Re: [OSM-dev] Modeling OSM sidewalk data

[Andy Townsend]

On 12/07/2016 08:42, Paul Norman wrote:
Does anyone have experience consuming OSM sidewalk data and have 
thoughts about what would be better to work with? 


Rather than having sidewalk data available in a column, I instead with a 
different highway value for "this road has a usable sidewalk".  We 
already have a plethora of highway values, so the extra work to support 
isn't high, and has the advantage that you don't need to mess about with 
the schema (though you do need to reload the database) if anything changes:


https://github.com/SomeoneElseOSM/SomeoneElse-style/blob/master/style.lua#L258

My initial goal was to be able to render sidewalks on unclassified and 
tertiary roads.  Note that there are more values in that list than 
you're probably expecting - that was what was found to be in the data 
when I looked locally.  Here's a diary entry, with a picture of what 
things look like


http://www.openstreetmap.org/user/SomeoneElse/diary/38136

(though I've tidied up the rendering a bit since then).

To address Komяpa's point, it's absolutely true that mapping sidewalks 
as separate ways is something that you need to consider in some cases 
(though I'd dispute that it is "only applicable to cities with a blocky 
structure" - sidewalk=both et al mostly work well in the UK).  In some 
places (e.g. in my experience Helsinki) sidewalks meander off on their 
own and then rejoin roads, and separate mapping makes sense.  Also, if 
you're mapping dropped kerbs etc. for wheelchair use then it's a lot 
easier to map sidewalks separately.


Unfortunately, the biggest problem with "sidewalks mapped as separate 
ways" is that they sometimes tend not to be done very well - often the 
mappers produce something that looks nice on a printed map but doesn't 
represent the topology (by no means always, but often enough to be an 
issue).  For example (not far from Komяpa's link) I very much doubt that 
this shows a correct foot route between the highlighted places:


https://www.openstreetmap.org/directions?engine=graphhopper_foot=53.94458%2C27.72384%3B53.94459%2C27.72419#map=19/53.94428/27.72472

The road that these are the sidewalks of 
https://www.openstreetmap.org/way/281594594 has no sidewalk information 
(though in this case, being a living_street, you should be safe walking 
along the middle of the road).


From the readme of https://github.com/ClearTables/ClearTables it sounds 
like you're trying to process OSM data as it actually exists, not "as it 
ought to be tagged", which means you might have to do something "clever" 
to detect poorly mapped examples like the one above, perhaps based on 
proximity.  I'd be interested to see what you come up with...


Best Regards,

Andy



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

Re: [OSM-dev] Modeling OSM sidewalk data

[Martin Koppenhoefer]

sent from a phone

> Il giorno 12 lug 2016, alle ore 11:15, Komяpa  ha scritto:
> 
> Both ways are only applicable to cities with blocky structure.
> In ex-USSR, architecture of cities is different, and footways / sidewalks 
> graph is more complicated than a replica of roads for cars, so all the 
> sidewalks have to be drawn separately.


There are sidewalks (alongside roads) and other footpaths, just like everywhere 
else (but in different proportions). You can decide to draw the sidewalks 
separately or not (like anywhere else), while you have in any case to draw the 
independent footways with their own geometry.

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

Re: [OSM-dev] Modeling OSM sidewalk data

[Komяpa]

Hi Paul,

Both ways are only applicable to cities with blocky structure.
In ex-USSR, architecture of cities is different, and footways / sidewalks
graph is more complicated than a replica of roads for cars, so all the
sidewalks have to be drawn separately.

See, for instance, https://www.openstreetmap.org/#map=16/53.9446/27.7170

вт, 12 июл. 2016 г. в 10:45, Paul Norman :

> I'm working on putting OSM sidewalk data into PostgreSQL with
> ClearTables (https://github.com/ClearTables/ClearTables/issues/39) and
> looking at two different ways, and wondering if anyone has experience
> modeling it.
>
> There are two obvious options
>
> 1. Have a sidewalk column with an enum type with both/left/right/neither
>
> 2. Have sidewalk_left and sidewalk_right boolean columns
>
> I can see advantages and disadvantages to both.
>
> Does anyone have experience consuming OSM sidewalk data and have
> thoughts about what would be better to work with?
>
> ___
> dev mailing list
> dev@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/dev
>
___
dev mailing list
dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/dev

[OSM-dev] Modeling OSM sidewalk data

[Paul Norman]

I'm working on putting OSM sidewalk data into PostgreSQL with 
ClearTables (https://github.com/ClearTables/ClearTables/issues/39) and 
looking at two different ways, and wondering if anyone has experience 
modeling it.


There are two obvious options

1. Have a sidewalk column with an enum type with both/left/right/neither

2. Have sidewalk_left and sidewalk_right boolean columns

I can see advantages and disadvantages to both.

Does anyone have experience consuming OSM sidewalk data and have 
thoughts about what would be better to work with?


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