Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

On Tue, Jul 18, 2017 at 9:24 AM marc marc  wrote:

> Le 18. 07. 17 à 16:01, Nick Bolten a écrit :
> >> All crossing between a sidewalk and a driveways I have tag have the same
> >> type of kerb on each side. It's why I use kerb=lowered without any need
> >> for left/right details, it is for the whole crossing.
> > I think I'm confused again
>
> highway=residential
> sidewalk=both
> kerb=raised
>
> highway=service
> service=driveway
> sidewalk=no
>
> on shared node
> kerb=lowered
>

I've seen some routers interpret kerb=* along a way as being a potential
barrier, though hopefully that is restricted to non-road ways. This would
be doubly confusing for a combined road+sidewalk way: is this kerb feature
encountered as one walks down the sidewalk? Both kerb=raised and
kerb=lowered are used on footway=sidewalk ways to describe changes in
displacement.


>
> >> Therefore I never needed to ask myself how I tag a useless mixed layout
> >> A crossing with a raised kerb on one side and a lowered kerb on the
> >> other side is as unusable for wheelchair as if it was raised on both
> >> sides.
> > Not true! Some wheelchair users are happy to go down a raised curb
> kerb=raised is defined on the wiki as "implies wheelchair=no"
> so it is used on kerb with a high height and wheelchair routing avoid it
> Of course you can also break this...
> Do you know a tool capable of making a wheelchair routing with
> kerb=raised usable in incline=down direction ?
> Or are you talking about a theoretical usecase that currently
> does not exist ?


Then that wiki note for kerb=raised isn't accurate.

I've done such a thing using pgRouting in the past. It actually required
guessing the 'high' and 'low' side, since incline=up/down is rare. Roughly,
given the kerb node, the previous way, and the next way:

cost AS CASE
  WHEN node.kerb = 'raised'
  THEN CASE
WHEN prevway.footway = 'sidewalk' AND nextway.isroad
THEN 0
ELSE Inf
  ELSE 0
END

Most routers don't yet have a notion of 'going from this way to that way'
available to their user-editable cost functions, which is more relevant to
pedestrians than street traffic (but they do hard-code similar logic for
elements of street traffic, e.g. OSRM and turn restrictions).

Also, this information is useful for demographics outside of people with
disabilities or other explicit mobility needs. A parent walking a stroller
cares about the availability of curb ramps and the direction of traversing
raised curbs. A parent with small children may want there to be curbs or
some other barrier to protect them from cars (that's their purpose, after
all). A cyclist may want to take a route that never involves riding along
the street - but also not up any raised curbs. Such a cyclist would also
often be fine with routes where one goes down a raised curb every so often.


> Keep foot routing working before thinking about exceptional cases...
> because exceptional cases 'll not work if you break many things !
>

I'm not sure how any of these points break foot routing...

But these are not exceptional cases, these are normal cases: unlike with
automotive traffic, pedestrian needs do not fit neatly into 1-3 categories,
so we must instead rely on identifying specific potential barriers / access
modifiers. Ask 20 people from the 10-20% of the population with specific
mobility requirements and you will get 20 unique (and often incompatible)
routing profiles, including their treatment of curbs. And this is before we
get into requirements beyond mobility, such as impaired vision or hearing,
or safety concerns. This is also why our maps are almost useless for a huge
portion of the population: they treat pedestrian access as a monolith that
essentially amounts to, "you can go on some streets and also these extra
footways".

Best,

Nick
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 18. 07. 17 à 14:29, Martin Koppenhoefer a écrit :
>> You can try to make a proposal that mean : those 2 way (street/sidewalk)
>> are only one for the routing.
>> maybe a relation like associatedstreet or that extend it.
> type=area does this somehow, as it defines (as default) that you can cross 
> anywhere on the area. 
> You can also define barriers between the members to say where you have an 
> obstacle, etc.

This proposal seems freezed and need some work.
4 reply on the talk page. 4 "don't understand"
I understand the idea, not how software 'll use it easily :)
Do you know where it is in test ?
maybe it is needed to first make pedestian area working before willing 
to make an relation between sidewalk/street/barrier working for routing
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 18. 07. 17 à 16:01, Nick Bolten a écrit :
>> All crossing between a sidewalk and a driveways I have tag have the same
>> type of kerb on each side. It's why I use kerb=lowered without any need
>> for left/right details, it is for the whole crossing.
> I think I'm confused again

highway=residential
sidewalk=both
kerb=raised

highway=service
service=driveway
sidewalk=no

on shared node
kerb=lowered

>> Therefore I never needed to ask myself how I tag a useless mixed layout
>> A crossing with a raised kerb on one side and a lowered kerb on the
>> other side is as unusable for wheelchair as if it was raised on both
>> sides.
> Not true! Some wheelchair users are happy to go down a raised curb
kerb=raised is defined on the wiki as "implies wheelchair=no"
so it is used on kerb with a high height and wheelchair routing avoid it
Of course you can also break this...
Do you know a tool capable of making a wheelchair routing with 
kerb=raised usable in incline=down direction ?
Or are you talking about a theoretical usecase that currently
does not exist ?

Keep foot routing working before thinking about exceptional cases... 
because exceptional cases 'll not work if you break many things !
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

On Sun, Jul 16, 2017, 1:55 PM marc marc  wrote:

> All crossing between a sidewalk and a driveways I have tag have the same
> type of kerb on each side. It's why I use kerb=lowered without any need
> for left/right details, it is for the whole crossing.
>

I think I'm confused again: is the kerb key being used on the node shared
by the street and driveway, or the one shared by the driveway and sidewalk,
or both? Or are you describing a different situation?

Therefore I never needed to ask myself how I tag a useless mixed layout
> A crossing with a raised kerb on one side and a lowered kerb on the
> other side is as unusable for wheelchair as if it was raised on both sides.
>

Not true! Some wheelchair users are happy to go down a raised curb but not
up one, or will cross all raised curbs below a specific height. There are
also non-wheelchair users with similar requirements.

For a T crossing (one street and a sidewalk), maybe kerb=raised because
> feature is the same (forget the lowered side as it is useless) I agree it
> is not perfect, this crossing also :-) But currently it is the only way to
> have a full working routing between 2 sidewalk hoocked to a way.
>

I'm having trouble visualizing this - maybe an example or diagram would
help?

Best,

Nick

>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Martin Koppenhoefer]

sent from a phone

> On 16. Jul 2017, at 20:18, marc marc  wrote:
> 
> You can try to make a proposal that mean : those 2 way (street/sidewalk) 
> are only one for the routing.
> maybe a relation like associatedstreet or that extend it.


type=area does this somehow, as it defines (as default) that you can cross 
anywhere on the area. You can also define barriers between the members to say 
where you have an obstacle, etc.

Cheers,
Martin 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

Interesting! Since the data is on a node on the street way, how do we
figure out if it's left/right side? kerb:left/right=*? Or do we figure it
out spatially (find the driveway way and do some math)?

On Sun, Jul 16, 2017, 1:14 PM marc marc  wrote:

> Le 16. 07. 17 à 20:38, Nick Bolten a écrit :
> >> There is no need to use so many section. A crossing is a node, not a
> >> section/way. So put one kerb=raised on the way and kerb=lowered on the
> >> node. It's done :-) You have the same number of section/tag in both
> cases.
> > Hmm, I'm not sure I understand. Which node would get kerb=lowered? Since
> > I'm talking about driveways, is it the node shared by the driveway and
> > street?
> yes
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 16. 07. 17 à 20:38, Nick Bolten a écrit :
>> There is no need to use so many section. A crossing is a node, not a 
>> section/way. So put one kerb=raised on the way and kerb=lowered on the 
>> node. It's done :-) You have the same number of section/tag in both cases. 
> Hmm, I'm not sure I understand. Which node would get kerb=lowered? Since 
> I'm talking about driveways, is it the node shared by the driveway and 
> street?
yes

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

> There is no need to use so many section. A crossing is a node, not a
section/way. So put one kerb=raised on the way and kerb=lowered on the
node. It's done :-) You have the same number of section/tag in both cases.

Hmm, I'm not sure I understand. Which node would get kerb=lowered? Since
I'm talking about driveways, is it the node shared by the driveway and
street?

On Sun, Jul 16, 2017, 11:09 AM marc marc  wrote:

> Le 16. 07. 17 à 01:29, Nick Bolten a écrit :
> > a block with 10 driveways would
> > actually need to be split into 10 lowered/flush curb sections and 11
> > raised sections, for a minimum of 21 segments for a single block.
>
> There is no need to use so many section.
> A crossing is a node, not a section/way.
> So put one kerb=raised on the way and kerb=lowered on the node.
> It's done :-) You have the same number of section/tag in both cases.
>
> Regards,
> Marc
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

> You can tag the curbs at each side of a crossing using left/right tags,
and you can find out the length by looking at the road's width (or estimate
it from the number of lanes). It's not perfect, but at least there are good
enough ways to deal with this

But you can't handle the directional features without forcing an
azimuth-style annotation: if you want to describe the curbs or APS at each
end, you have to have a notion of start/end, so you also need a direction.
Both the azimuth and the width are ways to compensate for representing a
linear feature as a node, and for good reason you'd be hard-pressed to find
any examples of that having been annotated. highway=crossing is really for
car routing (where it causes a delay penalty) and maybe sometimes
rendering, and does not factor into any current pedestrian routing
solutions outside of some toy examples. There really aren't good enough
ways to deal with this, and no one really tries to route the totality of
pedestrians: the data model doesn't support it and the data largely doesn't
exist yet.

> The same cannot be said for the problems caused by sidewalks mapped as
separate ways. Drawing a sidewalk way next to the road produces just that:
Two ways with no machine-readable semantic relationship to each other.

This is an important point that should be addressed via tagging
improvements: by default, separate sidewalks have no metadata-based
relationship to their street, which can be useful. This is basically the
only thing that's inherently better by labeling streets with sidewalk
information, and also basically the only thing anyone can currently use as
a result. This can be addressed in a few ways that probably fall outside
the scope of this thread: (1) just keep the sidewalk=right/left/both/no
tags and use them for this sole purpose, (2) use an associated street tag
(how many buildings are coded to streets), or (3) an associated street
relation (how many other buildings are coded to streets).

Also, with that said, associating sidewalk lines with street lines is a
core part of workflows I use, so it's not totally impossible.

> From personal experience: I'm rendering sidewalks in a 3D renderer. It
works pretty well with lane-like sidewalk tagging, but separate ways
prevent pretty much everything I would like to do with them: I can't draw
the kerbs, because I cannot easily determine whether I'm on the left or the
right of the street. I can't make sure that the sidewalk is at comparable
elevation to the road on hilly terrain. And the sidewalk and road will
always have random gaps or overlap each other, because they are never
really accurately drawn.

As part of our workflow, we also figure out left/right associations (some
straightforward vector stuff), but it would be much better to use one of
the 3 solutions above. I'm not sure if I understand the elevation issue -
is there any chance you're not interpolating? And the issue of overlapping
streets/sidewalks indicate data errors, not tagging errors: either the
sidewalks are drawn in the wrong place or (more commonly) the street widths
are inaccurate. I've fixed so many incorrect-unit-width streets.
> Other applications have similar problems:

> Want to make sidewalks into a part of the road's line style? You can't –
instead you end up drawing them as separate lines, which looks ok at
detailed zooms, but stops
working as you zoom out. Want to show routing instructions such as "follow
the right sidewalk of Foobar Road"? Doesn't work, because we don't know
which road this sidewalk is the sidewalk of.


All good points that could hopefully be addressed by those 3 options.

> Want your pedestrian router to cross smaller roads anywhere, which is an
integral part of how pedestrians typically navigate street networks? Again,
you
can't – you need to hope the mapper has drawn "virtual crossings" at
semi-regular intervals (or even at all).

This was addressed earlier in the thread, and the crossings aren't virtual,
they're just unmarked (and are part of current tagging schemas).

> Yes, separate sidewalk geometries allow for nice details like telephone
poles on a sidewalk. But they lose information that's a lot more
fundamental: The semantic connection with the road.

That's not a detail, it's a fundamental barrier to anyone in a wheelchair.

> That's only a problem if you map this lowering of the kerb as a property
of the highway way. I believe it may be feasible to map this as the
property of the junction and crossing nodes instead.

I don't see how this would work. It also seems like an argument against
sidewalk:*:kerb in general?

> Plus, drawing separate ways only seems easier because you are omitting
essential information, as described above. If you actually were to
express the relationship between the sidewalk and the road with a
relation for each section of sidewalk, it would very quickly stop being
simple.

But it would nicely separate concerns - and a relation isn't the only
option.

On Sun, Jul 16, 

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 16. 07. 17 à 14:24, Svavar Kjarrval a écrit :
> I do think that the routers should
> be programmed to evaluate when it's safe to suggest to the user to cross
> the street without a specifically mapped crossing.
it is already the case
use sidewalk tag on the way when you can cross the street
use a separated way when you can't cross

> I'd see myself as a vandal if I were to start
> deleting footways en masse, for that purpose.
You can try to make a proposal that mean : those 2 way (street/sidewalk) 
are only one for the routing.
maybe a relation like associatedstreet or that extend it.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 16. 07. 17 à 01:29, Nick Bolten a écrit :
> a block with 10 driveways would 
> actually need to be split into 10 lowered/flush curb sections and 11 
> raised sections, for a minimum of 21 segments for a single block.

There is no need to use so many section.
A crossing is a node, not a section/way.
So put one kerb=raised on the way and kerb=lowered on the node.
It's done :-) You have the same number of section/tag in both cases.

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Svavar Kjarrval]

On lau 15.júl 2017 11:13, marc marc wrote:
>
> Your demonstration is only that a wrong map create sometimes a wrong 
> routing :-)
> What will your reaction be when Mapzen tell you to cross a road where it 
> is impossible ? However this is exactly the current map for [2]
> You would not agree that the routing would make you drive from one road 
> to another because they are close and it save 1km. Therefore I do not 
> understand why you would want the routing to do this when you are walking.
> If the map is wrong, first fix the map, not the routing engine.
My example was rather the inconsistency in the assumptions the routers
are making based on the current data. I do think that the routers should
be programmed to evaluate when it's safe to suggest to the user to cross
the street without a specifically mapped crossing. Due to the potential
legal and logistics issues, I do assume the routers would need more map
data than is currently provided. The data isn't innocent. Some routers
assume living streets have that property but one (currently) can't tell
the router it may assume the same regarding a specific segment of
another highway type. Is a part of the solution to implement such a tag?
I don't know.
>
>> but GraphHopper directs the user to take a complicated path
> Complicated because the map is wrong in this case.
> GraphHopper will not make you cross a road where a mapper tell 
> (unintentionally but erroneously) it is impossible.
> Is it bad? IMHO no, it is the best reply.
> I think that is the problem. You would like the routing to guess errors 
> and guess where we can jump from one path to the other in the absence of 
> connection.
> But the simplest/efficient/only solution is to fix the map.
To continue from my point before. GraphHopper is ready to determine the
user has reached the destination when on the street itself, therefore
assuming no barriers from that location to the destination, but is not
ready to (properly) take into account the same situation from the
starting point to the street itself. If the routing engine had done that
from the beginning, it would've suggested the user to cross the street
instead of the much longer route. In another case I tested, Mapzen
suggested that I would start on a footway on the other side of the house
across the street, therefore starting my journey on foot by jumping over
a house. There is a street between the houses, which could lead to the
destination albeit by a little longer route, so it wasn't absolutely
forced in its decision. When it comes to start and end positions
compared to the calculated path between them, there is a big routing
bias favouring the former.

I think it's unavoidable for routers to guess mapping errors, although
we should always aim to limit the type of guesses they need to make,
especially when it comes to the aforementioned jumps. If the routing
software is mainly written to work efficiently in areas with perfect
mapping data, they start to become much worse where the data is
incomplete/incorrect. The aim is, of course, to have perfect mapping
data, but it's not realistic to expect that to apply everywhere at any time.

There has been some effort to detect where mapping data has the
potential to cause problems in routing calculations, and I'm all for
that. Then there are situations where the cause is based on our own
limits, like the availability of officially accepted tags to deal with
certain situations. There are official accepted tags to indicate tha
location of a barrier but none (that I know of) to indicate the absence
of them. The router doesn't *really know* there are no barriers in the
space between; for all it knows, the footway and the street could be
separated by a local black hole. In the GraphHopper case, it's more
afraid of local black holes when it leaves no. 73 than when it suggests
stopping on the street in front of no. 42. So, how can we help the
routers deal with this? Deleting the footways is not a realistic action
in this case, IMHO.
>> Just to be clear: Is it valid, in your opinion, to connect the end of a
>> footway along a street, directly to the street itself?
>> I'm not objecting to such a method, I've just been hesitant to apply it
>> without approval by documentation or the community.
> Guidelines for roads is very easy: split a road in 2 when you can NOT 
> switch from one to the other (for example a road with a island). and 
> create connection where you can switch. But do NOT cut a road into 2 if 
> you switch everywhere from one to the other (for example a street with 2 
> lanes must be keep as one street, not 2).
> Just do the same with the sidewalk as if the sidewalk was a lane 
> reserved for pedestrians. I never see a problem with that.
The data is already there. I'd see myself as a vandal if I were to start
deleting footways en masse, for that purpose. There might also be valid
future cases where it becomes necessary to keep them separete, which
would then urge the community to 

Re: [Tagging] Formally informal sidewalks

[Tobias Knerr]

On 15.07.2017 19:06, Nick Bolten wrote:
[...]
> It can't properly describe
> crossings, since they've been condensed into a node, but important
> information like length, the curbs at each side (direction of
> traversal + curb type both matter), APS directionality, etc, are all
> essentially linear features.

You can tag the curbs at each side of a crossing using left/right tags,
and you can find out the length by looking at the road's width (or
estimate it from the number of lanes). It's not perfect, but at least
there are good enough ways to deal with this.

The same cannot be said for the problems caused by sidewalks mapped as
separate ways. Drawing a sidewalk way next to the road produces just
that: Two ways with no machine-readable semantic relationship to each other.

From personal experience: I'm rendering sidewalks in a 3D renderer. It
works pretty well with lane-like sidewalk tagging, but separate ways
prevent pretty much everything I would like to do with them: I can't
draw the kerbs, because I cannot easily determine whether I'm on the
left or the right of the street. I can't make sure that the sidewalk is
at comparable elevation to the road on hilly terrain. And the sidewalk
and road will always have random gaps or overlap each other, because
they are never really accurately drawn.

Other applications have similar problems: Want to make sidewalks into a
part of the road's line style? You can't – instead you end up drawing
them as separate lines, which looks ok at detailed zooms, but stops
working as you zoom out. Want to show routing instructions such as
"follow the right sidewalk of Foobar Road"? Doesn't work, because we
don't know which road this sidewalk is the sidewalk of. Want your
pedestrian router to cross smaller roads anywhere, which is an integral
part of how pedestrians typically navigate street networks? Again, you
can't – you need to hope the mapper has drawn "virtual crossings" at
semi-regular intervals (or even at all).

Yes, separate sidewalk geometries allow for nice details like telephone
poles on a sidewalk. But they lose information that's a lot more
fundamental: The semantic connection with the road.

> If the curb is raised, every time a driveway
> or alley intersects the sidewalk, the curb changes to lowered or flush.
That's only a problem if you map this lowering of the kerb as a property
of the highway way. I believe it may be feasible to map this as the
property of the junction and crossing nodes instead.

Plus, drawing separate ways only seems easier because you are omitting
essential information, as described above. If you actually were to
express the relationship between the sidewalk and the road with a
relation for each section of sidewalk, it would very quickly stop being
simple.

> Hope this makes some sense... it feels a bit ranty.

Yeah, I know the feeling. In fact, I couldn't resist the urge to write a
rant of my own in response. ;)

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

> no, it isn't a pedestrian way, it is a street with sidewalk, it is not the
same for routing.

There is certainly a dedicated pedestrian (and maybe cycling) way there:
the sidewalk. If the sidewalk:right* keys are meant to only describe
features of the street, then they are complementary to, rather than an
alternative to, highway=footway, footway=sidewalk.

> well... every way is a spatial inaccuracy because we use a way in stead of
an area..  (...)

For sure! There's always a sacrifice of spatial accuracy in mapping
abstractions. But there's also a point where they get stretched too far -
and in these cases, we're trying to shoehorn separated pedestrian ways and
features encountered along them into street centerlines, with all the
downsides and awkwardness I mentioned.

> Or not, the street and the sidewalk do not change often here (> 10
years), the sidewalk is often the same type/surface/kerb ... for the whole
street.

Sure, but more data gets added as people map more features. And I
underestimated the number of segments: a block with 10 driveways would
actually need to be split into 10 lowered/flush curb sections and 11 raised
sections, for a minimum of 21 segments for a single block. This is ignoring
curb ramps (each adds 2 more), and other features meant to split road ways,
like turn lanes, lane numbers, parking lanes, etc. And then someone comes
along and wants to say, 'the surface of the sidewalk from here to here is
asphalt' and sees a bajillion segments. This is with the additional loss of
specificity and visibility of the features, and the only gain is that a
pedestrian network is shoehorned into a street network.

> Before advising people to describe as separate path when this is not the 
> casethe
routing must first understand that our 2 way are a sidewalk connected to a
street.If not, you break the routing and the only advantage seems to move
sidewalk 3m away.

Not sure if I understand what you mean, I'm not sure how this breaks
routing.

> Read me again, I did not say you can walk on the road if a sidewalk
exist. I stated the rule used to know when a road must be single or
splited. (...)

I think there might be a misunderstanding, because I didn't say otherwise...

> By keeping this rule for sidewalks, we avoid a lot of routing problem.

But I'm pointing out that by that same rule, sidewalks should usually be
separated. Your rule of thumb was this: "Guidelines for roads is very easy:
split a road in 2 when you can NOT switch from one to the other (for
example a road with a island)."

The trouble is that unlike roads, where vehicular access falls into very
few category restrictions, pedestrians have a huge diversity of
restrictions. So for many (10-20% of the population), you "can NOT switch"
from the sidewalk to the street due to things like curbs or bollards or
safety concerns.

And finally, for routing to actually handle street crossings as nodes and
things like driveways, they need to essentially recreate the topology of
separated sidewalks anyways (and none do this). These features are actually
mostly ignored for pedestrian routers and used instead for routing
motorized vehicles.

But we're getting a bit off-topic. In terms of the original question, I
think using highway=forward, footway=crossway is the least-bad option for
making sure separate sidewalk ways are well-connected to the street grid,
including when sidewalks terminate.

Best,

Nick

On Sat, Jul 15, 2017 at 1:53 PM Martin Koppenhoefer 
wrote:

>
>
> sent from a phone
>
> > On 15. Jul 2017, at 08:13, Marc Gemis  wrote:
> >
> > I think adding sidewalks might benefit pedestrian routing
>
>
> adding driveways benefits pedestrian routing as well, because you can
> consider all those little crossings as potentially dangerous, and route
> people (e.g. kids) through places with fewer driveway crossings. Plus you
> see for the gates whether cars can pass.
>
> Admittedly, it really depends on the situation whether I'd map all of them
> or refrain for the moment because it seems too tedious and too little to
> gain.
>
> Cheers,
> Martin
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Martin Koppenhoefer]

sent from a phone

> On 15. Jul 2017, at 08:13, Marc Gemis  wrote:
> 
> I think adding sidewalks might benefit pedestrian routing


adding driveways benefits pedestrian routing as well, because you can consider 
all those little crossings as potentially dangerous, and route people (e.g. 
kids) through places with fewer driveway crossings. Plus you see for the gates 
whether cars can pass.

Admittedly, it really depends on the situation whether I'd map all of them or 
refrain for the moment because it seems too tedious and too little to gain.

Cheers,
Martin 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 15. 07. 17 à 19:06, Nick Bolten a écrit :
>  > marc marc wrote:
>  > For wheelchair routing.
>> If all crossing have a lower kerb, it is maybe enough to add
>> sidewalk:both:wheelchair=yes to the street.
> wheelchair=yes should be used sparingly,
you are right, i use it only when there is no barrier, when all crossing 
are lowered, enough width , good surface, no incline
In case of doubt, I do not put this tag.

>>> easier to add characteristics for wheelchair routing to them.
>> you can add those tag to the street as documented here
>  > https://wiki.openstreetmap.org/wiki/Wheelchair_routing#Sidewalks
>  > for example I use the following tag on the street :
>  > sidewalk=right
> Yes, but that's a fairly unnatural way to describe a pedestrian way: 
no, it isn't a pedestrian way, it is a street with sidewalk, it is not 
the same for routing.

> leads to spatial inaccuracy
well... every way is a spatial inaccuracy because we use a way in stead 
of an area.. and in real live and routing, it doesn't care, everything 
work fine except that the map doesn't show the real width.
See area:highway proposed features pending for a few years :-)

> it becomes a real maintainability nightmare.

Or not, the street and the sidewalk do not change often here (> 10 years),
the sidewalk is often the same type/surface/kerb ... for the whole street.
But if you cut the sidewalk away off the street, routing problem will 
often occur.
Before advising people to describe as separate path when this is not the 
casethe routing must first understand that our 2 way are a sidewalk 
connected to a street.If not, you break the routing and the only 
advantage seems to move sidewalk 3m away.

>  > Guidelines for roads is very easy: split a road in 2 when you can NOT
>> switch from one to the other (for example a road with a island). and
>> create connection where you can switch. But do NOT cut a road into 2 if
>> you switch everywhere from one to the other (for example a street with 2
>> lanes must be keep as one street, not 2).
>> Just do the same with the sidewalk as if the sidewalk was a lane
>> reserved for pedestrians. I never see a problem with that.
> Ah, but it's almost never valid to switch from a sidewalk 'lane' to a 
> street 'lane', 
Read me again, I did not say you can walk on the road if a sidewalk exist.
I stated the rule used to know when a road must be single or splited.
By keeping this rule for sidewalks, we avoid a lot of routing problem. 
Routing will be able to choose where to cross the street or what todo 
when sidewalk stops because tags show that the sidewalk is connected to 
the street without obstruction between them.

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

> marc marc wrote:
>
> For wheelchair routing.
> If all crossing have a lower kerb, it is maybe enough to add
> sidewalk:both:wheelchair=yes to the street.

wheelchair=yes should be used sparingly, because it's making an editorial
decision on behalf of wheelchair users, who actually have a wide range of
accessibility preferences. Instead, we can mark potential barriers and let
them (or routing software) decide. For example, one person may consider a
rolled curb accessible (many manual wheelchair users) while another may
consider that inaccessible (certainly many powered wheelchair users). I
personally treat it as an override, such as when an implied barrier isn't
really a barrier, or when there are no obvious alternatives yet, such as
building entrances.

>> easier to add characteristics for wheelchair routing to them.
> you can add those tag to the street as documented here
> https://wiki.openstreetmap.org/wiki/Wheelchair_routing#Sidewalks
> for example I use the following tag on the street :
> kerb=raised
> sidewalk=right
> sidewalk:right:smoothness=excellent
> sidewalk:right:surface=asphalt
> sidewalk:right:tactile_paving=no
> sidewalk:right:wheelchair=yes

Yes, but that's a fairly unnatural way to describe a pedestrian way: as one
of two lanes of a street. It makes those footways invisible by default,
coded into a left/right from forward schema, leads to spatial inaccuracy
(the sidewalk is 30 meters away, there is a telephone pole blocking part of
the path...), etc. It can't properly describe crossings, since they've been
condensed into a node, but important information like length, the curbs at
each side (direction of traversal + curb type both matter), APS
directionality, etc, are all essentially linear features. Finally, it
rapidly explodes street lines with sidewalk information. If the curb is
raised, every time a driveway or alley intersects the sidewalk, the curb
changes to lowered or flush. If there are 10 driveways along a street, the
street must now be a minimum of 11 separate ways due solely to annotating
sidewalk information. Add in other features that may split the way
(smoothness, surface, barriers, street features like lanes, parking lanes,
etc) and it becomes a real maintainability nightmare.

> Guidelines for roads is very easy: split a road in 2 when you can NOT
> switch from one to the other (for example a road with a island). and
> create connection where you can switch. But do NOT cut a road into 2 if
> you switch everywhere from one to the other (for example a street with 2
> lanes must be keep as one street, not 2).
> Just do the same with the sidewalk as if the sidewalk was a lane
> reserved for pedestrians. I never see a problem with that.

Ah, but it's almost never valid to switch from a sidewalk 'lane' to a
street 'lane', or vice versa. You can cross the street, but simply walking
in the middle of it is at least unsafe, and sometimes illegal, unless it's
a pedestrian street. When pedestrian routers fall back to streets, it's
implied/hoped that you'll do the safe thing and stay out of car traffic. In
addition, most sidewalks have curbs or bollards or some other form of
barrier to some large minority of people. So, by this rule, we should
almost always draw separate pedestrian ways.

Hope this makes some sense... it feels a bit ranty.

Best,

Nick

On Sat, Jul 15, 2017 at 6:30 AM John Willis  wrote:

>
>
> > On Jul 15, 2017, at 7:04 PM, Svavar Kjarrval  wrote:
> >
> > Just to be clear: Is it valid, in your opinion, to connect the end of a
> > footway along a street, directly to the street itself?
>
> If the street becomes the route, I say yes, especially if there is no
> reasonable barrier to prevent pedestrians from continuing.
>
> In some cases, there are barriers to stop to stop any kind or traffic. In
> other cases, one side ends and the other continues, meaning peds would
> reasonably cross there to access the other sidewalk (providing there is no
> reasonable zebra crossing nearby, a situation the occurs in my
> suburban-rural area frequently), which would the. Require an unmarked
> crossing.
>
>  But If you are walking on the sidewalk and it abruptly ends (especially
> if it makes some affordance to transition you onto the shoulder, like a
> taper or a break in the curb to allow cyclists free passage) - then by all
> means! The road is now the way you are traveling on, not the sidewalk -
> which up until 3 meters before, could have been a curb, hedge, and fence
> separated way!
>
> Javbw.
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[John Willis]

> On Jul 15, 2017, at 7:04 PM, Svavar Kjarrval  wrote:
> 
> Just to be clear: Is it valid, in your opinion, to connect the end of a
> footway along a street, directly to the street itself?

If the street becomes the route, I say yes, especially if there is no 
reasonable barrier to prevent pedestrians from continuing. 

In some cases, there are barriers to stop to stop any kind or traffic. In other 
cases, one side ends and the other continues, meaning peds would reasonably 
cross there to access the other sidewalk (providing there is no reasonable 
zebra crossing nearby, a situation the occurs in my suburban-rural area 
frequently), which would the. Require an unmarked crossing. 

 But If you are walking on the sidewalk and it abruptly ends (especially if it 
makes some affordance to transition you onto the shoulder, like a taper or a 
break in the curb to allow cyclists free passage) - then by all means! The road 
is now the way you are traveling on, not the sidewalk - which up until 3 meters 
before, could have been a curb, hedge, and fence separated way! 

Javbw. 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Andre Engels]

On Sat, Jul 15, 2017 at 12:04 PM, Svavar Kjarrval  wrote:

>>> where the footway ends
>>> prematurely, the routing software doesn't know it may suggest such a
>>> "jump" onto the street or not,
>> the end of the footway must be connected to the street if you are
>> able/allowed to switch to the street by foot.
>> If needed, cut the road : one segment with sidewalk=left/righ, second
>> segment with sidewalk=no
> Just to be clear: Is it valid, in your opinion, to connect the end of a
> footway along a street, directly to the street itself? That is, to
> continue the footway but make a direct connection from the end of the
> footway directly to the street.
> I'm not objecting to such a method, I've just been hesitant to apply it
> without approval by documentation or the community.

If 'follow the footway to the end, then move onto the street and
continue walking on the street' is a logical (or even possible)
walking route, then connecting the end of the footway to the street in
my opinion is not only valid, but actually the *only* correct way of
mapping that has the footway separately.


-- 
André Engels, andreeng...@gmail.com

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 15. 07. 17 à 12:04, Svavar Kjarrval a écrit :

> This point is demonstrated in my quoted example [2]. 
> Mapzen assumes the user can jump over the road
>  (or assume the user is already there) and walk a few steps, 
Your demonstration is only that a wrong map create sometimes a wrong 
routing :-)
What will your reaction be when Mapzen tell you to cross a road where it 
is impossible ? However this is exactly the current map for [2]
You would not agree that the routing would make you drive from one road 
to another because they are close and it save 1km. Therefore I do not 
understand why you would want the routing to do this when you are walking.
If the map is wrong, first fix the map, not the routing engine.

> but GraphHopper directs the user to take a complicated path
Complicated because the map is wrong in this case.
GraphHopper will not make you cross a road where a mapper tell 
(unintentionally but erroneously) it is impossible.
Is it bad? IMHO no, it is the best reply.
I think that is the problem. You would like the routing to guess errors 
and guess where we can jump from one path to the other in the absence of 
connection.
But the simplest/efficient/only solution is to fix the map.

>>> where the footway ends prematurely, the routing software
>>>  doesn't know it may suggest such a "jump" onto the street or not,
>> the end of the footway must be connected to the street if you are
>> able/allowed to switch to the street by foot.
>> If needed, cut the road : one segment with sidewalk=left/righ, second
>> segment with sidewalk=no
> Just to be clear: Is it valid, in your opinion, to connect the end of a
> footway along a street, directly to the street itself?
> I'm not objecting to such a method, I've just been hesitant to apply it
> without approval by documentation or the community.
Guidelines for roads is very easy: split a road in 2 when you can NOT 
switch from one to the other (for example a road with a island). and 
create connection where you can switch. But do NOT cut a road into 2 if 
you switch everywhere from one to the other (for example a street with 2 
lanes must be keep as one street, not 2).
Just do the same with the sidewalk as if the sidewalk was a lane 
reserved for pedestrians. I never see a problem with that.

> [2] A link with navigation routing:
> http://www.openstreetmap.org/directions?engine=mapzen_foot=64.08769%2C-21.90140%3B64.08802%2C-21.90113#map=19/64.08791/-21.90122

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Svavar Kjarrval]

On fös 14.júl 2017 11:08, marc marc wrote:
> Le 14. 07. 17 à 12:20, Svavar Kjarrval a écrit :
>> A street with a sidewalk on either side but no marked crossings:
>> http://www.openstreetmap.org/#map=17/64.08800/-21.89846
>> (Sidenote: If one tries to route from no. 73 to 42,
>> GraphHopper suggests a long route while Mapzen assumes the user is
>> already on the other side of the street)
> It is a fault (and in my opinion a mistake) to tag a sidewalk separated 
> from the road where it is not!
> there is only one point that the maper create to connect the sidewalk 
> and the road https://www.openstreetmap.org/node/2673312760
> of course routing can only use this point, luckily !
>
> A sidewalk really isolated from the road (= by a barrier) does not allow 
> crossing outside a crossing. This is the current situation of your example.
> This is not specific to the sidewalk, the same happens with roads:
> If you cut a road with 2 lanes into 2 road without any link between 
> them, routing will not allow you to jump from one lane to the other.
The routing engine seems to have made the mistake of latching too much
onto the footway and ignoring the road as a possible point of travel.
I've noticed the behaviour in some routing engines using OSM data that
when it finds a footway close by which has a possible route to the
destination, it tends to ignore the nearby roads completely or place too
little value on them. Some routing algorithms mistakenly ignore a
seemingly expensive first link even though it might lead to a much
cheaper overall path. My previous routing example [1] seems to
demonstrate that.

I do acknowledge those limits on routing engines, although it's mostly
in the assumptions the programmers are prepared to make. This point is
demonstrated in my quoted example [2]. Mapzen assumes the user can jump
over the road (or assume the user is already there) and walk a few
steps, but GraphHopper directs the user to take a complicated path via
the footway in front of the house (totalling 1.1 km), only to end the
directions by suggesting to the user to walk on the street itself to the
front of the house with the starting pin. Mapzen seems to be prepared to
assume that there are no barriers between the starting pin and the
footway on the other side of the street, but GraphHopper does not do
that initially but is, later on, prepared to assume that there are no
barriers from the street itself and to the destination point.

Routing engines aren't perfect and it's a major balancing act when it
comes to performance and cost of operations. That's why I'm interested
in knowing if there's something we can do, data-wise, to help the
routing engines perform such tasks quickly and cheaply, with the end
result being directions which would make sense to the user. At least
while avoiding tagging purely for the sake of the router.

>
>> where the footway ends
>> prematurely, the routing software doesn't know it may suggest such a
>> "jump" onto the street or not,
> the end of the footway must be connected to the street if you are 
> able/allowed to switch to the street by foot.
> If needed, cut the road : one segment with sidewalk=left/righ, second 
> segment with sidewalk=no
Just to be clear: Is it valid, in your opinion, to connect the end of a
footway along a street, directly to the street itself? That is, to
continue the footway but make a direct connection from the end of the
footway directly to the street.
I'm not objecting to such a method, I've just been hesitant to apply it
without approval by documentation or the community.
>
>> I haven't been able to find any tag or method to do it
> a road with not-separed sidewalk should be taged as such :-)
Is there documentation or guides on how to apply that in different
situations and how to solve some of the gray areas?
>
>> the "common sense approach" would expect.
> routing doesn't know "common sense approach" :)
> if 2 sidewalk or roads are taged as "separated without any link", 
> routing can't guess that a connection exists.
Yep, we're not that far yet. :P
But until it does, it might not hurt to help it along by aiding it make
such decisions. While I do understand that need, I don't want to apply
and advocate measures which might damage the data quality needlessly.

[1]
https://www.openstreetmap.org/directions?engine=graphhopper_foot=64.14793%2C-21.96048%3B64.14875%2C-21.96216#map=18/64.14809/-21.96170
[2] A link with navigation routing:
http://www.openstreetmap.org/directions?engine=mapzen_foot=64.08769%2C-21.90140%3B64.08802%2C-21.90113#map=19/64.08791/-21.90122

With regards,
Svavar Kjarrval


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 15. 07. 17 à 08:13, Marc Gemis a écrit :

>> On Jul 14, 2017, at 11:32 PM, Nick Bolten  wrote:
>>> --> need to add all driveways?
> On Fri, Jul 14, 2017 at 11:56 PM, John Willis  wrote:
>> This is generally a good idea - and to make sure they share a node.
> Why ? What is the benefit of adding driveways of 3-5 meters long ?
For wheelchair routing.
If all crossing have a lower kerb, it is maybe enough to add 
sidewalk:both:wheelchair=yes to the street.
But if some crossing have a raised kerb, wheelchair routing can't work 
without a crossing with any existing way including service.

> I experimented with it in my neighborhood and the only thing it does
> is confuse navigation programs.
I have the opposite :-)
Maybe connect the entrance of the house with a path to driveway.
Maybe map only the beginning of the driveway where the crossing is located.

>> To me, if you are considering adding sidewalks, you’ve already committed to
>> adding the service roads/tracks/etc.
> I think adding sidewalks might benefit pedestrian routing
See the first question, create a sidewalk path separately in osm from 
the street where the sidewalk is in fact hooked to the street create 
currently more routing problem that it solve.

> easier to add characteristics for wheelchair routing to them.
you can add those tag to the street as documented here
https://wiki.openstreetmap.org/wiki/Wheelchair_routing#Sidewalks
for example I use the following tag on the street :
kerb=raised
sidewalk=right
sidewalk:right:smoothness=excellent
sidewalk:right:surface=asphalt
sidewalk:right:tactile_paving=no
sidewalk:right:wheelchair=yes

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[John Willis]

> On Jul 15, 2017, at 4:18 PM, Nick Bolten  wrote:
> 
> sort (not unlike a *_link for roads)

This was my reasoning for highway= footway_link earlier, perhaps 
highway=footway_routing might be a more accurate tag. 

=} 

Javbw. 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[John Willis]

> On Jul 15, 2017, at 3:13 PM, Marc Gemis  wrote:
> 
> My neighbour's driveway is longer than
> mine (it's a company) and now OsmAnd insists on taking his, because it
> comes closer to my house.

Admittedly, I don’t map a lot of residential driveways (because most 
residential in Suburban Japan has no sidewalks), AND I do not use OSM for 
routing, but wouldn’t a footway linking your building to your driveway solve 
this? 

When I map retail buildings and I know where the doors are, I map the entrance 
and add the mappable footway to the entrance node from the parking driveway. 

Would doing the same for your location solve a routing problem? 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

To Marc:

> Why ? What is the benefit of adding driveways of 3-5 meters long ?
I experimented with it in my neighborhood and the only thing it does
is confuse navigation programs. My neighbour's driveway is longer than
mine (it's a company) and now OsmAnd insists on taking his, because it
comes closer to my house.

In my case, it's for pedestrians: it is often a de facto ramp from the
sidewalk to the street (and vice versa), which is a practical route for
many wheelchair users. It can also imply an uneven surface as you go from
sidewalk to driveway to sidewalk, which can be a barrier to some.

It doesn't really add much in terms of routing cars, like you note.

On Sat, Jul 15, 2017 at 12:18 AM Nick Bolten  wrote:

> To John:
>
> Those are all very good points. This one is particularly interesting:
>
> >An example of this issue is where a road with no sidewalks meets another
> road with sidewalks, but does not cross it (and is not in an urban environ,
> so there is no real paint to show a crossing=zebra) . Do you add a
> crossing=unmarked that goes from the sidewalk to the node of the road’s T
> junction? People on the sidewalk far side of the T junction will expect to
> be able to cross the street there and continue on the road.
>
> I don't think we really have adequate tags to describe that situation, so
> everyone makes due by either doing what you suggest (a half-crossing) or
> connecting footways/sidewalks directly to roads. Neither makes perfect
> semantic sense: it's not really a road crossing and it's also not really a
> sidewalk, it's just a change of path that a pedestrian would realistically
> need to make. It should probably use an entirely new tag for a pedestrian
> transition of some sort (not unlike a *_link for roads), but that would of
> course need to be hashed out in a separate proposal. In the meantime, I
> also tend to use highway=footway, footway=crossing, crossing=unmarked to
> connect an ending sidewalk to the road.
>
> On Fri, Jul 14, 2017 at 11:46 PM Andre Engels 
> wrote:
>
>> My strategy in this kind of case is to add those driveways and virtual
>> crossings that are useful for routing purposes. So if there is a
>> junction, if there is a driveway opposite it, I will add that driveway
>> (or maybe just the part of the driveway upto the sidewalk), if there
>> is none, but people can cross there (in the case you describe: If
>> there is an interruption in the hedge), I add a footway from the
>> sidewalk to the junction, if neither is the case, I add the driveway
>> or crossing point that is closest to the junction (on both sides if
>> necessary).
>>
>> On Fri, Jul 14, 2017 at 3:48 PM, Marc Gemis  wrote:
>> > Another typical case
>> >
>> > - no explicitly marked crossings
>> > - sidewalk parallel to road
>> > - kerb separating sidewalk from road
>> > - hedge, interrupted for each driveway and at the junctions, placed on
>> > sidewalk, parallel with road.
>> >
>> > --> need to add all driveways ?
>> > --> need to draw virtual crossings at junctions ?
>> >
>> > m
>> >
>> >
>> > On Fri, Jul 14, 2017 at 3:41 PM, Mike N  wrote:
>> >> On 7/14/2017 8:14 AM, Marc Gemis wrote:
>> 
>>  but merge sidewalk with the road where the is no space/barier between
>>  them.
>> >>
>> >>
>> >>> and that's were the discussion starts. When I asked when one has to
>> >>> draw a separate sidewalk a few weeks ago on this mailing list someone
>> >>> answered: as soon as there is a kerb.
>> >>
>> >>
>> >>   Similarly, I have been combining sidewalks with roads where there is
>> no
>> >> separation.   But when there is a small grass separation from the
>> roadway,
>> >> they are drawn separately.  For those cases, it is usually allowed to
>> cross
>> >> the grassy separation and the road to get to the opposite sidewalk.
>> >>
>> >>   Throwing out the R word here - what about a relation that defines
>> which
>> >> disconnected ways could be walked to or across from any point on a
>> current
>> >> way?   That would also include the road since there would be no
>> barrier.
>> >>
>> >>
>> >>
>> >> ___
>> >> Tagging mailing list
>> >> Tagging@openstreetmap.org
>> >> https://lists.openstreetmap.org/listinfo/tagging
>> >
>> > ___
>> > Tagging mailing list
>> > Tagging@openstreetmap.org
>> > https://lists.openstreetmap.org/listinfo/tagging
>>
>>
>>
>> --
>> André Engels, andreeng...@gmail.com
>>
>> ___
>> Tagging mailing list
>> Tagging@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/tagging
>>
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

To John:

Those are all very good points. This one is particularly interesting:

>An example of this issue is where a road with no sidewalks meets another
road with sidewalks, but does not cross it (and is not in an urban environ,
so there is no real paint to show a crossing=zebra) . Do you add a
crossing=unmarked that goes from the sidewalk to the node of the road’s T
junction? People on the sidewalk far side of the T junction will expect to
be able to cross the street there and continue on the road.

I don't think we really have adequate tags to describe that situation, so
everyone makes due by either doing what you suggest (a half-crossing) or
connecting footways/sidewalks directly to roads. Neither makes perfect
semantic sense: it's not really a road crossing and it's also not really a
sidewalk, it's just a change of path that a pedestrian would realistically
need to make. It should probably use an entirely new tag for a pedestrian
transition of some sort (not unlike a *_link for roads), but that would of
course need to be hashed out in a separate proposal. In the meantime, I
also tend to use highway=footway, footway=crossing, crossing=unmarked to
connect an ending sidewalk to the road.

On Fri, Jul 14, 2017 at 11:46 PM Andre Engels  wrote:

> My strategy in this kind of case is to add those driveways and virtual
> crossings that are useful for routing purposes. So if there is a
> junction, if there is a driveway opposite it, I will add that driveway
> (or maybe just the part of the driveway upto the sidewalk), if there
> is none, but people can cross there (in the case you describe: If
> there is an interruption in the hedge), I add a footway from the
> sidewalk to the junction, if neither is the case, I add the driveway
> or crossing point that is closest to the junction (on both sides if
> necessary).
>
> On Fri, Jul 14, 2017 at 3:48 PM, Marc Gemis  wrote:
> > Another typical case
> >
> > - no explicitly marked crossings
> > - sidewalk parallel to road
> > - kerb separating sidewalk from road
> > - hedge, interrupted for each driveway and at the junctions, placed on
> > sidewalk, parallel with road.
> >
> > --> need to add all driveways ?
> > --> need to draw virtual crossings at junctions ?
> >
> > m
> >
> >
> > On Fri, Jul 14, 2017 at 3:41 PM, Mike N  wrote:
> >> On 7/14/2017 8:14 AM, Marc Gemis wrote:
> 
>  but merge sidewalk with the road where the is no space/barier between
>  them.
> >>
> >>
> >>> and that's were the discussion starts. When I asked when one has to
> >>> draw a separate sidewalk a few weeks ago on this mailing list someone
> >>> answered: as soon as there is a kerb.
> >>
> >>
> >>   Similarly, I have been combining sidewalks with roads where there is
> no
> >> separation.   But when there is a small grass separation from the
> roadway,
> >> they are drawn separately.  For those cases, it is usually allowed to
> cross
> >> the grassy separation and the road to get to the opposite sidewalk.
> >>
> >>   Throwing out the R word here - what about a relation that defines
> which
> >> disconnected ways could be walked to or across from any point on a
> current
> >> way?   That would also include the road since there would be no barrier.
> >>
> >>
> >>
> >> ___
> >> Tagging mailing list
> >> Tagging@openstreetmap.org
> >> https://lists.openstreetmap.org/listinfo/tagging
> >
> > ___
> > Tagging mailing list
> > Tagging@openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/tagging
>
>
>
> --
> André Engels, andreeng...@gmail.com
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Andre Engels]

My strategy in this kind of case is to add those driveways and virtual
crossings that are useful for routing purposes. So if there is a
junction, if there is a driveway opposite it, I will add that driveway
(or maybe just the part of the driveway upto the sidewalk), if there
is none, but people can cross there (in the case you describe: If
there is an interruption in the hedge), I add a footway from the
sidewalk to the junction, if neither is the case, I add the driveway
or crossing point that is closest to the junction (on both sides if
necessary).

On Fri, Jul 14, 2017 at 3:48 PM, Marc Gemis  wrote:
> Another typical case
>
> - no explicitly marked crossings
> - sidewalk parallel to road
> - kerb separating sidewalk from road
> - hedge, interrupted for each driveway and at the junctions, placed on
> sidewalk, parallel with road.
>
> --> need to add all driveways ?
> --> need to draw virtual crossings at junctions ?
>
> m
>
>
> On Fri, Jul 14, 2017 at 3:41 PM, Mike N  wrote:
>> On 7/14/2017 8:14 AM, Marc Gemis wrote:

 but merge sidewalk with the road where the is no space/barier between
 them.
>>
>>
>>> and that's were the discussion starts. When I asked when one has to
>>> draw a separate sidewalk a few weeks ago on this mailing list someone
>>> answered: as soon as there is a kerb.
>>
>>
>>   Similarly, I have been combining sidewalks with roads where there is no
>> separation.   But when there is a small grass separation from the roadway,
>> they are drawn separately.  For those cases, it is usually allowed to cross
>> the grassy separation and the road to get to the opposite sidewalk.
>>
>>   Throwing out the R word here - what about a relation that defines which
>> disconnected ways could be walked to or across from any point on a current
>> way?   That would also include the road since there would be no barrier.
>>
>>
>>
>> ___
>> Tagging mailing list
>> Tagging@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/tagging
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging



-- 
André Engels, andreeng...@gmail.com

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Marc Gemis]

On Fri, Jul 14, 2017 at 11:56 PM, John Willis  wrote:
>
> On Jul 14, 2017, at 11:32 PM, Nick Bolten  wrote:
>
>> --> need to add all driveways?
>
> This is generally a good idea - and to make sure they share a node.
>
>

Why ? What is the benefit of adding driveways of 3-5 meters long ?
I experimented with it in my neighborhood and the only thing it does
is confuse navigation programs. My neighbour's driveway is longer than
mine (it's a company) and now OsmAnd insists on taking his, because it
comes closer to my house.


> To me, if you are considering adding sidewalks, you’ve already committed to
> adding the service roads/tracks/etc.

I think adding sidewalks might benefit pedestrian routing, and makes
it somewhat easier to add characteristics for wheelchair routing to
them. I only added driveways when I thought it might be useful for
routing (e.g house is not near main roads so routing needs some help
to properly work).

regards

m

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[John Willis]

> On Jul 14, 2017, at 11:32 PM, Nick Bolten  wrote:
> 
> > --> need to add all driveways?
> 
> This is generally a good idea - and to make sure they share a node.

To me, if you are considering adding sidewalks, you’ve already committed to 
adding the service roads/tracks/etc. 

Adding the hidden crossing (crossing=unmarked) IMHO is the thing to be 
discussed.  

An example of this issue is where a road with no sidewalks meets another road 
with sidewalks, but does not cross it (and is not in an urban environ, so there 
is no real paint to show a crossing=zebra) . Do you add a crossing=unmarked 
that goes from the sidewalk to the node of the road’s T junction? People on the 
sidewalk far side of the T junction will expect to be able to cross the street 
there and continue on the road. 

In my region, even on major national roads sidewalks abruptly stop, let alone 
on tertiary roads. Usually a road is being brought up to a modern standard 
section by section, but the surrounding roads are not. A building project 
forced the adjacent roads to be upgraded, but the beginning and end of those 
roads are still the older narrow versions, such as this tertiary road here:

https://www.openstreetmap.org/#map=19/36.42388/137.87330

Because where the sidewalk abruptly ends is dangerous for peds, I put in an 
unmarked crossing to the other side, and linked the sidewalk to the road where 
it ends to the east. 

I am committed to mapping all sidewalks as separate ways, because they often 
have routing completely separate from roads in Japan, and the nature of them 
appearing and disappearing be mapped with a separate way is the best way to 
show this - but how peds “rejoin” the road when it does end is what is not 
documented in OSM.


Javbw ___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 14. 07. 17 à 15:41, Mike N a écrit :
> when there is a small grass separation from the 
> roadway, they are drawn separately.  For those cases, it is usually 
> allowed to cross the grassy separation and the road to get to the 
> opposite sidewalk.
you can add a access tag like foot=permissive on the grass area but 
routing right now seems to not support pedestrian area.
you should look for the state of the proposals on this subject.

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

If those two footways make up a reasonable continuing path, that's a good
case for using the unmarked crossing tagging schema. It communicates all of
the features actually being traversed (footway -> crossing the street ->
footway) and is extensible: you can easily add curb and surface information.

On Fri, Jul 14, 2017 at 7:35 AM Martin Koppenhoefer 
wrote:

>
>
> sent from a phone
>
>
> On 14. Jul 2017, at 13:16, Marc Gemis  wrote:
>
>
> But what if there are no crossings marked? Do we have to invent
>
> crossings then ? (e.g. near each junction)
>
> It is not uncommon to have such a network of sidewalks without
>
> "zebra"-crossings.
>
> People are allowed to cross everywhere then.
>
>
>
> I wouldn't map the sidewalks as proper footways in this case, but you
> could also "connect" them with a relation, e.g. type=area. The footways
> inside the block should connect with the road, e.g. here
> http://www.openstreetmap.org/?mlat=64.08781=-21.89969#map=19/64.08781/-21.89969w
>
> cheers,
> Martin
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Martin Koppenhoefer]

sent from a phone

> On 14. Jul 2017, at 13:16, Marc Gemis  wrote:
> 
> But what if there are no crossings marked? Do we have to invent
> crossings then ? (e.g. near each junction)
> It is not uncommon to have such a network of sidewalks without
> "zebra"-crossings.
> People are allowed to cross everywhere then.


I wouldn't map the sidewalks as proper footways in this case, but you could 
also "connect" them with a relation, e.g. type=area. The footways inside the 
block should connect with the road, e.g. here 
http://www.openstreetmap.org/?mlat=64.08781=-21.89969#map=19/64.08781/-21.89969w

cheers,
Martin ___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Nick Bolten]

> --> need to add all driveways?

This is generally a good idea - and to make sure they share a node.

> --> need to draw virtual crossings at junctions?

These aren't totally artificial/virtual. You can consider them 'unmarked
crossings' and there's already tags on the wiki: highway=footway,
footway=crossing, crossing=unmarked.

Most, but not all, 'weird' routes can be avoided by adding unmarked
crossings. One exception is routing between two POIs on opposite sides of
the street, but:
  - This is not a primary use case for end user routing. If foot traffic
can reasonably be expected to cross anywhere and it's just across the
street, the end user can figure out their actual path easily. Users that
aren't comfortable figuring out that path should most likely not be
crossing at arbitrary locations, and would prefer crossing at junctions
(and preferably marked crossings).
  - Routers can always fall back to using streets / ignoring footways for
short distances. Routers that bake costs into their graph will need to
create an extra graph, but that's an inherent limitation for extending that
approach.

> Throwing out the R word here - what about a relation that defines
> which disconnected ways could be walked to or across from any point on a
> current way?   That would also include the road since there would be no
> barrier.

This is potentially tricky and should probably go in its own thread
eventually (OpenSidewalks people and I have been meaning to do a good
overview and RFC related to it). Example:

- The left side of the street has 1 sidewalk line (sidewalk A)
- The right side of the street has 2 sidewalk lines (sidewalks B, C), split
right in the middle of the block

Drawing:


---
BCC

I think this would imply needing 2 relations. And if the relation only
contains the info 'sidewalk A', 'sidewalk B', 'street', the simple
interpretation implies that one can always get from sidewalk A to sidewalk
B via the street. However, not all crossings from A to B are orthogonal:
the diagonal crossings may be impractical. So routers will need to
incorporate spatial logic as well, essentially just adding unmarked
crossings at various points. Totally doable, but not super simple, either,
and that strategy doesn't actually require the relation.e

A natural reaction is to revert to treating sidewalks as metadata of
streets, but this is no good for accessible routing, and is also not a good
description of the pedestrian transportation network. But for the most
part, I have yet to see an example of a really bad route when unmarked
crossings are added at appropriate locations.

On Fri, Jul 14, 2017 at 6:50 AM Marc Gemis  wrote:

> Another typical case
>
> - no explicitly marked crossings
> - sidewalk parallel to road
> - kerb separating sidewalk from road
> - hedge, interrupted for each driveway and at the junctions, placed on
> sidewalk, parallel with road.
>
> --> need to add all driveways ?
> --> need to draw virtual crossings at junctions ?
>
> m
>
>
> On Fri, Jul 14, 2017 at 3:41 PM, Mike N  wrote:
> > On 7/14/2017 8:14 AM, Marc Gemis wrote:
> >>>
> >>> but merge sidewalk with the road where the is no space/barier between
> >>> them.
> >
> >
> >> and that's were the discussion starts. When I asked when one has to
> >> draw a separate sidewalk a few weeks ago on this mailing list someone
> >> answered: as soon as there is a kerb.
> >
> >
> >   Similarly, I have been combining sidewalks with roads where there is no
> > separation.   But when there is a small grass separation from the
> roadway,
> > they are drawn separately.  For those cases, it is usually allowed to
> cross
> > the grassy separation and the road to get to the opposite sidewalk.
> >
> >   Throwing out the R word here - what about a relation that defines which
> > disconnected ways could be walked to or across from any point on a
> current
> > way?   That would also include the road since there would be no barrier.
> >
> >
> >
> > ___
> > Tagging mailing list
> > Tagging@openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/tagging
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Marc Gemis]

Another typical case

- no explicitly marked crossings
- sidewalk parallel to road
- kerb separating sidewalk from road
- hedge, interrupted for each driveway and at the junctions, placed on
sidewalk, parallel with road.

--> need to add all driveways ?
--> need to draw virtual crossings at junctions ?

m


On Fri, Jul 14, 2017 at 3:41 PM, Mike N  wrote:
> On 7/14/2017 8:14 AM, Marc Gemis wrote:
>>>
>>> but merge sidewalk with the road where the is no space/barier between
>>> them.
>
>
>> and that's were the discussion starts. When I asked when one has to
>> draw a separate sidewalk a few weeks ago on this mailing list someone
>> answered: as soon as there is a kerb.
>
>
>   Similarly, I have been combining sidewalks with roads where there is no
> separation.   But when there is a small grass separation from the roadway,
> they are drawn separately.  For those cases, it is usually allowed to cross
> the grassy separation and the road to get to the opposite sidewalk.
>
>   Throwing out the R word here - what about a relation that defines which
> disconnected ways could be walked to or across from any point on a current
> way?   That would also include the road since there would be no barrier.
>
>
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Mike N]

On 7/14/2017 8:14 AM, Marc Gemis wrote:

but merge sidewalk with the road where the is no space/barier between them.



and that's were the discussion starts. When I asked when one has to
draw a separate sidewalk a few weeks ago on this mailing list someone
answered: as soon as there is a kerb.


  Similarly, I have been combining sidewalks with roads where there is 
no separation.   But when there is a small grass separation from the 
roadway, they are drawn separately.  For those cases, it is usually 
allowed to cross the grassy separation and the road to get to the 
opposite sidewalk.


  Throwing out the R word here - what about a relation that defines 
which disconnected ways could be walked to or across from any point on a 
current way?   That would also include the road since there would be no 
barrier.



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Marc Gemis]

> but merge sidewalk with the road where the is no space/barier between them.

and that's were the discussion starts. When I asked when one has to
draw a separate sidewalk a few weeks ago on this mailing list someone
answered: as soon as there is a kerb.

m.

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

>>> A street with a sidewalk on either side but no marked crossings:
>> These are (IMHO) mapping errors. You can't draw isolated footway islands and
>> expect a router to magically understand those are sidewalks which you can
>> cross without a connection.
> It is not uncommon to have such a network of sidewalks without
> "zebra"-crossings.
> People are allowed to cross everywhere then.
if you can cross everywhere, sidewalks are not separated but are part of 
the road (like a lane)
I tag : highway=residential + sidewalk=both without any separated 
sidewalk path
I think that your sidewalk must be review :
keep isolated sidewalk as there are right now.
but merge sidewalk with the road where the is no space/barier between them.

 > There isn't even always a lowered kerb for people in wheelchairs. If
 > the are lucky, they can use the connection of a driveway with the
 > street, as those usually have a lowered kerb.
For wheelchairs routing, I put kerb=raised on the highway and a 
highway=crossing crossing=unmarked kerb=lowered where wheelchairs can cross.

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Marc Gemis]

On Fri, Jul 14, 2017 at 1:08 PM, marc marc  wrote:

>> the "common sense approach" would expect.
> routing doesn't know "common sense approach" :)
> if 2 sidewalk or roads are taged as "separated without any link",
> routing can't guess that a connection exists.
>


Somewhere on the wiki there is a page that explains theoretically how
routing in such a case can be done.
Of course, I can't find it right now

regards

m.

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Svavar Kjarrval]

On fös 14.júl 2017 10:51, Martin Koppenhoefer wrote:
>
>
> 2017-07-14 12:20 GMT+02:00 Svavar Kjarrval  >:
>
>
> A street segment with no sidewalks on either side:
> http://www.openstreetmap.org/#map=19/64.12876/-21.90466
> 
>
>
>
> This is an urban example, but probably you don't have sidewalks in
> most of the country (rural areas), and it likely isn't a problem for
> routing engines.
>
Don't know how the comment about it being an urban example contributes
to this discussion, as this is a real life case needing a solution. More
than half the population of the country lives in the capital area, from
which all the examples are located. Therefore, that area has received
the most focus of the local OSM community.
>
>  
>
> A street with a sidewalk on either side but no marked crossings:
> http://www.openstreetmap.org/#map=17/64.08800/-21.89846
> 
> (Sidenote: If one tries to route from no. 73 to 42,
> GraphHopper suggests a long route while Mapzen assumes the user is
> already on the other side of the street)
>
>
>
>
> These are (IMHO) mapping errors. You can't draw isolated footway
> islands and expect a router to magically understand those are
> sidewalks which you can cross without a connection. E.g this:
> http://www.openstreetmap.org/way/102907998
> There aren't even footway subtags like footway=sidewalk, but even if
> there were I wouldn't expect working routing from this graph.
The example was provided as a mean to visualise, not an example of a
routing error (per my comment regarding it).
>
>
>  
>
> A street segment where the paved sidewalk ends prematurely (same as I
> described, except they do widen the street in that case):
> 
> http://www.openstreetmap.org/?mlat=64.11777=-21.84680#map=19/64.11777/-21.84680
> 
> 
>
>
>
> no immediate problem for routing, as they are connected
Same as above, provided for visualisation.
>
>  
>
> (Sidenote: I do wonder if it would be alright to put a sidewalk
> talk on
> the road segment at the end of that street)
>
>
>
> the properties will always refer to the whole object, so if a part of
> the road has a sidewalk, another part has not, you have to split the
> road and add different tags.
>
> I wonder how all those tags have come into OSM, and what their meaning
> is? Has this pile of cryptic, undocumented abbreviations really made
> it through the import process?
> http://www.openstreetmap.org/way/92639788
The import process was organised by the local community. The references
are mostly in Icelandic words or abbreviations thereof. I'm not saying
it was perfect, and in retrospect, I think it would've probably been
better if they had been translated to English before the import.
>
>
>
> Routers seem to
>
> have a hard time knowing when it's alright to suggest the user "jump"
> onto the sidewalk from the road or vice versa if there isn't a footway
> such as ones used for crossings. 
>
>
>
>
> you should assume that routers never "jump" from one way to the other
> without an explicit connection.
Indeed I do, and I do understand (some of) the reasoning for it. Part of
the issue is the lack of data for the router to realise that there is a
connection or make it understand that it such a "jump" would be ordinary
in certain circumstances.
>
>
> Cheers,
> Martin
>
With regards,
Svavar Kjarrval


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Marc Gemis]

On Fri, Jul 14, 2017 at 12:51 PM, Martin Koppenhoefer
 wrote:
>
>
> 2017-07-14 12:20 GMT+02:00 Svavar Kjarrval :
>>
>>
>> A street with a sidewalk on either side but no marked crossings:
>> http://www.openstreetmap.org/#map=17/64.08800/-21.89846
>> (Sidenote: If one tries to route from no. 73 to 42,
>> GraphHopper suggests a long route while Mapzen assumes the user is
>> already on the other side of the street)
>>
>
>
>
> These are (IMHO) mapping errors. You can't draw isolated footway islands and
> expect a router to magically understand those are sidewalks which you can
> cross without a connection. E.g this:
> http://www.openstreetmap.org/way/102907998
> There aren't even footway subtags like footway=sidewalk, but even if there
> were I wouldn't expect working routing from this graph.
>

But what if there are no crossings marked? Do we have to invent
crossings then ? (e.g. near each junction)
It is not uncommon to have such a network of sidewalks without
"zebra"-crossings.
People are allowed to cross everywhere then.

There isn't even always a lowered kerb for people in wheelchairs. If
the are lucky, they can use the connection of a driveway with the
street, as those usually have a lowered kerb.

m.

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 14. 07. 17 à 12:20, Svavar Kjarrval a écrit :
> A street with a sidewalk on either side but no marked crossings:
> http://www.openstreetmap.org/#map=17/64.08800/-21.89846
> (Sidenote: If one tries to route from no. 73 to 42,
> GraphHopper suggests a long route while Mapzen assumes the user is
> already on the other side of the street)
It is a fault (and in my opinion a mistake) to tag a sidewalk separated 
from the road where it is not!
there is only one point that the maper create to connect the sidewalk 
and the road https://www.openstreetmap.org/node/2673312760
of course routing can only use this point, luckily !

A sidewalk really isolated from the road (= by a barrier) does not allow 
crossing outside a crossing. This is the current situation of your example.
This is not specific to the sidewalk, the same happens with roads:
If you cut a road with 2 lanes into 2 road without any link between 
them, routing will not allow you to jump from one lane to the other.

> where the footway ends
> prematurely, the routing software doesn't know it may suggest such a
> "jump" onto the street or not,
the end of the footway must be connected to the street if you are 
able/allowed to switch to the street by foot.
If needed, cut the road : one segment with sidewalk=left/righ, second 
segment with sidewalk=no

> I haven't been able to find any tag or method to do it
a road with not-separed sidewalk should be taged as such :-)

> the "common sense approach" would expect.
routing doesn't know "common sense approach" :)
if 2 sidewalk or roads are taged as "separated without any link", 
routing can't guess that a connection exists.

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Martin Koppenhoefer]

2017-07-14 12:20 GMT+02:00 Svavar Kjarrval :

>
> A street segment with no sidewalks on either side:
> http://www.openstreetmap.org/#map=19/64.12876/-21.90466
>
>

This is an urban example, but probably you don't have sidewalks in most of
the country (rural areas), and it likely isn't a problem for routing
engines.




> A street with a sidewalk on either side but no marked crossings:
> http://www.openstreetmap.org/#map=17/64.08800/-21.89846
> (Sidenote: If one tries to route from no. 73 to 42,
> GraphHopper suggests a long route while Mapzen assumes the user is
> already on the other side of the street)
>
>


These are (IMHO) mapping errors. You can't draw isolated footway islands
and expect a router to magically understand those are sidewalks which you
can cross without a connection. E.g this:
http://www.openstreetmap.org/way/102907998
There aren't even footway subtags like footway=sidewalk, but even if there
were I wouldn't expect working routing from this graph.




> A street segment where the paved sidewalk ends prematurely (same as I
> described, except they do widen the street in that case):
> http://www.openstreetmap.org/?mlat=64.11777=-21.84680#
> map=19/64.11777/-21.84680
>


no immediate problem for routing, as they are connected



> (Sidenote: I do wonder if it would be alright to put a sidewalk talk on
> the road segment at the end of that street)
>


the properties will always refer to the whole object, so if a part of the
road has a sidewalk, another part has not, you have to split the road and
add different tags.

I wonder how all those tags have come into OSM, and what their meaning is?
Has this pile of cryptic, undocumented abbreviations really made it through
the import process?
http://www.openstreetmap.org/way/92639788



Routers seem to

> have a hard time knowing when it's alright to suggest the user "jump"
> onto the sidewalk from the road or vice versa if there isn't a footway
> such as ones used for crossings.




you should assume that routers never "jump" from one way to the other
without an explicit connection.


Cheers,
Martin
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Svavar Kjarrval]

On fim 13.júl 2017 13:49, Andy Townsend wrote:
> Perhaps a few links to photos would help?
>
> It'd make it a lot easier for other people to visualise.

Don't think I have such photos on me and I'm fairly sure some people
wouldn't want links to copyrighted photos in Google Street View. I'll do
the next-best thing and provide links to OSM locations. If people check
them out on Google Street View or via other such sources, it would be
their business. The areas are mainly picked for visualisation, not
because I've found any specific routing issues.

A street segment with no sidewalks on either side:
http://www.openstreetmap.org/#map=19/64.12876/-21.90466

A street with a sidewalk on either side but no marked crossings:
http://www.openstreetmap.org/#map=17/64.08800/-21.89846
(Sidenote: If one tries to route from no. 73 to 42,
GraphHopper suggests a long route while Mapzen assumes the user is
already on the other side of the street)

A street segment where the paved sidewalk ends prematurely (same as I
described, except they do widen the street in that case):
http://www.openstreetmap.org/?mlat=64.11777=-21.84680#map=19/64.11777/-21.84680
(Sidenote: I do wonder if it would be alright to put a sidewalk talk on
the road segment at the end of that street)

On fim 13.júl 2017 14:08, John Willis wrote:
> In places complicated enough to warrant separate footpaths, then assuming 
> they *cannot* cross the street wherever they want (and forced to go to 
> crosswalks or signals) is by far the best choice. But where this complicated 
> sidewalk tagging ends, and the minor, residential, and service roads without 
> sidewalks begin interests me greatly. Is there a “footway_link” ? Not a 
> traditional _link road, but a logical link to when sidewalks end - do they 
> need some kind of “link” to the adjacent road so Routing continues on?
That's one of the issues I've been wondering myself. Routers seem to
have a hard time knowing when it's alright to suggest the user "jump"
onto the sidewalk from the road or vice versa if there isn't a footway
such as ones used for crossings. In cases where the footway ends
prematurely, the routing software doesn't know it may suggest such a
"jump" onto the street or not, and would be likely to give up on that
segment. I do understand there are probably justified reasons for them
not to do it without positive data allowing them to, so we might need to
input some type of data (like a link) telling the routing software that
such a connection is fine in that case. Sadly, I'm not sure what method
I'd be allowed to use since I haven't been able to find any tag or
method to do it.

On fim 13.júl 2017 14:17, marc marc wrote:
> can you give an exemple ? I never see this problem.
> I just test GraphHopper and Mapzen on 2 streets without sidewalk without 
> any routing problem.
This is, of course, not a problem when finding routes between two houses
on the opposite side of the street. There are problems where the routers
discover a footway nearby but that footway leads to a much longer route
when in fact it would be much quicker to walk on the street itself, as
the "common sense approach" would expect.
Here is an example of that:
https://www.openstreetmap.org/directions?engine=graphhopper_foot=64.14793%2C-21.96048%3B64.14875%2C-21.96216#map=18/64.14809/-21.96170
(Sidenote: If one moves the points much closer to the street of
Aflagrandi, the routers will finally "get it", but these are not the
coordinates people would utilise when looking up the starting point and
the destination point.)

With regards,
Svavar Kjarrval


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[marc marc]

Le 13. 07. 17 à 15:28, Svavar Kjarrval a écrit :
> the local public transport authority started utilising OSM
good :-)

> 1. Sometimes streets don't have formal sidewalks (no markings on the
> street nor signs) but there is an "common sense expectation" that
> pedestrians are allowed to traverse on the edge of that street, and also
> cross it anywhere using caution. 
> This can cause problems for routing software
can you give an exemple ? I never see this problem.
I just test GraphHopper and Mapzen on 2 streets without sidewalk without 
any routing problem.

Regards,
Marc
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[John Willis]

> On Jul 13, 2017, at 10:28 PM, Svavar Kjarrval  wrote:
> 
> when the "common sense" approach would be to "just go
> across the street".

This is a question I have too, and I’m wondering if this is something you solve 
at the tagging or engine level. 

Afaik, this is why the roads have the sidewalk=left/right/both tag - so the 
sidewalk is considered “part of the road” and routing engines do not have to 
think about what side of the road you are on. 

Places with complicated and separate footpaths need Highway=footway, but this 
is the downside to that. 

I wonder if adding foot=yes tag to the roads without sidewalks and and foot=no 
to ways where it is dangerous would help. 


In places complicated enough to warrant separate footpaths, then assuming they 
*cannot* cross the street wherever they want (and forced to go to crosswalks or 
signals) is by far the best choice. But where this complicated sidewalk tagging 
ends, and the minor, residential, and service roads without sidewalks begin 
interests me greatly. Is there a “footway_link” ? Not a traditional _link road, 
but a logical link to when sidewalks end - do they need some kind of “link” to 
the adjacent road so Routing continues on? 

Currently, in some situations I link them to the adjacent road segment or 
across an intersection with crossing+crossing=unmarked, but I am unsure if this 
is necessary or proper. 

Javbw 



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Martin Koppenhoefer]

sent from a phone

> On 13. Jul 2017, at 15:28, Svavar Kjarrval  wrote:
> 
> 1. Sometimes streets don't have formal sidewalks (no markings on the
> street nor signs) but there is an "common sense expectation" that
> pedestrians are allowed to traverse on the edge of that street, and also
> cross it anywhere using caution. The maximum speed is considered low
> enough, but they are not technically living streets. This can cause
> problems for routing software since it generally doesn't have a basis to
> assume this behaviour is alright.


I don't know which jurisdiction you are talking about, but walking on the edge 
of the street seems perfectly legal in many places (in absence of sidewalks or 
if you are carrying big loads that would be a nuisance to other pedestrians on 
a sidewalk), with the exception of motorways and motorroads. The same for 
crossing anywhere, provided you are farther away than x meters of a signed 
crossing (where x around here is 100 meters if I recall correctly, and 0-30-100 
meters in other European countries) and it is not a road restricted to motor 
vehicles (motorway etc.) and you believe it is safe to do so. Actually the only 
countries I am aware of restricting pedestrians (jaywalking) from crossing 
anywhere (when "reasonably far" away from traffic lights and crossings) are the 
US, Canada and Singapore.

I don't know how current routers handle this case, and it is clearly preferable 
walking on decent sidewalks than on the road, but in absence of sidewalks the 
router should still be able to lead pedestrians over a road.

Cheers,
Martin 




___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Formally informal sidewalks

[Andy Townsend]

Perhaps a few links to photos would help?

It'd make it a lot easier for other people to visualise.

Best Regards,

Andy


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging