Perhaps some refinement of obstacle? I know there are initiatives like this from Living Streets - Cut the Clutter to identify this, some of which might be suitable for OSM.
Where footway parking is regulated, we might have parking:$side=on_kerb or parking:$side=half_on_kerb tagged on the road, which could give a useful hint to routers if the sidewalks aren’t mapped separately. Some way of tagging the presence of footway parking together with existing tagging for effective width?
Now imagine someone at $GiantCorporation came in and remapped this to a separate sidewalk way
I recently mapped a sidewalk obstructed by dumpsters:
I originally just added obstacle=dumpsters but that felt rather like troll tagging since the dumpsters block the entire sidewalk and appear to be there all the time. Another mapper pointed me to some previous discussion on this and I’ve since also changed the separate sidewalk way to disused:highway=footway + disused:footway=sidewalk and put disused:sidewalk:left[1] on the central street way.
It feels a little odd to tag the disused:* prefix because the sidewalk is “in use”, it’s just in use as a space for dumpsters rather than a space for walking. So applying it to the footway and sidewalk tags feels appropriate since the way clearly isn’t used for walking on the side as it stands.
way points toward camera ↩︎
I think this is actually insightful - what I’m complaining about in part is tagging as pavement ways what are actually “kerbs” - spaces alongside the road reserved to stop vehicles colliding with walls, often used for storage or parking. They were never for foot traffic, except maybe to be used as refuges.
Tagging disused:footway=sidewalk on a kerb is like tagging disused:leisure=pitch on a grass verge - but at least they don’t break routing so I don’t actually mind if we standardise on that.
The example in the photo I posted looked to me like it had been used as a sidewalk at one point (it’s at least wide enough to have been), but it’s hard to know. Perhaps it has always been dumpster storage and has never been used as a sidewalk. If that were known to be the case, not:highway=footway + not:footway=sidewalk would be a better fit, communicating to the next mapper “yes there is a thing that looks like it might be a sidewalk here, and no it is not actually a sidewalk and should not be mapped as one”. This works as an exception in an area that is otherwise full of correct separately mapped sidewalks. In an area that mostly has lots of these “not-sidewalk” lookalikes, it would be nice to not have to add a separate way for each one just to hopefully prevent a future mapper from adding a sidewalk way where there shouldn’t be one.
I suppose my follow-up to this would be “well what are we doing then" - we’re not making the map more useful for anyone and we’re degrading it for some people. I think a footway should be traversable, otherwise it’s not a footway.
I don’t agree, in that I don’t think that a map should be a catalogue of all things that exist in space, but I don’t really mind people adding more things to the database so long as they don’t make it worse.
Would you be open to the suggested not:footway=sidewalk (for the kerb example) and disused:footway=sidewalk f(or pavements that can’t currently be used)?
Or some value of area:highway between the barrier=kerb way and the wall, together with sidewalk:$side=no on the road? Using not:footway=sidewalk suggests a linear feature to me.
I don’t remember whether I shared this before or elsewhere, but here’s some “food for thought” with respect to partial mapping.
Let’s say we have a situation on-the-ground like this:
To start, we only have the houses mapped:
With this data, a router can only make routes “as the crow flies” - not perfect, but functional and reasonably accurate:
Now, let’s say someone comes in and adds only these roads, which are in fact present on-the-ground, but the mapping is incomplete:
Our router’s output, now, is technically valid (it’s doing the best it can) but produces worse results than the no-street-mapping case we saw earlier:
Of course, were the roads fully mapped, we’d get the most-correct route:
So, the question is: Should the snake-like road mapping be deleted because its accurate-but-incomplete mapping results in worse routing?
My answer to that prompt is “No.”
In line with that, I object to the removal of as-termed “decorative” sidewalks.
This conclusion does not apply to “non-sidewalks” - the very small (essentially an extended curb top surface) areas shown in some images posted earlier are not functionally or structurally sidewalks, and should not be tagged as such. Edge cases of barely-usable sidewalks are better solved by tagging details like width. Sidewalks that do exist and are permanently unusable should be tagged in a way that indicates that status, such as with a lifecycle prefix.
I fully agree, but where are people mapping footways which are not usable en mass?
There are only 3-4 footways mapped in my town that you can’t walk down with ease, and both of these are marked with horrible smoothness, and intermediate to bad visibility.
Disused footway is a good suggestion, especially places where pavements have “legally” be en adopted as something else, or the Biffa bin example was quite good… but only should exist if a pavement existed before (and still exists under the new use).
I have been for some time mulling in my mind though about pedestrian routing and classification. For roads it is easy, both the classification of the road, speed limits etc. If you want to do a 100 mile trip, a router will know to connect you to a faster major road, vs as the crow flies through town roads.
This doesn’t really work for pedestrian routing. A sidewalk is a sidewalk, and classification can be muddled. 40MPH city centre roads vs 20mph city centre residential roads, its likely the 40mph road would have much better pedestrian infrastructure associated with it, vs the minefield of parked cars and bins on the residential road, but a router may think it “friendlier” to route a pedestrian down a quieter street.
Mapping pedestrian infrastructure as separate along these high volume route could be beneficial to move people along safer routes…. yet at the same time other people may not class that as safer, and extracting the pavement data from that road may make it less favourable in other routers.
Surely that should only be for only one day?
The rule here is that they are only put out on collection day.
Unfortunately not, it depends on the location. Similarly, there are sidewalks that are essentially permanently blocked by parked cars.
No. The “dumpsters" around the back of shops tend to stay there all the time, and on many streets of terraces the “wheelie bins" also tend to get left out permanently (because the only other option is to drag them all the way down to the alleyway, which might not even be usable anymore anyway).
(On the continent they use on-street shared bins but we left dammit 
)
Manchester, and it’s a big problem that many people have commented on.
So my proposal above (which I split out into a separate thread) was based on “permeability" of a road - that is, how easy it is to cross between the carriageway, pavement, etc. A proposal to base it on the road classification would be more conservative than my own, but I would support it if it had consensus.
That’s the law, but there some local authorities which refuse to use their enforcement powers under s. 137 Highways Act 1980 or s. 46 Environmental Protection Act 1990. On roads with a lot of HMOs in Newham, wheelie bins are permanently (and almost certainly illegally) kept on the footway at all times.
[EDIT: There are powers to make exceptions, which would probably be used in some of the cases which @JamesBelchamber mentions]
The same authority also isn’t very keen on enforcing the prohibition of footway parking when it’s school run related.
I’m not sure it’s the same everywhere.
A pain finding one with that isn’t overflowing on Fuerteventura and no apparent recycling option.
okey, so this is another lets-post-a-lot thread, which any sane person throws up into AI and lets it summarise it.
I know hallucinations, but really i risk less with those than to keep arguing the same thing over and over.
Sidewalks are not the problem.
Real problem is incomplete mapping
So i was pointing numerous times that crossing a street at any point from one sidewalk to another is simply impossibe via OSM mapping.
But is it? Turns out that noone ever helped me to understand this problem in depth there.
AI did by digging previous talks here and there and that subject.
So turns out routings could route from one separatelly mapped sidewalk to the other just fine. The only reason they dont, knowing sidewalks and street highway widths and having the generated area that its proximity is just fine for routing, is that stuff like barriers, or difference in levels, possible embarkments, retaining walls are not mapped.
There was a separation proposal to solve exactly this routing problem.., but nowadays community keeps disagreeing and argue to no end rather than face the issue and solve it.
I’m very worried that OSM needs a team that will try to solve a problem in depth and vote it internally with practical coordination with a mapping service team, so things start making sense and we can have actual issues solved.
People keep worry about separate sidewalks, i usually mix that with cycleways… but the real problem are actually wide streets for car traffic with very complicated lane widths, their merge, individual geometry, areas excluded from traffic with paint, overtaking ability…
People keep saying that its the area:highway that should be for all of that, but then it really is not and not only it causes editing clutter and problems, noone ever uses it. A real issue is sidetracked and no working solition is given.
I hope this post is going to help someone, sry for writing this much.
Any sane person sees that red flag and thinks: AI;DR
so much feedback.
Better solve the routing issue of routing to door entrances by a sidewalk to door entrances on the other side of the street by the other sidewalk.
EDIT> But i gues removing all the sidewalks & not solving any of those issues is going to help, pointing AI red flag even more is.
i knew it’s pointless, but i responded anyway, gnight
I’ve added a lot of sidewalks and I’ve never suggested removing all of them. A deliberate misrepresentation is even more tedious than inflicting AI slop on an otherwise useful discussion.
