Why not ask your question on the new OpenStreetMap Community Forum?

I've been tidying up some of the features (e.g. paths, open areas) in a part of the New Forest that I'm familiar with. I came across paths marked as bridleways within the forest which seemed doubtful to me, so I fired off some questions to the New Forest National Park Authority to which I received useful information. It may be useful for anyone else working on this area so it would be worth saving. Perhaps just including it here is sufficient.

Adam, a National Park Access Ranger said:

Regarding the bridleways on the forest. I would refer to the HCC definitive map to get a sure answer on whether a path is a public bridleway or not. That map can be found at https://row.hants.gov.uk/. There are a number of bridleways present in the open forest and most of them were added across the railway line to ensure that bridges were built to allow the freedom of movement for stock and people across the open forest. It’s also worth bearing in mind that since 1968 it’s been legal to use a bicycle on a bridleway (provided you give way to other users). So actually having more bridleways across the forest would very much help cyclists.

If a track isn’t marked as a public right of way then it’s just a track on private land. Most of the private land in the middle of the New Forest (crown land) is under a piece of legislation dating back from 1925 which gives the right to roam on foot and horseback. Cyclists can use bridleways but they haven’t got any rights under the 1925 legislation so bridleways certainly aren’t superfluous for those riding bikes.

From looking at the definitive map referenced above, rights of way inc. bridleways are pretty much restricted to where the forest meets human habitation. I see none in what I'd call the forest proper, so any found there are probably an error.

Since the whole of the Forest (or large swathes of it) have a right to roam enshrined in law, I'm not sure if this classes as permissive access or not. And is it possible to indicate whatever access it is on a bounding area to save having to add the tag to every path? Guidance please!


asked 15 Nov '19, 14:48

ceperman's gravatar image

accept rate: 0%

(15 Nov '19, 21:31) andy mackey

I intended to put this under eteb3's comment but for some reason i can only answer a question then convert it to a comment and placement is a problem. why as it been changed?

(15 Nov '19, 21:35) andy mackey

why as it been changed?

It hasn't (for me at least). Have you changed your browser?

(15 Nov '19, 22:04) SomeoneElse ♦

Thanks works ok with firefox but not chrome.

(15 Nov '19, 22:12) andy mackey

My recollection is that the New Forest is something of a special case (although it's a very long time since I've been there) - there's a legal right of access that isn't because of the presence of particular public bridleways. In a sense, the situation is similar to foot access on CRoW (Countryside and Rights of Way) Act land - in OSM terms that is "foot=yes" implicitly due to the CRoW Act, irrespective of any footpaths that might happen to run across it, though there may be time conditions on the access (e.g. not when grouse shooting). Obviously when your park ranger says "bridleways" he means "designation=public_bridleway" not "highway=bridleway" in OSM terms.

I'd have no hesitation tagging your "horse access paths" in the New Forest as:

  • highway=bridleway (or highway=track if wide enough for a tractor)
  • horse=yes (legal right as mentioned above)
  • foot=yes (legal right as mentioned above)
  • bicycle=permissive (if cycling is generally allowed)

If no idea if there's an apporpriate designation for this - I'm guessing not. For footways on CRoW Act land I've sometimes added a source:access or source:foot explaining why it's "yes" not "permissive".

permanent link

answered 15 Nov '19, 16:41

SomeoneElse's gravatar image

SomeoneElse ♦
accept rate: 16%

There are tags newforest:pathtype and newforest_pathtype which I have used which I hope bestow on the path (which is just tagged highway=path) those properties appropriate to it being in the New Forest i.e. foot=permissive and horse=permissive, without it having to be stated explicitly. I've found them used elsewhere but can't find them documented. Is this sufficient? Further, I was hoping I could just tag the enclosing area with the permissive property and have the contained paths "inherit" it. After all, the permission covers the whole area and not just the paths within. Tagging the paths individually could be seen as misleading.

(19 Nov '19, 07:49) ceperman

I have to say, I'm a bit unconvinced about "newforest:pathtype". Looking at taginfo, the values seem to be things that are actually best suited to other attributes ("path" is meaningless, "cycleway" is probably expressed as a mixture of access+descriptive tags, "track" whould probably be a highway tag in its own right, "gravel" is a surface tag and "narrow" should be handled by "width").

I'd suggest asking whoever it was who used them before to describe in a bit more detail what they're supposed to indicate.

(19 Nov '19, 15:17) SomeoneElse ♦

OK, I can do that. But the question of inheritance remains - does such a concept exist?. Whichever permissive tags are used, is there some way of avoiding having to tag each path individually? As I mentioned, the permissive nature applies to the area and not just to the paths within it. And this would apply to other permissive areas (such as my local council-owned wood), it's not unique to the New Forest.

(21 Nov '19, 10:42) ceperman

But the question of inheritance remains - does such a concept exist?.

I think the answer is "it depends". There's certainly the concept of default access rights (a highway=motorway can reasonably be assumed to be motor_vehicle=yes, for example). Also a town within a country doesn't need an "is_in:country=blah" tag since that is derivable using a geographic database.

What I'm not convinced of is whether anyone would expect to inherit access rights on a road or track from a surrounding area. To take an example that I added recently, "dog=no" on https://www.openstreetmap.org/way/263486696 was added explicitly to the track because I don't expect people to infer it from the surrounding access land. OSM tends to use a "principle of list surprise" in such cases, so it made sense to explicitly tag the track.

(21 Nov '19, 10:57) SomeoneElse ♦

For roads the only place you might expect a concept of inheritance to work across most data consumers is at the country level: one can assume things like default speeds on classes of roads, which side of the road is used, what standard signage exists, whether roundabouts are clockwise or anticlockwise. Within a country pretty much the only inheritance which works is that of administrative boundaries. So I'd say tagging each way explicitly is the way to go (and is probably a general trend anyway, e.g. what the National Trust are planning to do).

(21 Nov '19, 12:56) SK53 ♦

Hi Chris

From looking at the definitive map referenced above, rights of way inc. bridleways are pretty much restricted to where the forest meets human habitation. I see none in what I'd call the forest proper, so any found there are probably an error.

Just checking you're aware of the bridleway/public bridleway distinction in England & Wales? Even where the definitive map shows no public bridleway, highway=bridleway (a way intended for use by horse riders) need not be an error, as that tagging combination is silent as to the legal right to make your way by horse. See here for more info, especially the section on E&W.

Reading the admittedly non-legal guidance from the New Forest National Park Authority here, I would tentatively suggest that bridleways inside the park perimeter should be tagged as rights of way (designation=public_bridleway), as there is a legal right of way there, despite that no path-specific legal designation on the definitive map is required. Having said that, if it's common knowledge that there's a right to roam, then adding the designated tag may confuse users, if it suggests the right to roam isn't so extensive as they thought and that the tag is special to that way rather than part of the general right.

Presumably the general right doesn't extend to cycles (unlike on a normal public bridleway): as long as my first suggestion above holds good, then where the Forestry Commission permits cycling inside the right-to-roam boundary, I presume foot=designated ; horse=designated ; bicycle=permissive would be best. ("Permissive" generally indicates "permission given which may be withdrawn", distinct from a right, but a time-served senior mapper told me this needn't be rigidly interpreted.)

There may well be other national parks where this nut has been cracked, so I'd wait for others' responses, but hope this helps a bit.

permanent link

answered 15 Nov '19, 16:25

eteb3's gravatar image

accept rate: 6%

edited 15 Nov '19, 16:40

PS, in case it needs saying, obviously copying from the definitive map is a no-no: https://wiki.openstreetmap.org/wiki/Copyright_information_for_UK_mappers (not that you said you were).

(15 Nov '19, 16:32) eteb3

My question was about not marking a way as a bridleway, but I'm not sure how you can make any determination about rights of way, even in a negative sense, without referring to what appears to be derived data. What a minefield!

(16 Nov '19, 08:10) ceperman

I wasn't aware of the distinction between bridleway and public (i.e. RoW) bridleway. How do you identify a non-public bridleway? Public bridleways are signed, non-public bridleway probably not, and how would you distinguish between a way intended for use by horse-riders and one that just happens to be used by horse-riders? In the forested parts of the New Forest there are paths everywhere, but the right to roam means that you're not restricted to these, you can walk and ride anywhere. I see paths there as a way of finding your way around, but not something you have to follow. So, back to my original question, a designation of bridleway on what appears to be just a.n.other path in the forest, where there is no more evidence on the ground of horse-riding on this particular path than anywhere else, seems meaningless. And even if there were, I'm sure it would fall into the just happens category, so still not a bridleway.

(16 Nov '19, 08:28) ceperman

How do you identify a non-public bridleway?

Usually a sign (or perhaps local knowledge). Most of the ones that I've mapped have a sign saying something to the effect of "permissive horse access" - https://www.openstreetmap.org/way/533061950 is one example.

but I'm not sure how you can make any determination about rights of way, even in a negative sense, without referring to what appears to be derived data

Typically also based on signage, although some councils have released their rights of way data under a licence compatible with OSM. See here and the links from there. For more general discussions, the talk-gb mailing list is a pretty good place to start.

(16 Nov '19, 08:53) SomeoneElse ♦
Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here



Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:


question asked: 15 Nov '19, 14:48

question was seen: 3,326 times

last updated: 21 Nov '19, 12:56

powered by OSQA