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

Andy Robinson (blackadder-lists) ajrlists at googlemail.com
Fri Jul 25 18:25:27 BST 2008

Frederik Ramm wrote:
>Sent: 25 July 2008 6:13 PM
>To: Charlie Echo
>Cc: talk
>Subject: Re: [OSM-talk] House numbers... One more suggestion
>> I added one suggestion to the "House Numbers" page. It's the eighth one.
>> http://wiki.openstreetmap.org/index.php/Proposed_features/House_numbers
>I find that very confusing.
>I'm using the "Karlsruhe Schema"
>and if you want to start mapping house numbers withtout endless
>discussions, I recommend you just do the same. The house numbers will
>even show up on Osmarender z17, albeit in a sort of "debug" view which
>will have to be improved.
>The Karlsruhe Schema has multiple facets. We assume that in the long
>run (long! run!) we will have all building outlines in our database
>anyway, so each building will simply be tagged with its house number
>and that's that. Until that happens, the Karlsruhe schema lets you
>place individual nodes for individual house numbers, or use
>interpolation between nodes placed at street corners. These
>interpolation rules are explicitly modeled as a new way connecting the
>nodes, which may look a but unusual at first but (a) it's going to go
>away once editors learn how to "dim" these lines and (b) it's going to
>go away once we have each house separate.
>Your suggestion of using "house_number:W_L" and such is asking for
>trouble IMHO.
>> - it will save a lot of time: entering street names and THEN numbers
>doubles the number of visits, and so on.
>This is not true, at least not generally. The initial mapping is often
>done by car or bike where you cannot be expected to record house
>numbers, even if you only want to record those at an intersection.
>House numbers typically require visiting the area on foot or at least,
>if by bike, then with a lot of extra time. Also, house numbers may
>be entered by people who live in the area and don't even have a GPS.
>So no need to rush.

+1 to this approach generally.



More information about the talk mailing list