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

[stevea]

On Nov 27, 2021, at 2:03 PM, JochenB <JochenB at wolke7.net> wrote:
> Am 24.11.2021 um 09:59 schrieb Yves via Tagging:
>> 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.
> 
> The aim of the proposal is not to define new relation types. It is only
> a matter of making existing relationships distinguishable by means of an
> additional property.
> 
> This information cannot be derived from the properties of the ways. You
> have to be mapped by us either to the ways themselves or to the relation.

Yet again, I "type out loud" here why this "additional property" cannot be captured by a widely-agreed-upon value (or "sub-value" might be more appropriate, if "value" is "roughly speaking" something like "the network at this nation/state/city's level") of the cycle_network=* key.

Start with the "design work" of taking every single network in Europe and craft a syntax for these values of cycle_network.  Sure, "on its outer layers" this will be geographic in nature (nation, state, locality...) because IT ALREADY IS (the way that the rest of world has chosen to use the namespace of cycle_network=* values).  Then, modify values WITHIN these "outer layers" to perhaps contain some "inner layers" which denote any 'basic_network' which exists in that region.  If that is at a state level, assign a "sub-layer" to that state's bicycle networks that is that state's bicycle BASIC network.  I'm repeating myself because this hasn't been addressed (or if it has, not to my understanding) as a workable solution.  I'm coming full-circle to why cycle_network can't be used as this "value for the additional property" you wish to assign:  it meets all your criteria noted above, AND it fits into the method by which the rest of the world has figured out how to do this and denote it in OSM.

Flips:  THAT is "why the f*** stevea is insisting so hard."  And, I think you are on the right track, as your suggestion does recognize that "certain routes belong to particular a (or multiple) particular NETWORK(s)."  That is exactly what the cycle_network=* key (already) does.  Choose values that everybody can agree upon:  done.  See, that's what CAN be difficult, or seems to be perceived as difficult, but isn't necessarily so difficult, it "merely" requires a good design of the cycle_network=* value namespace for your areas of the world as they need to map your areas of the world's networks.  Yes, lots of Contributors will have to agree among yourselves what those values are to be, but that's called "good design results in consensus."  There are plenty of "proofs of existence" of this without inventing new key extensions, it's all done in cycle_network, because good design was done in the namespace of its values.