Tod Fitch wrote: > This thread has been quite amazing to me. My impression is that it > starts with some routers (a.k.a data consumers, a.k.a. “renderers”) > treating a “no” as a “maybe” and now people are looking for a new > term to indicate that “we really, really, mean NO!”. This is worse > than tagging for the render, it is obsoleting a straight forward > and explicit tag value for a broken renderer.
No, you have got that the wrong way round, and it would be kind for you to be a bit surer of your facts before throwing around accusations of brokenness. People have been using bicycle=no to tag footways where cycling is banned, but where you may push a bike, since the very earliest days of OSM. Here's an instance from 2006: https://www.openstreetmap.org/way/2606296/history . I'm pretty sure there weren't _any_ OSM routers in existence then. The reason that routers will sometimes route via such a path, with an instruction to dismount, is that this tagging practice has always been widespread. It doesn't "start with some routers". It started with the tagging. Fairly obviously, if the users of a particular router complain to the router's authors that they're being prevented from plotting a viable route, then the authors are pretty obviously going to change the router so they stop getting complaints. So either fix the existing instances in OSM of bicycle=no being used to mean bicycle=dismount, or introduce a new tag. Richard cycle.travel
_______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
