Peter,

Very interesting work you have done here, the CSVs are particularly 
interesting.  Where did you source the Dept of Transportation data ?

We  (CloudMade) look at similar stats on a county level in the USA - and at a 
full country level in Europe and the rest of the world, taking a check point 
each week so we can look at trends.

I'd certainly be interested in what you are doing, and helping with any 
challenges you have.

All the best,

Jim

Jim Brown - CTO CloudMade


From: [email protected] 
[mailto:[email protected]] On Behalf Of Tristan Thomas
Sent: 27 July 2009 22:23
To: [email protected]
Subject: Re: [Talk-GB] Progress on estimating UK coverage

I'm new so haven't seen that before, but that's really good.  Now all you need 
to do is split it up even more!

On 27/07/2009 22:13, Peter Reed wrote:
Last week I posted a note about my efforts to compare actual road lengths (or 
at least Department for Transport statistics)  against the lengths of road that 
are now in the map for each local authority area in England.

This is a quick follow-up on the discussion that resulted, and an update on 
progress since.

Thanks to all the useful suggestions, I've now managed to find and fix a number 
of areas where my analysis was double counting.
I've also improved the way different highway tags are translated so that I 
match the DfT statistics more accurately.
I've discovered and cleared a few problems with the process that resulted in 
data getting lost along the way, so my figures are now more complete.

However, I still haven't managed to pick up all the admin boundaries that I 
should be getting, and I'm still using some ceremonial boundaries in place of 
the administrative ones.
That skews the results in places like Buckinghamshire and Kent, where roads 
within a unitary authority are still counted inside the neighbouring county.

Alongside fixing things, I've tried to tidy up my extract and load so that they 
are more robust and easier to manage (my steep learning curve on Postgresql and 
Postgis continues).

There is a first crude attempt at plotting it all here - 
http://www.reedhome.org.uk/Documents/OSMCover.png

This isn't a definitive result, of course - or even close to one. There are 
still gaps that shouldn't be there, and more checking to do.

Meanwhile, thanks for the encouragement, Pete








________________________________






_______________________________________________

Talk-GB mailing list

[email protected]<mailto:[email protected]>

http://lists.openstreetmap.org/listinfo/talk-gb


_______________________________________________
Talk-GB mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/talk-gb

Reply via email to