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

Andy Robinson (blackadder-lists) ajrlists at googlemail.com
Tue Jul 29 18:48:58 BST 2008


Karl Newman [mailto:siliconfiend at gmail.com] wrote:
>Sent: 29 July 2008 6:17 PM
>To: Andy Robinson (blackadder-lists)
>Cc: Jan-Benedict Glaw; Frederik Ramm; talk
>Subject: Re: [OSM-talk] House numbers... One more suggestion
>
>On Tue, Jul 29, 2008 at 4:06 AM, Andy Robinson (blackadder-lists)
><ajrlists at googlemail.com> wrote:
>
>
>	<snip>
>
>	The reason I don't like ideas like this is that the data you are
>adding to
>	the way (or as Frederik pointed out possibly also the intersections)
>is not
>	actually anything to do with the physical feature. There are no
house
>	numbers or other references on a road, only the name of the street
>and
>	perhaps a road reference number. Accepted it's a nice easy way to
>make
>	routing work more easily but that doesn't make it right for our
>dataset. If
>	we keep and maintain the simple ideal that what we map is what we
see
>then
>	it keeps it all very simple. Routing algorithms may need to be more
>complex
>	as a result but that doesn't give us an excuse to corrupt our data
>with
>	misleading information.
>
>	Cheers
>
>	Andy
>
>
>
>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 page, and the first using relations) was to have a relation
>that marked one node with left and/or right addresses at that point in a
>particular street. Having only a single value instead of a range makes it
>more resistant to breakage if the way is split or merged or if nodes are
>changed. It's reasonably easy for mappers, too, because it's only one point
>and number to mark, and if more detail is desired later, more nodes,
>numbers and relations can be added in between existing points.

But it's still putting data on the street that has no relation to the street
itself other than thorough the non-physical "address" association. I accept
the argument for "relating" the street to the buildings along it within an
address but I doubt we will be able to get basic contributors to go to the
level of work that any relationship editing requires.

>
>Andy, you say "it's a nice easy way to make routing work more easily but
>that doesn't make it right for our dataset". What is the purpose for adding
>house numbers, then? I hope it's for more than drawing numbers on the map,
>which really has limited usefulness (numbers at intersections might be
>helpful, and would be less clutter on a map).
>

The purpose of adding house numbers is to reference the object being added
to the database. If we have 100 buildings along a street users may very well
want to know what they are.

I haven't given the address relationship a lot of thought yet. Arguably an
address structure is a true relationship between the place, street and
building. A relationship for each address would I guess be a correct model,
but until we have simple tools to select the place and street level data I
don't see much benefit in adding stuff.

Cheers

Andy





More information about the talk mailing list