Yes, I just found that out. So I apologize for the spamming here. Verdy_p in the wiki though presents a scenario in which it would be difficult to know when passengers can really aboard/alight at a stop. If one goes from points A to B to C to D, then back to C, and passengers can only board/alight at C on their way back but not on their first pass, this cannot be currently expressed. Using roles wouldn't solve this though, only interspersing stops and ways in the member ordering.
On Tue, Jan 9, 2018 at 7:40 PM, Selfish Seahorse <[email protected]> wrote: > On 9 January 2018 at 19:57, Fernando Trebien <[email protected]> > wrote: >> I was about to fix a mistake I caused in the map due to these >> contradictions in the wiki, then I found a problematic case [1]. >> >> According to PTv2, this route needs to be broken into two, one per >> direction, and a route_master relation must be created for them. >> Without the forward/backward roles, I believe applications will not be >> able to easily find out the direction of travel of either since the >> route contains a single way, unless they parse the from and to tags, >> which are not required to match station names. > > The course of the route is determined by the order of the stops in the > route relation. Therefore forward/backward roles would be redundant. > > Regards > > _______________________________________________ > Tagging mailing list > [email protected] > https://lists.openstreetmap.org/listinfo/tagging -- Fernando Trebien +55 (51) 9962-5409 "Nullius in verba." _______________________________________________ Tagging mailing list [email protected] https://lists.openstreetmap.org/listinfo/tagging
