On Mon, Jan 11, 2010 at 20:19, Frederik Ramm <[email protected]> wrote: > Ævar Arnfjörð Bjarmason wrote: >> >> Aside from doing some minor tweaking on highway classifications (e.g. >> motorway -> trunk) this dataset is pretty much ready for importing >> into the live database. Any comments on that? > > Austria has not had the best experience with the kind of "shadow tagging" > you are suggesting here. The import was ages ago and still the database is > littered with "highway:plan.at=*" objects. > > (Also, they mungled the highway tags but left the name tags alone, leading > Mapnik to render ghost roads like on this picture: > http://osm.org/go/0JIs9Kh2z-)
I've munged the name=* tags too, everything but source=* in fact is a non-std tag. > I'm just saying - if a speedy cleanup is desired, I would import everything > with proper highway tags and accept that the map looks a bit chaotic for a > while, rather than have the stuff lying around forever. I'd rather not break the rendering / garmin maps / routing for any amount of time with lots on duplicated data. Anyway I've finished everything I think including tests to the api06.dev server (& some improvements to the bulk_upload_sax.py script). Here's the data as I expect to upload it tomorrow (aside from the munged ids): http://v.nix.is/~avar/noindex/ofp-final.osm.bz2 Any last comments ?:) _______________________________________________ Imports mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports
