On Wed, Oct 29, 2008 at 11:07 AM, Dave Hansen <[EMAIL PROTECTED]> wrote:
> On Tue, 2008-10-28 at 16:29 -0700, Alan Brown wrote: > > > > When that data is imported - how will you know that a missing road was > > intentionally deleted from OSM, or simply missing from the dataset? > > If they reappear, it would be rather irritating to me. It's not > > "overruling" me, in the sense that the old geometry no longer exists > > for you to see that I've editted it. > > You could merely import changes to TIGER - that would be part of being > > "paranoid" - and that would address this situation. > > That's a really good, and really functional idea. > > We could create convert old TIGER->OSM and new TIGER->OSM. Then, create > a JOSM-style diff and apply it to the database. If it applies cleanly, > commit it. Otherwise, throw it away and move on. > > -- Dave > > > _______________________________________________ > Talk-us mailing list > [email protected] > http://lists.openstreetmap.org/listinfo/talk-us > I would suggest a little more than that. The above eliminates what is unchanged and need not be imported. You will need to also diff the old tiger with tiger tagged data in OSM. Only what matches is ok to delete and replace with the new import. So you end up with: Same in old and new tiger = don't import as it is un-improved Not in old tiger = New data import, but may already be mapped by hand way names are in both, but do not match location = Improved data Compare old tiger with OSM tiger tagged data -If old tiger and OSM match unedited, replace with new -If the way was deleted or edited in OSM, discard the new tiger data in favor of the had edited. -- Dale Puch
_______________________________________________ Talk-us mailing list [email protected] http://lists.openstreetmap.org/listinfo/talk-us
