[Tagging] Subject: Feature Proposal - RFC - highway=social_path

Greg Troxel gdt at ir.bbn.com
Tue Jun 14 12:28:21 UTC 2016

Mark Wagner <mark+osm at carnildo.com> writes:

> I've been mentally trying to apply this to the parks I've mapped, and
> it's just not working.
> Palisades Park (2.5 sq. km) has two trails that are clearly "main".
> However, they're both maintained as access roads for brush-fire trucks,
> so I've mapped them as "highway=track" plus appropriate access tags.
> Riverside State Park (30+ sq. km) has the headline Centennial Trail
> that runs the length of the well-known part of the park.  However, it's
> usually not the best (or most popular) way to reach any given section of
> the other 90+ km of hiking trails.  It's also a "highway=cycleway,
> surface=asphalt, foot=yes" rather than a hiking trail.
> Slavin Conservation Area (2.5 sq. km) was intended to have a main loop
> trail, but the wetland restoration project was a little too successful
> and a section of the loop is now under water.  The trail network that
> developed as a result does clearly have a "main" section, but it
> doesn't reach the parking lot -- there are three routes of roughly
> equal popularity to cover the kilometer from the parking lot to the
> main trail.
> I'm just getting started at Mount Spokane State Park (50+ sq. km), but
> the official trail network is looking like a hub-and-spoke system,
> where most of the hubs are parking lots (or former parking lots).
> There are some trails that are "main" in the sense that I'll use them
> when giving directions to a lost hiker, but that's due to their "can't
> miss them" status rather than being the best, shortest, or most popular
> route to go somewhere.
> "trail=main" might work as a concept for smaller parks with a few
> high-attraction features and a well-designed trail system, but for
> larger parks, especially where the trail network evolved rather than
> being designed, it doesn't.

It sounds like it doesn't fit for those trail networks, but I think that's
because they are all main trails.

So perhaps we need a minor tag, not a main tag.

The areas I'm thinking of have a loop and access trails that are
officially maintained, and then some lesser trails that may be shown on
official maps, but the maintainers probably wish didn't exist.  They are
usually unnecessary in terms of being overly dense, and tend to be

My point is really that these lesser trails should have some way of
being distinguished tagging wise so that they can get some lesser
rendering (vanishing faster at lower zoom, thinner).

Another way to think about this is to look at other trail maps,
particularly those generated by humans instead of db/render.  I have
often seen thicker lines for more signficant trails and thinner for less
significant ones.

As to how to tell, often you cannot directly observe this just looking
at a trail.  But after you have learned the entire network and the
social conventions of the area, and looked at official posted
signs/maps/boards, it usually is fairly clear.

All that said, properties that are seriously managed may eradicate the
kinds of trails I am talking about, so that they don't tend to exist.

Here is an example of a place with too many trails (there are actually
too many in the woods):


the ones labeled unblazed are typically the ones that I would call

Here's a trail map put out by the local land trust (apologies for the


You can see that many of the trails I want to label minor aren't on that

But it's more than official/not.  Often there's a trail that is
important, and a side loop that is less important, even if it's

Overall, I'm convinced that main=yes is a mistake, because that's
arguably the normal case.  I know think we need some tag to indicate
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 180 bytes
Desc: not available
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20160614/fe629cdd/attachment.sig>

More information about the Tagging mailing list