[OSM-talk] OSM cycle map - ?excessive focus on long-distance routes

[Richard Mann]

On Wed, May 9, 2012 at 3:09 PM, Richard Fairhurst <richard at systemed.net>wrote:

> But as yet I haven't understood what point you're trying to make in this
> thread. Without trying to be obtuse... can you explain?
>
> cheers
> Richard
>
That there are legitimate ways of classifying cycle routes other than for
touristic purposes (and it's not just me; it seems to be a known, if
unresolved, distinction in Utrecht).

OSM tagging of cycle routes seems dominated by the touristic approach, and
this limits the usefulness of the data if you're more interested in utility
cycling.

Looking at the Dutch guidance, they define a main cycle route as one that
has more than 2000 cyclists per day (other countries might settle for a
lower threshold!). These account for about 20% of the lanes/tracks, but
about 80% of the distance cycled. At that sort of volume, signposting is a
bit irrelevant; it's more down to observing the dominant flows of cyclists
(typically reinforced by above-average facilities, though not always). In
an ideal world, you'd do proper counts and derive the data from bottom up,
but given that it's usually pretty obvious, I think a certain amount of
duck-tagging is appropriate.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20120510/d1896dd8/attachment.html>