Informal Consultation - National Trust Path mapping pilot

At National Trust we would like to have an inventory of all paths on National Trust managed land. These paths would include statutory designations (Footpaths, Bridleways, Restricted Byways, Byways) and permissive paths both on publically accessible land and on National Trust pay-for-entry estates. The aim is to enable management of the network, aid the creation and promotion of trails based on the path network and to publicise how National Trust enables access to the countryside.

After reviewing the datasets available to create this inventory we have concluded that Open Street Map would be the best source of data. The main obstacle is the consistency of tagging of paths / tracks particularly in relation to designations.

After reviewing we propose to use a hybrid approach to tag highways as summarised below:
(I will also add a page in the wiki with a formatted table and x-ref from this post when I have done so).

                           Designation                       Access      Highway         Foot           Horse         Bicycle    Vehicle     prow_ref 

Footpath public_footpath yes footway/track designated no no no
Bridleway public_bridleway yes footway/track designated designated yes no
Restricted Byway restricted_byway yes track designated designated designated no
Byway byway_open_to_all_traffic yes track yes yes yes designated
Permissive path permissive_footway permissive footway/track yes no no no -
Permissive Cycle trail Permissive_bridleway permissive footway/track yes/no yes/no yes/no no -

We propose that we would:

• Review existing highway data on all National Trust managed land.
• Utilise our ranger network / property based staff to use on the ground knowledge to verify data.
• Create any highways that do not currently exist – tag as above table.
• Where required split existing tracks to create a path network.
• Update any existing tags that contradict the above tagging (not remove any other tags).
• Create tags on existing highways where they do not currently exist.

In the short term we propose to edit areas where National Trust have land and are outside National Parks and are not covered by data on to understand the input required and to act as example areas. We will publicise where and when we have done this so OSM editors can see what we are proposing. At this stage we would use individual OSM accounts for the editing and annotate the change files to show it is done as part of the proposed process.

After completing these areas we will make a decision as to whether to start a programme to edit all paths on National Trust managed land.

If we proceed with this stage we will review how we set-up editors to ensure we are open as to whom is editing the data.

N.B. We have read and we would open with our editing programme and contactable in case of queries.

By using our network of rangers and countryside staff we can use aerial photos and local knowledge to capture the data thus ensuring we are compliant with copyright rules.

We are contacting the Open Street Map community to seek comments on this proposal.

Kind regards
Ian Dawes
National Trust Conservation Information Manager

Excellent stuff - it’d be great to get some of the gaps filled in (e.g. places where the NT knows the designation and for some reason it’s not in OSM yet).

By the way - if you’re wondering why it seems a bit quiet here it’s because for historical reasons there are more GB-related posts over at than in here. I think that this sort of “standard tags for footpaths” has been discussed over there a couple of times (for example, a web search for “permissive path site:” will turn up some old list threads).

Looking at the actual table, I’d tend to map some things a little differently to how you have them - I wouldn’t assume a footpath as bicycle=no and horse=no, as I can think of public footpaths that are signed as allowing permissive access by both, and there are plenty of constructed “multi user trails” that are also public footpaths - it depends on the individual example. Also as I understand things restricted byways are technically “vehicle=yes; motor_vehicle=no” (though in many cases the physical access restrictions would prevent e.g. a carriage and four from using one, even if the legal access restrictions don’t). I’d also tend to tag public bridleways as highway=bridleway as well as designation=public_bridleway, if they’re different enough on the ground (frequent horse access tends to mean they tend to be less overgrown with wider access, gates rather than stiles, etc.)

You’re absolutely correct that missing designations are probably the biggest issue. There was quite a bit of discussion about this over the summer (see e.g. , and before and after).

I don’t think that anyone at all will object to your using NT rangers to update OSM data. By their very definition they’re familiar with the local area and how it might have changed in recent years (a couple of bad winters or a river eroding a bank can easily make a perviously accessible path unusable, for example). The “Directed Editing Policy”* was designed among other things to allow local mapper feedback to editors who are not local but are perhaps “paid by the yard” to perform a task like “map all the service roads in a large city” - sometimes in those cases quality control can be lacking, mainly because of lack of familiarity with the local area. It’s still under discussion because OSM is a fairly broad church containing people with many different usecases for OSM data, and people have complained that it is either not strong enough at doing what it needs to or too strong, restricting how people can contribute to OSM data. In your case if the person editing the data says “I’m an NT ranger from XYZ updating paths in this area” and engages with any community questions such as “are you sure that the best tag for path X on so-and-so estate is Y” I don’t think there’ll be any problems.

For info there are plenty of ways of viewing OSM data that show the information that you’re interested in - is one obvious one that is designed to help people contribute missing designations among other things. (disclaimer - that’s a site I created) is another - if you zoom in you’ll see different designations of public access displayed differently.

Any other questions, please don’t hesitate to ask.

Best Regards,


  • full disclosure - I’m a member of OSM’s Data Working Group who was tasked with creating the policy, alothough it hasn’t been me personally who’s been working the sausage machine for it.

I’d like to support all of SomeoneElse’s comments. This is very thorough and well set out information. I should also add that this represents an exciting development: OSM being useful enough for an organisation to want to use it.

Like Andy (SomeoneElse) I see no problem at all with NT folk (Rangers or otherwise) contributing new data or updating existing data. Doing this in a more consistent fashion across the country is also beneficial. OSM has tended to get established in particular areas and often this means that tagging conventions can be fairly localised. In the East Midlands I think we have a pretty much common approach to mapping paths and tracks, but we do tend to get together at least once a year which has been good for sharing different approaches.

I’d like to add my personal perspective of mapping paths on NT land, which, I think, reinforces the idea that the NT itself can improve & enhance what already exists.

In my early OSM days I spent a couple of mornings walking as many paths as I could on the Cookham and Maidenhead Commons. Not only is there a network of varying width paths on the open areas around Pinkneys Green and Cookham Rise, but the woodlands of Maidenhead Thicket, Pinkneys Green and the Old Brick Works have very many paths too. Even now, nearly 10 years later, I am conscious that I may not have mapped every path in this area (let alone the whole of this NT estate) and that the paths as mapped could do with more information. On Pinkneys Green proper some of the paths are broad rides which are used by riders, others are so narrow that people cannot walk abreast. Similarly the woods by the car park on Pinkneys Drive have paths which are broad rides (e.g.,, wide paths, and narrow trods.

The access, width and surface of all these paths could undoubtedly be improved. In particular I never resolved which paths horse riders were welcome to use. This is reflected in the inconsistent use of highway=footway and bridleway. In other words there is great room for improvement. The detailed local knowledge of NT rangers or volunteers is very much to be welcomed.

In summary: this is a great initiative and I am willing to offer what support I can.


aka SK53

Not much to add other than to echo support for this! Looks like a great initiative.

If you can add appropriate ‘surface’ tags, that’s always a great help, particularly for cyclists.

I’d recommend against adding access=yes to paths. This is sometimes interpreted as “access permitted for all vehicle classes”, and there are routers which will therefore send cars down highway=footway, access=yes (see for an example). It doesn’t really give any extra information that foot=yes doesn’t, so you can safely leave it off.

The only things I can think of to add to the above discussion are:

In addition to surface=* tags on paths and tracks, tracktype=* might be useful for cyclists and users of mobility aids.

Some NT properties have recommended and waymarked(?) trails, e.g. Stourhead Park Hill Camp walk ( ). Would it be worth considering adding these as route relations (see )?