[talk-au] Fwd: [OSM-talk] Proliferation of path vs. footway

[Liz]

SUMMARY
----------  Forwarded Message  ----------

Subject: [OSM-talk]  Proliferation of path vs. footway
Date: Tue, 11 Aug 2009
From: Lauri Kytömaa <lkytomaa at cc.hut.fi>
To: talk at openstreetmap.org

Nop wrote:
>I think we should step back one step.
>The discussion here seems about to fall victim to the same mechanisms


Trying to keep my comment general at first to find what are the needs:
what should be in the highway tag and what are "local factors". This
turned into a stream of thoughts but hopefully coherent enough to
breed some more refined thoughts.


Things that all agree on:

highway=footway:
Something, where walking is allowed and possible for someone.
(walking might be and is allowed and possible elsewhere, too)

highway=cycleway:
something, where cycling is allowed and possible
(even a German dedicated/signposted cycleway fits that description,
i.e. it's not a oneway dependency - not all things tagged
highway=cycleway are german signposted cycleways). Pedestrian access
undefined - might be country dependent but not supported (yet), so
there has about always been a suggestion in the wiki to always tag it
with foot=no/yes/designated.

highway=path:
something not wide enough for four wheeled vehicles OR where
motorvehicles are forbidden (unless otherwise indicated by
snowmobile/agricultural=designated or similar).

Anything with
wheelchair=no: unsuitable for wheelchair users or other mobility
impaired

Anything with
highway=footway + foot=no (+ snowmobile=yes) would be silly

highway=track
implies that it's wide enough for a small motorcar to drive on,
even if it's illegal.



Things that people don't agree on:

1) Is a highway=cycleway + foot=yes any different from a
highway=footway + bicycle=yes
2) Is it significant if there signs read "footway + bicycle allowed"
or "combined foot and cycleway" (presumably a difference in the legal
"maxspeed" at least in Germany)
3a) is a forest trail any different from a paved sidewalk
3b) is a forest trail any different from an unpaved but built footpath
4) is a constructed way with the traffic sign "no motorvehicles" any
different from a constructed way with the traffic sign "combined foot
and cycleway" (or with a cycleway-signpost in the UK)



User needs:
Pedestrian / Cyclist / Horse rider / Urban planner / Statistician /
Safety engineer / Accessibility analyst / Crime investigator ...

A pedestrian considers mostly the surface and the build quality of the
ways _allowed_ to him. A trail in an urban forest (picture 1), formed
by repeated use only, is not usable for an average pedestrian, even if
a normally fit person in sneakers would go for a walk there sometimes,
even if only to walk the dog. A mountain trail is effectively the same,
even if more difficult to use. Just about every person, even in (very)
high heels would walk down (picture 2) if the way hasn't turned into a
puddle of mud. And a western world way constructed for walking usually
doesn't deteriorate that much. Then there's the third variant
in-between (3), which some would use and other's wouldn't.

1) http://wiki.openstreetmap.org/wiki/Image:06072009(045).jpg
2) http://wiki.openstreetmap.org/wiki/Image:Path-motorcarnohorseno.jpg
3) http://wiki.openstreetmap.org/wiki/Image:Path-footyes.jpg

Some cyclist disregard access rights and consider the surface and hills
only, while others would want to drive on dedicated cycleways only; on
those where only cyclists are allowed. Most common cyclist probably
don't care if there are pedestrians involved, they just wan't to use
legal and properly built ways and avoid driving amongst the cars.


Horse riding is something to think about, too.

For signposted bridleways it's quite unambiguous, even if a British
bridleway allows pedestrians and cyclists, too, whereas the German
(and Finnish) legally signposted bridleways allow neither.

But on a built way signposted as "no motor vehicles" horse riding might
be legal, but if it's signposted as a footway, cycleway or the
"combined foot and cycleway" (picture 4), horse riding is not allowed.
4) http://wiki.openstreetmap.org/wiki/Image:Path-lighttraffic.jpg

On the forest trails (picture 1 again) horse riding might again be
legal or private/permissive. If the picture 2 didn't have the "no
horses" sign, I'd think around here that it's legal to ride a horse
there.

City planners possibly need to consider if the way is signposted for
combined use or with a "no motor vehicles" - first ones the city might
have to keep in good walking condition to avoid expenses when someone
breaks his bike because of the unfixed potholes but the latter ways
don't possibly carry such limitations. On the other hand that doesn't
usually interest the cyclists at all even if it is so.

This can and does have implications when dedicing where to build the 
light traffic ways in the next suburb to be built - or where to add new 
cycleways to improve the percentage of cycling commuters.

Statisticians and safety engineers could want to know whether
(un)segregated shared use paths have more fatalities or broken legs
(or wild angry goose or ice cream eaters) than some other ways allowed
to cyclists and/or pedestrians. They're interested in all the details:
surface, lit, hills, signs, segregation line, potholes etc. so that
they can calculate the significance factor for each variable.

For accessibility analysis the various signs (the shared use/only
cyclists) and such matter because of the implied allowed speeds
(they don't have specific numeric limits, but a shared use way
requires the cyclists to "drive carefully" (or something to that
effect) whereas the dedicated cycleways have no limits. Yet other
factors affect speeds, too.

Crime investigation could consider the legal status of the way after a
crash, what is or was the access restriction, as in was the pedestrian
allowed to be there or could he have been expected to be there - but
they'd use the city planning authorities data anyway or check the place
themselves. Analysis of potential escape routes could consider physical
obstructions only - where could the thieves have gone - but is unlikely
done on osm data (yet).

Conclusion:

Some users care most about whether it's a built way or not, others want
to know what the sign was (are there likely users of other transport
methods) and some care only "Am I allowed or not?"

Alv


_______________________________________________
talk mailing list
talk at openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

-------------------------------------------------------