If a public footpath is obstructed, that doesn't change it from being foot=designated + designation=public_footpath

What's the significance of the field being private (ownership) here? Most PRoWs traverse private land.

Thanks for adding these. If you want to make it clear that they're permissive, you can also add foot=permissive

If they're permanent, it might also be worth replacing highway=path + informal=yes with highway=footway + designation=permissive_footpath

I've restored the highway=residential tag to the part of Hullbridge Road behind the gate and corrected it to access=private (no would mean that residents can't drive or walk to their houses).

I've also moved the position of the barrier=gate node back from the intersection with the PRoW, as gates on intersection nodes tend to break routing.

changeset/187587318

Thanks for adding this. For a vegan restaurant, you almost certainly want diet:vegan=only rather than diet:vegan=yes

Thanks - fixed in changeset/187212695

This non-existent crossing was added by you about 10 metres from a crossing node which was mapped in 2015 and the markings of which are clearly visible in the Bing aerial imagery.

node/13848654141

How, exactly, does adding a fictitious crossing improve my experience of using pedestrian data for routing in London?

I am not entirely convinced that drawing the eastern sidewalk of Vallance Road within the fenced boundary of Weaver's Fields is entirely helpful.

Fixed in changeset/187173174

"I acknowledge and thank you for the feedback."

Yet you still did noting to remedy your error.

How likely do you think it is that the sidewalk of a road would cross a navigable canal *in a tunnel* via a ford? The layer=-1 tag on the footway wasn't your fault, but accepting a suggestion which is so obviously wrong is unacceptable.

Pedestrian infrastructure needs to be mapped by people who know and care what they are doing, otherwise it is at best useless.

Both issues fixed in changeset/187172534

Fixed in changeset/187172534

Fixed in changeset/187172534

Also, what was the benefit to pedestrian routing of adding a sidewalk without connecting it to this crossing?

node/4677251400

The sidewalk of the Kilburn High Road bridge over the North London (railway) Line does not cross the line via a pedestrian level crossing, which is absolutely clear from the aerial imagery.

Please DO NOT accept the Rapid editor's suggestions to resolve a potential issue unless you understand what it is asking.

The fact that the carriageway of Kilburn Road crosses the railway line by a bridge should give some indication of how the sidewalk crosses the railway line.

As is immediately obvious from the aerial imagery, the sidewalk of Brunswick Court passes under the railway viaduct out of London Bridge. Tagging it as bridge=yes + layer=1 is clearly incorrect.

The result of you doing this was that the iD editor then displayed a yellow box warning you of a potential issue, as the footway and railway appeared to cross on the same layer. The idea of iD's warning and suggested fixes is that you actually read the text and understand what the effect of each does before selecting one. Just clicking on the first one added pedestrian level crossings across an 11 track railway out of an extremely busy station would be obviously wrong to anyone who knew or cared what they were doing.

Uber's claim with this undiscussed organised edit was that it wanted to improve pedestrian routing. What effect on routing along this edge do you think that the added cost of multiple railway crossings would have?

Thanks for adding these.

If you're adding address details to nodes rather than the building polygons, please could you also move the ref:GB:uprn=* tag (where present) to the node as well, as this should usually be on the same object which has the address.

The sidewalk of the Holloway Road bridge over the Gospel Oak - Barking Railway line does not cross the line via a pedestrian level crossing, which is absolutely clear from the aerial imagery.

Please DO NOT accept the Rapid editor's suggestions to resolve a potential issue unless you understand what it is asking.

The fact that the carriageway of Holloway Road crosses the railway line by a bridge should give some indication of how the sidewalk crosses the railway line.

How likely do you think it is that the sidewalks of a road passing under an extremely busy 11 track railway line would do so via level crossings?

If you don't know or care what you're doing, please DO NOT do it.

I wouldn't change the approach taken by an active mapper without asking first and respecting the response I get. I don't want to start something like the edit war around Exeter - https://community.openstreetmap.org/t/mapper-is-batch-converting-addr-tagged-ways-to-new-nodes/145899

In London, it seems to be more common to have the address for a building which only has a single address on the polygon, but on entrances where there are more addresses (e.g. a house no. 13 converted into flats with separate doors for 13 and 13A).

My preference while mapping is to put the tags on the largest object to which they unambiguously apply. That's usually the building as we don't really do place=plot in the UK. I'm not going to claim that it's the One True Way to do it, because it clearly isn't.

When I import UPRNs and postcodes, the ref:GB:uprn=* and addr:postcode tags get added to whatever already has address tags, whether that's the building polygon, an entrance node, or a node inside the polygon, as that respects the original mapper's preference.

Hi Harry,

Around this part of N4, you added the address details to the entrance=main node on terraced houses between 2016-2020.

Where there is just one entrance and address for a house, would you have any objection to me moving the address tags to the building polygons and changing the node to just entrance=home ?