[Talk-us] Separate relations for each direction of US & State highways.

[Martijn van Exel]

That is a really good and interesting point, Chris. I just confirmed
that iD leaves the north / south / east / west roles alone when
reversing a way. Neither does Potlatch (although Potlatch does not
seem to actually allow you to reverse a way, just to flip the oneway=
between 1 and -1).
At least iD would need to be patched for this, I would say. I
submitted one: https://github.com/systemed/iD/pull/2004

On Thu, Nov 21, 2013 at 1:27 AM, Chris Lawrence <lordsutch at gmail.com> wrote:
> On Wed, Nov 20, 2013 at 2:32 PM, Martijn van Exel <martijnv at telenav.com>
> wrote:
>>
>> Reading through this I see that most are in favor of avoiding dividing
>> relations more than necessary: no separate relations for directions,
>> especially not if the geometry is the same for both.
>>
>> That leaves the question of how to tag cardinal directions? As I said
>> before, I think the role tag is suitable for this. It's already used
>> pretty widely (with around 100k ways having a cardinal direction as
>> the role tag in a relation), documented
>> (https://wiki.openstreetmap.org/wiki/Route#Members) and it replaces
>> the forward / backward role values that are not particularly useful or
>> meaningful.
>> I created a stub of a wiki page that we could use to describe the
>> preferred tagging in more detail:
>>
>> http://wiki.openstreetmap.org/wiki/Highway_Directions_In_The_United_States.
>> (The current content is geared towards using the member role tag, but
>> this can be changed depending on the outcome of this discussion.)
>
>
> The only problem I can anticipate with this tagging scheme is that it's
> possible some editors don't understand anything other than
> left/right/forward/backward (I think), we could end up in data loss
> situations fairly easily.
>
> For example: way X pointing east is marked in relation Y as "east"
> (presumably we could assume that "east" = forward and the opposite cardinal
> direction "west" is backward). User reverses way X. Now the relation role is
> potentially backward.  JOSM seems to understand at least north/south and
> east/west and offers to fix it (see
> http://josm.openstreetmap.de/browser/josm/trunk/src/org/openstreetmap/josm/corrector/ReverseWayTagCorrector.java);
> no idea if iD or Potlatch do.
>
> We'd also need to make the validation tools smarter to recognize lossage
> (for example, realizing that the route is unbroken only if the chain of role
> tags once you account for the directions of the underlying ways is
> monotonic),
>
>
> Chris
> --
> Chris Lawrence <lordsutch at gmail.com>
>
> Website: http://www.cnlawrence.com/
>
> _______________________________________________
> Talk-us mailing list
> Talk-us at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>



-- 
--
Martijn van Exel
OSM data specialist
Telenav
http://www.osm.org/user/mvexel
http://wiki.openstreetmap.org/wiki/User:Mvexel
http://hdyc.neis-one.org/?mvexel