On 2nd April 2021, 70628 cross-country / nordic ski ways were members of one of 5849 route relations.

Among them, for 27423 ways, there is a piste:difficulty tag on the relation. I guess that the mapper estimated that the complete trail were ‘novice’, ‘easy’ or whatever.

But OSM is made by humans. We make mistakes. And we are very creatives. So, inevitably, there is 5625 ways where the difficulty of the way is not the same as the difficulty of the route, here are those differences:

Route difficulty Way difficulty Number of ways
novice easy 70
novice intermediate 15
easy novice 195
easy intermediate 463
easy advanced 12
intermediate novice 198
intermediate easy 2607
intermediate advanced 60
intermediate expert 7
advanced novice 53
advanced easy 984
advanced intermediate 766
advanced expert 3
expert easy 47
expert intermediate 64
expert advanced 2
extreme easy 8
extreme intermediate 2
extreme advanced 8

What about those 12 easy trails with ‘advanced’ parts, which are, according to the documentation ‘Steep sections […], with narrow steep passages or sharp turns in steep passages, often icy pistes.’ Somebody is up for a bad surprise.

Unless I’m missing something, there is nothing much we can do with this data. In any case, ways that are member of a relation are member of this relation, so if we want to know the most difficult section or the easiest, the information is already there. So please, take the time to put the piste:difficulty tag on ways, were it belongs.

Comment from gileri on 4 April 2021 at 21:00

Routes in general ways have an “average difficulty” and a maximum difficulty. An intermediate skier (same for any sport really) may accept a short section of higher difficulty on a longer way, easier way.

I see the relation difficulty as the average for the route, and its maximum can be derived by its ways. Isn’t it possible to use ways difficulties with a fallback on the relation difficulty ?

Comment from yvecai on 5 April 2021 at 09:01

Fallback on relation tag is technically feasible, of course. But given the way nordic difficulty is defined, I won’t dare sending everyone including 70+ year old on a trail if Joe mapper didn’t mapped each and every way difficulty. It seems to me that taking into account the ways difficulty only is to stay on the safe side. Also, average and maximum difficulty can be determined by looking at the members way tags, so why look at the relation difficulty at all?

Comment from gileri on 5 April 2021 at 09:17

I don’t see the problem then. If you believe that only ways difficulties should be used in your tool so be it. That way both the risk-averse people would be satisfied.

People that want more “aggregate” difficulty ratings can look at the relation. Both are useful imo.

Also, if a mapper only know the “average/aggregate” difficulty of the route, wouldn’t it be better that they add it to the relation instead of every members ? If they added it on every way, that would infer incorrect precision, likely preventing future mappers to define each way’s difficulty. And that would put people like you described at risk.

Comment from yvecai on 7 April 2021 at 08:12

You are certainly right. Maybe the best thing to do is to document this usage of aggregate / average difficulty on relations.

Login to leave a comment