[Tagging] Roundtrip and closed loop in relations
francians at gmail.com
Sun Dec 22 12:53:49 UTC 2019
Il dom 22 dic 2019, 11:59 Peter Elderson <pelderson at gmail.com> ha scritto:
> For PT, roundtrip is not an attribute of the route, it's a type of ticket
> or it's what you use the transport for. You can do a roundtrip on a
> circular line, but also on non-circular lines or mostly non-circular with a
> loop at the end, whatever. To express that a PT route is circular, I think
> the term circular would be better than roundtrip.
Agreed. My intention was to avoid Roundtrip as sinonym of Circular.
> For hiking|foot routes, exception is the rule when it comes to branches,
> alternatives, excursions, approaches and shortcuts. For me roundtrip on a
> walking route relation means: when you keep following the main route
> markings, it takes you back to where you begun. This does not exclude any
> alternatives, such as optional extra loops or a common approach/exit route
> at a starting point.
If following the route marking you will get back to start... It's a
As previously stated you could find marking on both directions and be a
single line straight and then reverse.
With old wiki definition this is Roundtrip=no... Now it is Roundtrip=yes
Seems sane to me.
Only roundtrip=yes is needed here, if not present assume it's not a
roundtrip. Note that many trails consist of a number of linear routes,
together making for a roundtrip. I tag roundtrip=yes only on the parent
route relation. Loop or circular would also be just fine, but I see no
reason to change existing tagging here.
Do you mean old definition or the new one?
> Question: who wants to know if a route is a circular route/loop/roundtrip?
> Is it the map user? No, (s)he can see it on the map.
Don't forget OSM is a database.. a new tag that helps to classify things
(without overtagging) can be useful. Can you extract easily all circular or
linear routes with current scheme?
Is it important for routing and navigation? I can't see how, but there are
> experts on this list who know more about this. So far I know of only one
> application: categorisation/filtering of trips in order to present the user
> a choice between roundtrip walks or linear walks. The roundtrips were
> actually meant to be daytrips, and linear walks were to be presented as "
> long distance walks", but a separate category long distance roundtrips
> could be deducted from the data, I guess.
> Question: who wants to know if a route is actually a closed loop without
> any branches?
> What do you need this information for?
Personally I would only be interested in the validation part. I already
wrote some rule in JOSM and Osmose... I don't currently plan to fix all
existing routes, but hopefully could help to figure how to do it.
So far, I know one application: if a route is tagged as a closed loop, e.g.
with closed_loop=yes, and it's not complete or interrupted somewhere, you
can detect that with a checking tool. It would be a sort of fixme, then.
Most routes I maintain would not profit from that.
> FrGr Peter Elderson
> > Op 21 dec. 2019 om 15:31 heeft marc marc <marc_marc_irc at hotmail.com>
> het volgende geschreven:
> > I always thought that routrip=yes was an alternative when there is no
> > start and end point to enter in from=* to=* key.
> > Otherwise circular routes with a known start/end point can enter
> > as from=A via=B to=A.
> > _______________________________________________
> > Tagging mailing list
> > Tagging at openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/tagging
> Tagging mailing list
> Tagging at openstreetmap.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Tagging