On 06/08/14 13:28, Pavlo Dudka wrote: > Lester, > I don't agree that ukrainian or other-language place names is secondary > information. > We should not extract this information to external data source. > Why don't you say "Let's remove population-tag. Values are changing, > let's integrate OSM with some service like > http://www.ons.gov.uk/ons/index.html";? > Reason is very obvious: implementation of integration between dozens > different services that use OSM-data and [Translation DB]/[Population > DB] is much more expensive than storing translations/population in OSM-DB.
Pavlo - you are miss understanding what I am saying ... The primary data in the heart of OSM would not have any 'language' embedded in it. That would come from tables of translated data which are still part of the database, but do not need to be accessed all the time. The main problem with having ALL language data spread all through database is that everybody gets it, while if you ONLY want Ukranian that would be a much smaller data set. For languages where there is no translations built in, a fall back to a third party tool could then be actioned. No doubt someone would use that to populate that language, and this is fine as long as mistakes are corrected. Certainly I would not be using wikipedia myself, but nominatim enhanced with a language selection facility ... still essentially contained within OSM! > "name"/"name:en" can't be the key for place names Totally agree. There should be a uniqueID for every object in OSM, not one that may change when people make mistakes changing other data :) -- Lester Caine - G8HFL ----------------------------- Contact - http://lsces.co.uk/wiki/?page=contact L.S.Caine Electronic Services - http://lsces.co.uk EnquirySolve - http://enquirysolve.com/ Model Engineers Digital Workshop - http://medw.co.uk Rainbow Digital Media - http://rainbowdigitalmedia.co.uk _______________________________________________ Talk-GB mailing list [email protected] https://lists.openstreetmap.org/listinfo/talk-gb
