[Talk-us] Removing US Bicycle Route tags

[KerryIrons]

Again, a number of points of clarification are needed.

First, there is a single body in the US for assigning numbers to US Bicycle
Routes.  AASHTO owns the process, just as they do for all federal highways
in the US.  There can be any number of state and local bicycle routes,
proposed or implemented, but those are not USBRs until AASHTO approves
designation.

The process for doing this can vary but it culminates with a state
department of transportation submitting an application to AASHTO to approve
proposed numbering.  Once AASHTO approves (they have never declined an
application) the route is officially a USBR.  While AASHTO encourages
signing of USBRs there is no signing requirement so a route can exist "on
paper" (and on the Internet) but not have any signs posted.  When a project
is initiated to get a section of a USBR approved within a state, the first
step is to define a proposed route, but there can be many revisions to that
route as it gains local jurisdiction approvals (required) for each route
section.  There is no problem with showing these proposed routes on OSM but
tagging them with USBR numbers can create significant work for the approval
process team due to "ruffled feathers" at the local jurisdiction level.

You can look at the USBR corridor plan at
www.adventurecycling.org/routes-and-maps/us-bicycle-route-system/national-co
rridor-plan/  The corridors are roughly 50 mile wide area in which a route
could be defined.  Just because a corridor exists does not mean that any
specific road/street/trail has been defined as part of the route.  On the
corridor map, a solid dark line means the route is approved by AASHTO, a
shadowed and colored line means that the corridor exists but no route is
defined, and a grey line means that a corridor could be added along that
path.  A corridor is a concept for future development of a route.  It is not
a route.

It should be noted that there is a lot of history to the USBRS that explains
the heretofore slow pace of route implementation.  It is inaccurate and
unfair to blame any one organization for that slow pace.  As of now there
are 5,600 miles of designated routes and many more are being developed.

As to whether the concerns I have raised are a mountain or a molehill, I
would simply say that those who want to ignore the political realities of
getting a route approved need to walk a mile in the shoes of those doing the
actual work.  Spending hours explaining why a route is not going through a
given community, even though there is a map somewhere showing that it does,
is not seen by a project team as a good use of their time.  Spending hours
trying to convince a community to accept a route when they feel it is being
shoved down their throat because it appeared on a map before they ever heard
about it is not a good way to spend time either.

My only goal here is to keep the OSM efforts in synch with the efforts of
various USBR project teams across the US.  There is no point in creating
extra work for the project teams or for OSM mappers.


Kerry Irons
Adventure Cycling Association

-----Original Message-----
From: Greg Troxel [mailto:gdt at ir.bbn.com] 
Sent: Wednesday, June 05, 2013 7:02 PM
To: Frederik Ramm
Cc: talk-us at openstreetmap.org
Subject: Re: [Talk-us] Removing US Bicycle Route tags


Frederik Ramm <frederik at remote.org> writes:

> An argument *against* having proposed routes is the verifiability - we 
> usually try to have data where someone on the ground could easily 
> check the correctness by looking at signs. Since proposed routes are 
> unlikely to be signposted, having them in OSM is questionable.

I see verifiability as having a broader sense.  In the case of officially
proposed USBR routes, someone who is local can look up the government
documents, meeting minutes, or whatever and determine if the route numbering
authority has in fact put the route into proposed status.  That's
essentially what Kerry is talking about.  That's beyond looking at signs,
but some things on the map aren't obvious from standing near them - official
names are a complicated mix of signs on the ground, meeting minutes from
naming authorities, 911 or tax databases, etc.  To me, the point is that one
can determine an answer by observing evidence, and reasonable people can
discuss the total evidence and come to rough consensus.

> On the other hand, I take exception at the original poster's apparent 
> insistence on "routes approved by AASHTO". Whether or not a certain 
> route has been approved by a certain third organisation is not usually 
> something that OSM would care about. The usual OSM approach would be

I don't see that at all. For a US highway, there is some part of the federal
bureaucracy that assigns highway numbers.  A road is a US highway if it's
officially been designated, and the signs are expected to keep up with that
offiical designation.  If there's a case where a road has been designated as
a US highway, and the locals know it, but there are no signs (Because
they've been stolen, or because there was no budget to put them up, or the
sign people are on strike, or they've all been knocked down in winter car
accidents, or whatever), then it's still proper to tag it as a US highway.

> that if a route is signposted, then it can be mapped - if not, then 
> not.

I do agree that tagging a highway because one wishes that it were otherwise
is bogus.  But as long as a local mapper is determing a form of reality by
relatively objective means, I don't see a problem.

> An AASHTO approved route that is not signposted would not normally be 
> mapped;

I think there may be a bit of terminology confusion: Kerry seems to mean
"approved" as "approved by the numbering authority as a proposed route which
has not yet been constructed/signed".  That's similar to "the government has
decided to extend I-101 on these 10 miles, but hasn't built it yet".  So
either it's ok to show it, or we should remove all highway=proposed.  But I
think it's useful to have highway=proposed, so that those who want can
render it.  highway=proposed is still subject to crowdsourcing editing and
quality control, and should mean that the cognizant naming authority has
published a specific plan.

I think this is the crux of Kerry's point - proposed cycle routes only make
sense if the authority that controls the relevant ref namespace has actually
proposed them.  So even from your verfiability concern viewpoint, I think if
people did as Kerry asked, there would be far fewer proposed routes in the
db, and all of them would be widely recognized as legitimately and actually
proposed.

> and a signposted route that is not approved by AASHTO has every right 
> to be mapped.

This is similar to what would happen if someone put up "US 99" signs on
their little side street, just because they were in the mood and had signs
and a hammer and nails.  That doesn't make it US 99 -- it's just simple
vandalism -- , if other evidence says it's not true.  This is really the
same situation.

Now if the guerilla route is not in an official namespace, and the signs
persist, then I have no issue with it being mapped.