[Talk-GB] Q3 2020 Quarterly project Cycle Infrastructure

Martin - CycleStreets list-osm-talk-gb at cyclestreets.net
Wed Jul 15 14:18:51 UTC 2020

On Tue, 14 Jul 2020, Adam Snape wrote:

> I have utmost respect for cyclestreets but that tagging guidance does 
> seem garbled at points.... 

Apologies; I think I was very tired when I wrote it. It was mainly intended 
as a starting point, to set out the ideal case of having those metadata 
tags present, but things like surface should have been better written.

I've fixed up the points noted, which I agree with. Obviously I hope others 
can enhance the section too.


> Since when has the segregated=yes/no tag on a cycleway referred to the 
> physical separation of cycle routes from the main carriageway rather than 
> the separation of cycles and pedestrians on the cycleway?

Sorry, yes, fixed.

> Quite agree, whilst harmless oneway=no seems a bit OTT, as tbh does 
> marking the surface on every single asphalt cycleway...

Have fixed this also. The intention was to ensure that the surface is 
considered when tagging - which is suprisingly still poor data in some 
places. A fair proportion of route feedback we get comes down to cases 
where routing has gone over a 'cycleway' that turns out to be some kind of 
muddy or badly-surfaced track. These are obviously easy to fix in OSM once 
the value is known.

My general feeling on surface is that, while asphalt is of course assumed 
by all routing engines I'm aware of, the amount of stuff in the UK that 
isn't asphalt makes it worthwhile putting the surface in explicitly. This 
demonstrates to future mappers that the value is actually known (rather 
than assumed/unknown/ambiguous).

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

More information about the Talk-GB mailing list