[Talk-us] Best practices for dealing with old TIGER tags?

[Kevin Kenny]

On 06/04/2016 09:12 PM, Harald Kliems wrote:

All these discussions are the reason why I almost never touch the
highway=* tag and rather add surface=* or other descriptive tags to
TIGER roads. There just isn't any consensus and many good reasons for
many positions about residential, unclassified, track, etc.

I go farther than that. I try to avoid adding subjective tagging
entirely. I'll add surface=, because I can tell asphalt from clay. But
tracktype=, smoothness=, mtb:scale=, sac_scale=, are all things I tend
to avoid. Two mappers are likely to come up with different answers,
which flies in the face of the
http://wiki.openstreetmap.org/wiki/Verifiability principle. And the
consequences are unlikely to be severe; subtle differences in
rendering, or tweaks in time estimates from a router, are all that I
really expect.

I find that there's an element of machismo in the assignment of the
scales. SAC scale is a really good example. Novices are likely to
overrate a route, because they lack confidence and also don't have the
experience of harder routes to compare against. Experienced
mountaineers are likely to underrate them. Someone who routinely
free-solos a route that's 5.6 on the Yosemite scale doesn't know what
a grade 4 route is! About the only people who can grade a route
reliably are experienced guides, who have an idea what to expect of
clients at various skill levels - and the description of the grades
has to be written in terms of, "what would you take a client on," or
they'll still underrate them.

I've seen trails with the steepness and exposure of
https://www.flickr.com/photos/65793193@N00/3183604309/ and
https://www.flickr.com/photos/65793193@N00/3183604743/ tagged as
anything from sac_scale=hiking (T1) to sac_scale=alpine_hiking (T4).
If subjectivity allows for that wide a range of classifications, the
scale isn't all that useful.