]
Amended my proposal: adding sum and division. author  TomT0m / talk page 18:24, 1 December 2015 (UTC)[reply]
Well, the mayor of London is not composed of London... and the description doesn't say that this property may express something like "composed of". It has a relatively good [ru] label, but looks strange in this case. Danneks (talk) 18:51, 1 December 2015 (UTC)[reply]
But a sum is the sum of its parts. author  TomT0m / talk page 19:53, 1 December 2015 (UTC)[reply]
I just always prefer to use specific properties. But if you are proposing a "division"/"is a quotient of" property (even thougn everything can be expressed without it), you need a way to distinguish between dividends and divisors. Danneks (talk) 21:06, 1 December 2015 (UTC)[reply]

Formulas II[edit]

Somehow the previous section disappeared in a transclusion. Here a short sample:

Rather than including the formula in one of the items, it's probably preferable to place it on a new item: Q21600451.

The property "mathematical formula" could be in ASCII if possible (for +-*/) or with <math> tag for anything else. --- Jura 06:38, 1 December 2015 (UTC)[reply]

How to include elements mentioned in the previous section?
@Jura1: Sorry I don't understand. My proposal and extensions of it seems more precise. author  TomT0m / talk page 18:17, 1 December 2015 (UTC)[reply]
I'm not sure how the approach would work with more complex formulas. Why not leave it to TeX as on Q21600451 (or ascii for simple cases).
P527 would enumerate what goes into the formula. --- Jura 21:18, 1 December 2015 (UTC)[reply]
@Jura1: My proposal is way better. Yours is structured but not really : "part of" does not explains the relations beetween the parts of the formula. My proposal to express formula just needs a few extra items to describe the parts with other primitives as needed AND explicits the relation beetween the parts : we can explain the formulas semantics, and not only weakly its structure (BTW we have a proposal in
   Under discussion
Data typeMISSING
Example 1MISSING
Example 2MISSING
Example 3MISSING
 since ages to link quantities items with formulas they occurs in. see https://www.wikidata.org/wiki/Wikidata:Property_proposal/Natural_science#formule_d.C3.A9crivant ) author  TomT0m / talk page 10:44, 2 December 2015 (UTC)[reply]
Probably, but I'm not convinced if that model can handle more complex operations. You might end up creating additional properties for every combination of operations.
Instead, TeX (or another new property) can give the relation. As formulas can be used in several ways, the explicit relation isn't necessarily needed in qualifiers. We could have a distinct property to list the operations. --- Jura 10:59, 2 December 2015 (UTC)[reply]
Of course it can, it's enough to add new properties as needed for other operation. Right now it can express rational functions (with a way to express variables, I think that part of your proposal is in the right path for this), polynomias, polynomias of any mathematical function in any domain number. A new propery would be needed for the usual Σ, ∏ . We have to think on derivatives. In any way, a tex string would be needed to explain the semantics of the symbols and function, while with propoerties we can define the semantics of the properties in their talk pages and render them as wanted. author  TomT0m / talk page 11:34, 2 December 2015 (UTC)[reply]
I suppose all approaches could be used on the same item, but I think it would work better if used on distinct items for the formulas.
Besides, for the sake of clarity, I'm not sure if units conversion should be combined into it. We already have a distinct property for that. --- Jura 17:14, 2 December 2015 (UTC)[reply]

I made a corresponding request at Wikidata:Property_proposal/Generic#Formule_math.C3.A9matique_.28fr.29. I also came across https://phabricator.wikimedia.org/T67397 (it doesn't seem to be worked on). --- Jura 11:54, 15 December 2015 (UTC)[reply]

The suggestion there is to use TeX instead. If its formatting is kept simple, I tend to agree with this. --- Jura 10:15, 29 December 2015 (UTC)[reply]

Some alerts[edit]

Hello, I cannot speak in english, so I cannot help you but I have some alerts HB (talk) 18:44, 25 February 2016 (UTC)[reply]

is really a rational number a set?[edit]

see rational number (Q1244890), irrational number (Q607728) , real number (Q12916), non-negative real number (Q18729403)...

Salut HB, alors l'idée des classes/instances dans Wikidata, c'est que la propriété instance of (P31) correspond à une relation d'appartenance. Donc techniquement on peut dire
⟨ 1 ⟩ instance of (P31) View with SQID ⟨ nombre naturel ⟩
, ce qui fait mécaniquement de nombre naturel une classe mathématique (class (Q217594)  View with Reasonator View with SQID), dont chaque nombre naturel est une instance. Donc oui, c'est assez naturel (haha) de considérer l'élément nombre naturel comme l'ensemble des nombres naturels, par construction. Sachant que instance de est analogue à l'appartenance, et que sous-classe de est analogue à la relation de sous-ensemble, axiomatiquement dans des langages comme OWL2 ou RDF, ça colle très bien.

Encore une fois, visiblement, je ne comprends rien à wikidata: si je lis la description de instance of (P31), je lis «Cet élément est un exemple spécifique de cette classe qui en précise la nature» ou «this item is a specific example and a member of that class». Si j'applique cette description à l'entier naturel cela donne « un entier naturel est un exemple spécifique de la classe des ensembles» ou « a natural number is a specific example of set» Ce qui est mathématiquement très faux aussi bien en français qu'en anglais. HB (talk) 09:49, 26 February 2016 (UTC)[reply]

Ce qui serait correct serait de créer un élément «Ensemble des nombres rationnels» (set of rational numbers) dont instance of (P31) serait bien set (Q36161) et divisible group (Q1782332) mais aussi field (Q190109) encore que...divisible group (Q1782332) c'est un ensemble avec une loi interne et field (Q190109), un ensemble avec deux lois internes. L'élément rational number (Q1244890) aurait pour instance of (P31) «Ensemble des nombres rationnels» (set of rational numbers). ( faux) HB (talk) 10:09, 26 February 2016 (UTC)[reply]

@HB: OK, alors prend la définition mathématique d'une classe : une classe est une collection d'élément dont les membres sont définis par une propriété logique. Un nombre naturel se définit très bien avec une propriété logique, dans toute théorie des ensembles avec une axiomatique qui correspond à l'axiomatique de Peano. Mathématiquement donc, la classe des nombres naturels se définit comme la classe des nombre qu'on peut construire par cette axiomatique (voir fr:construction des nombres naturels). Si tu regardes les fr:logique de description, les concepts sont très similaires à des classes mathématiques, dans le sens ou ils correspondent très précisément à des ensemble d'individus, et on peut les définir par des propriétés logique dans le langage de ces logiques. Ces logiques servent de base à la sémantique de langage comme fr:RDFS (ou les concepts sont explicitement nommés classe). Du coup, quand on donne une définition à un concept, ici celui d'entier naturel, on obtient dans ces logique assez naturellement l'ensemble des individus qui correspondent à cette définition, donc ici l'ensemble des entiers naturels. Grâce au punning classe/individu, une fonctionnalité d'OWL pour considérer les concepts parfois comme des individus (ce que les logique de description de base interdisent, l'ensemble des individu doit être disjoint de l'ensemble des concepts), il est fort commode de considérer la définition des entiers naturels comme l'ensemble des entiers naturels. C'est long, écrit vite fait et pas forcément clair, n'hésite pas à demander des détails. author  TomT0m / talk page 10:20, 26 February 2016 (UTC)[reply]

Butée je suis et je reste. En sémantique ou dans une base de donnée comme wikidata, tu peux, peut-être, je n'en sais rien, associer classe et individu et les considérer comme équivalent mais pas en math. J'en conclus fort logiquement que la logique de wikidata m'échappe et n'est pas mathématiquement compatible. Si tous les contributeurs de WikiProject Mathematics sont d'accord avec cette vision, je ne peux pas adhérer à ce projet. Je compte arrêter là la discussion. HB (talk) 10:33, 26 February 2016 (UTC) PS : je t'avoue qu'à partir de «Grâce au punning classe/individu ...» j'ai cessé de considérer que tu pouvais t'adresser à moi.[reply]

Attend un peu avant de te braquer (c'est pas toujours facile la discussion). Autre angle d'attaque: quel sens tu donne à "entier naturel nature de l'élément ensemble des entiers naturels" ? Ça n'a pas vraiment de sens. Soit c'est un nombre précis qui est membre de l'ensemble des entiers naturels, 1∈N ; 2∈N mais "entier naturel"∈N n'a aucun sens. J'insiste pour dire que la propriété "instance de" (nature de l'élément est sûrement mal choisi) est très similaire à l'appartenance à un ensemble. N∈"la classe des ensemble" a contrario a tout son sens. author  TomT0m / talk page 10:53, 26 February 2016 (UTC)[reply]

Sur ce point je suis d'accord et je revenais justement modifier ma proposition: L'élément rational number (Q1244890) aurait pour instance of (P31) number (Q11563). je cherchais vainement une propriété liant un individu à sa classe mais cela ne semble pas exister. Toujours mathématiquement, dire qu'un entier naturel est un élément de l'ensemble des entiers naturels est un tautologie mais n'a rien d'absurde, mais cela ne correspond pas à P31 HB (talk) 11:09, 26 February 2016 (UTC)[reply]

Heureusement que ça n'a rien d'absurde :) On est dans l'ordre de la définition avec P31 (OWL permet la définition de classe en extension, donc en donnant la définition sous forme de prédicat logique pour donner l'ensemble des membres. Donc effectivement dire que l'ensemble des entiers naturels est l'ensemble des nombres qu'on peut construire est tautologique, comme tout théorème). Par contre j'aurai tendance à dire que c'est à toi de montrer que ça ne correspond pas à P31. Tu n'as pas produit d'éléments convancant en ce sens jusqu'à présent, et surtout rien qui ne soit pas dans l'affirmation gratuite et qui réponde ou questionne en quoi que ce soit les éléments que j'ai fourni. author  TomT0m / talk page 12:43, 26 February 2016 (UTC)[reply]

(Je n'ai rien compris à ta parenthèse)

Pour en revenir à P31 : la définition est claire. Quand on dit [un entier naturel] est un exemple spécifique de la classe des [nombre]s. Tu relis deux éléments (entier naturel et nombre) par une inclusion de classes (la classe des entiers naturels est incluse dans la classe des nombres). Quand tu définis une relation d'appartenance, tu relis deux éléments de nature différentes [un individu] appartient à [un ensemble]. On peut écrire [un entier naturel] appartient à [la classe des nombres] mais tu vois bien qu'à droite tu n'as pas un individu (ou instance) mais une classe (ou ensemble).

Je sais, toi tu trouves légitime d'écrire [la classe des entiers naturels] appartient à [la classe des ensembles] (opérant ainsi une confusion dommageable entre élément et classe) mais ton interprétation ne résiste pas à l'utilisation classique de P31. Pour un exemple ailleurs qu'en math : je lis que France (Q142) a pour instance of (P31) country (Q6256) et country (Q6256) a pour instance of (P31) administrative territorial entity type (Q15617994)

  • Avec mon interprétation de P31 ça colle les deux fois : [la France] est un exemple spécifique de la classe des [Pays] et [Un pays]est un exemple spécifique de la classe des [type de division administrative]
  • Avec la tienne ça bug assez vite : pour voir une notion d'appartenance il faut que tu vois dans l'élément pays non pas un élément mais une classe. [la France] appartient à [la classe des pays] mais alors ça bug dans la seconde affirmation [la classe des pays] appartient à [la classe des types d'administration administrative]. Tu vois bien que là il faudrait une inclusion.

Quant on mélange individu et ensemble on court ce genre de risque. HB (talk) 13:42, 26 February 2016 (UTC)[reply]

Il n'y a pas de confusion entre élément et classe, un ensemble peut très bien contenir un autre ensemble, comme un ensemble des parties. Une partie d'un ensemble est un élément de l'ensemble des parties. Pour le reste, je suis moteur dans la distinction "division"/"type de division" sur Wikidata, et j'ai fait ça sur les bases de ce que j'énonce ici, donc je trouve assez cocasse qu'on vienne me sortir cet exemple comme motivant que ce j'avance ne tient pas la route. author  TomT0m / talk page 13:50, 26 February 2016 (UTC)[reply]
J'ajoute que pour éviter tout soucis de fondation, j'ai proposé de fonder les ensembles sur le principe de la "distinction type/jeton" ( en:Type–token_distinction en anglais mais c'est définitivement un articles à écrire/traduire en français ) avec les objets de bases qui sont les "jetons", les classes des ensembles de jetons, des classes de niveau 2 qui sont des classes de classes, etc. Avec "instance de" qui correspond à l'appartenance, et sous-classe de qui veut dire "sous-ensemble" (c'est standard). Pour les maths, la distinction marche pas très bien vu que les nombres et autres objets mathématiques sont des objets abstraits, mais je pense qu'on peut arbitrairement décider que les nombres sont des jetons, par exemple, ça n'a pas trop d'importance pratique. author  TomT0m / talk page 14:23, 26 February 2016 (UTC)[reply]

(pas compris grand chose à ton ajout). Je ne répons donc qu'à la première partie;

Et tu es suivi sur les autres pages où tu défends ton acception de P31?... Visiblement pas sur l'exemple que je donne ni sur le suivant : je lis que Oscar II of Sweden (Q52924) a pour instance of (P31) monarch (Q116) et monarch (Q116) a pour instance of (P31) position (Q4164871)

  • Avec mon interprétation de P31 ça colle les deux fois : [Oscar II de Suède] est un exemple spécifique de la classe des [monarque ] et [Un monarque ]est un exemple spécifique de la classe des [positions politiques].
  • Avec la tienne ça bug toujours [Oscar II de Suède] appartient à [la classe des monarques] et [la classe des monarque] appartient à [la classe des positions politiques]. Si je suis d'accord pour dire qu'un ensemble peut bien appartenir à un autre ensemble (n'oublie pas que tu parles à une matheuse) , je persiste à dire que dans les exemples que je donne, il s'agit d'une erreur de logique.

J'arrête là les frais. Si tu essaies de défendre ton idée sur d'autres pages, à moi seule je n'arriverai pas à te convaincre. Et si tu défends l'idée qu'il est bon de confondre l'individu (monarque) et la classe (classe des monarques), j'aurais toujours personnellement beaucoup de mal à te suivre. Je viendrai lire de temps en temps ici la réaction d'autres matheux. HB (talk) 14:36, 26 February 2016 (UTC)[reply]

Je peux alors t'aiguiller sur https://www.w3.org/2007/OWL/wiki/Punning . C'est le principe du "punning" et c'est pas moi qui l'ait inventé, c'est parce que c'est quelque chose d'assez pratique. Par ailleurs ça ne me pose pas de problème particulier d'assimiler la notion de "monarque" à l'ensemble des personnes qui l'ont exemplifiées. C'est exactement comme une défition : "un monarque est une personne qui blablabla" ... qui correspond assez naturellement à l'ensemble des personnes qui correspondent à cette définition. À tel point que souvent on définit la relation entre les notions de "monarque français", qui sont des exemples particuliers de monarques, comme une "sous-notion" de la notion de "monarque" si et seulement si l'ensemble des monarques français est un sous-ensemble de l'ensemble des monarques. Et c'est très pratique pour la modélisation qu'on trouve des occurences de ce schéma jeton/type/ensemble de type vraiment partout (quelques exemples sur Help:Classification). author  TomT0m / talk page 14:46, 26 February 2016 (UTC)[reply]
Donc pour résumer mon point de vue, le "punning" (un calembourg en anglais) est un outil qui nous permet rigoureusement, pour un élément donné, de considérer un élément wikidata quand ça nous arrange soit comme une définition pour une notion, soit comme l'extension de cette définition (au sens d'une définition par extension d'un ensemble). On note
⟨ exemple ⟩ instance of (P31) View with SQID ⟨ classe ⟩
si exemple est membre de l'extension. Si au contraire on préfère considérer la définition elle même, par exemple dire que "perroquet" est un taxon, on note
⟨ classe ⟩ instance of (P31) View with SQID ⟨ classe de classe ⟩
. Du coup la classe de classe peut être vue comme un type de définition particulière, ou comme un ensemble d'ensemble au choix, sans problème de rigueur et de manière assez pratique du point de vue de l'utilisation. author  TomT0m / talk page 15:03, 26 February 2016 (UTC)[reply]

can a natural number be nul ?[edit]

see natural number (Q21199) and the talk page : in french the response is «yes», in english, the response is «no». So we are not talking about the same concept but the interwiki are necessary.

Alors il y a un moyen simple de traîter ces cas d'interwikis compliqués, c'est d'utiliser les interwikis vieille école. Il y a un projet pour recencer tous ces cas : WD:XLINK, tu peux y laisser un message si tu veux. author  TomT0m / talk page 21:54, 25 February 2016 (UTC)[reply]
Non. Je signale, c'est tout. Je ne me lance pas dans des discussion en anglais avec des termes techniques qui visiblement me sont étrangers. HB (talk) 09:52, 26 February 2016 (UTC)[reply]
Je te demande juste si tu veux reporter ton cas d'interwiki compliqué sur cette page qui les compile. Tu peux le faire en français, il n'est pas impossible qu'elle soit déjà traduite en français d'ailleurs. author  TomT0m / talk page 10:24, 26 February 2016 (UTC)[reply]

can you look at Q140645 ?[edit]

I thing, there is a lot of mess. Thanks. Done by ‎Oliv0 HB (talk) 20:40, 25 February 2016 (UTC)[reply]

Disambiguation of some natural phenomenon and the theory describing it[edit]

Take hypercycle (Q4138754) - currently an instance of both a (mathematical) theory and an autocatalytic set (of molecules). I think this would have to be split into two items (e.g. as for evolution (Q1063) vs theory of evolution (Q17128217)), but then, it's not always obvious which Wikipedia articles to link to, since they typically cover both the mathematical and the non-mathematical aspects. As this problem affects a good number of items, I'd like to invite thoughts on how to address this in a more systematic fashion. --Daniel Mietchen (talk) 06:33, 3 March 2016 (UTC)[reply]

Mathematics Subject Classification[edit]

Opensofias
Tobias1984
Arthur Rubin
Cuvwb
TomT0m
Physikerwelt
Lymantria
Bigbossfarin
Infovarius
Helder
PhilMINT
Malore
Lore.mazza51
Wikisaurus
The Anome
The-erinaceous-one
Daniel Mietchen
Haansn08
Xenmorpha
John Samuel
Jeremy Dover
Toni 001
Bocardodarapti
Duckmather
HTinC23
fgnievinski

Notified participants of WikiProject Mathematics

Would it make sense to add a MSC property? I started to fill the proposal template [Wikidata:Property_proposal/MSC].--Physikerwelt (talk) 12:21, 26 July 2016 (UTC)[reply]

@Physikerwelt: I'd oppose. instance of (P31) is enough with the MSC class as value, and each MSC class tagged as
⟨ an MSC class ⟩ instance of (P31) View with SQID ⟨ MSC class ⟩
. Using the generic property should be a good practice, it's what many ontologies to. author  TomT0m / talk page 16:38, 26 July 2016 (UTC)[reply]

Consistent naming convention[edit]

A student of mine is currently developing a prototype for math aware question answering system.

I was testing his system with the question 

What is the are of a circle?

The system was not able to anser. Therafter I changed the property area of circle in https://www.wikidata.org/w/index.php?title=Q17278&oldid=417738145 and the system worked as expected.

My assumption is that a circle is

And that the area enclosed by that circle i.e.
is the surface integral, of a parametrisation of this set. For example

Now, I'm wondering why my edits have been reverted over and again. --Physikerwelt (talk) 10:43, 22 December 2016 (UTC)[reply]

@Physikerwelt: Do you understand the difference between circle (Q17278) and disk (Q238231)? There is even an item area of a disk (Q4115331). If you insist that "circle" should have non-zero area, then you should remove English sitelink from the first item and create some special (but strange) item for it. --Infovarius (talk) 21:06, 23 December 2016 (UTC)[reply]
If you want to be able to answer any ambiguous question that can be spoken then you should be able to convert them into disambiguous one. For example, I've added a synonim to the entity which has non-zero area. So you should be able to find the very "circle" you want and distinguish from other use cases of this word. --Infovarius (talk) 21:08, 23 December 2016 (UTC)[reply]
"What is the area of a circle?" area (Q11500) ("quantity that expresses the extent of a two-dimensional surface or shape, or planar lamina, in the plane") - 2D, circle (Q17278) ("simple curve of Euclidean geometry") - 1D. --Fractaler (talk) 08:47, 26 December 2016 (UTC)[reply]
@Infovarius: @Fractaler: w:Area of a circle describes it quite well
Although often referred to as the area of a circle in informal contexts, strictly speaking the term disk refers to the interior of the circle, while  circle is reserved for the boundary only, which is a curve and covers no area itself. Therefore, the area of a disk is the more precise phrase for the area enclosed by a circle.
Even though it's perfectly clear to me that I think it's worthwhile to determine a strategy how to match the common sense and the mathematically precise definition and to make wikidata usable for a broad audience. Especially slightly different meanings in different languages might cause extra confusion. For instance the German word 'Kreisfläche' which consists of 'Kreis'=circle (Q17278) and 'fläche'=area (Q11500) links to disk (Q238231).
One possible solution which would maintain the intuitive meaning but also keep the mathematical correctness, would be to broaden the area definition and allowing also for Jordan curves, which seems to be common and is used for instance in the definition of Osgood curve (Q7106547).--Physikerwelt (talk) 17:33, 29 December 2016 (UTC)[reply]
Good example with Osgood curve (Q7106547)! There are curves which have area themselves (and can have area of their interior), so how would you interprete "area of" in that case? --Infovarius (talk) 12:32, 23 June 2017 (UTC)[reply]

Modelling proofs and properties of some objects[edit]

I'm atm trying to add statements for object like algorithms, classes of algorithms like randomized algorithms, properties of some algorithms (termination for example.) And also proof-classes or methods like entropy compression (Q22908168)  View with Reasonator View with SQID. And also fields of study of those objects like Termination analysis (Q7702793)  View with Reasonator View with SQID. And also about proofs like termination proof (Q28122552)  View with Reasonator View with SQID

Quite a lot of objects type actually. And an interesting use case because there is a well known morphism beetween proofs and algorithm : Curry–Howard correspondence (Q975734)  View with Reasonator View with SQID.

I have a few problems to submit to community : how to link properties item to the class of mathematical object it defines ? Do we need different items for example for both the class of agorithms that terminate and the "termination" property ? I'd tend to say "yes". This would lead to statements like

  • ⟨ terminating algorithm ⟩ has quality Search ⟨ termination ⟩

This would reflect the set-theoretic notion of "intensionally defined set" which link the property itself to the set of objects it allows to define (terminating algorithm).

The disjoint union of (P2738) View with SQID property allows to say that an algorithm either always terminates or not :

The studies Search and study of Search pair of properties allows to link academical fields to the objects of study. I think we lack properties like "demonstrate" and stuffs to model morphism. Ideas/comments ?

User:TomT0m, that sounds interesting! I've been thinking about some related things, but not sure how to represent them in Wikidata. I'm interested to follow the discussion and might have some ideas to add... Arided (talk) 18:52, 23 January 2017 (UTC)[reply]

Proof property[edit]

I'm new to WikiData and I don't understand why there isn't a "(mathematical) proof" property. Is it because a theorem can have multiple proofs? Thank you --Malore (talk) 01:52, 24 October 2017 (UTC)[reply]

There is computes solution to (P2159). --Infovarius (talk) 15:24, 26 October 2017 (UTC)[reply]
Yes, but what I'm looking for is a property that allows to specify all the steps that leads to the demonstration of the theorem. --Malore (talk) 18:18, 6 November 2017 (UTC)[reply]
Wikidata is still a toy, has a lot of lacks. You can help out of the toy to make a non-toy. --Fractaler (talk) 17:32, 12 November 2017 (UTC)[reply]

model elements described by the mathematical markup language mathml[edit]

I was trying to model the MathML symbol for factorial (!). But I'm not sure if there is a more straightforward way to express that https://www.wikidata.org/w/index.php?title=Q120976&type=revision&diff=614980278&oldid=603506886 Any suggestions? --Physikerwelt (talk) 11:53, 29 December 2017 (UTC)[reply]

Graphs[edit]

I read recently about the Rado graph (Q7281501), so I watched the item out of curiosity. It seems we don’t really have a good model and properties for graph … to denote the edge set of the graph, the property used is based on (P144) View with SQID … Alsu an interesting example would be, how to model that the Rado graph is univesal, that is any graph is isomorphic to a subgraph of it ? author  TomT0m / talk page 10:17, 14 November 2018 (UTC)[reply]

Base elements[edit]

I came across two items that look very similar to me, but not sure if they are realy the same. It's about basis element (topology) (Q44924150) and base (Q810214). Greetings, Q.Zanden questions? 00:30, 4 May 2019 (UTC)[reply]

A possible Science/STEM User Group[edit]

There's a discussion about a possible User Group for STEM over at Meta:Talk:STEM_Wiki_User_Group. The idea would be to help coordinate, collaborate and network cross-subject, cross-wiki and cross-language to share experience and resources that may be valuable to the relevant wikiprojects. Current discussion includes preferred scope and structure. T.Shafee(evo&evo) (talk) 02:36, 26 May 2019 (UTC)[reply]

"Categorization" of mathematical concept[edit]

Please see Wikidata:Project_chat#Mathematical_terms. Wikisaurus (talk) 17:09, 6 December 2019 (UTC)[reply]

What is an invariant[edit]

Hi! I wonder how should one consider invariants, like root multiplicity (Q77886165), degree of a polynomial (Q1473607), degree of a continuous mapping (Q306564), group order (Q18408315), order of a group element (Q54555759), cardinality (Q4049983), dimension of a vector space (Q929302), determinant (Q178546) (of a matrix)? Are they instances or subclasses of invariant (Q188211)? Are they subclasses of natural number (Q21199) (integer (Q12503), real number (Q12916), etc)? Wikisaurus (talk) 15:32, 9 December 2019 (UTC)[reply]

Firstly, one may tell about invariants only when transformations are defined. If the group of transformations is trivial, then everything is an invariant. Secondly, cardinality and dimension are not natural if they are infinite. Moreover, is 0 a natural number? Determinant of a matrix has nothing to do with natural numbers, at all. Incnis Mrsi (talk) 17:57, 15 December 2019 (UTC)[reply]

Importing comprehensive knot property data from knotinfo[edit]

https://knotinfo.math.indiana.edu/ is a fabulous knot theory resource containing comprehensive data about almost 3000 knots. At the moment, we only seem to have a very few knot properties -- Alexander polynomial (P5350), Conway polynomial (P5351), Jones polynomial (P5352), and Alexander–Briggs notation (P6432) -- and remarkably few articles on specific mathematical knots. defining formula (P2534) is also used for the braid word of the knot.

What do other editors here think about the practicality of importing the complete content of Knotinfo into Wikidata? This would add perhaps 90 more properties to Wikidata, and almost 3000 new items. This seems like an excellent candidate for Wikidata: knots have so many different naming schemes that assigning Wikidata items to the simplest couple of thousand knots sounds like a great way to unify the naming schemes (pace https://xkcd.com/927/ , of course).

I think mathematical knots are a sufficiently fundamental kind of object that they surely all pass the notability criteria: I think the nearest equivalent to this is the several hundred items listing polyhedra, or the listings of thousands of individual chemical compounds. At the same time, this is not an unbounded set: knots with crossing number up to 12 are about as far as most mathematicians want to enumerate the knots individually.

Importing the data is trivial -- everything is downloadable from the site as a single spreadsheet in a simple format, and I can munge the data and use a bot to add it in a few hours -- but the preliminaries of adding several dozen new properties seems quite onerous, and we would also need to be sure that we are on firm grounds as regards copyright and crediting. (We would, of course, be acknowledging knotinfo as a reference for every single property of every item.)

I can also take care to merge in the entries for knots Wikidata has with the items that already exist, so there wouldn't be any duplication. As a final bonus, we could also link new items to MathWorld ID (P2812) for those knots that have them.

See User:The Anome/knot properties for a very provisional list of the properties, as yet neither cleaned up nor checked for consistency with existing properties.

Any thoughts? -- The Anome (talk) 23:48, 9 June 2020 (UTC)[reply]

Extension of Alexander-Briggs notation property to composite knots, to match common usage[edit]

At the moment, Alexander–Briggs notation (P6432) has only been used to define prime knots and links. However, the notation has been extended in practice to cover composite knots using their knot sums (see https://onlinelibrary.wiley.com/doi/10.1002/anie.201702531 , http://people.maths.ox.ac.uk/lackenby/csk24089.pdf and https://mathworld.wolfram.com/KnotSum.html for details). For example, granny knot (Q5596009) has an A-B notation of , and square knot (Q7582069) has notation . I suggest we extend the regular expression and class constraints to take this into account. -- The Anome (talk) 18:12, 19 June 2020 (UTC)[reply]

Group of related knot-related property proposals[edit]

Opensofias
Tobias1984
Arthur Rubin
Cuvwb
TomT0m
Physikerwelt
Lymantria
Bigbossfarin
Infovarius
Helder
PhilMINT
Malore
Lore.mazza51
Wikisaurus
The Anome
The-erinaceous-one
Daniel Mietchen
Haansn08
Xenmorpha
John Samuel
Jeremy Dover
Toni 001
Bocardodarapti
Duckmather
HTinC23
fgnievinski

Notified participants of WikiProject Mathematics

I've added a bunch of knot- and link-related proposals: see

The intent of these proposals is not to supplant other sources of information about knots, but to bring these different schemes together in Wikidata to act as a cross-namespace hub and resource discovery. The naming of knots has been a vexed issue for some time, both because of historical errors in numbers, and also because multiple competing naming and notation systems have evolved to try to resolve this. None of the online sources, with the exception of KnotInfo, have taken a fully systematic approach to naming, and even KnotInfo's approach does not address all use cases. More details can be found in the proposals themseleves. If approved, I should be able to fill in a lot of these preperties using my bot. Hopefully, doing things this way can avoid the https://xkcd.com/927/ problem. -- The Anome (talk) 14:52, 25 June 2020 (UTC)[reply]

Property proposal: Dowker-Thistlethwaite name[edit]

Opensofias
Tobias1984
Arthur Rubin
Cuvwb
TomT0m
Physikerwelt
Lymantria
Bigbossfarin
Infovarius
Helder
PhilMINT
Malore
Lore.mazza51
Wikisaurus
The Anome
The-erinaceous-one
Daniel Mietchen
Haansn08
Xenmorpha
John Samuel
Jeremy Dover
Toni 001
Bocardodarapti
Duckmather
HTinC23
fgnievinski

Notified participants of WikiProject Mathematics

Hello! I'm still trying to get approval for my proposal at Wikidata:Property proposal/Dowker-Thistlethwaite name. If granted, this would be an extremely useful property for items about w:prime knots, as it would give each an unambiguous identifier, something which is not possible with any other knot naming system. I've also got a bot run ready to add this identifier to some 2900+ items. This would also provide a unique key for database comparisons for these objects, and also a key into an authoritative external database.

I've put a lot of work into getting all this ready -- data complitation and cleaning, manual merging, and bot writing -- and it would be a shame for the project to stall half way through.

At the risk of being a bore, I'd very much like to get this property through the approvals process, so I can finish the project. Would anyone like to take another look at it? -- The Anome (talk) 13:04, 7 July 2020 (UTC)[reply]

Now done. Many thanks to the editors who helped get this created! I'll get right on with adding the data. -- The Anome (talk) 22:23, 7 July 2020 (UTC)[reply]

Classification of Numerical Methods fo Differential Equations[edit]

I have been working on improving the classification of numerical methods for solving ODEs. For instance, I've recently created backward Euler method, Adams–Moulton methods, and Adams–Bashforth methods.

I'm having trouble, however, creating statements that represent the convergence rate and the step count using existing properties.

  • For the rate of convergence, I think we need two new properties in order to encode the information: <order of convergence> and <with respect to>. This would allow us to take a sentence such as "the Euler method has a rate of convergence of local truncation error with respect to step size of O(h^2)" and translate into "<order of convergence> <of> <local truncation error> <with respect to> <step size> <is instance of> <quadratic>." The <order of convergence> and <with respect to> properties would be useful in other contexts, as well, such as when encoding the convergence rate of optimization algorithms.
  • step count should be simpler, and might even be possible with existing properties, but I haven't found a good way to do it, so I think it would be useful to add a "step count" property. Then, we could write that a method is <step count> <2>.

Does anybody have any suggestions for alternative methods or comments in support of/opposed to adding the suggested properties?

The-erinaceous-one (talk) 05:26, 25 July 2020 (UTC)[reply]

@The-erinaceous-one: "with respect to" is of (P642). --Infovarius (talk) 21:04, 25 August 2020 (UTC)[reply]
@Infovarius: Ah, thank you. This post is a bit out of date, I figured out that a decent way to model order of convergence like this: Euler method (Q868454)has characteristic (P1552)order of convergence (Q97940482)of (P642)global error (Q98343167)relative to (P2210)step size (Q98398288) Then to enter the value, another qualifier numeric value (P1181)"1". It seems like it works prett well ¯\_(ツ)_/¯ The-erinaceous-one (talk) 21:25, 25 August 2020 (UTC)[reply]

Userbox: Mathematics[edit]

We now have a userbox for WikiProject Mathematics: you can insert {{User mathematics}} on your user page or include it in your babel like this {{#babel:en|mathematics}}. Please feel free to choose a different icon! I didn't know what to pick, so I went with Euler's identity, but it would be better to have something that is square.

This user is a member of
WikiProject Mathematics

The-erinaceous-one (talk) 09:35, 6 September 2020 (UTC)[reply]


Finite projective planes[edit]

I have a database with just over 10,000 finite projective planes that I have compiled over several years, and I would like to move it into a more durable and usable format. It seems to me that this might be the right repository, but would like the assessment of more experienced users, and am definitely open to ideas on improving the ontology. For each plane in this set, we may have the following data elements:

  • Order: integer between 2 and 128, but could be larger
  • Lenz-Barlotti class: a brief (6 char) string identifier
  • Fingerprint: a longer string invariant of the plane, can be thousnads of characters, but likely not tens of thousands
  • Lines: a list of the lines of the plane in ASCII format, can be up to about 12MB. Enables creation of the plane.
  • Automorphism group: definitely need to record the order of the automorphism group, but would like to include generators as well (less than 1MB, typically)
  • Linear codes associated with plane: a set of two integers with a characteristic and a rank, all known planes only have 1 element
  • Construction Method: a plane may be constructed from a variety of known techniques. Need to include the technique name (which would be its own data element), and any data used by that technique to construct this specfic plane
  • Transform Method: a plane may be transformed into another plane from a variety of known techniques. Need to include technique name and (perhaps) data.
  • Description Method: "sporadic" ways in which this plane has been found...may be a computer search, or a one-off academic paper.
  • Substructures: some substructures such as ovals/hyperovals and configurations have are interesting solely as point-sets in a plane...they will have a stabilizer group and may have some other properties. Other substructures (e.g. unitals) have an existence outside the plane, but may or may not be interesting in that context.

I should note that my interest here is not in creating as many planes as possible (others are capably doing that), but rather to codify what is known about projective planes of relatively small order in a public, searchable fashion.  – The preceding unsigned comment was added by Jeremydover (talk • contribs) at April 25, 2021‎ (UTC).

Jeremydover Importing some (if not all) of the project planes you have in your database sounds like a meaningful addition to Wikidata, but I don't think we can include any large associated data (more than a few hundred kilobytes). Which data elements would require new Wikidata properties and which can be modeled using existing properties? — The Erinaceous One 🦔 15:51, 17 May 2021 (UTC)[reply]

about Coxeter–Dynkin diagram[edit]

This wikiProject has a goal to gather information from infoboxes. For Example, infobox in article en:Cube has a field "Coxeter diagram". Should we create a property for en:Coxeter–Dynkin_diagram? see Wikidata:Project_chat/Archive/2021/10#datatype_for_en:Coxeter–Dynkin_diagram? and en:Wikipedia_talk:WikiProject_Mathematics#about_Coxeter–Dynkin_diagram.

Opensofias
Tobias1984
Arthur Rubin
Cuvwb
TomT0m
Physikerwelt
Lymantria
Bigbossfarin
Infovarius
Helder
PhilMINT
Malore
Lore.mazza51
Wikisaurus
The Anome
The-erinaceous-one
Daniel Mietchen
Haansn08
Xenmorpha
John Samuel
Jeremy Dover
Toni 001
Bocardodarapti
Duckmather
HTinC23
fgnievinski

Notified participants of WikiProject Mathematics. --[雪菲🐉蛋糕🎂] >梓< [娜娜奇🐰鮮果茶☕]（☎️·☘️） 07:53, 25 October 2021 (UTC)[reply]

Absolutely yes, that's a really good idea. -- The Anome (talk) 20:29, 1 November 2021 (UTC)[reply]

Theorem and proof[edit]

Hello,

I do not find a specific property relating a fact (theorem or lemma) with a proof for it (something like is proved by proof 1, proof 2), and also the other way round: For a mathematical proof the simple property what the theorem is which it proves.

E.g. Q1506253 (infinitude of primes) does not refer to a proof for it like Q902630 (I did not find other items for this statement).

A similar missing relation is between a mathematical exercise and a mathematical solution for it. Both things might not be very relevant for Wikipedia, but they are for the mathematical courses like the ones on Wikiversity.

Bocardodarapti (talk) 12:28, 14 October 2022 (UTC)[reply]

@Bocardodarapti: I am not aware of "has proof" property that connects a theorem with its proof (I support creating one), but to connect mathematical problems to their solution, use solution to (P9030) on the solution item. — The Erinaceous One 🦔 02:10, 17 October 2022 (UTC)[reply]
P9030 refers to solution in the sense of a mathematical object solving a mathematical equation (like a solution to a differential equation), here I mean a written solution to an exercise (homework or in an exam). Like Q114401612 is a solution for Q114057106. Bocardodarapti (talk) 17:32, 17 October 2022 (UTC)[reply]
I think it would be better to use solution to (P9030) than create a new property. The two concepts are so similar that trying to use two properties will just cause confusion. — The Erinaceous One 🦔 08:18, 18 October 2022 (UTC)[reply]
Still I think there is a big semantic difference, and I do not see for who this might be confusing. Bocardodarapti (talk) 17:37, 18 October 2022 (UTC)[reply]
What would you call the property you are thinking of that would make it clearly different from solution to (P9030)? — The Erinaceous One 🦔 07:19, 19 October 2022 (UTC)[reply]
the property is whether the solution is a mathematical object (like a function, a tuple, a subvectorspace,...) or a mathematical text containing an argument or a computation.Bocardodarapti (talk) 08:30, 19 October 2022 (UTC)[reply]
I understand the distinction your making, but I don't think that adding a new property is necessary. The type of solution in the sense "a mathematical text containing an argument or a computation" could be called a "step-by-step solution". Are there any examples where using the existing "solution to" property would cause ambiguity or confusion when it is used on a "step-by-step solution" item? — The Erinaceous One 🦔 03:14, 20 October 2022 (UTC)[reply]
For me proof is a particular case of solution (another one is counter-example). So I agree with using P9030. --Infovarius (talk) 07:48, 19 October 2022 (UTC)[reply]
On a philosophical level there is a certain analogy, I agree. But as a working mathematician it looks completely different. No mathematician thinks when he or she writes a proof 'Oh, I have find a solution'. In a textbook, say, the theorems are accompagnied by a proof and for some of the exercises there are also solutions given. For me the question is whether Wikidata can express a clear specific semantic which is everywhere in a specific science.Bocardodarapti (talk) 08:40, 19 October 2022 (UTC)[reply]
I agree that it makes sense to have separate properties for "has proof" and "has solution", but I don't think we should have separate properties for solutions in the sense of <mathematical object>is a solution to<mathematical equation> vs. <step-by-step text>is a solution to<mathematical problem>. — The Erinaceous One 🦔 03:08, 20 October 2022 (UTC)[reply]

Scope and interwikis (copy from Wikidata_talk:Wikiversity)[edit]

I would like to discuss an activity of User:Bocardodarapti. He imports and creates a structure in Wikidata about numerous Wikiversity pages created for his own course (see contribs). I see a couple of problems here: 1) are those pages all notable for Wikidata? See e.g. v:de:Kategorie:Theorie der bestimmten Integrale and all its subcategories and leaves, esp. items like Q114345261, Q114766310, Q114769143 and hundreds (thousands?) similar. 2) his tendence of changing main category items to newly created items about particular "theories" (also arguable matter) like this change. Please discuss. Infovarius (talk) 20:35, 19 October 2022 (UTC)[reply]

Hello,
two short answers: ad 2) the last mentioned change follows the pattern/model of Q12479 (number theory) and Q7217289 (category:number theory) with the direct link via main category and subject of the category. The subject of the category number theory is number theory and not numbers. Numbers are studied by number theory. This seems to me a reasonable model for a mathematical theory/subfield.
ad 1) some statistics: the mathematical content (which is built up in a modular way) of the German wikiversity consists of roughly 3200 facts (same amount of proofs), 2400 definitions, 1200 examples, 500 remarks, 12 000 exercises (about 500 have an English version on the English Wikiversity, linked together by a Wikidata item), 3500 solutions. They are used in 20 mathematical courses at universities (covering bachelor degree and say the algebraic and geometric side of master degree) and are categorized in 1800 theories/subfields (to compare: the mathematical subject classification (MSC) has about 80 main entries, each has roughly about 10 letters and each letter has about 10 numbers, gives 8000. This classification is considered within themathematical community as quite rude, therefore one adds key words). My understanding of Wikidata is that it should provide a framework for organizing a science, and science is huge.
beside that, I am sure I do make mistakes. For background: I want to make an English version of one of my courses (where there are some 'international students') next year and for that I want to use the interwikilinks via Wikidata. That was the starting point of my work here. Bocardodarapti (talk) 09:06, 20 October 2022 (UTC)[reply]
2) There is difference. I agree with Q12479-Q7217289 relation but all sitelinks in Category:Rational numbers (Q7028361) are about rational numbers, not about theory of rational numbers (except 2 of Wikiversity). 1) Just to understand: is it (German wikiversity course) your personal work?
to a large extent yes, the topology course is from a colleague. Everything with some additions, corrections and improvements by assistents and students and other people; 4 courses were used by colleagues. Most of the theorems etc. are of course common mathematical knowledge, as you would expect from a mathematical course or a textbook. Bocardodarapti (talk) 07:34, 21 October 2022 (UTC)[reply]
I don't see anything wrong with Bocardodarapti's edits.
Regarding item 1), linking together pages across various Wikis is the original intent for WikiData. It looks like the items he is creating are often linked to Wikiversity pages in multiple languages, which adds links to the "In Other Languages" section of the sidebar on each of those pages. This is great, in my opinion. One suggestion I would make, Bocardodarapti, is that if you are automatically importing of items, then, maybe populate the descriptions in English and German to say what the item is (e.g., "Wikiversity mathematical exercise"), so that it is easier for editors to understand what they are. Perhaps we could also make a new class for Wikiversity exercises?
Regarding item 2), I tend to agree with Infovarius regarding Category:Rational numbers (Q7028361) because the name of the category is Category Rational Numbers not Category Rational Number Theory, but I don't have strong feelings either way. This seems more like a debate to be had on that item's talk page than here. — The Erinaceous One 🦔 03:36, 23 October 2022 (UTC)[reply]

P.S. Mathematical subject classification (MSC) with the source provided would be good info for Wikidata. As well as UDC and other classifications. --Infovarius (talk) 20:17, 20 October 2022 (UTC)[reply]

I agree that it makes sense to attach different sitelinks on the same author-defined topic to a single item for convenience, but I do think that those items should be excluded from statements in mainstream items. Statements like cardinality (Q4049983)studied in (P2579)theory of cardinality (Q114705928) (See Q4049983#P2579) just makes no sense, since there is no such thing as "theory of cardianalities". 慈居 (talk) 12:38, 22 August 2023 (UTC)[reply]

Community[edit]

Just a question: how active is this community?Bocardodarapti (talk) 09:31, 20 October 2022 (UTC)[reply]

@Bocardodarapti: the number of active users is pretty small. You've already met Infovarius and me. In the past, I've also seen Toni 001 and TomT0m be pretty engaged. You can see a few others that have participated in the discussions above. — The Erinaceous One 🦔 03:40, 23 October 2022 (UTC)[reply]

polyhedra and polytopes[edit]

I work on a software project [1]https://www.polymake.org and we are planning to insert the IDs of wikidata items in our documentation and file format in an effort to make our data more independent from our software and easier to parse. I started a discussion at Q172937 . The general definition of polyhedron in my part of the mathematical community is as the intersection of finitely many affine halfspaces. The mentioned item restricts to the 3dim case and also seems to assume boundedness via its connection with 4polytope. The other definition Q747980 will also not work, since this is definitely bounded, nevertheless this is fine, since we usually say that a polytope is a bounded polyhedron. Can you give me advice on how to proceed? Should I just add a new item and relabel the old one? Lmkastner (talk) 12:36, 12 December 2022 (UTC)[reply]

To indicate over which field something is a vector space, etc.[edit]

According to Property:P642#P2559 the use of of (P642) should be replaced by its refining properties in the long term. The way I can think of to indicate that something is a vector space over a particular field in Wikidata is via statement somethingsubclass of (P279)vector space (Q125977)of (P642)field of rational numbers (Q6585992). How about creating new properties

  • <vector space over field>
  • <module over ring>
  • <group ring of group>
  • <group ring over ring>

etc.? Then we could replace potentially ambiguous property of (P642) by using statements

慈居 (talk) 10:55, 11 August 2023 (UTC)[reply]

definition domain (P1568) and codomain (P1571) as qualifiers of instance of (P31)?[edit]

There is a related discussion in Property talk:P31. 慈居 (talk) 18:18, 18 August 2023 (UTC)[reply]

Symmetries in physics[edit]

I opened a discussion about symmetries in physics theory and how to model this but as this is definitely relevant more generally in maths I also put a link here, please participate in the topic on the project page author  TomT0m / talk page 13:59, 19 March 2024 (UTC)[reply]

Proofs[edit]

We have items about proofs but I think not much properties to model them, for example proof of the Euler product formula for the Riemann zeta function (Q4117082). What would be a set of properties to model them ?

  • [proof] proves/demonstrates [formula/theorem]
  • [theorem] has proof [proof]
  • [proof] uses [math technique / theory / formula] ?

May we use generalization of (P7719) View with SQID for that, as the demonstration is a set/sequence of formulas that in the end implies the theorem ? author  TomT0m / talk page 15:15, 17 April 2024 (UTC)[reply]

The three properties you propose seems desirable to me. A proof, as you've commented, is a sequence of formulas that ends with the desired theorem and such that each formula is a consequence of applying rules of inference to axioms and preceding formulas. Since a proof itself is not a theorem, I think generalization of (P7719) does not apply to them. 慈居 (talk) 11:16, 20 April 2024 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=Wikidata_talk:WikiProject_Mathematics&oldid=2133481039"