"width" on streets: Time for a recommendation

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
44 messages Options
123
Reply | Threaded
Open this post in threaded view
|

"width" on streets: Time for a recommendation

Supaplex

Hey all,

again and again there are discussions about which parts of a street (sidewalks and cycle paths, parking lanes, carriageway) should be considered when determining the width of a street. There does not seem to be a consensus and therefore information on street widths is difficult to interpret or is not even mapped. The following variants are common/are discussed:

1) Width of the actual carriageway, without parking lanes and sidewalks
2) Width between curbs / edges of the road without sidewalks, but with parked cars when they are on street
3) Width including sidewalks / roadside paths

I tend to option 2):
- The width can be clearly defined and measured
- The width of the actual carriageway can be determined by using "parking:lane" scheme correctly (or alternatively/supplementarily by specifying the width of parking lanes). "width:carriageway" (or "width:lanes", if there are marked lanes) also could be used to map this width directly.
- The width of roadside paths can optionally be specified with "sidewalk:width" etc.

Wouldn't it be time to document a recommendation in the Wiki to reduce further ambiguities? Which variant is the most recommendable? Anyway, the width of a street is a significant value to evaluate its suitability or safety for certain modes of transport or to determine the speed that can be expected there.

Thanks for your comments,
Alex


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

dieterdreist
Am Mo., 14. Sept. 2020 um 20:37 Uhr schrieb Supaplex <[hidden email]>:
again and again there are discussions about which parts of a street (sidewalks and cycle paths, parking lanes, carriageway) should be considered when determining the width of a street. There does not seem to be a consensus and therefore information on street widths is difficult to interpret or is not even mapped.


indeed, mapping the width generally requires measuring the width, and it is often not practical (unless you are willing to spend a lot of time or have very good aerial imagery at hand).

 
The following variants are common/are discussed:

1) Width of the actual carriageway, without parking lanes and sidewalks
2) Width between curbs / edges of the road without sidewalks, but with parked cars when they are on street
3) Width including sidewalks / roadside paths

I tend to option 2):
- The width can be clearly defined and measured
- The width of the actual carriageway can be determined by using "parking:lane" scheme correctly (or alternatively/supplementarily by specifying the width of parking lanes). "width:carriageway" (or "width:lanes", if there are marked lanes) also could be used to map this width directly.
- The width of roadside paths can optionally be specified with "sidewalk:width" etc.




I agree that 2 could be a reasonable definition for urban areas, what I can see could be brought up against it:
the tags should generally apply to the mapped object. As we see a highway=* to include the sidewalks, it would be somehow odd to not include them in width. But I agree, through definition, we could define it to mean only the road (including parking alongside), and we are already pursuing a similar approach with regard to lanes (only car lanes, no bike lanes or sidewalks counted).

when there aren't kerb stones, how would you suggest to proceed? (my suggestion: measure from the middle of the wide lateral boundary lines if there are, otherwise measure the paved width, on unpaved roads, measure the extent of the maximum width that vehicles actually use, on a medium to narrow part of the highway (i.e. do not add the smallest width to a long stretch of highway if it only occurs for a short part, rather split the highway in this case of tag the narrow exceptions explicitly while using a medium value for longer stretches).

I would definitely not include widths of separated ways (e.g. cycleway=track) in the highway width on the main way. For these, properties like cycleway:width or footway:width could be added (or map the separately, avoiding too many splits of the main highway)

Wouldn't it be time to document a recommendation in the Wiki to reduce further ambiguities?



yes

Cheers
Martin

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Tobias Zwick
In reply to this post by Supaplex

Absolutely high time! Thank you for bringing this up.

I was under the impression that the wiki already defined it like 2). 1) is not practical because parking lanes can be informal or can change quickly, 3) is also not practical because sidewalks + additional greenery/space between road and sidewalk can vary a lot.

I plan to soon implement a "What is the width of this road" quest in StreetComplete where the user can measure the width of the road using his or her smartphone (similar to the app Measure from Google [1]). The app will need to instruct the user very clearly what should be measured.

The instruction "curb to curb" is pretty clear. However, there is one more problem to solve, what about if there are no curbs? For example, track-like roads that just consist of either one strip of asphalt surface or are not paved at all? I see two possible definitions:

  1. Width of the paved surface (if paved)
  2. Usable width of the road

1 has the advantage that there is no room for interpretation, but falls short of what to do with unpaved roads. 2 leaves some room for interpretation but also covers cases where the usable width of the road is much different from the width of the paved part of the road.

Tobias

[1] https://play.google.com/store/apps/details?id=com.google.tango.measure

On 14.09.20 20:34, Supaplex wrote:

Hey all,

again and again there are discussions about which parts of a street (sidewalks and cycle paths, parking lanes, carriageway) should be considered when determining the width of a street. There does not seem to be a consensus and therefore information on street widths is difficult to interpret or is not even mapped. The following variants are common/are discussed:

1) Width of the actual carriageway, without parking lanes and sidewalks
2) Width between curbs / edges of the road without sidewalks, but with parked cars when they are on street
3) Width including sidewalks / roadside paths

I tend to option 2):
- The width can be clearly defined and measured
- The width of the actual carriageway can be determined by using "parking:lane" scheme correctly (or alternatively/supplementarily by specifying the width of parking lanes). "width:carriageway" (or "width:lanes", if there are marked lanes) also could be used to map this width directly.
- The width of roadside paths can optionally be specified with "sidewalk:width" etc.

Wouldn't it be time to document a recommendation in the Wiki to reduce further ambiguities? Which variant is the most recommendable? Anyway, the width of a street is a significant value to evaluate its suitability or safety for certain modes of transport or to determine the speed that can be expected there.

Thanks for your comments,
Alex


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

voschix
On Tue, 15 Sep 2020 at 10:34, Tobias Zwick <[hidden email]> wrote:
I plan to soon implement a "What is the width of this road" quest in StreetComplete where the user can measure the width of the road using his or her smartphone (similar to the app Measure from Google [1]). The app will need to instruct the user very clearly what should be measured.

With the satellite photos that OSM can legally use, we usually have a resolution problem. In most cases it will be tricky to get to a precision better than one metre, which in my view is not enough.
Alternative: use est_width=  or width= plus source.width=estimated
Another problem are the contrast enhancement filters that most satellite images are subject to. They oftenwiden the roads to some extent or create an artificial limiting line, depending on the filtering algorithm.
In case there are Mapillary Mapillary photos you can, in many countries, count the stripes of zebra crossings (in Italy they are 0.5m wide) to measure the width of urban roads.
 

Virus-free. www.avast.com

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

dieterdreist
In reply to this post by Tobias Zwick
Am Di., 15. Sept. 2020 um 10:34 Uhr schrieb Tobias Zwick <[hidden email]>:
I was under the impression that the wiki already defined it like 2).


If it were like this it would be fortunate, because we already have nearly 1,9 million highways tagged with "width", and if we could reasonably expect that these are following the definition, we would maybe not have to resort to yet another key like width:carriageway

Looking at width tag variants:

here are those with more than 1K uses (ignoring yh:WIDTH which is leading even before the main tag, but clearly not a recommendable format for tags, as it isn't self explaining and has uppercase letters, and ignoring those that do not seem to refer to roads):

cycling_width
cycleway:est_width
shoulder:width
width:lanes
width:street
sidewalk:width
sidewalk:right:width
width:lanes:forward and backward
cycleway:width
crossing:width
cycleway:left:width
sidewalk:left:width
width:shoulder
sidewalk:both:width
width:average

Cheers
Martin



_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Tagging mailing list
In reply to this post by Supaplex



Sep 14, 2020, 20:34 by [hidden email]:

Hey all,

again and again there are discussions about which parts of a street (sidewalks and cycle paths, parking lanes, carriageway) should be considered when determining the width of a street. There does not seem to be a consensus and therefore information on street widths is difficult to interpret or is not even mapped. The following variants are common/are discussed:

1) Width of the actual carriageway, without parking lanes and sidewalks
2) Width between curbs / edges of the road without sidewalks, but with parked cars when they are on street
3) Width including sidewalks / roadside paths

I tend to option 2):
- The width can be clearly defined and measured
- The width of the actual carriageway can be determined by using "parking:lane" scheme correctly (or alternatively/supplementarily by specifying the width of parking lanes). "width:carriageway" (or "width:lanes", if there are marked lanes) also could be used to map this width directly.
- The width of roadside paths can optionally be specified with "sidewalk:width" etc.

Wouldn't it be time to document a recommendation in the Wiki to reduce further ambiguities? Which variant is the most recommendable? Anyway, the width of a street is a significant value to evaluate its suitability or safety for certain modes of transport or to determine the speed that can be expected there.

Thanks for your comments,
Alex

I would also expect (2)

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Jan Michel
In reply to this post by Supaplex
I expect the "width" of a way to be the actual width of the object it
represents.
This obviously changes depending on the mapping style applied, e.g.

- if it's a highway with sidewalk and cycleway tags, it's the width of
all of it

- if it's just a road with footways mapped as separate ways next to it,
then it's the width of just the road.

If you want to tag how much space there is for some kind of vehicle
moving in some direction, there are the specific width tags like
width:lanes, sidewalk:width, cycleway:width, shoulder:width, verge:width
and so on.

Jan


On 14.09.20 20:34, Supaplex wrote:
> 1) Width of the actual carriageway, without parking lanes and sidewalks
> 2) Width between curbs / edges of the road without sidewalks, but with
> parked cars when they are on street
> 3) Width including sidewalks / roadside paths


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Jan Michel
In reply to this post by dieterdreist
On 15.09.20 10:52, Martin Koppenhoefer wrote:
> Looking at width tag variants:
> https://taginfo.openstreetmap.org/search?q=width
>
> here are those with more than 1K uses

Please be careful with such a list. E.g. width:shoulder stems
from a user in Finland and from something that looks like an old
(un-)organized edit in Nepal that used several other uncommon tags as
well. The documented tag is 'shoulder:width'


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Mark Wagner
In reply to this post by dieterdreist
On Tue, 15 Sep 2020 10:09:17 +0200
Martin Koppenhoefer <[hidden email]> wrote:

> on unpaved
> roads, measure the extent of the maximum width that vehicles actually
> use, on a medium to narrow part of the highway (i.e. do not add the
> smallest width to a long stretch of highway if it only occurs for a
> short part, rather split the highway in this case of tag the narrow
> exceptions explicitly while using a medium value for longer
> stretches).

In my experience, unpaved roads don't have a well-defined width.
Typically, you've got the following options, roughly from widest to
narrowest:

1) Obstacle-free width: the distance that's clear of fences, trees,
ditches, brush, boulders, and other obstacles.  Not well-defined for
roads in farm country, which may be obstacle-free all the way
to the next road, and misleading in ranch country, where the nearest
obstacle is usually the fence marking the edge of the right-of-way.

2) Vegetation-free width: the distance clear of any vegetation.
Usually the easiest to spot and measure, but may include things such as
spoil from road maintenance which are unsuitable for driving on.

3) Maintained width: the distance that's kept smooth, level, and firm
through regular maintenance.  This is probably the closest to the
"curb-to-curb" or "edge-to-edge" width of a paved road.  This has the
problem that it can change from year to year as graders take different
routes eg. around curves, and is difficult to tell apart from the
vegetation-free width. It also has the problem that many roads are not
maintained except for removal of fallen trees and other obstacles.

4) Driven path: the portion of the road regularly used by drivers.
Unpaved roads frequently develop a well-defined set of ruts that are
easy to measure.  However, this width varies rapidly with road
conditions.

Which one is "the" width of the road?
--
Mark

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

dieterdreist


sent from a phone

> On 15. Sep 2020, at 21:39, Mark Wagner <[hidden email]> wrote:
>
> Which one is "the" width of the road?


not only from year to year but also with the seasons...

Cheers Martin

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Supaplex
In reply to this post by Jan Michel
I expect the "width" of a way to be the actual width of the object it represents.
It depends on how we define "highway" in the OSM sense. You could also assume that sidewalks etc. are "sticking" on the highway merely for pragmatic reasons. Depending on the point of view, sidewalks and highways represent different entities. (There is no law definition here, I only find a German court decision that deals with street widths and thus means the distance between the curbs, with carriageway and parked vehicles, so as definition 2 above.)

But I agree that it would be better to always specify which width is meant exactly when mapping widths on streets (especially to use "width:carriageway" for the rating of traffic suitability). Nevertheless, a default, which meaning of "width" is meant without a prefix/suffix, would still be helpful. Fun Fact: On the wiki highway page - in contrast to what is discussed here - it says since 2012 that "width" means the width of the carriageway (but it does not look like this paragraph has ever been discussed): https://wiki.openstreetmap.org/wiki/Highways#Surface.2C_width_and_lighting

Alex

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

dieterdreist
In reply to this post by Jan Michel


sent from a phone

> On 15. Sep 2020, at 19:05, Jan Michel <[hidden email]> wrote:
>
> If you want to tag how much space there is for some kind of vehicle moving in some direction, there are the specific width tags like width:lanes, sidewalk:width, cycleway:width, shoulder:width, verge:width
> and so on.


following your initial statement (all parts), you would include the verges in the width?


Cheers Martin
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Jeroen Hoek
In reply to this post by dieterdreist
On 15-09-2020 10:09, Martin Koppenhoefer wrote:
> Am Mo., 14. Sept. 2020 um 20:37 Uhr schrieb Supaplex
> <[hidden email] <mailto:[hidden email]>>:
> indeed, mapping the width generally requires measuring the width, and it
> is often not practical (unless you are willing to spend a lot of time or
> have very good aerial imagery at hand).

It will vary a lot per country and the available resources. In the
Netherlands we are blessed to have both yearly 25cm satellite imagery,
and municipal topography outline layers available. Those two, combined
are really quite nice to have.

In JOSM it looks like this:

http://jeroenhoek.nl/temp/josm-dutch-bgt.png

Measuring the width of a street (option 2) is often trivial with these;
in this case it seems to be a neat 6m.

I wonder if it is feasible to have JOSM render the width, optionally, as
a sort of semi-transparent background beneath the way-line. It would
make aligning these to the middle of the street even easier, and tagging
the width less error prone too due to the visual feedback.

Jeroen Hoek

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Jan Michel
In reply to this post by dieterdreist
On 16.09.20 10:30, Martin Koppenhoefer wrote:
>> On 15. Sep 2020, at 19:05, Jan Michel<[hidden email]>  wrote:
>>
>> If you want to tag how much space there is for some kind of vehicle moving in some direction, there are the specific width tags like width:lanes, sidewalk:width, cycleway:width, shoulder:width, verge:width
>> and so on.
>
> following your initial statement (all parts), you would include the verges in the width?

That's a difficult point. I'm not a friend of the 'verge' tag at all (as
opposed to the 'shoulder' which is an area traffic can make use of). It
was somehow invented and put to use in 2016, but as far as I know never
discussed. I definitely prefer to tag verges as separate objects - even
one of the examples in the Wiki shows a ~5m wide park like area with
trees to be tagged as verge. I think it's really far-fetched to call
this a part of the highway.

To answer your question - no, because I don't see verges as part of the
road.



_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Grzegorz Szymaszek
In reply to this post by Jeroen Hoek
On Wed, Sep 16, 2020 at 07:02:26PM +0200, Jeroen Hoek <[hidden email]> wrote:
> I wonder if it is feasible to have JOSM render the width, optionally, as
> a sort of semi-transparent background beneath the way-line.

JOSM can already do it:
https://josm.openstreetmap.de/wiki/Styles/Lane_and_Road_Attributes

--
Grzegorz

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

signature.asc (849 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Taskar Center
In reply to this post by Supaplex
Hi,

This is yet another example why "sticking" the sidewalks onto the highway (as a tag) rather than mapping them as separate ways is appearing to be less and less practical. Please see our sidewalk schema proposal from several years ago.

I think @Mark brings up really relevant width distinctions, and I believe that once we agree that sidewalks require their own geometry, we should have a similar discussion about the interpretation of width in the sidewalks context. 

I look at this issue from the perspective of routing. Routers are interested in functional width (which would be Mark's 'driven path' option). Even with the consideration of transiency of both of the last two of Mark's definitions, 'maintained' and 'driven path' width, this is a much better approximation for additional considerations than routing- it can be an indicator of traffic stress, it can provide information for the 'slow streets' movement, it can also provide a means of reconciling improper imports that labeled all roads as 'primary' when they should not. 

My last comment has to do with the separation of sidewalks from streets- in that in many locales the responsibility of street maintenance falls on a different entity than sidewalk maintenance (for example, in Seattle, the sidewalk is the responsibility of the homeowner, rather than the municipality who IS responsible for the street infrastructure). So it is actually advantageous to have these mapped as separate entities so we can keep track of infrastructure maintenance.

Best regards,

Anat



Sent from my mobile. Please excuse brevity and typos.
On Wed, Sep 16, 2020 at 1:23 AM Supaplex <[hidden email]> wrote:
I expect the "width" of a way to be the actual width of the object it represents.
It depends on how we define "highway" in the OSM sense. You could also assume that sidewalks etc. are "sticking" on the highway merely for pragmatic reasons. Depending on the point of view, sidewalks and highways represent different entities. (There is no law definition here, I only find a German court decision that deals with street widths and thus means the distance between the curbs, with carriageway and parked vehicles, so as definition 2 above.)

But I agree that it would be better to always specify which width is meant exactly when mapping widths on streets (especially to use "width:carriageway" for the rating of traffic suitability). Nevertheless, a default, which meaning of "width" is meant without a prefix/suffix, would still be helpful. Fun Fact: On the wiki highway page - in contrast to what is discussed here - it says since 2012 that "width" means the width of the carriageway (but it does not look like this paragraph has ever been discussed): https://wiki.openstreetmap.org/wiki/Highways#Surface.2C_width_and_lighting

Alex
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Alan Mackie


On Thu, 17 Sep 2020, 01:37 Taskar Center, <[hidden email]> wrote:
Hi,

This is yet another example why "sticking" the sidewalks onto the highway (as a tag) rather than mapping them as separate ways is appearing to be less and less practical. Please see our sidewalk schema proposal from several years ago.
This is all well and good for roads without tree cover in areas where the imagery is good. At other times a tag on the road is the best option if you don't want to just make up geometry.

I think @Mark brings up really relevant width distinctions, and I believe that once we agree that sidewalks require their own geometry, we should have a similar discussion about the interpretation of width in the sidewalks context. 

I look at this issue from the perspective of routing. Routers are interested in functional width (which would be Mark's 'driven path' option). Even with the consideration of transiency of both of the last two of Mark's definitions, 'maintained' and 'driven path' width, this is a much better approximation for additional considerations than routing- it can be an indicator of traffic stress, it can provide information for the 'slow streets' movement, it can also provide a means of reconciling improper imports that labeled all roads as 'primary' when they should not. 

My last comment has to do with the separation of sidewalks from streets- in that in many locales the responsibility of street maintenance falls on a different entity than sidewalk maintenance (for example, in Seattle, the sidewalk is the responsibility of the homeowner, rather than the municipality who IS responsible for the street infrastructure). So it is actually advantageous to have these mapped as separate entities so we can keep track of infrastructure maintenance.

Best regards,

Anat



Sent from my mobile. Please excuse brevity and typos.
On Wed, Sep 16, 2020 at 1:23 AM Supaplex <[hidden email]> wrote:
I expect the "width" of a way to be the actual width of the object it represents.
It depends on how we define "highway" in the OSM sense. You could also assume that sidewalks etc. are "sticking" on the highway merely for pragmatic reasons. Depending on the point of view, sidewalks and highways represent different entities. (There is no law definition here, I only find a German court decision that deals with street widths and thus means the distance between the curbs, with carriageway and parked vehicles, so as definition 2 above.)

But I agree that it would be better to always specify which width is meant exactly when mapping widths on streets (especially to use "width:carriageway" for the rating of traffic suitability). Nevertheless, a default, which meaning of "width" is meant without a prefix/suffix, would still be helpful. Fun Fact: On the wiki highway page - in contrast to what is discussed here - it says since 2012 that "width" means the width of the carriageway (but it does not look like this paragraph has ever been discussed): https://wiki.openstreetmap.org/wiki/Highways#Surface.2C_width_and_lighting

Alex
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

dieterdreist
In reply to this post by Taskar Center
Am Do., 17. Sept. 2020 um 02:37 Uhr schrieb Taskar Center <[hidden email]>:
This is yet another example why "sticking" the sidewalks onto the highway (as a tag) rather than mapping them as separate ways is appearing to be less and less practical.


why should these be mutually exclusive alternatives ("rather than")? The sidewalk tags on the road are a property of the road that tell if there is a sidewalk, and on which side. This does not prevent you from mapping the sidewalk explicitly as highway=footway and footway=sidewalk. 

Cheers
Martin

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: "width" on streets: Time for a recommendation

Tordanik
In reply to this post by Taskar Center
On 17.09.20 02:35, Taskar Center wrote:
> This is yet another example why "sticking" the sidewalks onto the
> highway (as a tag) rather than mapping them as separate ways is
> appearing to be less and less practical. Please see our sidewalk schema
> proposal
> <https://wiki.openstreetmap.org/wiki/Proposed_features/sidewalk_schema>
> from several years ago.

Your sidewalk proposal unfortunately doesn't really address the crucial
shortcoming of separately mapped sidewalks: The lack of a reliable
mechanism for figuring out which section of road a given sidewalk way
belongs to.

I agree that we should be able to give sidewalks their own geometry, but
we _also_ need the relationship between sidewalk and road. So far, all
the proposals attempting to support the former end up sacrificing the
latter.

There have been some promising discussions recently around the
sidepath_of idea, but that's still just brainstorming. Until a practical
solution is found and actually used in the database, sidewalk mapping
will remain a choice between two options that are broken in different ways.

As for the main issue of the thread: I would welcome a clear definition
for the meaning of width. In my own mapping and when writing the
relevant code in OSM2World, I have counted sidewalks etc. as part of the
road's width if they are mapped as tags on the main way. But I would of
course change that if there finally was a documented and widely
agreed-upon recommendation. I don't care so much which one it is - but
we need one.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Linking Sidewalks to Highways Was: Re: "width" on streets: Time for a recommendation

Alan Mackie


On Fri, 18 Sep 2020 at 21:35, Tobias Knerr <[hidden email]> wrote:
On 17.09.20 02:35, Taskar Center wrote:
> This is yet another example why "sticking" the sidewalks onto the
> highway (as a tag) rather than mapping them as separate ways is
> appearing to be less and less practical. Please see our sidewalk schema
> proposal
> <https://wiki.openstreetmap.org/wiki/Proposed_features/sidewalk_schema>
> from several years ago.

Your sidewalk proposal unfortunately doesn't really address the crucial
shortcoming of separately mapped sidewalks: The lack of a reliable
mechanism for figuring out which section of road a given sidewalk way
belongs to.

I agree that we should be able to give sidewalks their own geometry, but
we _also_ need the relationship between sidewalk and road. So far, all
the proposals attempting to support the former end up sacrificing the
latter.
Was this meant to be one of the purposes of associated street relations?

There have been some promising discussions recently around the
sidepath_of idea, but that's still just brainstorming. Until a practical
solution is found and actually used in the database, sidewalk mapping
will remain a choice between two options that are broken in different ways.
I hadn't heard this one. Do you have a link to the discussion? I would personally prefer sidewalk_of or walkway_of if we were to go this route though. Sidepath sounds like something that's branching to me.

Both associated street and sidepath_of still have the issue of when you're allowed to jump from one to the other, kerbs can be stepped over by most, railing less so (they're often to keep pedestrians out of blindspots). It must be difficult to tell if a sidewalk is separated specifically because the transition from one to the other is more thoroughly blocked and not simply as an added level of detail  with no more than the normal impediment to foot traffic.  The only thing I've seen discussed that might work for this was in a talk about way and street areas.

As for the main issue of the thread: I would welcome a clear definition
for the meaning of width. In my own mapping and when writing the
relevant code in OSM2World, I have counted sidewalks etc. as part of the
road's width if they are mapped as tags on the main way. But I would of
course change that if there finally was a documented and widely
agreed-upon recommendation. I don't care so much which one it is - but
we need one.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
123