Dem Edit ist einige Diskussion vorausgegangen
u.a. mit denjenigen, die die Wiki-Seite erstellt haben.
Pls. don't do that because the way currently hiking routes are tagged is broken.
Yes, there has been a lot of discussion and from this it should have been clear that there is no unanimous support for the proposed changes and many questions were left unanswered. What is it that is currently broken?
One argument was that in regions with few named guideposts, too many relations would have to be created with the established guidelines. But now a lot of base network hiking routes have been imported in Graubünden based on these "new" guidelines with an extreme fragmentation of relations that often consist only of a single way of a few hundred meters [1]. How is that an improvement? It's easy to create data this way when copying from the cantonal GIS, but this is difficult to maintain through on-the-ground surveys because of the lack of metadata.
It also shows a contradiction of the proposal: from= and to= tags were proposed to identify the start and end point of a route, which I support. But how would these be used in the case of the stub relations in Graubünden [1]? Should names be invented and how would these then be distinguished from actual guidepost names? If the proposal was strictly followed, routes would almost never start at a named guidepost because it is very common that multiple routes start at a named guidepost and then split off shortly after that at an unnamed one as we saw in Glarus [2]. Something mentioned again and again is that a way should not be in more than one base network relation, so I ask once more, why is that a problem? Ways can be members of many other relations anyway and if multiple routes follow that way it can certainly be mapped like that.
As I said before, I'm not strictly against adapting the tagging guidelines for regions in which the signposting is not yet done according to the federal guidelines [3], e.g. Graubünden. But this must not involve invalidating the current tagging for other regions where it works and is widely applied already.
If some of you prefer to have a personal discussion on this topic, we can arrange a time for that, although I'd prefer decisions to be made publicly through the mailing list or wiki.
Best,
Enno