Desirable, certainly. That is, if the source is deemed reliable. I would suggest adding a tag to keep track of the producer of these routes, because here in France there has been an incident last year where our national geographic service had to retire a ski tour map after discovering dangerous bugs in it (their automated method was immature). As said previously, this cannot be verified formally on the ground (exactly like most place names in the mountains), and all boils down to the credibility of sources.
The source is reliable and the routes are updated with input from the alpine club on a yearly basis to account e.g. for changes in wildlife protection zones and changes in glacier topography (climate changeâŠ)
There is also no advantage in having them in OSM over just using the tour data provided by the SAC as an overlay over an OSM map. They usually donât refer to existing ways, so donât need to be exactly in sync with the underlying OSM data. Importing them just has the disadvantage that they are likely to become outdated. And ski route data is really something you donât want to become stale.
I just found the coverage to be rather limited and thought it could be cool to do a bulk import for the routes to be displayed in a unified way (like on https://openskimap.org world wide) rather than a country- or source specific overlay
They are often not really precise, they are more of a corridor in which you pass - not only depending on the snow and avalanche situation but also as a matter of taste. (I looked at a few that I take a few times every winter).
They can be outdates quite fast due to the climate-change (melting glaciers) or changing protection-areas (or rules in them), so one would need to update them before every skiing season.
We already have the discussions about the unmarked mountain-routes - do we need a winter-version as well?
Yes, I agree. One of my motivations for this question was that given backcountry ski routes are principally mapped in OSM, wouldnât it be nice to have officially published and regularly updated ones rather than user-supplied ones of mixed accuracy/quality?
But the regularly updated only applies to the SAC dataset there is nobody that is going to continously update it in OSM.
This isnât a new issue, it is just far more critical for these routes as there is nothing on the ground that would tell a user that they are literally going to fall off a cliff because they are following a route that was last updated a decade ago.
The correct action would be to remove the existing routes of this kind.
As I understand Key:piste:type - OpenStreetMap Wiki there is no requirement for these routes to be signposted (outside of ski resorts I have never seen a signposted ski route). As far as I can see the routes recommended by an alpine club fit well into this category
There is an overall requirement for on-the-ground verifiability for all data in OSM unless an exception has been agreed upon. So unless the wiki page explicitly mentions that routes that are not signposted or otherwise verifiable on the ground are okay, the unwritten default is that they have to be marked.
Given that they are mapped for ages (for the worse or the best, no opinion) ( osm tag history ), and the the fact that it seems quite impractical to signpost on snow in remote mountainous area, I think we can safely assume the default is ânot signpostedâ
What if that data is backed by Stravaâs heatmap? Also, wouldnât it be possible to update the dataset annually with the provided information from the SAC?
As it has been mentioned by various users above one of the points why they should actually not be mapped at all (Which I support) is that they can not be verified on the ground.
All that could be used is the the SAC dataset (which not allways is very accurate) and as you mention Strava data - which I would not use at all as we do not know the conditions when the person was out, nor how much risk they were willing to take.
Alpin skitouring routes are in my eyes often something that can only be decided while you are there, and they can change quite a bit from one day to another. Of course there are stretches that are âallwaysâ the same - but mapping only those makes not much sense and how should a user see where they can not follow the route but have to make their own decision.
So mapping them and even more just importing them is a bad idea.
