Relation: Coast to Coast Walk

Hi all! Being a Coastie myself, I have been working on the C2C mapping on OSM during the last months, aiming to have the C2C mapped in detail sufficiently enough for walking it using OSM mapping only. I think this has been achieved since – try it out :slight_smile:

My worry is the number of alternative routings given in the relation. First, Relation:route does not offer “alternative” or “excursion” roles at all, supposing any route to be a single line from start to end point. Second, any alternative is messing up the relation as soon as it’s converted into a GPX track (in order to save it to an outdoor navigation device for highlighting the route). There were numerous alternatives given, splitting the whole walk into some 20-25 separate tracks in the GPX file. I found this rather annoying.

So, I yesterday took the liberty to remove some (certainly nice and recommendable) alternatives off the relation (some of which I had added myself previously), leaving only these alternatives pointed out by Wainwright himself:

  1. The “High Stile” alternative from Ennerdale YHA to Brandreth
  2. The “Gibson Knott / Calf Crag” alternative from Wythburn to Grasmere
  3. The “Great Tongue” alternative routes from Grasmere to Grisedale Hause (which are rather short, the eastern one could be taken away too, though I find it nicer to walk)
  4. The two “Helvellyn” and “St Sunday Crag” alternatives from Grisedale Tarn to Patterdale
  5. The Swaledale alternative from Keld to Reeth (though this is not given detailed in Wainwright’s book; the route shown has been taken of the Trailblazer guide)

Furthermore, there are the 3 seasonal permissive routes over Nine Standards Rigg, which are still members of the relation.

I think this should be enough of alternatives (still messing up the GPX file). Of course, there are many more alternatives which might be chosen in poor weather, but any walker should be able to work them out for himself IMHO, if necessary.

Now I wonder if this relation ought to be split into two, one giving the “main” route, the other containing all alternatives. Technically, this would be a misuse of the route relation too, since the alternatives do not make up a whole route, only summoning some separate tracks. The benefit would be the relation to convert into one single track showing the “official” route (I know there is no such thing on the C2C, call it the recommended route), leaving any liberty to set up one’s own excursions.

What do you think?


It’s better to ask this sort of thing on talk-gb: traffic in the forum tends to be very light. I’ll post to talk-gb to draw attention to this subject.

Thanks! I might be a bit spoiled by the German forum which is really lively. Maybe I ought to subscribe to talk-gb too.


If by “not offer” you mean an editor doesn’t supply a prescribed option it doesn’t mean one of your choice can’t be added. Just outside the top ten >

I don’t think data should be deleted from OSM just because of GPX’s limitations or a specific user’s desire for a linear route. This is an equivalent of tagging incorrectly to suit renderer.

Are these alternate routes signed on the ground? If so they should definitely be kept. If they’re purely mentioned in books then it’s arguable if they should be tagged. Personally I’m against adding book only routes. With the increase in self publication & websites/blogs it’s difficult to know where to stop. I’ve already seen a few ‘Where I take my dogs for a walk’.

If you’re after a linear route for GPX there are a few routing websites with a GPX download option. Some even have real time dragging to customise your walk. Try here:

Certainly. I did not mean to alter the relation to suit my personal desires. I meant to free it from proliferation. (Strictly, I did not delete any data. I can add the alternatives back into the relation if the community thinks it’s useful.)

The C2C ist a (nearly) unmarked route, so there is hardly anything of it signed on the ground. The most official representation is Wainwright’s guide, the alternatives of which I left in.

Question is: Is it recommendable to leave these alternatives in, roled as “alternative”, or is it better to take them out into a separate relation?

That’s what I am trying to avoid. Some of the “private” extensions (I’ve never heard of) did add even more climbing to an already strenuous walk. Whoever likes to do such alternatives is encouraged to do so but that does not mean they belong into the relation.


On more general question about what should or shouldn’t go in the relation, I’d actually only add any routes that are explicitly signed on the ground - we shouldn’t need people to buy AW’s book to be able to verify anything in OSM.

We have to draw the line somewhere and that’s a good a place as any; otherwise we’d have all sorts of made-up “routes” in OSM that someone wrote a book about once or described on a web page. I’m sure I’ve seen signposting for the C2C up on the North Yorks Moors though (though not sure how complete it is).

If you were really keen you might want to add any signposted “alternatives” as separate relations, much in the way that the National Byway loops are done (and if it’s getting too big, perhaps split into smaller alternative and non-alternative sections and create a super-relation like the NB is, but that will confuse newbie mappers).

I don’t know about the North Lakes but bits of North Yorkshire certainly suffer from lack of on-the-ground editing - based on what I’ve seen up towards Gargrave, some of the paths that have been added a few years ago are “within a field or two” rather than being accurately recorded. That’s changing though - bits of the Dales are now very well mapped; I was in Dentdale earlier in the year and almost nothing needed adding. Obviously you’ve helped with that, so thanks!

(edit: spelling)

I would argue that a route in a guidebook is a database in copyright law, so can’t be used as a source, unless the book is out of copyright, or there is an explicit licence grant. (It seems pretty obvious that intellectual property would be recognized in such things.)

The transport people have route_master relations to deal with variant routes, although even in the transport context they don’t account for the combinatorial problem.

I decided it’s time to show some innovative effort.

IMHO, when setting up hiking routes as relations, there must be a way to tag members as main or alternative routings. Any hiking route will develop its alternatives, be it for avoiding high ground in poor weather or circumnavigate more difficult parts for inexperienced walkers.

So, I applied the role “main” to the route described in the book as such, and respectively the role “alternative” to any other routing (“alternative” has already been in use before my edit). Some day, we should write a proposal to get these roling into official state on hiking routes – and others too, I don’t think cycling routes differ much in this respect.

In most places, Wainwright’s description is clear about which route he considers the main one. The only confusing part is the Nine Standards traverse, where my copy of his book describes the “blue” route as the official one while the “red” route is tagged as “original route, now obsolete”, and the “green” route is not given in detail at all. Since these routes are clearly signposted on the ground (they are all permissive routes, changed by season on the landowner’s behalf), they all have to show up in the relation of course. I put the red route into “main” here and the two others as “alternative” (though there really is no option for walkers here, they have to take the active route for their month of walking).

Officially, the C2C is unmarked. However, in some place there has some signage been put up by landowners, in order to avoid walkers get lost and trespass on private ground. I have seen some signposting on the NYM moorlands too. Some more can be found in Lakeland and on the limestone plateau west of Kirkby Stephen. In the Yorkshire Dales there is one single C2C signpost (at a spot where you can hardly get wrong). In most other places you have to know your route (C2C to the right here, along the bridleway).


Should ‘the’ coast to coast walk be mapped as an established route at all? Wainwright certainly didn’t envisage in that way, and his use of the indefinite article in his title "a coast to coast walk’ was deliberate, as he was writing it as inspiration to walkers to devise long distance routes of their own. In his own words:-
“I want to encourage in others the ambition to devise with the aid of maps their own cross-country marathons and not be merely followers of other people’s routes: there is no end to the possibilities for originality and initiative.”
Maps exist, in part, to give you that freedom, and not to just slavishly follow a ‘sanctioned’ route. So some irony involved if people ‘merely follow’ his published route.