slightly related - a discussion on the Dutch forum: http://forum.openstreetmap.org/viewtopic.php?id=32342&p=1

@trigpoint thanks.

@roburite One of the reasons why a "single way" isn't often used for large features such as a national park boundary is that OSM ways are limited to 2000 nodes, and it's easy to exceed that with even a small level of detail. There are "relationship checker" tools that can identify small gaps that might not be immediately humanly visible - we can use one of those to see what the problem is, and alternative renderings that can show if it's just a problem with one map style. If "something has gone a bit wrong" it's often easiest to undo the changes as a whole and then try and work forward from there rather than trying to fix things piecemeal - it's simply a quicker way to get to the destination.

It should "show as a block" on OpenStreetMap-carto at some zoom levels (see the parks at osm.org/#map=7/53.998/-2.098 for example), and at higher zoom levels to z9 there's a green background tinge. After that you get a name on the border, like osm.org/#map=16/54.2322/-1.2253 . That's not working now because the relation is not complete: relation/86909#map=10/51.1574/-3.6076

Personally, I'd map railway access tracks as "highway=track" as for example at way/30010081 , but obviously it depends on what they look like - some are more like service roads, and there are lots that are just footpaths. Here it's difficult to tell from just the imagery. I guessed highway=track mostly because that's what it was originally mapped as by "barb-andmike" - from the imagery it could just as easily be a highway=service. Given that it only goes to a house and some stuff likely connected with the airport, it's likely not an unclassified road. As a walker what I'd be most interested in of course would be the general access rights there (public footpath, bridleway, whatever) - but that's not viisble from imagery.

Looks like geocaching.com (if that's what you're talking about) offer lots of different background styles, but don't host any themselves, so you'd just have to lobby for one of those to support it; not necessarily openstreetmap-carto.

Hi - thanks for fixing. If you don't mind me asking, what was the "example data" from that got uploaded? I'm wondering if we could perhaps do something like https://github.com/hotosm/learnosm/issues/335 to try and prevent other people uploading the same test data (a dozen or so people have also done so), but in order to do that we'd need to know where it comes from.
Thanks - andy.

Hi Arun - are you sure that way/329032015 is really an unclassified road? It doesn't particularly look like one from the imagery. What I suspect it is is a farm track with a "public right of way" over it (given that someone's named in "leeds country way"). My guess is that highway=track is correct, and the access rights are at the very least "foot=yes; designation=public_footpath" (but that'd need to be verified by survey or by talking to someone local). Given that it is part of relation/5012105 , I suspect that the name of this road isn't "leeds country way" either. The best locals to talk to are probably the first mapper of this object (this looks like one of the first things they mapped in OSM after they joined - they've done lots more since) or chippy, who's from Leeds, and mapped the bit just to the west.

@sladen - thanks. Did they log a bug in whatever satnav doesn't like the previous route as mapped?

@sladen - are you OK to revert this? way/341629727 looks pretty implausible to me.

There's a discussion of what should go in the name tag at osm.wiki/Names#Name_is_the_name_only . I don't believe that "cliff" makes sense here.

Can you please explain what you were trying to achieve by this edit? What was the problem with the data as it was before, and what you believe you have fixed? Also, similarly, changeset/33276302 .

What benefit do you see for adding "name:hu=Nottingham" when "name" is already "Nottingham"? What is your source for this information? I've never seen it signed locally.

What does "correction suite analyse" mean?

I can't claim to be familiar with this area, but after a survey of an area east of Goole last year, most of the OSSV-derived data added by a similar changeset where "wrong", where "wrong" was either duplicate, in the wrong place or unverifiable.

Hi - just for info, I got a chance to check the area tonight. I've fixed some of the stuff already (the doubling-back footpath and the designations) and will add more details (fences etc.) when I upload and map my GPS traces.

Thanks for tidying here. Did you get a chance to see whether the doubling-back footpath at osm.org/#map=19/53.20930/-1.07179 could possibly make sense? Also is the area of the woodland at the west correct - it doesn't quite match the imagery (but the GPS traces suggest that the imagery might be offset, and there are no long east-west traces here to confirm)?

You're more local to the area than me, so if you reckon that "Staffordshire Moorland Walks" are worth mapping as a route I'd go for it. I did initially try keeping track of SMW signs as I do for e.g. Staffordshire Way ones, but couldn't get any sense of where the route "went". I did recently fill in a bit of a gap in the Staffordshire Way near there and do a bit more Churnet Way - now that and the Sabrina Way are definitely distinct routes that need tracking down on the ground and adding to OSM :)

It's a tricky one - I've always found the OS's inclusion of "Staffordshire Moorland Walks" rather odd (they don't do it with the equivalent "Derwent Valley Walks" in Derbyshire, for example). There's no suggestion that the indvidual footpaths along which the "Staffordshire Moorland Walks" signs point individually each have that _name_, any more than the individual roads and paths forming the Staffordshire Way do. You could perhaps argue that it's a "route" - but adding it would cause the whole area around http://waymarkedtrails.org/en/?zoom=13&lat=52.99604&lon=-2.03081&hill=0# to light up with Staffordshire Moorland Walks "route". For more info, see osm.wiki/Names#Name_is_the_name_only

Excellent - thanks!

Hello, as I mentioned on changeset/33199691 , neither of these sources are likely to be licence-compatible (unless you've obtained explicit permission). If you haven't, and both sources are licence-incompatible, then these changes will have to be undone. Can you please confirm what steps you've taken to ensure that the licence of the data that you are continuing to add is suitable for adding to OpenStreetMap?