[OSM-talk] highway=cycle&footway

David Earl david at frankieandshadow.com
Tue Mar 24 08:17:11 GMT 2009


On 24/03/2009 07:25, Ed Loach wrote:
>> I think it is perfectly obvious in the UK, it's a cycleway if
>> it has the
>> blue cycle sign indicating that a surface is permitted for
>> cycling when
>> it otherwise not be.
> 
> My highway code defines that sign with a round background as "Route
> to be used by pedal cycles only" - so I'd have thought
> highway=cycleway, foot=no. The rectangular background version
> "Recommended route for pedal cycles" is a little less clear, though
> I'd probably also use highway=cycleway depending what else the route
> was like.

I wasn't being precise enough in specifying which signs I meant (and 
again, this only refers to the UK).

I mean the first 3 signs at the top of page 36 of:

http://www.dft.gov.uk/pgr/roads/tss/trafficsigns.pdf

which includes the one you're referring to, but also (perhaps 
especially) the two shared use ones. It is true that one is "route for 
pedal cycles only" but except in the case of cycle tracks (see 
yesterday; there are virtually none of these), these are now used almost 
exclusively to indicate by-passes to no entry signs, and usually have a 
footway adjacent, just as you would on any ordinary road where you also 
have a footway and wouldn't mark one specially. So in nearly all cases, 
cycleway can include pedestrians.

Paths away from roads need not always have the round signs because the 
same blanket rules don't apply, so if something is explicitly signed as 
per top left of p113 and similar, I'd mark a cycleway unless there are 
additional signs indicating other classification (such as a bridleway).

My point then was that it is only in rare cases you need a plethora of 
tags to indicate exceptions to this general rule expressable as a single 
tag. highway=cycleway covers nearly all cases on its own; just 
occasionally extra information is needed. Sometimes the evidence on the 
ground is contradicted by local knowledge of the official status (as per 
the discussion about the A420, was it?, also in Oxford), in which case 
designation is appropriate.

In other jurisdictions, what is and isn't allowed would be different, 
just as the classes of vehicle allowed to use a road marked 
highway=motorway differ, but the same principle applies.

It's what we've all been doing, explicitly or implicitly, for a long 
time and I see no good reason to change it.

Obviously I haven't seen the original evidence on the ground in Oxford, 
so I'm not qualified to say. From what Richard originally said, it 
sounded like it was ksigned as a bridleway, but he now says there is 
other evidence. If that's just tyre tracs, or that lots of people do 
cycle on it, that doesn't for me override the sign. But whatever the 
conclusion, if it is marked as a cycleway, it clearly needs extra tags 
to note the exception for horses, whereas if a bridleway it doesn't.

Surface quality and so on are entirely orthogonal attributes. Personally 
I'd use them again where it is unexpected: bridleways aren't usually 
hard surfaced, so I'd be inclined to mark one that is; vice-versa of 
bridleways.

But in the end it's all down to personal preference, and where opinions 
differ on the ground then people are going to have to negotiate. 
Personally I'd like to see tighter definitions, but given the way this 
list operates, I see no prospect wahtsoever of being able to achieve 
that, so the best we can hope for IMO is a broad consensus with 
haphazard exceptions especially in the detail. The trouble is, the more 
people go down their own favourite tagging schema, the harder it is for 
data consumers to make sense of the data.

David




More information about the talk mailing list