Re: [talk-au] Path versus Footway

[Warin]

On 4/2/22 10:05, Graeme Fitzpatrick wrote:





On Thu, 3 Feb 2022 at 19:32,  wrote:

I assume these National parks where different rules are in effect
have a boundary relation.

In which case it would be possible to either:

a) tag a def: directly on that boundary relation with the rules
that apply
or (maybe better in this case)
b) create a type=defaults relation “Tasmania National Parks
Defaults” with all the defaults that apply in national parks, then
add that relation to any national park boundary relation where it
applies as member with the role of defaults

(b) is basically following the defaults proposal exactly, and
allows to define the defaults once and the re-use them for all
national parks.


A problem with that would be that in a number of cases that I know of, 
you can ride a bike along the roads /into/ the National Park, but you 
can't then take your bike onto the walking tracks, so a Park-wide 
default may not work?



Around me .. it varies. While there are 'general rules' (that I'd term 
'a guide') the on the ground situation is confusing to say the least!


The tracks - wide enough for a 4WD seem to be open for MTBs.. but paths 
- not wide enough for 4WDs are closed to MTBs... that is not the 
situation in all National Parks, and in some National Parks only some 
tracks are open while others are closed.



Personally I'll continue to tag them individually ...

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

Well, the advantage of that approach, ***if it were supported by data 
consumers***, is that you could just classify your ways using some tag, and 
then have whatever consequences are of legislation apply to them. If the 
legislation changes, you just make the change to these default definitions, and 
they apply everywhere.

 

Whereas if you explicit tag the access (and whatever else) on each way, then:
a) you can’t tell if a certain access value on the way comes from general 
legislation, or from an explicit sign

b) if the legislation changes, you have to find and change all the ways that 
are affected, and at that time figure out a) because you can’t just blindly 
change all tags if there might be explicit signs in place.

 

 

From: Graeme Fitzpatrick  
Sent: Friday, 4 February 2022 14:12
To: osm.talk...@thorsten.engler.id.au
Cc: OSM-Au 
Subject: Re: [talk-au] Path versus Footway

 

Thanks, both!

 

Yep, get's very messy very quickly :-(

 

Thanks

 

Graeme

 

 

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Graeme Fitzpatrick]

Thanks, both!

Yep, get's very messy very quickly :-(

Thanks

Graeme
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

def: can have conditions:

 

“def:highway=footway[walking_track=yes];access:bicycle”=no

 

(I’m not proposing this particular tagging scheme, this is just an example)

 

So if there is any tag on your walking track footways, or paths, or whatever 
that can be used to distinguish them, you could use that as a condition for the 
default value.

 

The issue with this is that we are now quickly moving into territory where it 
would be essential that these def tags are read and interpreted by data 
consumes for the data to make any sense… so that would be a problem.

 

It’s a difficult balancing act between “don’t tag local law” and “this is 
exactly the explicitly specified access for this way, even if it’s derived from 
local law instead of any physical signs or barriers”.

 

 

From: Graeme Fitzpatrick  
Sent: Friday, 4 February 2022 09:06
To: osm.talk...@thorsten.engler.id.au
Cc: OSM-Au 
Subject: Re: [talk-au] Path versus Footway

 




 

 

 

On Thu, 3 Feb 2022 at 19:32, mailto:osm.talk...@thorsten.engler.id.au> > wrote:

I assume these National parks where different rules are in effect have a 
boundary relation.

 

In which case it would be possible to either:

 

a) tag a def: directly on that boundary relation with the rules that apply
or (maybe better in this case)
b) create a type=defaults relation “Tasmania National Parks Defaults” with all 
the defaults that apply in national parks, then add that relation to any 
national park boundary relation where it applies as member with the role of 
defaults

(b) is basically following the defaults proposal exactly, and allows to define 
the defaults once and the re-use them for all national parks.

 

A problem with that would be that in a number of cases that I know of, you can 
ride a bike along the roads into the National Park, but you can't then take 
your bike onto the walking tracks, so a Park-wide default may not work?

 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

Hi Graeme, 

 

That’s correct – however the defaults can be set on very specific tags 
(def:highway=footway;access:bicycle=no) and if absolutely necessary even down 
to within a single park/reserve/area/track rather than a blanket ruling. It 
would get very fiddly at that degree of detail.

 

That’s why my initial enquiry was very specific about ‘tracks for exclusive use 
by walkers’ as I knew that it quickly gets murky when you start and get into 
bikes, horse, 4WD’s etc, let alone ebike and escooters (don’t even go there!!) 

 

Cheers - Phil

 

From: Graeme Fitzpatrick  
Sent: Friday, 4 February 2022 10:06 AM
To: osm.talk...@thorsten.engler.id.au
Cc: OSM-Au 
Subject: Re: [talk-au] Path versus Footway

 




 

 

 

On Thu, 3 Feb 2022 at 19:32, mailto:osm.talk...@thorsten.engler.id.au> > wrote:

I assume these National parks where different rules are in effect have a 
boundary relation.

 

In which case it would be possible to either:

 

a) tag a def: directly on that boundary relation with the rules that apply
or (maybe better in this case)
b) create a type=defaults relation “Tasmania National Parks Defaults” with all 
the defaults that apply in national parks, then add that relation to any 
national park boundary relation where it applies as member with the role of 
defaults

(b) is basically following the defaults proposal exactly, and allows to define 
the defaults once and the re-use them for all national parks.

 

A problem with that would be that in a number of cases that I know of, you can 
ride a bike along the roads into the National Park, but you can't then take 
your bike onto the walking tracks, so a Park-wide default may not work?

 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Graeme Fitzpatrick]

On Thu, 3 Feb 2022 at 19:32,  wrote:

> I assume these National parks where different rules are in effect have a
> boundary relation.
>
>
>
> In which case it would be possible to either:
>
>
>
> a) tag a def: directly on that boundary relation with the rules that apply
> or (maybe better in this case)
> b) create a type=defaults relation “Tasmania National Parks Defaults” with
> all the defaults that apply in national parks, then add that relation to
> any national park boundary relation where it applies as member with the
> role of defaults
>
> (b) is basically following the defaults proposal exactly, and allows to
> define the defaults once and the re-use them for all national parks.
>

A problem with that would be that in a number of cases that I know of, you
can ride a bike along the roads *into* the National Park, but you can't
then take your bike onto the walking tracks, so a Park-wide default may not
work?

Thanks

Graeme
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

Thanks Thorsten,

 

That would work however there has been a fairly recent regulation change that 
does allow bicycle use within signed areas, but those signed/specified/leased 
areas could also get the same defaults relation. At the moment I am not aware 
of any areas within National Parks but some in other reserve types.

 

Maydena Bike Park is one such example, however I would need to obtain the 
leased/licenced area boundary to create the required area - 
https://www.openstreetmap.org/way/478043164#map=15/-42.7664/146.6467

 

https://www.maydenabikepark.com/

 

The relevant bit of legislation if anyone is interested

 

https://www.legislation.tas.gov.au/view/html/inforce/current/sr-2019-076?query=((PrintType%3D%22act.reprint%22+AND+Amending%3C%3E%22pure%22+AND+PitValid%3D%40pointInTime(2022020300))+OR+(PrintType%3D%22act.reprint%22+AND+Amending%3D%22pure%22+AND+PitValid%3D%40pointInTime(2022020300))+OR+(PrintType%3D%22reprint%22+AND+Amending%3C%3E%22pure%22+AND+PitValid%3D%40pointInTime(2022020300))+OR+(PrintType%3D%22reprint%22+AND+Amending%3D%22pure%22+AND+PitValid%3D%40pointInTime(2022020300)))+AND+Title%3D(%22parks%22)=Document+Types%3D%22%3Cspan+class%3D%27dq-highlight%27%3EActs%3C%2Fspan%3E%2C+%3Cspan+class%3D%27dq-highlight%27%3EAmending+Acts%3C%2Fspan%3E%2C+%3Cspan+class%3D%27dq-highlight%27%3ESRs%3C%2Fspan%3E%2C+%3Cspan+class%3D%27dq-highlight%27%3EAmending+SRs%3C%2Fspan%3E%22%2C+Search+In%3D%22%3Cspan+class%3D%27dq-highlight%27%3ETitle%3C%2Fspan%3E%22%2C+All+Words%3D%22%3Cspan+class%3D%27dq-highlight%27%3ENational+Parks+and+Reserves+Management+Regulations+2019%3C%2Fspan%3E%22%2C+Point+In+Time%3D%22%3Cspan+class%3D%27dq-highlight%27%3E03%2F02%2F2022%3C%2Fspan%3E%22#GS22@EN

 

Cheers - Phil

 

 

From: osm.talk...@thorsten.engler.id.au  
Sent: Thursday, 3 February 2022 8:27 PM
To: 'OSM-Au' 
Subject: Re: [talk-au] Path versus Footway

 

I assume these National parks where different rules are in effect have a 
boundary relation.

 

In which case it would be possible to either:

 

a) tag a def: directly on that boundary relation with the rules that apply
or (maybe better in this case)
b) create a type=defaults relation “Tasmania National Parks Defaults” with all 
the defaults that apply in national parks, then add that relation to any 
national park boundary relation where it applies as member with the role of 
defaults

(b) is basically following the defaults proposal exactly, and allows to define 
the defaults once and the re-use them for all national parks.

 

Cheers,

Thorsten

 

From: Phil Wyatt < <mailto:p...@wyatt-family.com> p...@wyatt-family.com> 
Sent: Thursday, 3 February 2022 18:38
To: 'Little Maps' < <mailto:mapslit...@gmail.com> mapslit...@gmail.com>; 
'OSM-Au' < <mailto:talk-au@openstreetmap.org> talk-au@openstreetmap.org>
Subject: Re: [talk-au] Path versus Footway

 

I probably should have qualified my comment as I am dealing solely with tracks 
within National Parks (at this stage). I know there are tracks outside of 
National Parks where such bike restrictions do not apply.

 

Cheers - Phil

 

From: Little Maps < <mailto:mapslit...@gmail.com> mapslit...@gmail.com> 
Sent: Thursday, 3 February 2022 7:19 PM
To: Phil Wyatt < <mailto:p...@wyatt-family.com> p...@wyatt-family.com>; OSM-Au 
< <mailto:talk-au@openstreetmap.org> talk-au@openstreetmap.org>
Subject: Re: [talk-au] Path versus Footway

 

Hi all, thanks for a really informative discussion. I’m puzzled by the comments 
I’ve copied below. I’m uncertain when legislative defaults apply (and hence 
explicit access tagging isn’t required) and when tagging is needed. In the 
instance mentioned below, bicycle = no should not be added to urban footways in 
Vic as routers etc should work that out for themselves based on state 
legislation. (Or they could look at the entry in the state’s boundary relation, 
but it seems agreed that few data consumers do that). 

 

On bushwalking tracks in Tassie, bikes are banned on walking paths because 
they’re classed as vehicles. Again this is legislated and, as I interpreted the 
comments below, it’s suggested that data users should know this from 
legislation, and hence not need explicit access tags for bikes, unless access 
on a specific path deviates from the legislation.

 

However, bikes are allowed on footpaths (footways) in Tassie, so the same 
features (highway=footways) is, I assume, subject to 2 different legislations 
in the same state, depending on whether it’s an urban footpath or a bushwalking 
track. I’m curious how a data consumer / router would know which role a footway 
(or a path) was playing unless access restrictions were added to all? 
(Especially if it’s agree that few if any consumers use the National or state 
access guidelines, as was stated earlier). Isn’t it impossible for them to draw 
any conclusion unless tags are added? Or is the consensus that urban footpaths

Re: [talk-au] Path versus Footway

[osm.talk-au]

I assume these National parks where different rules are in effect have a 
boundary relation.

 

In which case it would be possible to either:

 

a) tag a def: directly on that boundary relation with the rules that apply
or (maybe better in this case)
b) create a type=defaults relation “Tasmania National Parks Defaults” with all 
the defaults that apply in national parks, then add that relation to any 
national park boundary relation where it applies as member with the role of 
defaults

(b) is basically following the defaults proposal exactly, and allows to define 
the defaults once and the re-use them for all national parks.

 

Cheers,

Thorsten

 

From: Phil Wyatt  
Sent: Thursday, 3 February 2022 18:38
To: 'Little Maps' ; 'OSM-Au' 
Subject: Re: [talk-au] Path versus Footway

 

I probably should have qualified my comment as I am dealing solely with tracks 
within National Parks (at this stage). I know there are tracks outside of 
National Parks where such bike restrictions do not apply.

 

Cheers - Phil

 

From: Little Maps mailto:mapslit...@gmail.com> > 
Sent: Thursday, 3 February 2022 7:19 PM
To: Phil Wyatt mailto:p...@wyatt-family.com> >; OSM-Au 
mailto:talk-au@openstreetmap.org> >
Subject: Re: [talk-au] Path versus Footway

 

Hi all, thanks for a really informative discussion. I’m puzzled by the comments 
I’ve copied below. I’m uncertain when legislative defaults apply (and hence 
explicit access tagging isn’t required) and when tagging is needed. In the 
instance mentioned below, bicycle = no should not be added to urban footways in 
Vic as routers etc should work that out for themselves based on state 
legislation. (Or they could look at the entry in the state’s boundary relation, 
but it seems agreed that few data consumers do that). 

 

On bushwalking tracks in Tassie, bikes are banned on walking paths because 
they’re classed as vehicles. Again this is legislated and, as I interpreted the 
comments below, it’s suggested that data users should know this from 
legislation, and hence not need explicit access tags for bikes, unless access 
on a specific path deviates from the legislation.

 

However, bikes are allowed on footpaths (footways) in Tassie, so the same 
features (highway=footways) is, I assume, subject to 2 different legislations 
in the same state, depending on whether it’s an urban footpath or a bushwalking 
track. I’m curious how a data consumer / router would know which role a footway 
(or a path) was playing unless access restrictions were added to all? 
(Especially if it’s agree that few if any consumers use the National or state 
access guidelines, as was stated earlier). Isn’t it impossible for them to draw 
any conclusion unless tags are added? Or is the consensus that urban footpaths 
(footways) don’t need access tags but bush walking paths (footways) do?

 

Hope this make sense, thanks again, Ian

“ > Mmm, certainly bikes are banned on walking tracks (they are classified as 
vehicles in tas and need to stick to 'roads') (from Phil)
 
Hi. This sounds a bit like the issue a couple of months ago with the User who 
wanted to tag all footpaths in Victoria with bicycle=no and the community 
consensus was that it wasn't OSM's role to document legislation, the data 
consumers could worry about what to do with cyclists and footpaths and OSM 
would concentrate on ground truth. Tony. “
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

I probably should have qualified my comment as I am dealing solely with tracks 
within National Parks (at this stage). I know there are tracks outside of 
National Parks where such bike restrictions do not apply.

 

Cheers - Phil

 

From: Little Maps  
Sent: Thursday, 3 February 2022 7:19 PM
To: Phil Wyatt ; OSM-Au 
Subject: Re: [talk-au] Path versus Footway

 

Hi all, thanks for a really informative discussion. I’m puzzled by the comments 
I’ve copied below. I’m uncertain when legislative defaults apply (and hence 
explicit access tagging isn’t required) and when tagging is needed. In the 
instance mentioned below, bicycle = no should not be added to urban footways in 
Vic as routers etc should work that out for themselves based on state 
legislation. (Or they could look at the entry in the state’s boundary relation, 
but it seems agreed that few data consumers do that). 

 

On bushwalking tracks in Tassie, bikes are banned on walking paths because 
they’re classed as vehicles. Again this is legislated and, as I interpreted the 
comments below, it’s suggested that data users should know this from 
legislation, and hence not need explicit access tags for bikes, unless access 
on a specific path deviates from the legislation.

 

However, bikes are allowed on footpaths (footways) in Tassie, so the same 
features (highway=footways) is, I assume, subject to 2 different legislations 
in the same state, depending on whether it’s an urban footpath or a bushwalking 
track. I’m curious how a data consumer / router would know which role a footway 
(or a path) was playing unless access restrictions were added to all? 
(Especially if it’s agree that few if any consumers use the National or state 
access guidelines, as was stated earlier). Isn’t it impossible for them to draw 
any conclusion unless tags are added? Or is the consensus that urban footpaths 
(footways) don’t need access tags but bush walking paths (footways) do?

 

Hope this make sense, thanks again, Ian

“ > Mmm, certainly bikes are banned on walking tracks (they are classified as 
vehicles in tas and need to stick to 'roads') (from Phil)
 
Hi. This sounds a bit like the issue a couple of months ago with the User who 
wanted to tag all footpaths in Victoria with bicycle=no and the community 
consensus was that it wasn't OSM's role to document legislation, the data 
consumers could worry about what to do with cyclists and footpaths and OSM 
would concentrate on ground truth. Tony. “
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Josh Marshall]

My 2¢ here, as both an avid runner/hiker and bike rider (in NSW). Most of my 
editing is along those lines, along with tracks through the bush when I go 
exploring. I’m particularly conscious of routing issues and fixing them if 
there’s an issue†, given I use a number of route planners that use OSM††.

The following is how I have been tagging with new ways. I don’t change existing 
ones unless there’s a good reason.

I have understood highway=footway as paths that are an alternative to a road 
that runs parallel, and are only for use by pedestrians for their safety and 
comfort. 

My stance here was most likely due to my initial exposure via iD’s "Foot Path”* 
so I associated them with the common definition. And footway is by default 
bicycles=undesignated, which suits Australia well with our differing laws on 
bicycles allowed on footpaths.^

Whereas once they diverge from being the "pedestrian lane” of roads and become 
a separate access route in their own right, I have been tagging as 
highway=path. Particularly as these are default access to all traffic except 
vehicular, and specifically bicycle=yes.

(highway=cycleway is also a little tricky in its overlap with =path. At least 
around me, it seems the major shared paths that form commuting links are tagged 
cycleway and foot=yes, so I’ve been happy to roll with that approach.)

Cheers,
Josh


Footnotes:
† A big one recently was the M1 on/off-ramps for both exits at Karuah, north of 
Newcastle… they all had bicycles=no, and the routing to get around that was so 
bad I had to fix it as 

†† Strava, Komoot, and the fabulous indie mobile app Footpath (nothing better 
for planning and quickly checking a route, imho) all had the same error with 
the Karuah ramps, so I knew the problem was with the underlying OSM tagging.

* Not quite as controversial as the highway=track drama, as far as I can tell.

^ I’m in NSW where it’s not legal for age>16 riders to go on the footpath… but 
there are some footpaths that are very short sections that add better bicycle 
connectivity, and would only ever be used by casual and commuting cyclists. For 
these I have added bicycle=yes to the footpath. Example: 
https://www.openstreetmap.org/way/1026269344 Use case: my newly high-schooling 
son riding to school.


> On 2 Feb 2022, at 11:24 pm, osm.talk...@thorsten.engler.id.au wrote:
> 
> 
> I rarely map things that aren’t urban footpaths.
>  
> So generally footway or cycleway. As I’m generally mapping in Queensland, 
> where there isn’t much if any legal distinction between general footpath and 
> a signed “shared path”, I’m using footway or cycleway depending on how cycle 
> friendly (wide enough, no low hanging branches, smooth enough surface, …) I 
> find the way, simply to get them to render differently in Carto, though the 
> legal access restrictions for routing purposes are identical.
>  
> In the rare cases where I did map paths “in the woods”, I’ve usually used 
> path (or track, depending…).
>  
> Cheers,
> Thorsten
>  
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Little Maps]

Hi all, thanks for a really informative discussion. I’m puzzled by the comments 
I’ve copied below. I’m uncertain when legislative defaults apply (and hence 
explicit access tagging isn’t required) and when tagging is needed. In the 
instance mentioned below, bicycle = no should not be added to urban footways in 
Vic as routers etc should work that out for themselves based on state 
legislation. (Or they could look at the entry in the state’s boundary relation, 
but it seems agreed that few data consumers do that). 

On bushwalking tracks in Tassie, bikes are banned on walking paths because 
they’re classed as vehicles. Again this is legislated and, as I interpreted the 
comments below, it’s suggested that data users should know this from 
legislation, and hence not need explicit access tags for bikes, unless access 
on a specific path deviates from the legislation.

However, bikes are allowed on footpaths (footways) in Tassie, so the same 
features (highway=footways) is, I assume, subject to 2 different legislations 
in the same state, depending on whether it’s an urban footpath or a bushwalking 
track. I’m curious how a data consumer / router would know which role a footway 
(or a path) was playing unless access restrictions were added to all? 
(Especially if it’s agree that few if any consumers use the National or state 
access guidelines, as was stated earlier). Isn’t it impossible for them to draw 
any conclusion unless tags are added? Or is the consensus that urban footpaths 
(footways) don’t need access tags but bush walking paths (footways) do?

Hope this make sense, thanks again, Ian
“ > Mmm, certainly bikes are banned on walking tracks (they are classified as 
vehicles in tas and need to stick to 'roads') (from Phil)

Hi. This sounds a bit like the issue a couple of months ago with the User who 
wanted to tag all footpaths in Victoria with bicycle=no and the community 
consensus was that it wasn't OSM's role to document legislation, the data 
consumers could worry about what to do with cyclists and footpaths and OSM 
would concentrate on ground truth. Tony. “___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Kim Oldfield via Talk-au]

Hi,

On 3/2/22 09:34, Phil Wyatt wrote:


…. and then work on getting the def:syntax incorporated as defaults 
into the database somehow?




That would be good, but I'm not sure how to do it.

One place to start would be to mention on 
https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions#Australia 
that these defaults have been documented in the state relations and 
include links to these relations. This at least raises the visibility of 
how we have tried to document them, and anyone writing a routing engine 
can see these defaults in machine readable form.


I am happy to edit that wiki, with community guidance, to at least 
show what is in the relations def:’s if that is acceptable. I also 
note other countries have notes in respect of some restrictions so if 
folks have links to relevant material I don’t think it hurts to add them.




Updating the WIKI with state variations would make sense.

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

Hi Thorsten,

 

Many thanks for that detailed explanation. 

 

Sounds like your request is to at least update the footway/bicycle restrictions 
in 
https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions#Australia
 to at least what is listed in the state relations, even though this has not 
been ‘endorsed’ by the community in any process (but appears to be the legal 
status of bikes on footpaths?), and may not be used by data consumers. 

 

…. and then work on getting the def:syntax incorporated as defaults into the 
database somehow?

 

I am happy to edit that wiki, with community guidance, to at least show what is 
in the relations def:’s if that is acceptable. I also note other countries have 
notes in respect of some restrictions so if folks have links to relevant 
material I don’t think it hurts to add them.

 

Cheers - Phil

 

From: osm.talk...@thorsten.engler.id.au  
Sent: Wednesday, 2 February 2022 11:13 PM
To: 'OSM-Au' 
Subject: Re: [talk-au] Path versus Footway

 

Tasmania:  <https://www.openstreetmap.org/relation/2369652> 
https://www.openstreetmap.org/relation/2369652

There seems to be only a single default key defined for Tasmania currently:

 

"def:highway=footway;access:bicycle"=yes

There are no default values defined on Australia:  
<https://www.openstreetmap.org/relation/80500> 
https://www.openstreetmap.org/relation/80500

 

Now, it’s worth pointing out that the proposal that this tagging scheme is 
based on:

 <https://wiki.openstreetmap.org/wiki/Proposed_features/Defaults> 
https://wiki.openstreetmap.org/wiki/Proposed_features/Defaults

 

a.  Never went through RFC or voting
b.  Envisions that the def tags are placed on a separate type=defaults 
relation which is then a member of role defaults of the boundary relation, 
instead of being applied directly to the boundary relation as we have done.

 

As such it is exceedingly unlikely that any type of data consumer is actually 
using them.

 

Nonetheless, that proposal represents the only attempt I’m aware of to actually 
define defaults inside the OSM database instead of simply throwing your hands 
up in the air and shout “Who knows? Whatever..”

 

So really, in reality, defaults are whatever the developer of every single data 
consumer decided.

 

Our choices come down to:

a) Just shrug and let all data consumers and mappers make up their mind on 
their own
b) At least attempt to somehow write down on the wiki what defaults mappers 
should assume, and data consumers hopefully accept
c) use (and extend use of) that somewhat unwieldy def: syntax to make our 
wishes in regards to defaults explicit in the database. It would at least allow 
us to point to it and say “see, we explicitly and in a machine readable form 
recorded our assumed defaults,” if any data consumer asks.

 

For the other states and territories, currently defined defaults are:

 

SA:  <https://www.openstreetmap.org/relation/2316596> 
https://www.openstreetmap.org/relation/2316596

 

"def:highway=footway;access:bicycle"=yes

WA:  <https://www.openstreetmap.org/relation/2316598> 
https://www.openstreetmap.org/relation/2316598

 

"def:highway=footway;access:bicycle"=yes

 

NT:  <https://www.openstreetmap.org/relation/2316594> 
https://www.openstreetmap.org/relation/2316594

 

"def:highway=footway;access:bicycle"=yes

 

Qld:  <https://www.openstreetmap.org/relation/2316595> 
https://www.openstreetmap.org/relation/2316595

 

"def:highway=footway;access:bicycle"=yes

 

NSW:  <https://www.openstreetmap.org/relation/2316593> 
https://www.openstreetmap.org/relation/2316593

 

"def:highway=footway;access:bicycle"=no

"def:highway=living_street;maxspeed"=10

"def:highway=residential;maxspeed"=50

 

Vic:  <https://www.openstreetmap.org/relation/2316741> 
https://www.openstreetmap.org/relation/2316741

 

"def:highway=footway;access:bicycle"=no

 

ACT:  <https://www.openstreetmap.org/relation/2354197> 
https://www.openstreetmap.org/relation/2354197

 

"def:highway=footway;access:bicycle"=yes

 

Jervis Bay Territory:  <https://www.openstreetmap.org/relation/2357330> 
https://www.openstreetmap.org/relation/2357330

none

 

From: Phil Wyatt < <mailto:p...@wyatt-family.com> p...@wyatt-family.com> 
Sent: Wednesday, 2 February 2022 20:46
To:  <mailto:osm.talk...@thorsten.engler.id.au> 
osm.talk...@thorsten.engler.id.au; 'OSM-Au' < 
<mailto:talk-au@openstreetmap.org> talk-au@openstreetmap.org>
Subject: RE: [talk-au] Path versus Footway

 

Hi Thorsten,

 

Is there somewhere to view those defaults for Tasmania? I assume its not 
usually editable by mappers?

 

Cheers - Phil

 

From:  <mailto:osm.talk...@thorsten.engler.id.au> 
osm.talk...@thorsten.engler.id.au < <mailto:osm.talk...@thorsten.engler.id.au> 
osm.talk

Re: [talk-au] Path versus Footway

[osm.talk-au]

As I mentioned in my previous post, it’s extremely unlikely any data consumer 
is making use of that information.

 

But, there is, as far as I’m aware, no other attempt at defining expected 
defaults in the OSM database. 

 

That, despite the fact that, as can be seen at 
https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions , 
pretty much every country has their own expectations of what the defaults are.

 

I think the best we can do is

a) Document the defaults we assume as part of the Australian Tagging Guidelines 
and also copy the relevant information into the table for Australia in the page 
linked above

b) Use the def: keys as described in the Defaults Proposal, either as that 
proposal envisioned on separate type=defaults relations, or as we are currently 
doing, directly on the boundary relations, to at least make an attempt at 
providing information about the assumed defaults in an easily machine readable 
format.

 

Without that, the current situation is that every data consumer has to figure 
something out on their own, for every country or just make general assumptions. 
And all mappers simply have to come up with their own ideas of what the 
defaults might be when they are deciding which values they are going to tag as 
“being different from defaults”.

 

Obviously, when, in lack of any authoritative source, the idea of what the 
defaults may be are different between individual mappers, and mappers and data 
consumers, the outcome is largely undefined and undefinable. Which is what the 
situation is right now.

 

 

From: Andy Townsend  
Sent: Wednesday, 2 February 2022 23:31
To: talk-au@openstreetmap.org
Subject: Re: [talk-au] Path versus Footway

 

On 02/02/2022 11:36, Kim Oldfield via Talk-au wrote:

On 2/2/22 21:45, Phil Wyatt wrote:



Is there somewhere to view those defaults for Tasmania? I assume its not 
usually editable by mappers?


See https://www.openstreetmap.org/relation/2369652

Specifically the tag: def:highway=footway;access:bicycle = yes

In OSM worldwide, that's only set for Australia:

https://taginfo.openstreetmap.org/keys/?key=def%3Ahighway%3Dfootway%3Baccess%3Abicycle#overview

Do any routers actually read that?

Best Regards,

Andy

 

 
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Andy Townsend]

On 02/02/2022 11:36, Kim Oldfield via Talk-au wrote:

On 2/2/22 21:45, Phil Wyatt wrote:


Is there somewhere to view those defaults for Tasmania? I assume its 
not usually editable by mappers?




See https://www.openstreetmap.org/relation/2369652

Specifically the tag: def:highway=footway;access:bicycle = yes


In OSM worldwide, that's only set for Australia:

https://taginfo.openstreetmap.org/keys/?key=def%3Ahighway%3Dfootway%3Baccess%3Abicycle#overview

Do any routers actually read that?

Best Regards,

Andy

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

I rarely map things that aren’t urban footpaths. 

 

So generally footway or cycleway. As I’m generally mapping in Queensland, where 
there isn’t much if any legal distinction between general footpath and a signed 
“shared path”, I’m using footway or cycleway depending on how cycle friendly 
(wide enough, no low hanging branches, smooth enough surface, …) I find the 
way, simply to get them to render differently in Carto, though the legal access 
restrictions for routing purposes are identical.

 

In the rare cases where I did map paths “in the woods”, I’ve usually used path 
(or track, depending…).

 

Cheers,

Thorsten

 

From: Phil Wyatt  
Sent: Wednesday, 2 February 2022 20:13
To: osm.talk...@thorsten.engler.id.au; 'OSM-Au' 
Subject: RE: [talk-au] Path versus Footway

 

So how do YOU decide which to use when the track is for ‘exclusively for foot 
traffic’ or do you just mix it up on a whim, change each week, go with whatever 
is similar around the object you are mapping?

 

Cheers - Phil

 

From: osm.talk...@thorsten.engler.id.au 
<mailto:osm.talk...@thorsten.engler.id.au>  mailto:osm.talk...@thorsten.engler.id.au> > 
Sent: Wednesday, 2 February 2022 8:58 PM
To: 'OSM-Au' mailto:talk-au@openstreetmap.org> >
Subject: Re: [talk-au] Path versus Footway

 

In the end, the only thing that counts is what is tagged on the objects in the 
database, and the OSM database API does not impose any restrictions about that.

 

I believe even iD allows you in the end to just freely specify any tags you 
like on any object?

 

I’m sure it’s possible to work out some tagging scheme that adequately 
describes the situation you linked to.

 

From: Graeme Fitzpatrick mailto:graemefi...@gmail.com> 
> 
Sent: Wednesday, 2 February 2022 17:29
To: osm.talk...@thorsten.engler.id.au 
<mailto:osm.talk...@thorsten.engler.id.au> 
Cc: OSM-Au mailto:talk-au@openstreetmap.org> >
Subject: Re: [talk-au] Path versus Footway

 

 

On Wed, 2 Feb 2022 at 16:54, mailto:osm.talk...@thorsten.engler.id.au> > wrote:

As far as I’m concerned, footway, cycleway, path(, and bridleway) are all 
essentially the same thing, a non-motor_vehicle path, just with different 
implied default access restrictions.

 

We should probably have a discussion about how appropriate the ones listed here 
are:

 

https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions#Australia

 

Yep!

 

How do we handle this: https://goo.gl/maps/x39C4ky1w6S7XoLUA when motorway says 
bicycle=no?

 

& similarly, you can't (at least in iD) add bike lanes to trunk roads.


 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

Tasmania: https://www.openstreetmap.org/relation/2369652

There seems to be only a single default key defined for Tasmania currently:

 

"def:highway=footway;access:bicycle"=yes

There are no default values defined on Australia: 
https://www.openstreetmap.org/relation/80500

 

Now, it’s worth pointing out that the proposal that this tagging scheme is 
based on:

https://wiki.openstreetmap.org/wiki/Proposed_features/Defaults

 

a.  Never went through RFC or voting
b.  Envisions that the def tags are placed on a separate type=defaults 
relation which is then a member of role defaults of the boundary relation, 
instead of being applied directly to the boundary relation as we have done.

 

As such it is exceedingly unlikely that any type of data consumer is actually 
using them.

 

Nonetheless, that proposal represents the only attempt I’m aware of to actually 
define defaults inside the OSM database instead of simply throwing your hands 
up in the air and shout “Who knows? Whatever..”

 

So really, in reality, defaults are whatever the developer of every single data 
consumer decided.

 

Our choices come down to:

a) Just shrug and let all data consumers and mappers make up their mind on 
their own
b) At least attempt to somehow write down on the wiki what defaults mappers 
should assume, and data consumers hopefully accept
c) use (and extend use of) that somewhat unwieldy def: syntax to make our 
wishes in regards to defaults explicit in the database. It would at least allow 
us to point to it and say “see, we explicitly and in a machine readable form 
recorded our assumed defaults,” if any data consumer asks.

 

For the other states and territories, currently defined defaults are:

 

SA:  <https://www.openstreetmap.org/relation/2316596> 
https://www.openstreetmap.org/relation/2316596

 

"def:highway=footway;access:bicycle"=yes

WA: https://www.openstreetmap.org/relation/2316598

 

"def:highway=footway;access:bicycle"=yes

 

NT: https://www.openstreetmap.org/relation/2316594

 

"def:highway=footway;access:bicycle"=yes

 

Qld: https://www.openstreetmap.org/relation/2316595

 

"def:highway=footway;access:bicycle"=yes

 

NSW:  <https://www.openstreetmap.org/relation/2316593> 
https://www.openstreetmap.org/relation/2316593

 

"def:highway=footway;access:bicycle"=no

"def:highway=living_street;maxspeed"=10

"def:highway=residential;maxspeed"=50

 

Vic: https://www.openstreetmap.org/relation/2316741

 

"def:highway=footway;access:bicycle"=no

 

ACT: https://www.openstreetmap.org/relation/2354197

 

"def:highway=footway;access:bicycle"=yes

 

Jervis Bay Territory: https://www.openstreetmap.org/relation/2357330

none

 

From: Phil Wyatt  
Sent: Wednesday, 2 February 2022 20:46
To: osm.talk...@thorsten.engler.id.au; 'OSM-Au' 
Subject: RE: [talk-au] Path versus Footway

 

Hi Thorsten,

 

Is there somewhere to view those defaults for Tasmania? I assume its not 
usually editable by mappers?

 

Cheers - Phil

 

From:  <mailto:osm.talk...@thorsten.engler.id.au> 
osm.talk...@thorsten.engler.id.au < <mailto:osm.talk...@thorsten.engler.id.au> 
osm.talk...@thorsten.engler.id.au> 
Sent: Wednesday, 2 February 2022 9:00 PM
To: 'OSM-Au' < <mailto:talk-au@openstreetmap.org> talk-au@openstreetmap.org>
Subject: Re: [talk-au] Path versus Footway

 

That table is just the suggested defaults.

 

We actually have default values specified on the state boundaries currently I 
think using the format specified here:  
<https://wiki.openstreetmap.org/wiki/Proposed_features/Defaults> 
https://wiki.openstreetmap.org/wiki/Proposed_features/Defaults I think.

 

Any use of explicit access tags will override defaults.

 

There isn’t really a fully accepted way used by all data consumers to specify 
defaults in OSM currently.

 

So at the end, it really comes down to whatever defaults any particular data 
consumer applies.

 

As long as you explicitly tag access, any type of path, foot/cycle/bridle-way 
can be made to reflect whatever you want.

 

From: Graeme Fitzpatrick < <mailto:graemefi...@gmail.com> 
graemefi...@gmail.com> 
Sent: Wednesday, 2 February 2022 17:32
To: Phil Wyatt < <mailto:p...@wyatt-family.com> p...@wyatt-family.com>
Cc:  <mailto:osm.talk...@thorsten.engler.id.au> 
osm.talk...@thorsten.engler.id.au; OSM-Au < <mailto:talk-au@openstreetmap.org> 
talk-au@openstreetmap.org>
Subject: Re: [talk-au] Path versus Footway

 




 

On Wed, 2 Feb 2022 at 17:24, Phil Wyatt mailto:p...@wyatt-family.com> > wrote:

 

So reading from that chart and in regard to my query about ‘tracks that are 
exclusively for foot traffic’ you would say it can ONLY be a footway?

 

By that list, yes?

 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Kim Oldfield via Talk-au]

On 2/2/22 21:45, Phil Wyatt wrote:


Is there somewhere to view those defaults for Tasmania? I assume its 
not usually editable by mappers?




See https://www.openstreetmap.org/relation/2369652

Specifically the tag: def:highway=footway;access:bicycle = yes

While it appears to be editable just like any other relation in OSM, 
changing the state defaults without first discussing on this list would 
be likely get your changes reverted.


To find the state relation and defaults zoom in on any location in 
Tasmania, click the Query features icon and click anywhere in Tasmania, 
eg 
https://www.openstreetmap.org/query?lat=-41.99883=146.57288#map=17/-42.8/146.57296 

then click on the "State Boundary Tasmania 
" link. The same process 
can be used to find the defaults for any other state or territory.


Regards,
Kim


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

Hi Thorsten,

 

Is there somewhere to view those defaults for Tasmania? I assume its not 
usually editable by mappers?

 

Cheers - Phil

 

From: osm.talk...@thorsten.engler.id.au  
Sent: Wednesday, 2 February 2022 9:00 PM
To: 'OSM-Au' 
Subject: Re: [talk-au] Path versus Footway

 

That table is just the suggested defaults.

 

We actually have default values specified on the state boundaries currently I 
think using the format specified here: 
https://wiki.openstreetmap.org/wiki/Proposed_features/Defaults I think.

 

Any use of explicit access tags will override defaults.

 

There isn’t really a fully accepted way used by all data consumers to specify 
defaults in OSM currently.

 

So at the end, it really comes down to whatever defaults any particular data 
consumer applies.

 

As long as you explicitly tag access, any type of path, foot/cycle/bridle-way 
can be made to reflect whatever you want.

 

From: Graeme Fitzpatrick mailto:graemefi...@gmail.com> 
> 
Sent: Wednesday, 2 February 2022 17:32
To: Phil Wyatt mailto:p...@wyatt-family.com> >
Cc: osm.talk...@thorsten.engler.id.au 
<mailto:osm.talk...@thorsten.engler.id.au> ; OSM-Au mailto:talk-au@openstreetmap.org> >
Subject: Re: [talk-au] Path versus Footway

 




 

On Wed, 2 Feb 2022 at 17:24, Phil Wyatt mailto:p...@wyatt-family.com> > wrote:

 

So reading from that chart and in regard to my query about ‘tracks that are 
exclusively for foot traffic’ you would say it can ONLY be a footway?

 

By that list, yes?

 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

So how do YOU decide which to use when the track is for ‘exclusively for foot 
traffic’ or do you just mix it up on a whim, change each week, go with whatever 
is similar around the object you are mapping?

 

Cheers - Phil

 

From: osm.talk...@thorsten.engler.id.au  
Sent: Wednesday, 2 February 2022 8:58 PM
To: 'OSM-Au' 
Subject: Re: [talk-au] Path versus Footway

 

In the end, the only thing that counts is what is tagged on the objects in the 
database, and the OSM database API does not impose any restrictions about that.

 

I believe even iD allows you in the end to just freely specify any tags you 
like on any object?

 

I’m sure it’s possible to work out some tagging scheme that adequately 
describes the situation you linked to.

 

From: Graeme Fitzpatrick mailto:graemefi...@gmail.com> 
> 
Sent: Wednesday, 2 February 2022 17:29
To: osm.talk...@thorsten.engler.id.au 
<mailto:osm.talk...@thorsten.engler.id.au> 
Cc: OSM-Au mailto:talk-au@openstreetmap.org> >
Subject: Re: [talk-au] Path versus Footway

 

 

On Wed, 2 Feb 2022 at 16:54, mailto:osm.talk...@thorsten.engler.id.au> > wrote:

As far as I’m concerned, footway, cycleway, path(, and bridleway) are all 
essentially the same thing, a non-motor_vehicle path, just with different 
implied default access restrictions.

 

We should probably have a discussion about how appropriate the ones listed here 
are:

 

https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions#Australia

 

Yep!

 

How do we handle this: https://goo.gl/maps/x39C4ky1w6S7XoLUA when motorway says 
bicycle=no?

 

& similarly, you can't (at least in iD) add bike lanes to trunk roads.


 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

That table is just the suggested defaults.

 

We actually have default values specified on the state boundaries currently I 
think using the format specified here: 
https://wiki.openstreetmap.org/wiki/Proposed_features/Defaults I think.

 

Any use of explicit access tags will override defaults.

 

There isn’t really a fully accepted way used by all data consumers to specify 
defaults in OSM currently.

 

So at the end, it really comes down to whatever defaults any particular data 
consumer applies.

 

As long as you explicitly tag access, any type of path, foot/cycle/bridle-way 
can be made to reflect whatever you want.

 

From: Graeme Fitzpatrick  
Sent: Wednesday, 2 February 2022 17:32
To: Phil Wyatt 
Cc: osm.talk...@thorsten.engler.id.au; OSM-Au 
Subject: Re: [talk-au] Path versus Footway

 




 

On Wed, 2 Feb 2022 at 17:24, Phil Wyatt mailto:p...@wyatt-family.com> > wrote:

 

So reading from that chart and in regard to my query about ‘tracks that are 
exclusively for foot traffic’ you would say it can ONLY be a footway?

 

By that list, yes?

 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

In the end, the only thing that counts is what is tagged on the objects in the 
database, and the OSM database API does not impose any restrictions about that.

 

I believe even iD allows you in the end to just freely specify any tags you 
like on any object?

 

I’m sure it’s possible to work out some tagging scheme that adequately 
describes the situation you linked to.

 

From: Graeme Fitzpatrick  
Sent: Wednesday, 2 February 2022 17:29
To: osm.talk...@thorsten.engler.id.au
Cc: OSM-Au 
Subject: Re: [talk-au] Path versus Footway

 

 

On Wed, 2 Feb 2022 at 16:54, mailto:osm.talk...@thorsten.engler.id.au> > wrote:

As far as I’m concerned, footway, cycleway, path(, and bridleway) are all 
essentially the same thing, a non-motor_vehicle path, just with different 
implied default access restrictions.

 

We should probably have a discussion about how appropriate the ones listed here 
are:

 

https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions#Australia

 

Yep!

 

How do we handle this: https://goo.gl/maps/x39C4ky1w6S7XoLUA when motorway says 
bicycle=no?

 

& similarly, you can't (at least in iD) add bike lanes to trunk roads.


 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

It can be anything you want, as long as you add enough explicit access tags.

 

From: Phil Wyatt  
Sent: Wednesday, 2 February 2022 17:20
To: osm.talk...@thorsten.engler.id.au; 'OSM-Au' 
Subject: RE: [talk-au] Path versus Footway

 

Thanks Thorsten,

 

So reading from that chart and in regard to my query about 'tracks that are
exclusively for foot traffic' you would say it can ONLY be a footway?

 

Cheers - Phil

 

From: osm.talk...@thorsten.engler.id.au
<mailto:osm.talk...@thorsten.engler.id.au>
mailto:osm.talk...@thorsten.engler.id.au> > 
Sent: Wednesday, 2 February 2022 5:51 PM
To: 'OSM-Au' mailto:talk-au@openstreetmap.org> >
Subject: Re: [talk-au] Path versus Footway

 

As far as I'm concerned, footway, cycleway, path(, and bridleway) are all
essentially the same thing, a non-motor_vehicle path, just with different
implied default access restrictions.

 

We should probably have a discussion about how appropriate the ones listed
here are:

 

https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions
#Australia

 

 

 

From: Phil Wyatt mailto:p...@wyatt-family.com> > 
Sent: Wednesday, 2 February 2022 11:00
To: OSM-Au mailto:talk-au@openstreetmap.org> >
Subject: [talk-au] Path versus Footway

 

Hi Folks,

 

I am contemplating a review of 'walking  tracks' tagging in Tasmania,
outside of urban areas. In my case I am starting with tracks that are
exclusively for foot traffic. My investigation has led me to what appears to
be a conflict within OSM of what is the correct tagging to use. 

 

https://wiki.openstreetmap.org/wiki/Tag:highway%3Dpath would suggest that
most could be a 'path' and this seems to be verified on existing data with
this styled overpass query (by bounding box)
https://overpass-turbo.eu/s/1fGX

 

*   Blue represents a path
*   Red represents a footway
*   Black represents steps

 

The path tag also considers extra tagging such as the sac_scale, visibility,
surface, operator etc etc which is useful extra information. Sac_scale and
operator are certainly used less frequently on footway.

 

The footway tagging
https://wiki.openstreetmap.org/wiki/Tag:highway%3Dfootway seems to have been
written with urban infrastructure in mind and as usual for OSM tagging does
not provide definitive detail (ie  it could have said 'used exclusively by
pedestrians', instead it say mainly or exclusively).

 

Of course there are always cases on the margins of both and an example would
be a high use, possibly with disabled access, tracks such as Russell Falls
in Tasmania (to highlight one that is likely known by many)
https://overpass-turbo.eu/s/1fGZ

 

So that brings me to the recently created Australian Walking Track page
https://wiki.openstreetmap.org/wiki/Australia/Walking_Tracks which gives the
options to use both tags (path and footway) but without any real
qualification about choosing between the two. This still seems to be in
conflict with the Australian tagging guidelines on Bushwalking (and cycling
tracks)
https://wiki.openstreetmap.org/wiki/Australian_Tagging_Guidelines#Bush_Walki
ng_and_Cycling_Tracks that definitively says 'Do not use highway=footway'.

 

So my question is - do you think we can come up with some criteria where a
footway ends and path commences or should we just go with the flow and stick
with OSM  <https://wiki.openstreetmap.org/wiki/Any_tags_you_like> 'any tags
you like'? My main goal is to make sure the two Australian wikis are not in
conflict with each other.

 

I am aware there is some controversy re footway/pathway and bikes but I
would like to ignore that in this context

 

Cheers - Phil

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

Thanks Tom - all opinions welcome and yours seems to partly equate with the
current reality in OSM (at least in Australia)

-Original Message-
From: Tom Brennan  
Sent: Wednesday, 2 February 2022 8:27 PM
To: talk-au@openstreetmap.org
Subject: Re: [talk-au] Path versus Footway

I suspect it might be hard to come up with definitive criteria, but I think
you could come close.

I agree that there do tend to be some edge cases - typically:
1. Dirt/roughly paved paths in urban areas - I prefer "path" for these, as
they might be less suited to people with mobility issues 2. Paved tracks in
national parks - I also prefer "path", as they tend to connect up to the
rest of the track system

I would be happy enough if footway just disappeared and we tagged all paths
with the relevant extra tagging (surface is probably the most important).
But that's probably going off topic...

cheers
Tom

Canyoning? try http://ozultimate.com/canyoning Bushwalking? try
http://bushwalkingnsw.com

On 2/02/2022 11:59 am, Phil Wyatt wrote:
> Hi Folks,
> 
>   
> 
> I am contemplating a review of 'walking  tracks' tagging in Tasmania, 
> outside of urban areas. In my case I am starting with tracks that are 
> exclusively for foot traffic. My investigation has led me to what 
> appears to be a conflict within OSM of what is the correct tagging to use.
> 
>   
> 
> https://wiki.openstreetmap.org/wiki/Tag:highway%3Dpath would suggest 
> that most could be a 'path' and this seems to be verified on existing 
> data with this styled overpass query (by bounding box) 
> https://overpass-turbo.eu/s/1fGX
> 
>   
> 
> * Blue represents a path
> * Red represents a footway
> * Black represents steps
> 
>   
> 
> The path tag also considers extra tagging such as the sac_scale, 
> visibility, surface, operator etc etc which is useful extra 
> information. Sac_scale and operator are certainly used less frequently on
footway.
> 
>   
> 
> The footway tagging
> https://wiki.openstreetmap.org/wiki/Tag:highway%3Dfootway seems to 
> have been written with urban infrastructure in mind and as usual for 
> OSM tagging does not provide definitive detail (ie  it could have said 
> 'used exclusively by pedestrians', instead it say mainly or exclusively).
> 
>   
> 
> Of course there are always cases on the margins of both and an example 
> would be a high use, possibly with disabled access, tracks such as 
> Russell Falls in Tasmania (to highlight one that is likely known by 
> many) https://overpass-turbo.eu/s/1fGZ
> 
>   
> 
> So that brings me to the recently created Australian Walking Track 
> page https://wiki.openstreetmap.org/wiki/Australia/Walking_Tracks 
> which gives the options to use both tags (path and footway) but 
> without any real qualification about choosing between the two. This 
> still seems to be in conflict with the Australian tagging guidelines 
> on Bushwalking (and cycling
> tracks)
> https://wiki.openstreetmap.org/wiki/Australian_Tagging_Guidelines#Bush
> _Walki ng_and_Cycling_Tracks that definitively says 'Do not use 
> highway=footway'.
> 
>   
> 
> So my question is - do you think we can come up with some criteria 
> where a footway ends and path commences or should we just go with the 
> flow and stick with OSM  
> <https://wiki.openstreetmap.org/wiki/Any_tags_you_like> 'any tags you 
> like'? My main goal is to make sure the two Australian wikis are not in
conflict with each other.
> 
>   
> 
> I am aware there is some controversy re footway/pathway and bikes but 
> I would like to ignore that in this context
> 
>   
> 
> Cheers - Phil
> 
> 
> 
> ___
> Talk-au mailing list
> Talk-au@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-au

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Tom Brennan]

I suspect it might be hard to come up with definitive criteria, but I 
think you could come close.


I agree that there do tend to be some edge cases - typically:
1. Dirt/roughly paved paths in urban areas - I prefer "path" for these, 
as they might be less suited to people with mobility issues
2. Paved tracks in national parks - I also prefer "path", as they tend 
to connect up to the rest of the track system


I would be happy enough if footway just disappeared and we tagged all 
paths with the relevant extra tagging (surface is probably the most 
important). But that's probably going off topic...


cheers
Tom

Canyoning? try http://ozultimate.com/canyoning
Bushwalking? try http://bushwalkingnsw.com

On 2/02/2022 11:59 am, Phil Wyatt wrote:

Hi Folks,

  


I am contemplating a review of 'walking  tracks' tagging in Tasmania,
outside of urban areas. In my case I am starting with tracks that are
exclusively for foot traffic. My investigation has led me to what appears to
be a conflict within OSM of what is the correct tagging to use.

  


https://wiki.openstreetmap.org/wiki/Tag:highway%3Dpath would suggest that
most could be a 'path' and this seems to be verified on existing data with
this styled overpass query (by bounding box)
https://overpass-turbo.eu/s/1fGX

  


*   Blue represents a path
*   Red represents a footway
*   Black represents steps

  


The path tag also considers extra tagging such as the sac_scale, visibility,
surface, operator etc etc which is useful extra information. Sac_scale and
operator are certainly used less frequently on footway.

  


The footway tagging
https://wiki.openstreetmap.org/wiki/Tag:highway%3Dfootway seems to have been
written with urban infrastructure in mind and as usual for OSM tagging does
not provide definitive detail (ie  it could have said 'used exclusively by
pedestrians', instead it say mainly or exclusively).

  


Of course there are always cases on the margins of both and an example would
be a high use, possibly with disabled access, tracks such as Russell Falls
in Tasmania (to highlight one that is likely known by many)
https://overpass-turbo.eu/s/1fGZ

  


So that brings me to the recently created Australian Walking Track page
https://wiki.openstreetmap.org/wiki/Australia/Walking_Tracks which gives the
options to use both tags (path and footway) but without any real
qualification about choosing between the two. This still seems to be in
conflict with the Australian tagging guidelines on Bushwalking (and cycling
tracks)
https://wiki.openstreetmap.org/wiki/Australian_Tagging_Guidelines#Bush_Walki
ng_and_Cycling_Tracks that definitively says 'Do not use highway=footway'.

  


So my question is - do you think we can come up with some criteria where a
footway ends and path commences or should we just go with the flow and stick
with OSM   'any tags
you like'? My main goal is to make sure the two Australian wikis are not in
conflict with each other.

  


I am aware there is some controversy re footway/pathway and bikes but I
would like to ignore that in this context

  


Cheers - Phil



___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Phil Wyatt]

Probably worth starting a routing thread rather than merge with a specific 
questions on foot traffic only thread

 

Cheers - Phil

 

From: Graeme Fitzpatrick  
Sent: Wednesday, 2 February 2022 6:29 PM
To: osm.talk...@thorsten.engler.id.au
Cc: OSM-Au 
Subject: Re: [talk-au] Path versus Footway

 

 

On Wed, 2 Feb 2022 at 16:54, mailto:osm.talk...@thorsten.engler.id.au> > wrote:

As far as I’m concerned, footway, cycleway, path(, and bridleway) are all 
essentially the same thing, a non-motor_vehicle path, just with different 
implied default access restrictions.

 

We should probably have a discussion about how appropriate the ones listed here 
are:

 

https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions#Australia

 

Yep!

 

How do we handle this: https://goo.gl/maps/x39C4ky1w6S7XoLUA when motorway says 
bicycle=no?

 

& similarly, you can't (at least in iD) add bike lanes to trunk roads.


 

Thanks

 

Graeme

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Graeme Fitzpatrick]

On Wed, 2 Feb 2022 at 17:24, Phil Wyatt  wrote:

>
>
> So reading from that chart and in regard to my query about ‘tracks that
> are exclusively for foot traffic’ you would say it can ONLY be a footway?
>

By that list, yes?

Thanks

Graeme
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Graeme Fitzpatrick]

On Wed, 2 Feb 2022 at 16:54,  wrote:

> As far as I’m concerned, footway, cycleway, path(, and bridleway) are all
> essentially the same thing, a non-motor_vehicle path, just with different
> implied default access restrictions.
>
>
>
> We should probably have a discussion about how appropriate the ones listed
> here are:
>
>
>
>
> https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions#Australia
>

Yep!

How do we handle this: https://goo.gl/maps/x39C4ky1w6S7XoLUA when motorway
says bicycle=no?

& similarly, you can't (at least in iD) add bike lanes to trunk roads.

Thanks

Graeme
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[Graeme Fitzpatrick]

On Wed, 2 Feb 2022 at 11:03, Phil Wyatt  wrote:

> some criteria where a footway ends and path commences
>

As I mentioned last week, I've started using path for just about everything
just to get away from adding foot=yes to bike paths & bike=yes to footpaths
:-)

Thanks

Graeme
___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Path versus Footway

[osm.talk-au]

As far as I'm concerned, footway, cycleway, path(, and bridleway) are all
essentially the same thing, a non-motor_vehicle path, just with different
implied default access restrictions.

 

We should probably have a discussion about how appropriate the ones listed
here are:

 

https://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access_restrictions
#Australia

 

 

 

From: Phil Wyatt  
Sent: Wednesday, 2 February 2022 11:00
To: OSM-Au 
Subject: [talk-au] Path versus Footway

 

Hi Folks,

 

I am contemplating a review of 'walking  tracks' tagging in Tasmania,
outside of urban areas. In my case I am starting with tracks that are
exclusively for foot traffic. My investigation has led me to what appears to
be a conflict within OSM of what is the correct tagging to use. 

 

https://wiki.openstreetmap.org/wiki/Tag:highway%3Dpath would suggest that
most could be a 'path' and this seems to be verified on existing data with
this styled overpass query (by bounding box)
https://overpass-turbo.eu/s/1fGX

 

*   Blue represents a path
*   Red represents a footway
*   Black represents steps

 

The path tag also considers extra tagging such as the sac_scale, visibility,
surface, operator etc etc which is useful extra information. Sac_scale and
operator are certainly used less frequently on footway.

 

The footway tagging
https://wiki.openstreetmap.org/wiki/Tag:highway%3Dfootway seems to have been
written with urban infrastructure in mind and as usual for OSM tagging does
not provide definitive detail (ie  it could have said 'used exclusively by
pedestrians', instead it say mainly or exclusively).

 

Of course there are always cases on the margins of both and an example would
be a high use, possibly with disabled access, tracks such as Russell Falls
in Tasmania (to highlight one that is likely known by many)
https://overpass-turbo.eu/s/1fGZ

 

So that brings me to the recently created Australian Walking Track page
https://wiki.openstreetmap.org/wiki/Australia/Walking_Tracks which gives the
options to use both tags (path and footway) but without any real
qualification about choosing between the two. This still seems to be in
conflict with the Australian tagging guidelines on Bushwalking (and cycling
tracks)
https://wiki.openstreetmap.org/wiki/Australian_Tagging_Guidelines#Bush_Walki
ng_and_Cycling_Tracks that definitively says 'Do not use highway=footway'.

 

So my question is - do you think we can come up with some criteria where a
footway ends and path commences or should we just go with the flow and stick
with OSM   'any tags
you like'? My main goal is to make sure the two Australian wikis are not in
conflict with each other.

 

I am aware there is some controversy re footway/pathway and bikes but I
would like to ignore that in this context

 

Cheers - Phil

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au