From the notes against the layer: http://linz2osm.openstreetmap.org.nz/data_dict/layer/canal_poly/info/ <http://linz2osm.openstreetmap.org.nz/data_dict/layer/canal_poly/info/>
I think the notes are very easy to access and relate directly to the layer in question. The tagging page clearly shows how we’ve mapped each source field to the output and any transformations we have done so it provides a good audit trail. The building_poly layer is a good example where complex mappings are easy to read - http://linz2osm.openstreetmap.org.nz/data_dict/layer/building_poly/tagging/ <http://linz2osm.openstreetmap.org.nz/data_dict/layer/building_poly/tagging/>. I take your point on the yearly review. If you look at our notes you will see that we did come up against inconsistencies in what the tagging should be. I think at the time what we choose was considered OK but there was an ongoing discussion about a new tagging scheme for rivers and waterways. Now it would seem that we should be updating the tagging to conform to the new tagging norms. Thanks Glen > On 24/04/2015, at 9:34 am, Christoph Hormann <chris_horm...@gmx.de> wrote: > > On Friday 24 April 2015, Glen Barnes wrote: >> >> I’m not sure where all this came from. We (including you) were >> instrumental in setting up the LINZ2OSM process years ago in NZ. We >> discussed and came up with tagging schemes and received agreement on >> the process on the nzopengis group which is where all of the >> discussions take place. This import/merge is business as usual as far >> as I am concerned. > > I don't have an opinion on this particular subject (i.e. aukland > buildings) but the way you manage the import with the linz2osm tool > seem to make it difficult for the normal mechanisms of imports in OSM > with advise and verification of the process by the larger international > OSM community to work. > > The tagging rules for example are fairly difficult to access within your > tool and the list on the wiki: > > http://wiki.openstreetmap.org/wiki/LINZ_attribute_matching > > looks very incomplete. > > To give an example: Canal polygons seem to be imported with > waterway=canal: > > http://www.openstreetmap.org/way/230078171 > http://linz2osm.openstreetmap.org.nz/data_dict/layer/canal_poly/tagging/ > > which is wrong (that tag is for the center line). I don't want to warm > up old stuff that is long done but if this import is still largely in > progress it might be a good idea to re-evaluate the initial decisions > made based on practical expecience and how mapping practice in OSM has > changed meanwhile. > > I would extend this to a general remark - all imports that run for more > than a year should IMO have yearly evaluations with update of the > documentation and a progress report and new RFC here on the mailing > list. > > -- > Christoph Hormann > http://www.imagico.de/ > > _______________________________________________ > Imports mailing list > Imports@openstreetmap.org > https://lists.openstreetmap.org/listinfo/imports
_______________________________________________ Imports mailing list Imports@openstreetmap.org https://lists.openstreetmap.org/listinfo/imports
