Chris Hill wrote: > I've been taking a look at the boundary data released as part of the > OS bundle. I've put together a little script that will extract a > named boundary as an OSM file ready for loading into JOSM. OS data > uses the OS projection and we use the WGS84 projection. I used > ogr2ogr to transform the data from one to the other, but there seems > to be a bug with its handling of the OS data projection. The > resulting OSM way is the correct size and shape, but is not in quite > the right place. The discrepancy in location varies across the UK, so > a fixed correction will not work. The variation between individual > nodes will be too small to matter in a parish boundary, but might be > a problem for the biggest counties.
I've had a degree of success with http://search.cpan.org/~toby/Geo-Coordinates-OSGB-2.04/ - I've used these packages in the past for rectification of OOC OS stuff and conversion of many-figure OS grid refs with a good degree of success. Do you get the same positional results with this? Does applying the OSTN02 transform too make any difference? -- Andrew Chadwick _______________________________________________ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb