It would optmise pedestrian navigation even more if you connected the separate sidewalks at clearly marked pedestrian crossings over main roads, like this one node/547514155

I don't doubt that there's a CCTV installation here, but do Flock operate in the UK and how do you know it's ALPR/ANPR?

It looks like you've got the letters of the postcode area transposed here, with AD rather than DA.

Two further questions:
1) Do you occupy the entire building?
2) You removed the tag building:levels=4 - was this incorrectly surveyed?

Drawing the crossing way from sidewalk to sidewalk, rather than kerb to kerb means that surveyors using tools like StreetComplete will not be prompted to check the kerb height or presence of tactile paving. Pedestrian mapping which does not take into account accessibility features is of little value.

Do it properly, or not at all.

Uber may have stopped editing in London, but what are they doing to rectify the problems caused? I'd suggest deleting all objects added by Uber mappers which are still at v1.

When you added separate sidewalks to Vaughan Road, they pass over a bridge which is really obvious in the aerial imagery.

The way you map a highway over a bridge is to add a positive layer=* and bridge=yes

They way you DO NOT map a highway over a bridge is to join it to all the ways it crosses, and accept iD's default (and stupid) suggestion to add a ford where it crosses waterways. Canals in the UK, even ornamental ones, are not crossed by fords. Fords on sidewalks where the parent highway does not cross the waterway by a ford are highly improbable (and in an urban area like London, infinitely improbable).

I actually live in London and use OSM-based routing software. Since Uber started to inflict paid "mappers" like you upon me, it is rare that I can plan a running or walking route without having to to make manual adjustments to my route and then repair OSM.

In this case, thanks to your edit, Komoot gave me a route which required me to travel 5 metres vertically upward through a bridge deck. This would give me a headache.

As an unpaid volunteer, you probably cannot understand how much I resent fixing the sloppy editing of paid corporate "mappers".

Please desist from mapping in London until you have received adequate training.

Thanks!

(Review requested)

Hi David,

In the UK, a highway=cycleway is generally assumed to be for both cyclists and pedestrians, unless it's got a foot=no access tag. It certainly doesn't hurt to add an explicit foot=yes on these.

Where there are sections of segregated cycle track, tagged with segregated=yes I would use highway=cycleway because it was built as one.

There are some shared pavements/sidewalks where the cycle "infrastructure" is just a blue shared path sign, some people prefer highway=footway + footway=sidewalk + bicycle=yes + segregated=no

Some people in the UK don't like highway=path because it can create ambiguity in routing and rendering. In other countries it's the de facto standard.

All the above is my own opinion. Some people will agree, others won't. You can feel free to ignore it, as the changes you have made are unlikely to break cycle or pedestrian routing.

The original mapping of these cycle tracks, done before you edited them, leaves a bit to be desired. They're not split at crossings and where the segregation ends and the "parent" streets like Horton Lane have incorrect or absent tags for sidewalk=* , cycleway=* and verge=* . There are even sections where the footway and cycle track seem to be physically separated by a grass strip rather than a painted white line, which I would probably map as a separate highway=footway and a highway=cycleway.

Rob

With overgrown footpaths, particularly public rights of way like this one, it's usually better to add descriptive tags rather than delete them. The PRoW still exists, even if it's impassable.

Tags which might be useful for this include: trail_visibility=* and obstacle=vegetation

It's also worth adding surface=* and sometimes smoothness=* and (for unpaved paths) tracktype=*

If you have the time, you could also report it to Herts CC at https://www.hertfordshire.gov.uk/services/highways-roads-and-pavements/report-a-problem/report-a-highway-fault/public-rights-of-way/public-rights-of-way.aspx

I've reinstated the way and added obstacle=vegetation

No problem, and thanks for double checking. Some of our highway authorities occasionally do a very good job of creating ambiguity!

If it helps, the traffic sign used on the approach to conventional roundabouts is the red triangular warning sign, TSRGD diagram 510. These aren't mandatory, but street level imagery usually helps where they are absent.
osm.wiki/Road_signs_in_the_United_Kingdom#510

The blue circular sign, TSRGD diagram 611.1, is used on the approach to mini roundabouts.
osm.wiki/Road_signs_in_the_United_Kingdom#611.1

The centre of a mini roundabout also has a white centre with three curved arrows around it, as shown in TSRGD diagram 1003.4 (not yet in the OSM wiki).
https://www.legislation.gov.uk/uksi/2016/362/schedule/9/made#tgp3-tbl3-tbd1-tr5-tc2

Hi,

The most recent Bing aerial imagery and a response to note/5359067 by a local mapper suggest that the junction of Plough Lane / Summerstown / Wimbledon road is a mini roundabout (with a painted traversable centre) rather than a roundabout. Do you have any information at TomTom which contradicts this?

Please don't add layer=-1 (suggesting that the footway area is below ground, when it obviously is not) just to make warnings from the iD go away. It doesn't solve the problem, but it does make it harder to find and correct later.

Please could you explain to me, as someone who actually lives in London and uses OSM data for pedestrian navigation, how this benefits me?

Separate sidewalks which are not connected at crossings, particularly those with road markings which are clearly visible in the aerial imagery, create unnecessary and artificial detours.

Sidewalk stubs which are only connected to other highways at one end serve no purpose other than for decoration.

Sidewalk stubs connected to crossings at the junctions of Stamford Hill with Portland Avenue and Dunsmure Road in changeset/184775834

Disconnected sidewalk stubs and missing crossings are less than useful for pedestrian routing. As I actually live in London and use OSM-based pedestrian routing, it's frustrating (1) to have to use workarounds because bad sidewalk mapping is worse than none at all and (2) having to fix the errors and omissions.

(Review requested)

That looks fine. It might be worth adding the address details for the other buildings on John Watkin Close, which might help reduce the potential for confusion.

Separate sidewalks on Brooke Road and Evering Road connected to Darville Road crossings in changeset/184642886

Are you sure that this CCTV installation is a Flock ALPR/ANPR camera?

If it's a normal CCTV installation, there's a general purpose app you can use to map them. If you install StreetComplete, you can use the "things" layer to add CCTV/surveillance and it won't add US- or Flock-specific tags. There's a YouTube video explaining it here https://www.youtube.com/watch?v=QcsC4d2hU5o

Sidewalk stubs on Blackstock Road and Rock Street connected to crossings in changeset/184431698

Welcome to OpenStreetMap.

A way tagged as highway=footway in the UK already implies no access for other transport modes, so adding bicycle=no is usually redundant.

An exception is where there is a sign explicitly prohibiting cycling on the footway ( traffic_sign=GB:951 ), in which case you can also add bicycle:signed=yes

There's a table of default access assumptions for different highway=* types in the UK at:
osm.wiki/OSM_tags_for_routing/Access_restrictions#United_Kingdom