Dave F. wrote:
In principle I understand what your saying & agree to some extent;
except that I think it's incorrect to assume that on ground surveying
is necessarily more accurate. GPS tracks are prone to being sent off
course by the surroundings such as heavy tree coverage & steep
topography. In these cases the maps are probably more accurate.
That's certainly true, but there are also (for whatever reason) errors
in the OSSV data too. The N carriageway of the A38 to the North of East
Midlands Designer Outlet is an example of this (around
http://www.openstreetmap.org/?lat=53.10942&lon=-1.31278&zoom=17&layers=B000FTF)
- edit in Potlatch and you can see that OSM has the carriageway slightly
the North of OSSV, but fetch the GPS traces in and you can see that they
all pretty much agree that both the OS and OSM have the road too far
south. This isn't an OSSV tracing problem as that road was last edited
by me reverting part of the RR8 "mass road renaming" in 2009.
Shuffle the background so that the underlying GPS traces match the OSSV
data and it's clear that some of the traced stuff on that portion of the
map is out by a 4-8m or so, but that doesn't mean that map isn't a good
one. However, the resulting map of East Midlands Designer Outlet (at
least partly done I think by the same contributor that we're talking
about) is now far more detailed and far more useful to someone actually
going shopping than it was before.
Full disclosure - some of the stuff N of the A38 and S of the EMDA was
added by me (and at least one service road traced off OSSV without
aligning the background first!)
_______________________________________________
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb
