[OSM-talk] Highway=trunk : harmonization between countries ?

Greg Troxel gdt at lexort.com
Tue Aug 22 23:20:36 UTC 2017


Richard <ricoz.osm at gmail.com> writes:

> On Tue, Aug 22, 2017 at 10:00:07PM +0200, Martin Koppenhoefer wrote:
>> 
>> 
>> sent from a phone
>> 
>> > On 22. Aug 2017, at 15:46, Richard <ricoz.osm at gmail.com> wrote:
>> > 
>> > called differently, but this is it:
>> > https://wiki.openstreetmap.org/wiki/Key:maxspeed:practical
>> 
>> 
>> yes, but practical maxspeed depends a lot on your equipment and
>> capabilities, and on other people driving in front of you, so this
>> tag will probably not be very uniform around the globe. Also, some
>> people are willing to risk a speeding ticket, others don't. With
>> regard to the latter, the situation in Italy is particularly
>> ridiculous: the authorities have to sign post speed controls ;-)
>> i.e. speeding tickets are kind of rare.

In most US states, there's a de facto limit higher than the signed limit
where there is very little risk of a ticket.  I'm thinking that
maxspeed:practical should be the 50th percentile of typical time actual
speeds.

> maxspeed:practical should take dense account or traffic jams into 
> account as good as possible. So far I am not aware of any router
> evaluating time based conditional restrictions but those could be
> used to take rush hours somewhat into account.

Agreed.  Or even live traffic.  But I agree with the notion that
maxspeed:practical should be a representative speed that's valid most of
the time.

> maxspeed:practical should not have any values above the legal speed
> limit.. and if it had routers should ignore such values anyway, at 
> least thats what I would expect from navigation software.
> Many years ago something like this was encouraged in the ancient 
> proposal but it is no longer in the description.. if there is any 
> remaining doubt I would explicitly state it in the wiki.

This seems unreasonable.  Maybe where you are people follow speed limits
(because they are enforced, or because speed limits are set by good
engineering practice instead of arbitrarily).  In that case, though,
maxspeed:practical will be essentially maxspeed anyway, and that's fine.
But in Massachusetts, uncongested traffic in clear weather essentially
always travels above the speed limit, and the delta varies by road type.
A typical city road posted 30 mph might move at 35 mph, and an
Interstate posted 65 mph might move at 80 mph.  But a particular road
that's almost Interstate (and correctly tagged trunk!) that is
inexplicably posted at 45 mph moves at 75 mph, because that's what all
the drivers think is the safe speed.

A router should be answering the question "If I take this route, what
will happen" as accurately as possible, as a first step in choosing a
route with a pleasing outcome.  Refusing to use a reasonable estimate of
traffic flow because it's below an arbitrary, known not to be enforced
limit, does users of the routing service a disservice.

(I don't know what Apple maps does, but I think they use speed estimates
from other apple users and do not clamp them to speed limits.  At least
it seems that way in that Apple computes routes that are in fact fast
but would be slower if speed limits were observed.)

Computing a route based on what's known to happen is not the same thing
as encouraging speeding -- it's more like admitting that it usually
happens.  And in all cases the driver is deciding how to drive.

So:

  maxspeed:practical should be able to have higher values than maxspeed

  routers should use those values, higher or not

and if that's not ok, then we need

  maxspeed:typical

which is defined to be what usually happens, regardless of what anybody
thinks about it.

Greg

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 162 bytes
Desc: not available
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20170822/cede68c0/attachment.sig>


More information about the talk mailing list