On Sun, 23 Aug 2009 17:48:58 +0100, Craig Wallace wrote:

> On 23/08/2009 17:15, David Paleino wrote:
> > On Sun, 23 Aug 2009 16:58:36 +0100, Craig Wallace wrote:
> >
> >> Why not place the stop sign node where the stop line / stop sign is
> >> physically located?
> >> Nothing arbitrary about that. You can measure the distance from the stop
> >> line to the centre of the junction of you want.
> >
> > You'd still need some kind of relationship for that to be effective (i.e. to
> > relate the highway=stop to the junction node) -- and AFAICT typical consumer
> > GPS units aren't that precise.
> 
> Why is it necessary to relate the highway=stop to the junction node?
> Isn't it obvious that if a highway=stop is within a few metres of a 
> junction, then its part of the same junction. It shouldn't affect 
> routing software etc anyway.

Well, one thing I could think of is "short" roads ~10m long or so. I've seen
quite of these, but "fortunately" none of them had a stop sign -- with the
current GPS accuracy (3m for my unit, at least nominally) it would've been a
problem taking a waypoint for a stop sign placed there.

Am I totally wrong/biased? :-)

(just out of curiosity: I've also seen roads 4-5m long)

> And it doesn't have to be very precise. Its easy to estimate the width 
> of a road, and how far away from the road edge the stop line is.

If we were to use highway=stop, it should be *on* the way (part of it), not on
one side.

> (and on a related note, is there any of mapping advanced stop lines (for 
> cyclists etc)?)

I believe not, but we could easily adapt highway=stop for this. Still, if we
deprecate it, we should think at something else :-)

Kindly,
David

-- 
 . ''`.  Debian maintainer | http://wiki.debian.org/DavidPaleino
 : :'  : Linuxer #334216 --|-- http://www.hanskalabs.net/
 `. `'`  GPG: 1392B174 ----|---- http://snipr.com/qa_page
   `-   2BAB C625 4E66 E7B8 450A C3E1 E6AA 9017 1392 B174

Attachment: signature.asc
Description: PGP signature

_______________________________________________
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Reply via email to