I've FINALLY gotten around to doing something with the StatsCan NRN data. Although this data is FAR too low resolution to get any useful road information, it does have address ranges for the left and right sides of streets, as well as street names.
I'm 90% done writing a python script that takes the StatsCan NRN data and generates the address interpolation ways for both sides of the street in OSM format. The workflow I'm imaging to use this data would be: 1) Generate address interpolation ways for LEFT side of streets in an area using python script. The ways are tagged to indicate which side of the road they should be put on, relative to the start of the street the interpolation way refers to. 2) Open OSM file in JOSM and use a combination of mapped roads and Yahoo satellite imagery to align the way start/end nodes, and add any required intermediate nodes to fit the way to the existing mapped date. 3) Repeat this process with the RIGHT side of the roads. There's still lots of cleanup work to do here, but I feel this is a little easier than just providing a WMS layer with the data and having someone re-enter is manually. The ways have the addr:housenumber of the start and end house numbers, as well as the addr:street to which they belong. The way is marked addr:interpolation odd/even, depending on the house numbers. My plan is to get everything finished and try it out for a small area here in Winnipeg, after which I'll post some links on here for people to review the results and see if they can find any issues. Tyler _______________________________________________ Talk-ca mailing list [email protected] http://lists.openstreetmap.org/listinfo/talk-ca
