[OSM-talk] Fwd: Nav4All navigation shut down by Navteq

Paul Houle paul at ontology2.com
Mon Feb 1 18:14:52 GMT 2010


Igor Brejc wrote:
>
> I have to agree with Nop, up to a point. OSM is a great project and I 
> invest a lot of my free time in it, but I still think it has a lot of 
> failure points. The first time I wanted to use OSM data for a 
> professional job, the data simply failed me. And I'm only talking 
> about generating a high-scale UK map, not some complex routing 
> application. Even drawing land borders between England, Scotland and 
> Wales proved to be big PITA because of different approaches to tagging 
> between the three regions (not to mention that England's regional 
> boundaries were tagged the same way as the border with Scotland). I 
> don't whether this has been improved in the meantime.
>
(1) My experience is that commercial and government GIS map data is less 
than perfect;  particularly if you're crossing international borders.  
Even if you're targeting North America,  it turns out the whole 
architecture of commercial information (generally built on top of the 
national census) is entirely different in Mexico,  Canada and the US. 
Project #2 in the pipeline is a fundamentally international GIS system 
that wouldn't be possible at all w/o dbpedia,  geonames,  freebase,  
Yahoo's shapes,  and Open Street Maps;  unless I had the kind of 
resources that Google has.

(2) A big part of the Web 2 -> Web 3 transition is going to be finding a 
way to clean up the tagging morass.  When it comes down to it,  tags 
suck for two reasons:  (i) they're a lot of work to create,  and (ii) 
people don't use them consistently.  There's are many approaches,  but 
it's one of the biggest problems that I see in front of me personally.




More information about the talk mailing list