Hi!
Concerning wikidata search: could we reuse the code for language search and trigger the search on the backend?
We probably could, though I'm not sure if a little hack we did would work with Wikidata - Wikidata overrides a lot of stuff that regular wikis do not, and our config hack may not be enough. Needs checking.
For shorter term, doing it client-side may be faster as Wdsearch does.
Also, what about performance considerations? Right now I imagine traffic for wikidata search and other APIs is less than what we get for en.wikipedia but if we make searches go to both will it be ok?
The JS code will send a query to wikidata.org (using action=query) and will generate a new search request on wikidata. Analyzing these logs will be hard because we won't be able to associate the original query and the query sent to wikidata.
Can't we mark them somehow, like with prov= or other things? Maybe this mark plus temporal proximity plus query term match should enable grouping them? Not sure how hard it is in practice, analytics people please correct.