[Tagging] Feature Proposal - RFC - value 'basic_network' for keys 'network:type', 'lcn' and 'lwn'

[JochenB]

Thanks Volker for the list of aspects.

My proposal actually only relates to one aspect. But it is already clear
to me that I have to revise it and that I will describe it in a more
abstract way.

In the proposal, I am referring to pedestrian traffic and cycling. I
think a definition of a basic network is not necessary for car traffic,
as there are no corresponding signs or they are already mapped on the
highway tags.

> We do not have a scheme for signposted non-touristic bicycle routes,
> _...__
> _
There is already a method of mapping these networks that is widely used.
They are mapped either in route or in network relations with
/'route=bicycle'/. In some regions a /'lcn=yes'/ is simply written to
the paths without any further information. The discussion here seems to
me at some points as if it didn't exist yet and we were in a green,
undeveloped meadow. But that's not the case where I'm on the move. There
are only two problems with this.

The first problem with the previous tagging schemes is that there are
two schemes for relations: Use of network relations (seldom, use has
largely fallen asleep) and route relations. Here my suggestion would be
to use route relations.

The second problem is that in route relations it has not yet been
possible to differentiate between route recommendations and the rest of
the hiking / cycling network. Hence the proposal to indicate to the ways
or relations of the hiking/cycling network that they belong to the
hiking/cycling network.

My proposal does not contain anything else at first.


> _We do have_ ways of tagging road infrastructure (ways, and nodes like
> obstacles and crossings) in such a way that a navigation/routing
> algorithm can find suitable routes, based on a set of technical
> criterions (road classification, surface, smoothness. track type,
> speed limits, ...). _But we do not have_ a tagging for generic "this
> is road suitable for bicycles" signs, as they exist e.g. in Germany,
> and in the USA.
> My understanding of the discussion is that the concept of the proposed
> "basic network", is a subset of roads that fulfill the criterion of
> being bicycle-friendly and are inter-connected.

I would like to highlight one difference. With the proposed tag, I want
to mark that a way is officially designated as part of the hiking /
cycling network. It's not about giving personal assessments as to
whether you can hike or cycle here. There are concrete measurable things
that can be used to determine whether a route / a relation should
receive this tag: the standardized signage.

One can, of course, expect these trails to be hiking / cycling-friendly.
But that doesn't mean that they really are. There are the individual
properties that describe this, e.g., /'surface'/ and /'smoothness'/.
I've already seen bike signs in the middle of the finest sand that I
couldn't even push the bike. As a rule, however, this expectation is met.


> _We do not have_ ways of explicitly tagging the "beauty" of a road
> (something like the green line on the once famous Michelin 1:200000 maps)

In the German bicycle signposting system, there are sometimes symbols
for this on the destination signposts. So there is something measurable
out there. This information can easily be tagged to a relation of the
basic network. I don't know anything like that in the network of hiking
trails.


> _We do not have_ an approved way of mapping cycle routes that are
> carrying you from A-town to B-town with consistently installed
> signposts "A-town>" and "B-town>", which we have also here in Italy
> (even though maintenance of such signposting is often uncertain)

I would have mapped it as a normal route relation with
/'route=bicycle'/. If they consistently identify the same destination,
they can be provided with /'from=*'/ and /'to=*'/. What are you missing
there?


> but I am still not convinced of the "basic-network" tagging approach
> proposed here.

What exactly does not fit? How would you differentiate the network with
signposting from the route recommendations?

I learned from other comments here that the tag /'network:type=*'/ is
unfavorable because /'type'/ does not contain any information about what
is classified here. That was the case before my proposal, but I can of
course propose a completely new tag like that:

    route:purpose=basic_network/route_recommendation

or simply:

    basic_network=yes

Is it better?

Regards,
Jochen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20211121/5dab30fa/attachment.htm>