[OSM-talk] FW: highway=cycle&footway

[Mike Harris]

Thanks Richard - my oops!


  _____  

From: Richard Mann [mailto:richard.mann.westoxford at googlemail.com] 
Sent: 26 March 2009 15:37
To: Mike Harris
Subject: Re: [OSM-talk] highway=cycle&footway


Oops, slipped off including "talk". I've forwarded my last to the list; you
may like to do so also...
 
Richard


On Thu, Mar 26, 2009 at 3:11 PM, Mike Harris <mikh43 at googlemail.com> wrote:


Richard - again helpful - after reading your comments I think the main area
of disagreement between Dave and me is the around the use of highway=path.
 
I am sure I have read somewhere (wiki, mailing lists?) a fairly strong plea
to minimise the use of highway=path whenever something more specific (such
as highway=footway) is available? Perhaps someone has a better memory than I
do?
 
This is one of three reasons why I have tended to favour highway=footway for
ways that are clearly unsuitable for more than pedestrian traffic.
 
By observation of the developing map itself - and also from the mailing
lists - a second reason might be that there seem to be two schools of
thought around the meaning of 'path': those who regard it as something less
well-defined on the ground than a 'footway' and those - apparently like
yourself - who see it as something 'more than just a footway'. I've taken
the middle course of avoiding it wherever possible -- at least where there
is an alternative tag for which there seems to be more consistency in
established practice - and keeping =path for vague paths that are 'there'
but are not public footpaths. Maybe I'm wrong! But who's right?
 
My third reason for avoiding highway=path is that someone could walk a route
one day and find that the path has not been reinstated across a ploughed
field or a crop. If this is a public footpath, pressure (and ultimately
legal action) will be used - sooner or later - by the highway authority. The
landowner may then reinstate and the path may become very clear indeed -
even a day or two later. Also, in many cases a farmer is allowed a grace
period (conditions too complex to matter here!)  before reinstating. So a
judgement based on lack of reinstatement - as Dave seems to suggest - while
objective, may be very ephemeral - and I'm hoping our maps are of lasting
value!
 
As I've already said, I'm in agreement with Dave on several of his points
and am pretty much in agreement with the points that you are now making
(other than on =path). I would certainly vote against highway=cycle&footway
as this can be done with foot= and bicycle= - as seems usually to be
existing practice. I would also probably vote against highway=cycleway +
cycleway=shared as I can expect arguments galore as to whether it is
highway=cycleway cycleway=shared (cycling viewpoint) or highway=bridleway
and bridleway=shared (equestrian viewpoint) etc. ad nauseam!
 
I still feel that cycleway is only well-defined in a limited set of cases
that I have mentioned earlier (with the usual grey area round the edges! -
of the definition, that is, not the cycleway (:>)) and that beyond these
cases the use of this tag does indeed tend to become somewhat subjective
according as the mapper is primarily a cyclist, walker or horse rider!
 
Mike (Cheshire)


  _____  

From: Richard Mann [mailto:richard.mann.westoxford at googlemail.com] 
Sent: 26 March 2009 12:58
To: Mike Harris 

Subject: Re: [OSM-talk] highway=cycle&footway


Before we all get too depressed, I think I agree with both of you (Dave /
Mike) that any changes to tagging should be backwardly-compatible, as far as
practical (or at least minimise the "wrongness" if the old tagging is
unchanged).
 
But we also need a scheme that is simple, effective and shows what's on the
ground, not just what's on the sign.
 
I think the nub of it is the tagging of path/bridleway/cycleway. I think
"path" serves a useful function for ways that are more than just footways,
but where usage/access for horses/mtb/bicycles is uncertain. I think
"bridleway" serves a useful function in those countries where access for
horses is well-established (and thereby is becomes a useful shorthand for
highway=path+designation=public_bridleway), but in practice there may be
little to distinguish a bridleway from a path (and there might be sense in
rendering them quite similarly). 
 
Whereas, highway=cycleway is an explicit assertion that the surface is
somewhat better than you might expect on a bridleway/path, without going
into the minefield of the multiple values that might be tagged for
tracktype/surface/smoothness.
 
I think I'm concluding that highway=cycle&footway is unnecessary; perhaps
highway=cycleway+cycleway=shared would be a better bet (and leave it to the
renderers whether they do anything with that). But if highway=cycleway is to
be used for shared cycleways, then the wiki definition will need to be more
inclusive than currently.
 
Richard (West Oxford)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20090326/20e88679/attachment.html>