[Talk-GB] DfT Cycling data - cycle lanes (quality)

[Martin - CycleStreets]

On Tue, 9 Oct 2012, Dave F. wrote:

> This is why I've largely ignored the DfT data. It just isn't accurate 
> enough to blindly copy into OSM ad hoc. I enter cycle lane data such as 
> this only when I have been to an area & actually seen it, which renders 
> most of the DfT info worthless.

The intention certainly is that people should be merging it only with local 
knowledge.

However, I've certainly found it pretty accurate (well above 90%) in the 
areas I know and have merged in. Cambridge for instance has a lot of 
infrastructure, and the data here is good, despite being one of the 
earliest to be surveyed 3-4 years ago.

Which area(s) have you found to be inaccurate, out of interest?


> I've even seen data for NCN 4 that's illegal where it suggests a right 
> hand turn but is explicitly denied by signs.

There aren't any turn restrictions in the data as far as I'm aware.


Martin,                     **  CycleStreets - For Cyclists, By Cyclists
Developer, CycleStreets     **  http://www.cyclestreets.net/