This is a proposal to create a special page which lists all lexemes containing forms with the same spelling (e.g. a page which would list all lexemes which have a lemma or form "orange" in different languages). This would provide a more Wiktionary-like view of the data: In Wiktionary, there is one page for each string of characters, which contains all the forms spelt the same. E.g. https://en.wiktionary.org/wiki/orange lists all forms spelt "orange" in various languages. It would be a good idea to have a special page which lists all lexemes containing forms with the same spelling (e.g.
The creation of a special page would mean we can use Cognate to link Wiktionary pages to Wikidata and vice versa and therefore make it possible to navigate between Wikidata and Wiktionary.
There should be links from lexemes to the special page, a page which would list all lexemes which have a lemma or form "orange" in different languages)so that people can navigate from lexemes to the special page. This would provide a more Wiktionary-like view of the data and it could also perhaps be used with Cognate to link Wiktionary pages to Wikidata and vice versa.When there are multiple lemmas or forms, That would make it easier for Wiktionary users to navigate to Wikidata, and if we link to the special page from the individual lexemes, also easier for Wikidata users to navigate to Wiktionarythere should be a link for each unique spelling.
It would be different from a search:
* It would return things in a consistent order (e.g. It would be different from a search because it would return things in a consistent order (by language then by part of speech, most likely). In particular, capitalisation and accents matter for Wiktionary,by language then by part of speech)
* It would pay attention to capitalisation and accents
* It would not include partial matches
E.g. but the search would ignore those. However, the entity suggester in the search bar could display this page as an entry point to all lexemes containing that form (as opposed as showing all forms independently in the entity suggesterhttps://www.wikidata.org/w/index.php?search=the&ns146=1 finds "the", which would take too much space). Additionally, when there are more than one form with the same string of characters, the string of characters of the form itself could link to this page,"thé" and "The quick brown fox jumps over the lazy dog"
There are several suggested ways to implement this:
* As a special page (e.g. so that the users can navigate from a form in one language to the homograph in another (or the same) language located in a different lexeme. This can be implemented in several ways: - A special pag"Special:ListLexemes/orange")
* As a namespace (e.g. "Special:ListLexemes/orange") - A namespace (e.g."Homograph:orange")
Since the page would be entirely generated and is not expected to be user-editable, "Homograph:orange") Using ns0 for listing homographs is not possible since it is reserved for itemsa special page seems most appropriate and it's not clear what benefit a dedicated namespace would have.