[OSM-talk] House numbers... One more suggestion

Karl Newman siliconfiend at gmail.com
Tue Jul 29 19:20:04 BST 2008

On Tue, Jul 29, 2008 at 11:02 AM, Andy Robinson (blackadder-lists) <
ajrlists at googlemail.com> wrote:

> Karl Newman wrote:
> >Sent: 29 July 2008 6:48 PM
> >To: Jochen Topf
> >Cc: talk
> >Subject: Re: [OSM-talk] House numbers... One more suggestion
> >
> >On Tue, Jul 29, 2008 at 10:27 AM, Jochen Topf <jochen at remote.org> wrote:
> >
> >
> >       On Tue, Jul 29, 2008 at 10:16:33AM -0700, Karl Newman wrote:
> >       > To be clear, I don't have a problem with tagging the actual
> >location of the
> >       > house or building. I think it's unnecessary, but the problem I
> have
> >with the
> >       > scheme is that it doesn't definitively link the node with the way
> >(what's a
> >       > house number without an associated street?) My suggestion (the
> >third on that
> >
> >
> >       Thats why you should not only tag the building/node with the house
> >       number, but with the full address. Yes, there is some duplication
> of
> >       data, but its still easier than relations and doesn't break as
> easy.
> >
> >
> >       Jochen
> >
> >
> >
> >I'm okay with data duplication. Could we do both, then--the Karlsruhe
> >schema where actual locations are tagged with full address, and a relation
> >to topologically link the house number with the closest node in the way?
> I really don't think you should link to nodes in the way, they get moved
> around all the time. By all means link with a relation to the way but
> linking to nodes is asking for trouble.
> Cheers
> Andy

Well, I'll accept that is a risk, but I don't know how much of a risk it
really is. Adding street numbers is an activity that I see as a refinement
of data, not necessarily something that will be grabbed on the first (or
second or third?) outing. Hopefully by the time numbers are added, the shape
of the street will have settled down. Additionally, I could foresee the
editors supporting something like where the nodes could be "soft-locked" in
place such that an additional action would need to be taken to move them.

I just thought of another risk of not having the nodes topologically
associated with the way: If the data is sliced into tiles for further
processing, it's entirely likely that a node will end up in one tile and the
associated street will end up in a different tile, making the number
unusable (for routing purposes) in both tiles.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20080729/8df2977d/attachment.html>

More information about the talk mailing list