Am 09.12.2010 02:49, schrieb Kenneth Gonsalves:
On Thu, 2010-12-09 at 08:59 +1100, Steve Bennett wrote:
(Personally, I would be arguing against it. "Don't do X because the
result would be less accurate than if you did Y" is an unhelpful kind
of perfectionism. The line makes the point that accuracy is important.
Well, coverage is also important. And you could argue that it's much
more efficient to map from aerial imagery first, then correct errors
with a local visit.)
I have absolutely no objection to map from aerial imagery first, then
correct errors with a local visit - as along as you are intending to
make that visit in the very near future. For example, we were to hold a
conference here:
http://www.openstreetmap.org/?lat=13.03175&lon=77.56565&zoom=17&layers=M
before the conference I did a rough sketch from satellite imagery. On
arrival at the spot I found that the ground reality was totally at
variance with the satellite imagery - and I got lost!
Seems the imagery got outdated.
If someone had perfectly mapped that area a few years ago and it got
completely outdated in the meantime - you would get into the exact same
problem. Taking your opinion further would mean we shouldn't map
anything at all because the map data might get outdated.
What your example really tells us is that you shouldn't "repair"
existing OSM data from (probably outdated) imagery without local knowledge.
what I object to is mapping a place one has no intention of visiting
Fine, seems you don't like the wiki principle ...
Regards, ULFL
_______________________________________________
talk mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/talk
