[OSM-talk] Cycle route improvements

[Dave Stubbs]

On Feb 5, 2008 4:08 PM, Andy Allan <gravitystorm at gmail.com> wrote:
> On Feb 5, 2008 2:47 PM, Ben Laenen <benlaenen at gmail.com> wrote:
> > Hi all, and Andy in particular,
>
> Hi Ben. An excellent selection of suggestions - I'll reply to them inline.

I'll translate into the relation equivalents...

>
> > as the past weeks went by while entering cycle routes for Belgium into
> > OSM, I've come across several issues. It's a big mash-up of needed tags
> > and improvements, so here goes...
> >
> >
> > * Tagging of alternate routes:
> >
> > Some routes have shortcuts, or in general alternate ways (e.g. for when
> > a route is blocked due to the weekly market day, it's possible to
> > follow another signed route, or sometimes there's an alternative for
> > cyclists who don't like sandy roads). These routes are generally marked
> > as a dashed line on most maps I've seen, to make sure it doesn't look
> > like the main route. I guess a tag as "route_special=alternate" added
> > to the route relation works for that to make that happen. I don't think
> > a "route_special=shortcut" is needed, after all, it's just an
> > alternate.
>
> We have ncn= yes / proposed and this could easily be extended to ncn =
> yes / proposed / connection / alternative / no without adding any more
> tags. I'm not sure if anyone will care too much about
> proposed-alternatives being different from proposed-main-routes, but
> this wouldn't be hard to solve anyway.

This is the "state" tag of a relation.

>> (that also happily solves the
>> current issue of not seeing the names for those routes which have no
>> reference numbers)
>
> That's a bug. Fixing it will also stop people using
> ncn_ref=Something-awfully-long-that-isn't-a-reference-really, which
> suggests the invention of an ncn_name= tag.

Or just use the "name" tag on a relation.
I know some of these already exist, but we've been ignoring them.