On 16/12/2025 00:00, Frederik Ramm wrote:
The problem that I, as an occasional armchair mapper of small towns in
England, have with this approach is that I have no idea how "set in
stone" these ancient way-of-rights are. I frequently encounter them
crassly mismatching aerial imagery. Here is a not-so-outrageous example
but I hope it still illustrates the problem I am facing:
https://www.openstreetmap.org/edit#map=20/51.0444084/-3.3114631
In the centre of the map, there's a "public footpath" that runs right
through a couple of trees and a concrete wall (even after correcting the
imagery offset to match other features).
In that case, I would map it along the actual route.
It *seems* obvious that the footpath should actually be a couple metres
to the East in this location. But I can't just move a right of way, can
I? For all I know, the right of way might indeed enshrine access to that
building's private garden, and what looks like the actual footway on the
picture could just be an attempt by the landowner to trick people out of
their rights!
What you've illustrated here is actually a common misconception among UK
mappers :-)
The reality is that public rights of way in England and Wales are
defined in law, historically, not as lines on a map but in descriptive
terms. In this particular case, the legal definition of the path (WG
15/4) is simply:
"North Street near Spring Gardens and runs in a north easterly direction
through the gardens to Style Road near Sewage Tank."
Originally, before people routinely started fencing their boundaries
between properties, this would literally have meant that there was a
right of way to walk past these houses to the rear, as well as past them
on the front. Over time, as people started to erect fences, the path
took on a more physically defined route as a separate, fenced off path
behind the gardens rather than across them. But the legal definition
hasn't changed. As long as it's still a path which runs approximately
north easterly behind people's houses, it's still the same path.
What that means in practice is that slight changes over time to the
on-the-ground route don't change the legal status, or mean that the
previously mapped route is authoritative. It's perfectly legitimate to
update the map data to reflect the current route of the path, so long as
it's clearly still the same path. It may be the Trigger's broom of
paths, but it's still the same path!
At this point, someone usually jumps in with the words "Definitive Map".
Now, it is true that local authorities have a legal obligation to survey
and map their rights of way, and publish that as the definitive map. But
the definitive map is a snapshot in time, and is subject to continual
review. If the on-the-ground route of a right of way has shifted
slightly over time, and that drift does not constitute a significant and
material amendment to the route, then the latest revision of the map
will simply show the updated route.
Going back to the path you mention, as it happens, Somerset's own Public
Rights of Way map does exactly that, and maps it along the physical
route behind the gardens rather than across them:
https://roam.somerset.gov.uk/roam/map
You'll need to search and zoom in for that, it doesn't seem to have the
ability to hotlink to a specific location or zoom level. But a search
for the path reference, WG 15/4, will give you the right place to home
in on. And if you then select the aerial imagery layer you can see that
the mapped path follows the physical route.
I admit that this example is a little silly because it's just such a
small offset but I have encountered more serious cases where a PROW
footway went through the shrubs on the east of a building, while a nice
asphalted footway was visible on the western side and I was just about
to make the map match the imagery when I saw that it was some sort of
PROW and I thought, well maybe I would be lying if I claimed the PROW
was on the western side.
How do you settle such cases when you encounter them?
I would map the on-the-ground route if visible, or the best
approximation to the described and signposted route if not.
There are cases where development causes an existing right of way to
creep sufficiently far from its original route that local campaigners
get upset about it and try to get the original route reinstated. There
are also cases where prospective development will clearly require a
visible diversion of a right of way and the highways authority will
therefore require the developer to submit an official change request for
the diversion. But none of that really matters to OSM mappers.
So long as we can see a path, and it matches the legal description of a
right of way, we can map it as it is on the ground and tag it with the
legal status. If a path does later become the subject of a local
dispute, then it's worth keeping an eye on how that progresses in order
to be prepared for any necessary changes. But we should not be slavish
adherents of previously published definitive maps. If it's obvious that
a path has slightly changed route since the definitive map was last
published, then we should update our records to show the current route
even if the local authority has yet to do so.
As an aside, there's some disagreement about whether the definitive map
is actually a valid source for OSM anyway, due to licencing issues.
Robert Whittaker has some useful advice on that, here:
https://osm.mathmos.net/prow/council-docs.html
But that's another reason why we shouldn't pin everything on the
definitive map. The legal status of a path is not subject to any
licensing restrictions, so we can always tag it correctly. But the
precise route on the ground will often need to be surveyed rather than
simply copied.
Mark
_______________________________________________
Talk-GB mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/talk-gb
