[Tagging] Feature Proposal - RFC - value 'basic_network' - cycle_network?

[Sebastian Gürtler]

Am 23.11.21 um 11:16 schrieb Volker Schmidt:
> One important thing (at least for me) that is emerging from this
> discussion, is an underlying, not documented, assumption that I am
> making, and that i shared most likely by all of my OSM cycling friends
> in Italy are sharing, but that some people in this discussion are
> explicitly not sharing:
> bicycle (and hiking ) routes in OSM are touristic routes. They are not
> geared towards "the safest and fastest connection".

It is not a matter of view of OSM users or cycling friends, it is from
the official guidelines. If you want to crop one homogeneous network
just to the "touristic" routes because OSM would only tag touristic
routes (why???) then it's exactly what Jochen tries to do, to get an
additional representation for the remaining routes.

I just have a source for North Rhine-Westfalia, published by the
ministry of transport of NRW, but they claim to cite the Germany wide
guidelines.
https://www.radverkehrsnetz.nrw.de/downloads/HBR_NRW_Kap03_Jul2019.pdf

"Gemäß dem Merkblatt zur wegweisenden Beschilderung für den Radverkehr
der FGSV (vgl. Kap. 1.3) kombiniert die Radverkehrswegweisung in NRW die
•zielorientierte und
•routenorientierte
Wegweisung miteinander. Sie wird daher sowohl den Bedürfnissen des
Alltags- als
auch des Freizeitradverkehrs gerecht (vgl. FGSV-Merkblatt, Kap. 2.1)"

"According to the FGSV's information sheet on signposting for cycling
(cf. chapter 1.3), cycle signposting in NRW combines the
- destination-oriented and
- route-oriented
signposting with each other. It therefore meets the needs of both
everyday and leisure cycling (cf. FGSV leaflet, chap. 2.1)".

This is not a matter of taste, it is just the situation we meet. Me, I'm
not very fond of splitting the network logically, I'd prefer to have a
unique tag for describing routes of the network as part of the official
network or compliant to the guidelines. They are characteristic and easy
to recognize on the ground, quite similar in whole of Germany. For this
I could accept easily to use a cycle_network=DE:official or similar. You
could tag also touristic routes if the follow completely the official
guidelines (which some don't!), and you could simply have some routes
without a name=* tag which were the "network:type=basic_network" routes.

The advantage of Jochens suggestion is to fill the established (yes, in
middle Europe that's true) key network:type with a second possible tag
(while if I look at taginfo, it seems to be an invitation to confuse it
with cycle_network=*). cycle_network is only rarely used worldwide at
all as I see. You have 173k of network type (167k of them node_network)
and only 19k with cycle_network=* at all.

And finally, according to the wiki cycle_network "indicates the specific
route network to which a route
<https://wiki.openstreetmap.org/wiki/Key:route>=bicycle
<https://wiki.openstreetmap.org/wiki/Tag:route%3Dbicycle> relation
belongs", not a characterization of the guideposting system. Tagging
cycle_network=DE:official seem to me like a bit abusing the tag for
another purpose.

Am 24.11.21 um 09:59 schrieb Yves via Tagging:
> Le 24 novembre 2021 05:10:37 GMT+01:00, Adam Franco <adamfranco at gmail.com> a écrit :
>> how about defining a simple tag/tags
>> that can be applied to highway ways that are officially recommended?
>> `bicycle=designated` is an access rule, but maybe something like
>> `bicycle_usage=recommended` or any other keys/values that get at the heart
>> of what is being recommed.
You won't be able to reconstruct that these ways are parts of specific
node to node connections in one network. And finally I prefer route
relations (not as collection but single linear routes) for maintaining
purposes. It is quite difficult to reconstruct the age of a route only
by changesets, for changes can be applied to other aspects of the ways
not only the overlying route. And finally bicycle=* tells something
about the legal possibility for cyclists which is another scope.
> Following this thread, I also fail to see why not. 'cycleway' comes to my mind.
cycleway=* is used completely differently, it has nothing to do with
bicycle routes but describing the material infrastructure concerning
bicycle traffic. That's not the topic in this discussion.
> First relationships in OSM are that ways are connected together. Any cycling tool out there can take this into account to build routes, networks, or whatever representation showing the data in a meaningful way.
> Not to say relations are not needed, but trying too hard to make a new relation type for each and every model fitting the views of a particular set of users or coordinators seems useless.
> Yves
It is not about routing one random way suitable for bicycles but to
describe a visible network as it is given on the ground, no matter the
specific quality of  this in each single case.

Sebastian

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20211124/170ba0a9/attachment.htm>