[OSM-talk] bulk loading towns/cities locations

Peter Miller peter.miller at itoworld.com
Sun Nov 4 10:06:57 GMT 2007


> 
> Message: 3
> Date: Sun, 04 Nov 2007 17:47:18 +1100
> From: "Brent Easton" <b.easton at exemail.com.au>
> Subject: Re: [OSM-talk] bulk loading towns/cities locations
> To: "Paul" <mylists at wilsononline.id.au>,	"OSM Talk List"
> 	<talk at openstreetmap.org>
> Message-ID: <200711041747180531.06C2F175 at smtp.nsw.exemail.com.au>
> Content-Type: text/plain; charset="us-ascii"
> 
> Hi Paul,
> 
> There are thousands of towns not loaded in Australia!
> 
> The good news is that there is a free set of data available from the
> Geographical Names Board that specifies every named location in Australia
> and specifies a Lat and Lon.
>


What about a bulk upload for the UK? It looks as though Wikipedia was
initially loaded with UK places from some suitable source. There is a
Wikipedia list of towns in the UK which all have grid references which came
from somewhere. http://en.wikipedia.org/wiki/Category:Towns_in_England


Has anyone considered doing a bulk upload for the UK (not just towns, but
all places? Is there a suitable source? Possibly one only loads places where
there isn't something in OSM with the same name within 10km already.

> The bad news is that the data is only recorded to the nearest minute (from
> memory) which means that each name can be up to about 1.8km from it actual
> location, and all location with 1.8km of another end up in the same place.
> 
> We can load this data up fairly easily. There is a script that was used to
> load the equivalent data from Ireland.
> 
> The question is how useful this is, given the (in)accuracy of the data. It
> depends on what the Australian Community think.
> 
> I can see some advantage to this (if the existing names in Australia where
> removed before the upload) to give a rough location of towns, but have
> always been hesitant because of the innacuracy. It also includes locations
> for all suburbs in cities, and these will just end up piled on top of each
> other .
> 
> Regards,
> Brent.





More information about the talk mailing list