--- On Thu, 27/8/09, Roy Wallace <[email protected]> wrote:
> My apologies, I misinterpreted your question "why should
> you care
> about needing this explicit information". To answer your
> question, I
> do not care personally. It will not benefit me personally
> whatsoever
> how a stop sign is marked. I'm just trying to contribute to
> OSM into
> the long-term future.
Ok, so your goal is to contribute to OSM in a meaningful way, that's fair
enough but lets not get emotive over an issue but look at the logical outcome
from simplest to most complex and test if they fail at all to address the
issue, we shouldn't just pick the most complex for the sake of it because it
might be more extensible than a simpler answer, which is only useful if that
extensibility is actually useful.
> I cannot. I can only contribute some hand waving and warn
> of
I don't have a problem with playing devils advocate, but to conduct these
things rationally and logically we need conditions to test against, if we can't
find a test that fails than it's a valid solution.
> If using a proximity search, it will be necessary to
> clearly define
> "nearest junction", though. And if a way travels through
> an
> intersection but has stop signs on both sides, it will need
> to be
> split and two stop nodes will need to be added. There may
> be issues
> like this that need sorting out that only arise when a full
> proposal
> is written up and/or it's in use. But no, I can't think of
> any fail
> examples right now.
That's just it, there is no need to split the way, you just put a stop sign
node either side of the junction, or on 3, 4 or more ways that intersect with
applicable stop sign nodes.
Any software needing to know junctions will already have this coded. Then it's
just a case of applying a proximity search when it parses stop signs to "know"
which junction a stop sign applies to.
_______________________________________________
talk mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/talk
