Rob The problem is that there are only just over 21'000 wikidata tags in OSM at this point in time (according to taginfo). Given the subject matter, it could well be that Edwards code could match 21'000 of them. And yes that would imply a lot of wasted effort, leading to my conclusions being just the opposite of yours....
Simon Am 28.08.2014 01:13, schrieb Rob Nickerson: > Simon, > > My understanding is that this is only looking at a very small subset of > wikidata (based on a list of wikipedia categories). As such an API would > only give you a partial view (70k) objects. We want to match as much as > possible and therefore people will continue adding wikidata tags to OSM > manually. At which stage you end up with some wikidata tags in OSM, an > API based on Edward's work, and no doubt lots of overlap and wasted effort. > > In summary I'm not sure an API is a good strategy. Instead I support the > proposed import. > > Regards, > Rob > > P.s. I consider myself part of the UK local community and, without > rereading the old thread, I felt there was some support and some > against, with the topic fizzling out. Its clear that a locality check > resolves some of the issues and I'm pleased to see this has been > included. Thanks for pulling this together :-) > > > > _______________________________________________ > talk mailing list > talk@openstreetmap.org > https://lists.openstreetmap.org/listinfo/talk >
signature.asc
Description: OpenPGP digital signature
_______________________________________________ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
