[OSM-talk] Merging data sets (was: web site map enhancement,l os angeles)

Putler, Dan Dan.Putler at sauder.ubc.ca
Fri Apr 7 18:02:18 BST 2006


A source identifier attributes makes a lot of sense. Although, I  
would make it a bit more generic, say <tag k="Source" v="TIGER" />.  
The reason for this is that other governments may release road  
network layers that could potentially be imported into OSM. There are  
actually two recently released, publicly available Canadian road  
network layers. One produced by Statistics Canada, which is similar  
in nature to the US TIGER/Line road layer, that uses a combination of  
Statistics Canada and Elections Canada data. The second, which is  
distributed by Natural Resources Canada, and uses data from local  
provincial and municipal governments along with NRC's own data. At  
some point, someone may consider uploading parts of one of these two  
layers into OSM, so a more general data source identifier would be  
useful.

Dan

On 7-Apr-06, at 3:01 AM, Etienne Cherdlu wrote:

> On 4/7/06, Tom Carden <tom at tom-carden.co.uk> wrote:
>> On 4/7/06, Dan Karran <dan at karran.net> wrote:
>>>
>>> I also believe it should be a manual process, but with the amount of
>>> data in question, that process is going to have to be simplified to
>>> avoid overwhelming users with masses of data. As Imi says, we need
>>> coders. If only I knew Java... (I sense the onset of a sharp  
>>> learning
>>> curve)
>>
>> Using the API, you could write a tool in whatever language you  
>> wanted.
>>  This is a specialised task, not one that the main OSM applet really
>> needs to be able to deal with all the time.
>
> If TIGER data was imported with, say, <tag k="TIGER" v="true"/><tag
> k="read-only" v="true"/> then it would be possible for the server to
> reject regular updates to these nodes and segments.  If there was then
> some easy way to copy elements from this 'layer' to the normal layer
> we could have the best of both worlds.
>
> Rendering clients could choose whether to display or ignore TIGER
> data, depending on what they are trying to achieve.
>
>> Regarding import of TIGER as a static 'gps-like' layer - do people
>> really think manually tracing and correcting TIGER on OSM would be a
>> useful application of their time?  I would think that even for areas
>> with a reasonable amount of existing data, TIGER would be at least an
>> order of magnitude more complete right now.  With respect to those
>> who've been working on OSM in the US, I doubt we'd be losing much  
>> (but
>> of course we should check).
>
> There is significant work that has been done in the Washington DC
> area.  Apart from that I've dabbled in California a bit.  I don't know
> of anything else.
>
> I doubt that much of the stuff that I did in California (canyon
> walking) will be in TIGER, although I'm sure all of the freeways will
> be (but are HOV lanes, for example, recorded separately in TIGER?).
>
> Etienne
> (Forgot to reply to the list again)
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20060407/824e49fd/attachment.html>


More information about the talk mailing list