Sorry to be picky, but `surface=paving_stones` is the commonly used tagging rather than `surface=flagstones`. "paving_stones" is one of the recognised "paved surface types" in the "standard" map.

For the Boots in Team Valley, it would probably be better to have a single `amenity=pharmacy` with `opening_hours:pharmacy`?
It will be confusing for data consumers to have two `amenity=pharmacy`.

Is level=1 correct for this ATM? It would imply it was on an upper floor, but the building looks to be a single level.

I think re-tagging the footways in the Cornmill centre from highway=footway to highway=corridor makes the map less usable - it it hard to understand the connectivity.

`highway=corridor` might make sense if mapping the corridors of a complex building, but is really part of a separate tagging scheme (simple indoor mapping).

Have these PRoWs been confirmed independently e.g. for Strava data? I'm looking at the N/S footway on Hedleyhope Fell and can't see how it relates to ground features.

I've corrected this to `amenity=cafe`. `shop=coffee` is literally a shop that sells coffee beans etc.

Is "Brook Lane Court" correct for the overall collection of buildings? The individual buildings are Broad Lane Court.

Good catch! Embarrassing as I originally added it...

It was confusing because the listing is "Barclays Bank", but refers to the building itself, not the bank brand.

Fixed.

I'm not sure about the name change here:
osm.org/edit#map=19/54.518700/-1.502995
from
"Morton Park Beefeater"
to
"Morton Park"

The latter is the branch name used on the website, but the restaurant itself would be identified as the "Morton Park Beefeater". "Morton Park" is the name of the area, and so it is a bit confusing for the two to have the same name. The visible signage has just "Beefeater".

If the car park is open, why not tag this as amenity=parking?

That's not what permissive means. The original tagging was:
motor_vehicle = no
motor_vehicle:conditional = yes @ (19:00-7:00)
which encodes the 7-7 restriction correctly.

I will revert for now, but it probably needs further checking if the sign has changed.

Conditional access is a very advanced topic and one that has big effects on routers etc. Best to steer clear or ask unless you are 100% sure.

Please also respond to my separate changeset query about not:brand changes.

"motor_vehicle=permissive" only makes sense on a private-owned highway, whereas this is a public highway. Access restrictions are complex, so I think you need to explain how exactly the sign has changed.

I'm not clear why wikidata:brand = Q37158 was changed to
not:wikidata:brand
This is the brand:wikidata used to identify Starbucks locations, for example, in this tool:
https://osm.mathmos.net/chains/Q37158/

I was confused by the NSUL postcode later, which has 8DU for all the buildings. It makes sense to go with 8DS. Fixed.

I'm a bit concerned about this edit.

Most importantly, we are not allowed to use Ordnance Survey maps as a source for OSM.

Secondly, we use `highway=unclassified` for public roads (adopted highways), even if a single lane, rather than `highway=service` (I'm looking here as Roughley Bank).

I really think the changes based on OS maps should be reverted.

Do the addresses added to building perimeters in this changeset correspond to entrances? It would make sense to add entrance=yes to capture this.

These cemetery sections should be tagged using "cemetery=sector" rather than as individual cemeteries. Although not rendered in Carto, cemetery sectors are shown in the Tracestack rendering e.g.
note/758921#map=17/53.426959/-3.046314&layers=PN

I've been able to complete the Wall Trail thanks to an old leaflet I found at home + a survey. These routes (which date from the 1990s) are much decayed, and only work if you have the leaflets (which I can't find online) - they don't link to information panels en route. To be honest, I would delete the fragments of the other trails and leave them to rest in peace. Paul H.

Alec, I'm honestly not picking on your edits, but I'm changing the alley behind Stowell Street back to `highway=service` and adding `service=alley`. I was going past this today, and it is a horrible skanky alleyway and no way a `highway=pedestrian`. I would be very cautious about making changes to highway classifications based on imagery.

I agree that suitability for bicycles is not a deciding factor for the suitability of of highway=bridleway. But this isn't the question; the question is whether a multi-user path intended for pedestrians / cycles / horses / wheelchairs is best tagged as a bridleway.

The accusation of tagging for the renderer is unfair - this implies using incorrect tagging to get a specific effect. The original tagging (not mine) was not incorrect. CyclOSM has an easy job, since it ignores the horse access tags. The interpretation of tags by Carto is more relevant since it needs to consider them all.
Here `highway=path` + `bicycle=designated` is rendered as `highway=cycleway` THEN (as a lower priority) `highway=path` + `horse=designated` is rendered as a bridleway, whether a way is intended for bicycles is a key factor.

Anyway, we are clearly not going to agree, so I'll open an issue on the UK Community Discussion Board.
https://community.openstreetmap.org/c/communities/uk/86