Odd though it sounds, we can't actually use wikipedia as a source for OSM as the licence is incompatible.
However, I'm sure that there are licence compatible sources (look at whatevev wikipedia used, for example) and also our of copyright sources (since I doubt that it has changed recently).

way/374455218#map=16/51.70150/-1.98438 still has horse=private and bicycle=private, which seems unlikely given that it is designated as a public bridleway?

(for the benefit of anyone stumbling across this) the access=private issue was addressed 5 years ago in changeset/78972463

Hello,
(also about way/374455218/history ). The changeset comment says "Correcting access / type where no ROW exists or where it is incorrect. Footpath only / driveways etc " which doesn't match the situation here.
It's tagged in OSM as a designated public bridleway (see https://map.atownsend.org.uk/maps/map/map.html#17/51.70149/-1.98439/H ) and the local authority also things it is one, too (see https://map.atownsend.org.uk/maps/map/map.html#17/51.70149/-1.98439/H/P ), Something that is designated as a public bridleway would be expected to be horse=yes unless it is closed for some reason (perhaps safety, after a landslip). Anything designated as a public bridleway is also bicycle=yes in terms of legal access tags, although it may not be suitable for regular cycling (and you can use other tags for that).
There's a discussion going on at https://community.openstreetmap.org/t/odd-access-no-tagging-of-uk-public-rights-of-way/132798 - perhaps pop in there and say hello? You can login to that site with your OSM ID.
Best Regards,
Andy

> I mistakenly introduced errors in this edit

Not really - if there is a gap in the trail (or we can't see where it goes) then there should be a gap in OSM, and if the WVW website isn't directly usable then for now we can only go with what is signed (or not) on the ground (specifically here in Symond's Yat East itself).
My recollection of the WVW (from further west) is that signage is very variable.
The email of the https://www.wyevalleywalk.org/ site suggests that it's part of the AONB, so the route might be available from them as OGL unencumbered by the OS mapping they've overlaid it on on their site.
Anyone feeling keen could ask (although if it was me, I'd probably think that was too much like hard work and stick to mapping the signposts!)

Hello,
It looks like this change has removed the section round the north of Huntsham Hill from the Wye Valley Walk: relation/61495#map=15/51.84537/-2.63507 . It was there a few days ago: https://overpass-turbo.eu/s/280C .
Does it need to be added back?
Best Regards,
Andy

Hello,
Just wondered whether there should be a gap in the non-NCN Mercian Way at relation/12015034#map=20/51.7439262/-2.2222510 ? (you can see two orange bits of the relation with a gap here).
Best Regards,
Andy

Hello,
way/1414082973 was newly added here, and you've added it to relation/19341660 (the "Ancient Trees Walk"), but I've also added it to the Way of Roses and 688 cycle routes as there was an obvious gap in those too.
Best Regards,
Andy

Hello,
Just wondered if the extra bit of NCN33 you can see at relation/76059#map=17/50.925816/-2.912251 is really still there, or is part of an old alignment?
Best Regards,
Andy

For info, the DWG got a nudge which mentioned this changeset discussion (not, I hasten to add, a complaint about it directly). I think a forum thread might make more sense (it'd be a bit more visible and allow easier linking) so I've created that at https://community.openstreetmap.org/t/separate-sidewalks-or-not-near-ealing/132613 .

Hello
You've written "added tags" in the changeset comment here, but that doesn't really help other OSM mappers understand what you changed and what the source was.
It would be great if you could use a bit more detail!
Best Regards,
Andy

It was a cockup on my part - I thought that I was changing the "outdoor_seating" tag on the pub, not the "outdoor_seating" tag on the outdoor seating...

This failed with "cannot upload changeset: 520 <none> at /home/ajtown/src/osm-revert-scripts/lib/Progress.pm line 104.", which isn't great.

The changeset comment was supposed to be "Revert undiscussed guessed-geometry import, see osm.org/user_blocks/18370"

Some iffy data remains, see e.g. https://overpass-turbo.eu/s/27wN . Also I've no idea where changeset/124097210 came from.

@deduce technologies : what was the source of this data?

This changeset reverts some or all edits made in changesets 168631271, 168631326, 168631334, 168631918, 168648673, 168675005, 168676597, 168676615, 168680892, 168680909, 168680937, 168693810, 168693861, 168693935.

This changeset reverts some or all edits made in changesets 168631271, 168631326, 168631334, 168631918, 168648673, 168675005, 168676597, 168676615, 168680892, 168680909, 168680937, 168693810, 168693861, 168693935.

This changeset has been reverted fully or in part by changesets 168708079, 168708082 where the changeset comment is: Revert undiscussed guessed-geometry import, see osm.org/user_blocks/18428

This changeset has been reverted fully or in part by changesets 168708079, 168708082 where the changeset comment is: Revert undiscussed guessed-geometry import, see osm.org/user_blocks/18428

This changeset has been reverted fully or in part by changesets 168708079, 168708082 where the changeset comment is: Revert undiscussed guessed-geometry import, see osm.org/user_blocks/18428

This changeset has been reverted fully or in part by changesets 168708079, 168708082 where the changeset comment is: Revert undiscussed guessed-geometry import, see osm.org/user_blocks/18428