[Tagging] cyclist profiles - was:Feature Proposal - RFC - value 'basic_network' - cycle_network?

Thu Dec 9 10:56:13 UTC 2021

For an example of which I think is bad mapping of an (on the road)
unverifiable collection of roads (though the operator calls it a
"functional cycling network"):
https://www.openstreetmap.org/relation/9857490

There is hardly any ground truth, there are no specific guideposts, the
streets have no recognizable attributes to indicate why they are chosen.
There is no system to guide the cyclists, so there are no routes. Without
external information, this "network" does not exist; it's just a collection
of roads chosen by some authority for a map and a dedicated app, and the
preference only exists on that map and for that app. They have made this
collection of ways type=route, only for this rendering and this routing
app.

IMO this does not belong in OSM. The collection and the external
information should be kept outside the OSM database and only combined with
OSM for the app and its rendering. *)

By contrast, the German "basic network" is easily recognized and verifiable
on the ground, and offers systematic guidance and channelling of cycle
traffic over verifiable dedicated chains of ways. This can be mapped and
tagged without any outside information (as usual you need some local
knowledge or understanding).

*) At least, according to the current OSM paradigma. I can see the added
value of mapping this kind of thing, and I can imagine a paradigma where
"value added objects" could be stored within OSM fo this kind of
application, where the requirement of ground truth is replaced by the
requirement of a verifiable data source and "truth operator". Just
brainstorming. I think it would then be fair to expect a regular donation
for this kind of usage.

Peter Elderson

Op wo 8 dec. 2021 om 00:00 schreef Sebastian Gürtler <
sebastian.guertler at gmx.de>:

>
> Am 06.12.21 um 09:56 schrieb stevea:
> > If true, what Martin says here seriously dilutes (in my mind) whatever
> it is that these signs denote.  Do the signs denote “roads which are
> suitable” (by pointing “away” from them)?  (For bicyclists and hikers?). Do
> they denote roads which are MORE suitable?  (By pointing “towards” them?).
> Is it really true that roads which are NOT so denoted are “LESS suitable”
> or “NOT suitable?”  (For bicyclists and hikers).  Or not?
> >
> > “Around here” (in my state and country), I have the right to demand an
> answer put directly to the (government, organization…) which erected a sign
> on a public roadway, to the question “What, exactly, does this sign mean?”
> And, "who put up these signs?"  And "what is it that THEY claim these signs
> actually mean?"  Is this written down anywhere?  (In my country, we have
> MUTCD, the “Manual on Uniform Traffic Control Devices.”  If your sign isn’t
> in that Manual, it is de facto a “non-standard” sign).  These signs in
> Germany either ARE or ARE NOT in compliance with Germany’s “signage
> standards,” whatever those are.  If they ARE in compliance, there is a
> concrete definition of exactly what it is they denote.  If possible /
> available, I’d like to read that here, please.
> >
> The manual what is needed is state dependent: e.g. for North
> Rhine-Westphalia (NRW): I just put into a translator as I am not able to
> translate this special legal German into the prop    er English terms
> (as I suspect the translator may also be not...). Published by the
> ministry of transport NRW.
>
> Actual rules here: https://www.radverkehrsnetz.nrw.de/rvn_hbr.asp
> "Hinweise zur wegweisenden Beschilderung für den Radverkehr" = "Notes on
> signposting for cycling traffic"
>
> You have the descriptions of the signs and the concepts here:
> https://www.radverkehrsnetz.nrw.de/downloads/HBR_NRW_Kap03_Jul2019.pdf
> ("Das Wegweisungssystem NRW" = "The signposting system NRW"), Contents
> e.g.: "Systematics of signposting: destination-oriented and
> route-oriented signposting", "Destination-oriented signposting: Signpost
> content, Signpost sizes, Signpost shapes, Selection and order of
> destinations", "Route-oriented signposting", "Supplementary orientation
> aids: Maps; Integration of local destinations; Guidance on a themed
> route; Route km marks on cycling fast links", "Interfaces to other
> signposting systems: Adjacent cycling networks, Node system, Mountain
> bike signposting"
>
> And on the legal status here:
> https://www.radverkehrsnetz.nrw.de/downloads/HBR_NRW_Kap04_Sep2017.pdf
> "StVO-Status der Radverkehrswegweisung in NRW" "'Road Traffic Act'
> status of cycling signposting in NRW"…
>
> here: Auszug  aus  dem  Erlass  des  nordrhein-westfälischen
> Verkehrsministeriums  vom 03.08.2000: "Zur  Ausschilderung  des
> landesweiten  Radverkehrsnetzes  sind  wegweisende Beschilderungen
> entsprechend dem von der Forschungsgesellschaft für Straßen- und
> Verkehrswesen (FGSV) herausgegebenen "Merkblatt zur wegweisenden
> Beschilderung für den Radverkehr" auszuführen.
> Gemäß  § 46 Abs.  2  der  StVO  erteilte  ich  die  Genehmigung,
> abweichend  von  den Regelungen des § 42 Abs. 8 StVO, die Radwegweisung
> künftig nach dem o.g. Merkblatt auszuführen. Als Regelfarbe für die
> Schrift ist Rot zu verwenden. Die im Merkblatt aufgeführten Wegweiser
> unterliegen damit den Regeln der StVO und bedürfen der Anordnung durch
> die Straßenverkehrsbehörden."
> Der Erlass bezieht sich damit nicht nur auf die Schilder des
> Landesweiten Radverkehrsnetzes, sondern auf die gesamte
> Radverkehrswegweisung (alle lokalen, regionalen und touristischen
> Routenbeschilderungen) in NRW mit entsprechend gestalteten Wegweisern.
> (Gilt für alle Wegweisungen nach Merkblatt)
>
> Excerpt from the decree of the North Rhine-Westphalian Ministry of
> Transport of 3 August 2000: "For the signposting of the state-wide
> cycling network, signposting is to be carried out in accordance with the
> "Leaflet on Signposting for Cycling" published by the Road and Traffic
> Research Association (Forschungsgesellschaft für Straßen- und
> Verkehrswesen, FGSV).
> Pursuant to § 46 (2) of the StVO, I hereby grant permission, in
> deviation from the provisions of § 42 (8) of the StVO, to implement the
> bicycle signposting in future in accordance with the above-mentioned
> information sheet. Red is to be used as the standard colour for the
> lettering. The signposts listed in the leaflet are thus subject to the
> rules of the StVO and require the order of the road traffic authorities."
> The decree thus refers not only to the signs of the state-wide cycling
> network, but to all cycling signposting (all local, regional and tourist
> route signposting) in NRW with appropriately designed signposts.
> (Applies to all signposts according to the leaflet).
>
> StVO=Straßenverkehrsordnung somewhat to be translated as e.g. "Road
> Traffic Act".
>
> ==
>
> I'll reduce my activity in this discussion for I just haven't that much
> spare time for participating adequately in it, still interested in
> possible results and consensus.
>
> And I just really recognize that the topic changed.
>
> I am by no means interested in tagging cyclist profiles but simply the
> routes I see on the ground. I tried to describe what can be found here,
> did some research on the official guidelines (which have much
> ambiguities concerning the purpose), and will mainly go on in correcting
> the existing mappings in a way that easily allows changing the tagging
> to any other scheme, without destroying any existent differentiating
> information.
>
> This is not possible with the taggings "network=bcn" for you have to
> remove preexisting "network=lcn/rcn" tags. It is also difficult with the
> tag cycle_network which is sometimes used but without a clear scheme,
> the adding of values with ";" is not really in use here, so it would
> raise discussions if I'd start it now.
>
> And - I don't want to wait, for I accumulate the information on the
> cycling network at this moment - and I won't ever introduce the
> information into the data base if I don't do it now. The map information
> is just a side effect of my cycling, nothing more. (Just looked it up:
> since 5/20 when I started doing that I rode about 4000 km on these
> network routes in my surrounding taking about 3000 photographs of
> guideposts and route markers. Not possible for me to go through the data
> afterwards...)
>
> A big part in my actual activity: deleting abandoned routes in osm and
> unify different taggings for the same kind of routes, which all need a
> visit on ground for there are no reliable sources elsewhere. That's my
> main reason for the creating of relations: I can tag them with a
> survey:date so it is visible for anyone what data is up to date and what
> is not. This is not possible (or too much work) with tagging at the ways
> for you need a complex scheme for the survey:date to tell which tags you
> have surveyed and which not. I usually have node(=branch) to node routes
> (sometimes complex sometimes simple) and a relation containing that and
> can tell exactly at which time it has been surveyed.
>
> There are a lot of road construction sites, new motorways etc. Anyone
> interested in keeping the network up to date can easily check whether
> the situation described in the data is already updated or is needing
> another survey.
>
> @steve: asking me for a concept... I didn't make the proposal, in fact:
> I even have some problems with the concept that the state follows with
> its guidelines and signposting, there are sometimes inconsistencies and
> additionally slight differences between the states. I think that's one
> of the problems that we have with mapping.
>
> But we have an easily visible fact on the ground: a germanwide
> recognizable set of specific signs for guiding cycle routes (which we
> don't have for hiking, motorized or other traffic), the main part is
> just the simple arrows in red or green (depending on the state) which is
> used for any official cycling route (and not for additional routes
> operated by other public and private institutions).
>
> I don't have a real good concept how to integrate the situation visible
> on ground into the osm database: It is ...
>
> ... difficult to transform the many preexisting different descriptions
> of the same thing into something more uniform
>
> ... difficult to integrate the different opinions and views on this (we
> have discussions ranging from political via linguistic just to simple
> technical aspects on it)
>
> ... difficult to cope with still visible overlapping old routes (e.g.
> older than 40 years) and the newer guideposting (starting about 20 years
> ago) and resulting inconsistencies.
>
> I think the most probable way to cope with it are relations with a tag
> in cycle_network that's based on a German wide consensus. (Big
> advantage: You won't need nested relations as they are in use in NRW now
> - these have in fact the character of collections: The whole network
> would be in https://www.openstreetmap.org/relation/33216, now aged 13
> years, and even then a first try to unify something older).
>
> Sebastian
>
>
> _______________________________________________
> Tagging mailing list
> Tagging at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20211209/478f7cda/attachment.htm>