Wikidata concepts are very useful for tagging content on 3rd party websites and other cases where a specific concept needs to be chosen. For this 3rd party developers shouldn't need to develop their own entity selector. We should provide something similar to what Freebase offered with Freebase Suggest.
As a current user of Freebase Suggest, which is going away apparently on June 30th, it would be super to have a Wikidata based alternative. I suspect (but do not know) that there are other Freebase Suggest users who will need to migrate to something, and having a path to Wikidata would be good for Wikidata entities being used outside of the Wikipedia ecosystem.
I see that the searchbox at www.wikidata.org is using this API call?
To be able to use that API call on another website I guess there would need to be a .js file that could be loaded from www.wikidata.org (same origin policy). Or perhaps there is a way to get it to work with the JSON-P callback parameter, or if CORS headers are set up appropriately from the wikidata API call?
An additional story: In an institutional website, a cataloger wants to create a metadata tag, that includes both a plain text and a URI based reference for what they are describing -- for fields created by the institutional website using some metadata standard. As an institutional website developer, I want to provide a suggestion box for that field, so that the cateloger uses our controlled vocabulary (based on Wikidata), instead of having to either a) copy and paste qids into the field or b) do the matching in a seperate ecosystem (say Google Spreadsheets or Open Refine). The output of the tool needs a plain-text label for human readable use of the catelogue/collection software, and the uri in a format appropriate for linked-open data schamas.
*Could we create a "model widget" that could be used by developers on common institutional platforms (Drupal, Dspace, Koah etc) , that provides access to an auto-fill tool, that is akin to what folks get when adding a property and value on Wikidata?
*What kinds of background/context would be needed to successfully develop this?