On 29/11/2014, Lukas Sommer <sommer...@gmail.com> wrote: > Okay. I’ll try a summary: [...]
Thanks for those insights. Note that this multiple-values discussion has gone beyond the alt_name key (which is itself a multiple-value key for the name key, making 'multiple alt_name values' kind of silly), so you should look at other keys in your analysis. > In this current thread, both solutions have had supporters, but there has > been some more support for the semicolon system. Dont assume the most vocally-supported option is the more popular one. It would be interesting to widen your analysis to other keys, but so far it points at numbered keys being more used in practice. > 1.) do a formal voting (I would propose the semicolon system) I dont think there's any need to obsolete one method or another at this stage. Like the addr:street vs associatedstreet debate, I dont see either scheme ever disapearing from osm (just like I doubt that alt_name will ever disappear in favor of multi-value name). If you want to continue the discussion, try to fix the semicolon scheme's technical issues instead (litteral ;, empty fields, false-positives...). You'll never get any scheme universally accepted unless it works universally, or at least in more cases than the alternative. > 2.) after that: ask for support for this in Nomination (and maybe other > software on which the HOT people rely) for the voted and accepted solution AFAIK Notinatim already supports semicolons, {alt,loc,old}_foo, and foo_<n>. > 3.) as mid-term goal: fade out the usage of the not-accepted solution > (maybe later even with a mechanical edit – not sure if this is easy) No (see previous comment). > 4.) as long-term goal: ask software to stop using the not-accepted > solution. I'd rather have ata-model support as a long-term goal. _______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging