[OSM-talk] Argh! - Wholesale deletion of foot|horse=yes tags

[Andy Allan]

On 11/3/07, Andy Street <mail at andystreet.me.uk> wrote:
> On Fri, 2007-11-02 at 17:20 +0000, Nick Whitelegg wrote:
> > Just noticed that in a very large area (about 10x10 miles) to the
> > northeast of Southampton, the foot/horse=yes tags have been removed from a
> > large number of footpaths and bridleways. Result is that Freemap is
> > showing them magenta (permissive) rather than red (official).
> >
> > Sorry to be picky but if this was intentional, please leave them in :-)
> > They're essential to distinguish between official and unofficial rights of
> > way.
>
> Yes, I noticed this too. I contacted the user concerned via the OSM
> website earlier in the week who informed me that they had removed the
> tags because footways were implied foot=yes (and likewise for
> bridleways) and therefore redundant.
>
> My understanding of the ROW tags has always been that if a tag is absent
> that we have no knowledge of the access permissions and the end user of
> the data should use a default which is sensible for their application.
> Is this correct? How do other mappers interpret these tags?

To be honest, I don't see how you can have a bridleway where anything
other than horse=yes makes sense - that's the very essence of a
bridleway. If the use of horses was either permissive, or forbidden,
then it's hardly a 'bridleway' in the first place, so I would expect
horse=permissive and horse=no to apply to other highway types, such as
footway.

But this is perhaps again an indication that we need to separate out
the physical and legal characteristics of a route more clearly.

Cheers,
Andy