[OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sat, Sep 19, 2009 at 11:43 AM, Martin Koppenhoefer <
dieterdreist at gmail.com> wrote:

> yes, but that's not the problem: straight parallel ways. The problem
> arises when they change (become one more or less), on intersections,
> etc. Try to imagine a situation like the one I posted above in a
> geometrically reduced system: it will get way too confusing. If we map
> lanes where they are, there is another benefit: positional correctness
> and ease of topological structure: you see what you do.
>

This can be done without resorting to mapping each lane separately.  If you
have a three lane road with no lane change restrictions or physical
barriers, you map it as one way, with three lanes, with the position as the
center of the three lanes.  When the road goes to two lanes, you map it as
one way, with two lanes, with the position as the center of the two lanes.

Perhaps there could be some sort of special designation for a way with 3
lanes at the beginning and 2 lanes at the end, which designates whether the
right or left lane ends, if you really want to get into the fine detail.

When a lane doesn't end, but becomes an exit only, it's even simpler.  You
pick a decision point where the (let's say right) lane becomes exit only.
You split the way at that point.  Before the decision point you have one
way, with three lanes, with the position as the center of the three lanes.
After the decision point you have two ways - the left way has two lanes with
the position as the center of the two lanes; the right way has one lane with
the position as the center of the lane.  Then when the right lane goes off
to exit, you just follow that geometry.
http://www.openstreetmap.org/?lat=27.944845&lon=-82.538208&zoom=18&layers=B000FTF
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20090919/261acf05/attachment.html>