I’m a cyclist who’s been adding a lot of cycle routes in Ontario, which has very little of its cycle routes mapped. The cycling system is very decentralized - it’s up to each municipality to design its own network and hardly any of them have a numbering system. Usually they’re signed as cycle routes and shown on municipal cycling maps but have no other way of identifying them. My practice so far has been to use relations for trails and routes that are either named or numbered, while using the lcn tag on routes without a name or number.
However, I’ve seen some examples of unnamed cycle routes that are part of a relation. This seems to be common on rural cycle routes along roads. Is this the preferred way of showing unnamed routes or are the lcn and rcn tags more appropriate?
If it’s an identifiable A-B route, yes, it’s generally best to create a route relation, even if there’s no formal name or ref.
If it’s a set of roads/paths around a city that don’t generally form particular A-B routes, but are signposted as being particularly suitable for cyclists, then yes, I’d agree that lcn=yes is appropriate.
Note that some cycle-specific renderers (e.g. Waymarked Trails and my own http://cycle.travel/map)) only render routes that form part of relations, though cycle.travel’s routing does give a slight uplift to roads that are tagged lcn=yes.
Thanks for the response. Looks like I’ll continue to tag routes the way that I’ve been doing it for the most part. I’m aware of Waymarked Trails and it’s great because you can see your edits almost immediately. For relations anyway. I don’t really know anything about cycle.travel, but it looks like it doesn’t update very often.
It doesn’t, no, because the routing takes about three or four days to generate on a very loud server and I value my hearing. But the target audience is cyclists rather than OSM mappers anyway.
But the target audience is cyclists rather than OSM mappers anyway.