> Ok, I hope this will be my final post in this long thread. I will try to > summarize what I understand from the discussion as the main issuesa and > what needs to be addressed to make it easier for mappers and data > consumers. > As said before, whatever you do to the tagging it will life more difficult for data consumers, because you cannot redo the existing hiking network in OSM. We may work on explaining better to mappers (wiki) how to use the *existing* zoo of tags.
> I would also suggest that instead of filling the inboxes of each and > everyone on this tagging list, we create a smaller "working group" that can > come up with a concrete suggestion to solve the major issues. What do you > think about that? Who would like to work with such a proposal? > I don't know if we need a working group, we should in any case document what we are doing in one of the existing formats: wiki discussion page; mailing list, forum. Working group sounds nice, but we would need to be extremely careful with documenting the discussion Your list below is missing the most important goal: compatibility with the existing data. It is extremely important to take into account that there are parts of the world where most of the hiking paths and tracks are already mapped. That mapping can be improved both from the geometry point of view as well as adding information, but always using the existing set of tags. I am happy to participate in the work. > *Major issues*, as I understand it: > > 1. How do we treat highway=path and highway=footway that has no > additional tags? > 2. Is highway=path a type of way (wilderness trail or whatever term we > use) or a way for non-specified/mixed use? That is, are we talking about > the physical characteristics of a way or its function? *Btw, this > would likely mean that 99 % of path/footway/cycleway in Sweden should be > path, if the latter interpretation is to be used.* > 3. #1 & #2 makes it really difficult for data consumers, they have to > depend on (often non-existing) subtags. > 4. Additional tags must be used to denote accessibility for > pedestrians/cyclists of ordinary ability, that is "this is NOT a hiking > trail/wilderness trail!. But which would these tags be? > 5. Additional tags must also be used to tell !this IS a wilderness > trail! (or whatever term we use). > > > *Subtags* > To specify the physical characteristics of a highway=path or > highway=footway we have a multitude of tags, with no particular > recommendation about which ones must or should be used (see #4 & #5 above): > surface, smoothness, width, trail_visibility, sac_scale, mtb:scale and > possibly incline. > > > *An additional issue:* > 6. sac_scale is currently the only tag (possibly together with mtb:scale) > to denote the difficulty of a hiking trail (that is, the way, not the > route). But it's very geared towards alpine trails and there is not enough > nuance in the lowest levels. Could the Yosemite Decimal System (YDS), > Australian Walking Track Grading System and others complement or expand on > sac_scale? > > > *What needs to be done?* > > 1. We have to rely on subtags... > 2. We need to decide what subtags to be used to tell this is an > accessible path or this is a wilderness trail. > 3. We need a way to better nuance hiking trails. > 4. Documention needs to be much more clear and specific, in order for > mappers and data consumers to really know when different kinds of highway > tags should be used and what subtags must/should be used. > 5. Editors need to be improved to encourage tagging that will make it > easier for data consumers. > 6. Better default rendering of non-urban paths, to encourage the use > of mentioned subtags. > > > Would this be a fair summary? What have I missed? Who is interestet in > continuing this work in a smaller group? Or should we continue to spam this > mailing list? > > /Daniel > > > > > > Den fre 29 maj 2020 kl 17:26 skrev Volker Schmidt <vosc...@gmail.com>: > >> Unfortunately it is more difficult to map properly the minor roads and >> ways, in comparison with the major roads. There much more variegated in >> appearance, in use, in rules ecc, and, at least in my part of the world >> there are also simply more in numbers. >> It is also correct that the available sets of tags of keys are not >> orthogonal, but whatever we invent in additional new tagging, won't make >> the existing tagging go away. So whatever we add, we make life for data >> consumers even more complicated. And redefining the meaning of the existing >> tags is also out of the question. >> What we can do is to improve the documentation, which is overlapping and >> dispersed abd, maybe, we can do better in documenting country-specific >> tagging traditions, but not more. >> Also when doing so we have to avoid absolutely anything that my appear to >> be wiki fiddling. >> >> >> >> >>
_______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
