| Alsee added a comment. |
I believe discussing this as a "Bonnie and Clyde problem" has led to flawed examination and flawed arguments, due to the example cited.
Wikidata was largely designed, and largely operated by that community, on the theory that there is supposed to be a 1-to-1 relationship between concepts and wikidata items, as well as a 1-to-1 relationship between wikidata items and the related articles.
The flaw is that it's an invalid model. The flaw is being swept under the rug with an implicit (and arrogant) presumption that English defines reality, that English takes precedence over other languages. For example:
English Wikipedia (and Wikidata) have items for "see-saw", a children's playground seat attached horizontally to a pivot point. English Wikipedia (and Wikidata) also have items for "swing", a children's playground seat attached horizontally to a pivot point. However some other languages do not consider vertical and horizontal arrangements to be different concepts. Those Wikipedia have a single article, under the single language-concept which covers both orientations.
The English article for see-saw and the English article for swing both need to link to the same foreign article. And that foreign article needs to link to both English articles. Wikidata's faulty insistence on 1-to-1 realtionships causes problems for both editors and readers.
An even more clear example is colors. Colors are literally continuous. It is impossible to objectively define one "right" answer for how many there are, and it is impossible to objectively define one "right" answer for where the dividing line is between colors. For example some languages do not distinguish between "blue" and "green". The English equivalent would be comparing "orangish-red" to "purplish-red". They are both "red", and the English language does not recognize them as distinct fundamental concepts.
And even within English, the dividing line between "concepts" is often arbitrary or fuzzy. Within English you get the "Bonnie and Clyde problem". Is the duo the significant concept? Or do you subdivide and deal with the one story in two different places? Different Wikipedia can reasonably have one article for the duo, or two articles for the individuals, or three articles for the duo and each individual. That is a judgement call, and the right answer may be different for different purposes and contexts.
The problem is fundamentally impossible to solve while enforcing a 1-to-1 relationship rule. Allowing Wikidata to link to redirects is better than nothing, but it is still a badly dysfunctional workaround. It still doesn't give readers the right interlanguage links. Having Wikidata support non 1-to-1 relationships makes things more complicated, but the world is complicated.
Cc: Alsee, Capankajsmilyo, Eugene, PokestarFan, ArthurPSmith, Pasleim, StudiesWorld, daniel, Metamorforme42, JEumerus, Harmonia_Amanda, Ash_Crow, DannyH, Agabi10, Choomaq, IKhitron, QZanden, thiemowmde, Toto256, Acer, Elitre, Sylvain_WMFr, Lea_Lacroix_WMDE, Schlum, TomT0m, Thryduulf, Rich_Farmbrough, Zppix, ChristianKl, Mike_Peel, Wittylama, Liuxinyu970226, SebastianHelm, MisterSynergy, Oliv0, JanusTroelsen, Blahma, MGChecker, MSGJ, Izno, Nnemo, bzimport, Unknown Object (MLST), DanielFriesen, Gymel, Denny, jeblad, Abraham, Addshore, SamB, Toru10, Wikidata-bugs, JAnD, Nemo_bis, He7d3r, -jem-, ValterVB, Filceolaire, Micru, JanZerebecki, matej_suchanek, Ricordisamoa, MZMcBride, Aklapper, Tgr, kaldari, Laddo, Lydia_Pintscher, Jane023, Ltrlg, JohnLewis, Fomafix, Zellfaze, Lahi, Gq86, GoranSMilovanovic, LawExplorer, aude, Mbch331
_______________________________________________ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
