At 12:19 AM 9/28/2007, Thom Shannon wrote: >Does anyone have any suggestions how to map a long tunnel? I know how to >tag the ways, the issue is where does the tunnel go. You can't exactly >get a GPS trace! I want to map the tunnels under the Mersey. The >queensway tunnel has 3 exits, all at slightly odd angles and there is a >junction in the middle of the tunnel. > >Do you think I could figure out some reasonable positions using a cars >odometer and a compass? Or does it not really matter that much? > >Another option just occurred, the tunnel was started in 1925, so any >maps from that era would be out of copyright. Perhaps a trip to the >library?
I feel that completeness of main features is more important than accuracy in these early days of OSM. I therefore simply join the tunnel mouths with as few nodes as possible (so that no one gets the idea I have a mysterious source of information) and mark the tunnel way source=interpolation. I sometimes place minimum nodes between the tunnel mouths if there is an obvious curve and if there are internal junctions on a best effort basis. Of course in the longer term better accuracy would be good, out-of-copyright maps would be your best bet I think. From experience in my UK home town, you may also find a local history museum or historian with a town planning map from the twenties - the features may, of course, not have been built or may have been adjusted but such maps were more widely distributed than actual engineering plans. Mike Stockholm [and Otley - Centre of the Known & Unknown Universe] _______________________________________________ Talk-GB mailing list [email protected] http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk-gb
