[Talk-us] Vegetation/landuse import

[Greg Williamson]

I'd be glad to help, to the extent that I can ... I am relatively new to OSM (lurker in the lists for a few months) and having trying to find something to get some traction with.

Recently laid off DBA (PostgreSQL, GIS stuff mostly) with fair PERL, rusty (but once upon a time good) C, some JAVA. On good days I can communicate in English so documentation is also a possible task I could address. I have a windoze box as my usual desktop, but have a linux of some flavor available, if that makes a difference for any of the tool sets.

Greg Williamson (Berzerkely resident)



----- Original Message ----
From: Apollinaris Schoell <aschoell at gmail.com>
To: Talk-us Openstreetmap <talk-us at openstreetmap.org>
Sent: Wed, December 2, 2009 9:42:33 AM
Subject: [Talk-us] Vegetation/landuse import

we have some very detailed data for vegetation in California and try to make a plan for an import.
Has anyone done something similar for US. Any locals interested to help? It's a huge dataset and needs some major work to be really useful. 
some challenges are
- simplification to reduce number of nodes. there are many redundant nodes and duplicates
- smoothing edges, resolution is ~5m grid and rendering looks ugly this way, some spline interpolation could really make it similar to USGS maps style
- areas duplicate nodes/lines in shape files, this should be converted to line segments and relations to avoid duplicate lines in osm. Still no simple solution for that as far as I know
- filter out useless features, urban, residential, …  areas are already covered in better ways and should be skipped
- propose tagging scheme, current tags don't support all the details in this data but should be used as much as possible to get rendering for most features
- find local experts to verify fix and upload the data.
- documentation on the wiki




_______________________________________________
Talk-us mailing list
Talk-us at openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us