RoadMatcher is a conflation tool that's part of OpenJump (more info: http://wiki.openstreetmap.org/wiki/Roadmatcher), I haven't personally used it so I can't comment on how hard/easy to use it is.
Let me prefix the rest of this by saying that I am strongly in favor of imports and I think it's great that a government agency is interesting in syncing data to OSM, however: Centerlines: Updating centerlines is a contentious issue. There have been a lot of discussions about updating the Tiger data to the 2010 census lines, and the consensus seems to be "damn this is hard". I think anything you do is going to involve some (probably large) degree of manual work to make sure the changes you're making are actually an improvement to the data quality. Spatial accuracy: When you say "The roads are often incorrect or do not intersect properly" do you have statistics on this? Lack of footpath connectivity is a pretty widespread issue in OSM, but are you finding major issues with streets? Relational accuracy: Another problem with importing data is that even if yours is more spatially accurate other data has been added that depends on the inaccurate positions. E.g. parks and shopping centers get their outlines drawn relative to the current streets, if you import new centerlines it you need to check that your data doesn't intersect existing areas and reposition the areas if it does. Automation: As a programmer I appreciate that doing things automatically is sexy, but I personally wouldn't import anything without at least doing by a hand comparison of all the changes you want to make vs current OSM vs satellite imagery. On Jan 13, 2011, at 11:40 AM, PJ Houser wrote: > Hi all, > > In Portland, Oregon, the region's transit agency Trimet > (http://www.trimet.org) is transitioning to an open-source trip planner > (http://maps5.trimet.org/otp/). The trip planner will use OSM data for a > multi-modal approach - walking, biking, driving - and Trimet's route data. > However, Portland data is not accurate enough nor complete enough yet in > OpenStreetMap, so my job is to help fix that with Trimet and the community. > The roads are often incorrect or do not intersect properly, making > consistently good routing impossible. Bike routes and off-road trails are > seriously lacking. > > Recently, local datasets (http://www.civicapps.org/) have been made public > domain, and Trimet would like to update Multnomah County and Clark County's > streets using Metro's (the regional agency) data. I'd love your input. We > have a shapefile of all the streets along with basic attributes, sufficient > for routing. I've started learning about bulk imports, but I don't want to > overwrite the community's input, particularly special tags or attributes that > Metro's data does not contain (like surface). > > Questions: > 1) How do you feel about some kind of automatic editing or (selective) > import? Other methods? > 2) How do I go about doing that? (I heard of something called RoadMatcher)? > 3) Any other input? > > I also posed these questions to the talk-us list. > > And for fair warning, I will soon be writing you all about importing bike > routes (as relations) and off-road trails. Both bike routes and off-road > trails are lacking in the Portland area and there seems to be less risk of > erasing users' work (as there isn't too much yet). > > Thank you for any help or input you can provide. > -- > PJ Houser > Trimet, GIS > [email protected] > _______________________________________________ > Imports mailing list > [email protected] > http://lists.openstreetmap.org/listinfo/imports
_______________________________________________ Imports mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports
