Commons talk:Categories

From Wikimedia Commons, the free media repository
Jump to: navigation, search


Way to organize and find files on the Commons[edit]

"The category structure is the primary way to organize and find files on the Commons" - Wikidata's items can be such way (and not only on the Commons): example 1, example 2 (by "Current location", "Artist", "Title", etc)? --Fractaler (talk) 13:29, 19 February 2018 (UTC)

Yes, there are several other ways. Others include galleries. Hence primary rather than only. LX (talk, contribs) 17:52, 19 February 2018 (UTC)
Thus, we have three models of this part of the world/universe: 1) using categories (primary), 2) using Wikidata's items, 3) galleries. For example, the current model for file:Retable de l'Agneau mystique.jpg - 1) Category:Ghent Altarpiece (->Religious paintings by Jan van Eyck;Renaissance panel paintings;Early Netherlandish paintings;Paintings in Sint-Baafskathedraal (Gent);Sept merveilles de Belgique;Visitor attractions in Ghent;15th-century paintings of Adam and Eve;15th-century religious paintings in Belgium;1432 paintings;15th-century paintings of Virgin Mary), 2) Q734834 (Ghent Altarpiece) -> polyptych;reredos 3) Jan van Eyck --Fractaler (talk) 07:04, 20 February 2018 (UTC)
And this page is about the primary one. Is there a particular change to the page that you're suggesting? LX (talk, contribs) 07:08, 20 February 2018 (UTC)
Yes. We have one part of the world model, hence the description of this part should be one, and not so, that according to one point this is so, for the second point it is so, for the third one somehow. Primary model said: "Ghent Altarpiece -> Religious paintings by Jan van Eyck;Renaissance panel paintings;Early Netherlandish paintings;Paintings in Sint-Baafskathedraal (Gent);Sept merveilles de Belgique;Visitor attractions in Ghent;15th-century paintings of Adam and Eve;15th-century religious paintings in Belgium;1432 paintings;15th-century paintings of Virgin Mary". Then Wikidata should say exactly the same: "Ghent Altarpiece -> Religious paintings by Jan van Eyck;Renaissance panel paintings;Early Netherlandish paintings;Paintings in Sint-Baafskathedraal (Gent);Sept merveilles de Belgique;Visitor attractions in Ghent;15th-century paintings of Adam and Eve;15th-century religious paintings in Belgium;1432 paintings;15th-century paintings of Virgin Mary". Ie, there must be a standard. One (1) standart, for all. Changed the model in the primary model (руку), the model automatically changed in the Wikidata. Fractaler (talk) 08:33, 20 February 2018 (UTC)
I'm sorry, but it's still not at all clear what change you're proposing to make to this page. LX (talk, contribs) 09:39, 20 February 2018 (UTC)
Now Category:Ghent Altarpiece has supercategories: Religious paintings by Jan van Eyck;Renaissance panel paintings;Early Netherlandish paintings;Paintings in Sint-Baafskathedraal (Gent);Sept merveilles de Belgique;Visitor attractions in Ghent;15th-century paintings of Adam and Eve;15th-century religious paintings in Belgium;1432 paintings;15th-century paintings of Virgin Mary. Wikidata does not have this. Ie, Wikidata has own "primary way to organize". Is it correct? Or commons way to organize is primary? Why does the Wikidata not repeat this primary way? --Fractaler (talk) 11:25, 20 February 2018 (UTC)
Wikidata and Wikimedia Commons are independent projects. This page is about how categories are used on Wikimedia Commons. How Wikidata is structured is a topic for another page (and another project). LX (talk, contribs) 11:56, 20 February 2018 (UTC)
I'm understood, thank you! --Fractaler (talk) 12:16, 20 February 2018 (UTC)
My guess is, some day the Wikidata structure will become as important to Commons as the current category structure. That is when we should drop the word "primary" but it won't happen soon. Perhaps sometime in the 2020s. Probably even later, but we can wait several years to begin worrying. Jim.henderson (talk) 05:00, 22 February 2018 (UTC)
Wikipedia, Commons and Wikidata try to classify (to model) the same world/universe. Wikipedia can no longer be considered (emphasis only on human readability and the suppression there of machine-readableness makes Wikipedia uncompetitive). Commons has nothing but categories (galleries are local, not global classification, they are not considered), so it remains to classify using categories (eventually we get a graph like a tree). If the categories are considered as sets (set theory, "subset -> set -> superset"), then, for example, inductively-deductive methods, transitivity testing, etc. can be applied to them. Wikidata models with Q* (also allows to get a graph like a tree, by Template:Item documentation) and P*. P*-approvals (local, relative classification) can be replaced with Q*-approvals (global, absolute classification), so P* can be queued to Occam with his razor. Commons has the advantage of having machine-readable format about subcategories (subcategories do not need to be specially written, since they appear automatically). The tools of the Wikidata do not have such automation, we can not immediately know about subsets, elements of the set of the set in question (they are written manually). But with the help of the Wikidata tool for the classification of world/universe objects, the language barrier is removed: Wikidata uses the universal identifier of the world object (Q*), which is not tied to a specific language. In the same way as, for example, the language of music, mathematics, etc. Hence the conclusion: a project should appear that will combine both advantages: 1) the visibility of subcategories, 2) the global, standart, universal ID of objects that have the corresponding language labels (for use not only by machines, but also by humans). --Fractaler (talk) 08:16, 22 February 2018 (UTC)