Am 08.06.2018 um 11:53 schrieb marc marc: > Le 08. 06. 18 à 11:34, Rory McCann a écrit : >> Replicate data from OSM, applying your data transformation programme > it's already what nearly all use of osm is doing. > it's named "preprocessor" or alias > it's waste that nearly all data use need to build their own list. > You are assuming that the data consumers doing pre-processing could/would agree on a specific way to normalize the data.
I envision very long discussions on the "normalizing" mailing list bike sheding what landuse=forest should be mapped to :-). Seriously, in a such attribute rich dataset as OSM, normalizing is never going to go away for data consumers and wouldn't go away even if we had a tagging czar and a completely regular tagging scheme. It would be nice if going forward we don't create more "weird" stuff, but I think that in general is the case. Simon
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging