Re: [Tagging] Cycling relation misuse

[John Willis via Tagging]

> On Oct 12, 2019, at 1:28 AM, Phyks  wrote:
> 
> Hi,
> 
> I've found similar issues in France recently. Cycling routes is too
> broad and diverse and covers various realities. From a rendering
> perspective (disclaimer: I'm one of the maintainer of the new CyclOSM
> rendering style, https://cyclosm.org ), it is very 
> often a nightmare to
> try to figure out which one are worth rendering and which ones are just
> "tag to render".


Similar to how bus routes are laid over existing road infrastructure, I think 
there should be a big distinction between the paths/crossings/roads that are 
assembled to make a cycling “road", and some route that people have come up 
with just for exercising that is just some generic road in rural area people go 
touring on. 

- Cycling roads/routes for travel/transportation with some kind of documented 
status with the government. 

- MTB routes, usually using off-road ways & infrastructure - documented by the 
maintainer of the route, whoever that is.

- roads used by cyclists for exercise/racing, with no documentation or signage 
- usually shared via online route-sharing sites.

if you are making a map of the cycling routes available, I would assume the 
first category is the most important, and possibly the only one that should be 
prominently rendered.

similar to how we render roads, the prominence of motorways pales to the 
prominence of lesser roads.  Please include them, but we would need tagging to 
show the purpose of the route, beyond “network” or what super-relation they 
belong to. 

This might be difficult, as the usage probably vary from region to region: MTB 
routes in Japan are negligible, and dedicated cycling roads abound. Whereas in 
San Deigo, there are zero “cycling roads” that are maintained by the 
government, and probably a lot of documented MTB routes in the wilderness parks.

but documenting & rendering any route that a cycle club enjoys cycling on the 
weekend? unneeded. a motorcycle club’s favorite route in the mountains is 
unworthy of a route relation as well.  

OSM is not an online route-sharing site. 

here is a “Nikko Loop” route made by some cyclist who enjoys cycling. 

https://ridewithgps.com/routes/31059198

This is the job of this other private website (ridewithgps.com 
) - document and share routes for cyclist users. But 
Nikko City has no documentation for such a route, and shouldn’t be included in 
OSM. 


Javbw

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

Re: [Tagging] Cycling relation misuse

[Warin]

On 12/10/19 03:28, Phyks wrote:

Hi,

I've found similar issues in France recently. Cycling routes is too
broad and diverse and covers various realities. From a rendering
perspective (disclaimer: I'm one of the maintainer of the new CyclOSM
rendering style, https://cyclosm.org), it is very often a nightmare to
try to figure out which one are worth rendering and which ones are just
"tag to render".

I'd say we either need subtags to precise and categorize the cycle
routes or some clear definition in the wiki.

Here are a few examples of what I mean by "too diverse":

* Some are racing routes, which have been added to OSM as a cycle route
but are by no means usable (no indication on the terrain, huge highways)
outside of the race. See
http://gis.19327.n8.nabble.com/Epreuve-Cyclo-randonnee-Paris-Brest-Paris-dans-OSM-td5924677.html
for instance (in French), now removed.

* Some are real roads with an official entity maintaining them (signs,
tourist maps, official documentation), with varying quality of
infrastructure but always a legal status. See for instance
https://www.openstreetmap.org/relation/2246847 (very bad infrastructure,
but official signs in the streets) or
https://www.openstreetmap.org/relation/6445738 (tourist road, official
organism in charge of maintaining it, dedicated and very good
infrastructure).

* Some are dedicated to a very particular category of cyclists, often
racing bikes. We have `route=mtb` for mountain bikes, we might have
`route=racing_bikes` for racing bikes? Typical example is
https://www.openstreetmap.org/relation/163266 (which might actually fall
into the tag to render category)

* Some have no official existence, but a practical one. Take
https://www.openstreetmap.org/relation/8664028 for instance, this is
just a (long) cycleway in Paris. There are no special signs, nothing
special, but everyone refers to it by this acronym, "REV", and it is
widely known. These are actually very hard to discriminate with the "tag
to render" and might easily fall back in this category.

So, in short, I think a clearer definition of what should be a cycle
route (with regards to an official entity, a widely used name or
anything else) and some tags for subcategorizing it further for special
uses (not made for any cyclist) could probaby help a lot!



While motor vehicles have road classifications to say which roads should be 
preferred cyclists have little.

There are a few 'official' maps (some by councils, cities and some even by 
cycling clubs) of usually disjointed routes but little to indicate what route 
to use from A to B. These could be for commuting, sightseeing or touring.

There are the 'racing routes' used by competitive cyclists both during a race 
and to practice for it.

So I too would like to see some additional tags for cycling routes.

P.S. It is now magpie season in Australia, cable ties are applied to bicycle 
helmets to keep them away from human flesh ..

https://www.magpiealert.com/






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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Markus]

On Fri, 11 Oct 2019, 22:00 Peter Elderson,  wrote:

> But where pedestrian crossing is not allowed at all, as in the case I
> described, two ways tagging does not give this routing problem.
>

No, but it's again not the only solution: the information that crossing the
road isn't permitted can also be added to the highway=* way (using
crossing=no, if i'm not wrong).
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Peter Elderson]

But where pedestrian crossing is not allowed at all, as in the case I 
described, two ways tagging does not give this routing problem.  These roads 
are not for pedestrians, there is no sidewalk, and if there is a separate 
footway or cycleway it is physically separated and cyclists/pedestrians will 
have to find the nearest crossing, as does the router. 

NB I am not pleading one way or the other. I just think there are cases where 
two-way tagging is a plausible and an acceptible solution, even when there is 
no physical vertical barrier. Oneway tagging might suggest crossing 
possibilities where in fact crossing is not feasible, even dangerous. That 
would be worse for a router than the opposite, because it might put people in 
danger. Routing a detour is the lesser evil.

Mvg Peter Elderson

> Op 11 okt. 2019 om 21:05 heeft Markus  het 
> volgende geschreven:
> 
>> On Fri, 11 Oct 2019, 11:21 Snusmumriken,  
>> wrote:
>> 
> 
>> That tag is about lane changing, I don't see how it could be applied to
>> my example
> 
> 
> If i understand the wiki page correctly,
> 
> lanes=2
> lanes:forward=1
> lanes:backward=1
> change:lanes:forward=not_left
> change:lanes:backward=not_left
> 
> would mean that on both lanes it isn't allowed to turn left onto the lane of 
> the oncoming traffic.
> 
> And for the case i misunderstand that tag, we can invent a new tag. Splitting 
> the road into two parallel ways isn't the only possible solution.
> 
>> My assumption is that pedestrian routing engine would stick to
>> sidewalks and crossings and not to tell the pedestrian to cross a
>> street where there is no crossing. The individual pedestrian can of
>> course make up his own mind what legal/physical risks are acceptable to
>> save a bit of time
> 
> 
> As Kevin already pointed out, there are many places without pedestrian 
> crossings. Therefore pedestrian routing wouldn't work where a road with 
> painted lane separation is mapped with two ways.
> 
> I wish you all a nice weekend
> 
> Markus
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Markus]

On Fri, 11 Oct 2019, 11:21 Snusmumriken, 
wrote:

>
> That tag is about lane changing, I don't see how it could be applied to
> my example
>

If i understand the wiki page correctly,

lanes=2
lanes:forward=1
lanes:backward=1
change:lanes:forward=not_left
change:lanes:backward=not_left

would mean that on both lanes it isn't allowed to turn left onto the lane
of the oncoming traffic.

And for the case i misunderstand that tag, we can invent a new tag.
Splitting the road into two parallel ways isn't the only possible solution.

My assumption is that pedestrian routing engine would stick to
> sidewalks and crossings and not to tell the pedestrian to cross a
> street where there is no crossing. The individual pedestrian can of
> course make up his own mind what legal/physical risks are acceptable to
> save a bit of time
>

As Kevin already pointed out, there are many places without pedestrian
crossings. Therefore pedestrian routing wouldn't work where a road with
painted lane separation is mapped with two ways.

I wish you all a nice weekend

Markus

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

Re: [Tagging] Cycling relation misuse

[Phyks]

Hi,

I've found similar issues in France recently. Cycling routes is too
broad and diverse and covers various realities. From a rendering
perspective (disclaimer: I'm one of the maintainer of the new CyclOSM
rendering style, https://cyclosm.org), it is very often a nightmare to
try to figure out which one are worth rendering and which ones are just
"tag to render".

I'd say we either need subtags to precise and categorize the cycle
routes or some clear definition in the wiki.

Here are a few examples of what I mean by "too diverse":

* Some are racing routes, which have been added to OSM as a cycle route
but are by no means usable (no indication on the terrain, huge highways)
outside of the race. See
http://gis.19327.n8.nabble.com/Epreuve-Cyclo-randonnee-Paris-Brest-Paris-dans-OSM-td5924677.html
for instance (in French), now removed.

* Some are real roads with an official entity maintaining them (signs,
tourist maps, official documentation), with varying quality of
infrastructure but always a legal status. See for instance
https://www.openstreetmap.org/relation/2246847 (very bad infrastructure,
but official signs in the streets) or
https://www.openstreetmap.org/relation/6445738 (tourist road, official
organism in charge of maintaining it, dedicated and very good
infrastructure).

* Some are dedicated to a very particular category of cyclists, often
racing bikes. We have `route=mtb` for mountain bikes, we might have
`route=racing_bikes` for racing bikes? Typical example is
https://www.openstreetmap.org/relation/163266 (which might actually fall
into the tag to render category)

* Some have no official existence, but a practical one. Take
https://www.openstreetmap.org/relation/8664028 for instance, this is
just a (long) cycleway in Paris. There are no special signs, nothing
special, but everyone refers to it by this acronym, "REV", and it is
widely known. These are actually very hard to discriminate with the "tag
to render" and might easily fall back in this category.

So, in short, I think a clearer definition of what should be a cycle
route (with regards to an official entity, a widely used name or
anything else) and some tags for subcategorizing it further for special
uses (not made for any cyclist) could probaby help a lot!

Best,
-- 
Phyks
Le 11/10/2019 à 13:52, Mateusz Konieczny a écrit :
> 
> 
> 11 Oct 2019, 12:18 by tagging@openstreetmap.org:
>>
>> Is there something Im not understanding? Can anyone make a route relation 
>> for any Way regardless if it is actually a designated oute by a city, 
>> signed, or publically documented?
>>
> Such tagging for rendering happens
> but is incorrect and should be deleted.
> 

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

Re: [Tagging] Cycling relation misuse

[Andy Townsend]

On 11/10/2019 12:51, Mateusz Konieczny wrote:


11 Oct 2019, 12:37 by rich...@systemed.net

(I have a fair few lines of code in cycle.travel's rendering and
routing
codes to blacklist certain routes in OSM which are made up or
otherwise
unsuitable.)

Can you list made-up lines that pollute OSM?

(I'm not Richard and these aren't cycle routes but) I've recently set a 
couple of walking routes to "name:signed=no" based on walking 
significant portions of them and never seeing a sign:


https://www.openstreetmap.org/relation/8450999

Incomplete "The Inn Way"; appears to be from an out of print book.

https://www.openstreetmap.org/relation/6367972

"Three Feathers Walk (Kilburn)", original source unclear but listed at LDWA.

I did wonder whether it was worth asking on talk-gb whether they should 
be deleted, but didn't bother in the end.



A couple of other examples that I have not seen signage for are:

https://www.openstreetmap.org/relation/7336319 (Wainwright's Coast to Coast)

https://www.openstreetmap.org/relation/1996318 (Lyke Wake Walk)

I was the last editor of both of those (editing path changes around Chop 
Gate), but only saw waymarks for the Cleveland Way. The second of these 
predates many of the national trails, the first is as well established 
as and probably walked more than many national trails.  Both are now 
much more than just "a favourite walk" or "something somebody created to 
sell a book".


Best Regards,

Andy


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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Peter Elderson]

Then you wouldn’t tag separate carriageways on that particular way. In my 
country, lots of roads have carriageways separated by two lines with a green 
paint band of 1 m in between. I understand this is a type of european lining. 
Sometimes there is grass in between for a stretch, or vertical barrier. 
Roundabouts are the favoured kind of crossings, the green band usually widens 
there and gets stripes, sometimes kerbs
and grassy areas, ideal for placing giant billboards or artistic objects.

I would not hesitate to map these as two ways. The sections approaching 
roundabouts already are mapped separate.

Mvg Peter Elderson

> Op 11 okt. 2019 om 15:27 heeft Kevin Kenny  het 
> volgende geschreven:
> 
> On Fri, Oct 11, 2019 at 5:21 AM Snusmumriken
>  wrote:
>> My assumption is that pedestrian routing engine would stick to
>> sidewalks and crossings and not to tell the pedestrian to cross a
>> street where there is no crossing. The individual pedestrian can of
>> course make up his own mind what legal/physical risks are acceptable to
>> save a bit of time
> 
> You surely don't live in my neighbourhood!  Aside from the fact that
> there are no sidewalks on the residential streets, I don't think there
> is any place where the tertiary road to the north has a marked
> crossing. The routing engine that you imagine would say, "you can't
> get there from here." I admit that the town isn't pedestrian-friendly,
> but I still walk to work daily.
> -- 
> 73 de ke9tv/2, Kevin
> 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Kevin Kenny]

On Fri, Oct 11, 2019 at 5:21 AM Snusmumriken
 wrote:
> My assumption is that pedestrian routing engine would stick to
> sidewalks and crossings and not to tell the pedestrian to cross a
> street where there is no crossing. The individual pedestrian can of
> course make up his own mind what legal/physical risks are acceptable to
> save a bit of time

You surely don't live in my neighbourhood!  Aside from the fact that
there are no sidewalks on the residential streets, I don't think there
is any place where the tertiary road to the north has a marked
crossing. The routing engine that you imagine would say, "you can't
get there from here." I admit that the town isn't pedestrian-friendly,
but I still walk to work daily.
-- 
73 de ke9tv/2, Kevin

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 14:59 by pelder...@gmail.com:

>
>
>> Op 11 okt. 2019 om 11:22 heeft Philip Barnes  het 
>> volgende geschreven:
>>
>> Not just the driver. Routing software can be used to determine which vehicle 
>> can give the quickest response.
>>
>> Phil (trigpoint)
>>
>
> I would never trust OSM data for emergency routing or any purpose requiring 
> high reliability, unless I had complete control and quality assurance of the 
> data. And since basic setup of OSM is that anyone can change data at any 
> time, I can be sure I don’t have guaranteed reliability. 
>
Sadly, sometimes all alternatives are even worse.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Peter Elderson]

> Op 11 okt. 2019 om 11:22 heeft Philip Barnes  het 
> volgende geschreven:
> 
> Not just the driver. Routing software can be used to determine which vehicle 
> can give the quickest response.
> 
> Phil (trigpoint)

I would never trust OSM data for emergency routing or any purpose requiring 
high reliability, unless I had complete control and quality assurance of the 
data. And since basic setup of OSM is that anyone can change data at any time, 
I can be sure I don’t have guaranteed reliability. 

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

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Fri, 2019-10-11 at 13:47 +0200, Mateusz Konieczny wrote:
> 
> Redefining stuff without very, very good
> reason seems to be a bad idea.

I don't see it as any kind of redefinition. I've been mapping like that
for many years. And as Florian Lohoff pointed out, so has many others
also.


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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Paul Johnson]

On Fri, Oct 11, 2019 at 5:38 AM Snusmumriken 
wrote:

> On Fri, 2019-10-11 at 11:21 +0200, Martin Koppenhoefer wrote:
> >
> >
> > Am Fr., 11. Okt. 2019 um 11:10 Uhr schrieb Snusmumriken <
> > snusmumriken.map...@runbox.com>:
> > > It is up to the driver. I think he can ignore most of the traffic
> > > laws
> > > in the cause of getting as fast and as safe to where he needs to
> > > go. So
> > > he would use his own judgment and not so much what a routing engine
> > > tells him what he can do.
> >
> >
> >
> > you are missing the point: when the emergency vehicle gets the call,
> > the routing engine will suggest a route to approach the place of
> > action from where it is now, and depending on the osm data (and other
> > data like traffic congestion, unaccessible roads, etc.) it may
> > suggest different routes. Of course you can dismiss this in general
> > and say: "the driver will know where to go" or "will use his own
> > judgement", i.e. would not use OSM data at all, but this is not the
> > reality, in reality, OSM is used more and more in emergency
> > scenarios. There are companies dedicated to provide OSM-data-based
> > infrastructure for use by emergency services. I have seen it.
>
>
> Thanks for clearing that out. I still think it is better to map for the
> 99.99% of drivers who need to follow the law strictly. Special tagging
> for different emergency vehicles could be applied.
>
> Just to be clear, I'm not advocating that legal separation MUST lead to
> way separation. Just that a rule that wouldn't allow it would be a very
> bad rule. What makes most sense based upon the ground truth should be
> followed.
>

I think you're asking for a new tag, or adding turn restrictions, not
physical separation.  It's pretty well established that two lines is two
roadways, for which crossing over is only really going to happen where
another way is crossing between the two, not "you can't cross this line on
the pavement".  It's not like the rest of the world doesn't have this
problem, the US frequently has flush medians (
https://i.imgur.com/st58ROv.png) that indicate that you can't turn across
them or use it like a lane.  About the only time these don't get mapped as
a single way is if the median is of a geometry to deal with two closely
adjacent intersections that the only reason there's not another curbed
island there is to deal with vehicle offtracking.  Or more rarely because
it gets a fire station driveway over the median, but then the emergency gap
gets tagged as such.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Cycling relation misuse

[John Willis via Tagging]

> On Oct 11, 2019, at 8:52 PM, Mateusz Konieczny  
> wrote:
> 
> Can anyone make a route relation for any Way regardless if it is actually a 
> designated oute by a city, signed, or publically documented?
> Such tagging for rendering happens
> but is incorrect and should be deleted.

This is what my gut told me. I’ll be careful as I go through them in the next 
few months. I research the designated routes by actually cycling them and 
looking up the governmental maps of their routes. so many of them incomplete.

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

Re: [Tagging] Cycling relation misuse

[Mateusz Konieczny]

11 Oct 2019, 12:18 by tagging@openstreetmap.org:
>
> Is there something Im not understanding? Can anyone make a route relation for 
> any Way regardless if it is actually a designated oute by a city, signed, or 
> publically documented?
>
Such tagging for rendering happens
but is incorrect and should be deleted.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Cycling relation misuse

[Mateusz Konieczny]

11 Oct 2019, 12:37 by rich...@systemed.net
> (I have a fair few lines of code in cycle.travel's rendering and routing
> codes to blacklist certain routes in OSM which are made up or otherwise
> unsuitable.)
>
Can you list made-up lines that pollute OSM?___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 12:38 by snusmumriken.map...@runbox.com:

> On Fri, 2019-10-11 at 11:21 +0200, Martin Koppenhoefer wrote:
>
>>
>>
>> Am Fr., 11. Okt. 2019 um 11:10 Uhr schrieb Snusmumriken <
>> snusmumriken.map...@runbox.com>:
>> > It is up to the driver. I think he can ignore most of the traffic
>> > laws
>> > in the cause of getting as fast and as safe to where he needs to
>> > go. So
>> > he would use his own judgment and not so much what a routing engine
>> > tells him what he can do.
>>
>>
>>
>> you are missing the point: when the emergency vehicle gets the call,
>> the routing engine will suggest a route to approach the place of
>> action from where it is now, and depending on the osm data (and other
>> data like traffic congestion, unaccessible roads, etc.) it may
>> suggest different routes. Of course you can dismiss this in general
>> and say: "the driver will know where to go" or "will use his own
>> judgement", i.e. would not use OSM data at all, but this is not the
>> reality, in reality, OSM is used more and more in emergency
>> scenarios. There are companies dedicated to provide OSM-data-based
>> infrastructure for use by emergency services. I have seen it.
>>
>
>
> Thanks for clearing that out. I still think it is better to map for the
> 99.99% of drivers who need to follow the law strictly. Special tagging
> for different emergency vehicles could be applied.
>
> Just to be clear, I'm not advocating that legal separation MUST lead to
> way separation. Just that a rule that wouldn't allow it would be a very
> bad rule. What makes most sense based upon the ground truth should be
> followed.
>
And we are doing this.

Maybe there is better choice than
"One highway line = carriageway",
but in my opinion healthy conservatives is a good idea.

Redefining stuff without very, very good
reason seems to be a bad idea.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 13:23 by f...@zz.de:

> Hi
>
> On Thu, Oct 10, 2019 at 10:59:53PM +0200, Mateusz Konieczny wrote:
>
>> > I had a quick 10 Minute Look at Mapillary and i have found 10s of
>> > examples of separate way although no physical barrier. 
>> >
>> It can be easily done for any kind of mistake.
>>
>> Have you tried comparing it to split way tagging?
>>
>> It may be a good argument is that tagging
>> is appearing often. Just because it is appearing at 
>> all is not very interesting, we know this.
>>
>
> I would call them mistaked. By intuition i would map
> these areas exactly as they are right now.
>
> And these stuff is not some ice road at the north pole. These
> are streets which have been touched by 1000s of mappers and
> you call all of them beeing inexperienced noobs making mistakes?
>
Or maybe people were too busy to fix it.

I recently reverted two large scale
copyright violations (one was on my 
todo list for months).

I have lists of about 100k automatically
detected mistakes, not found by JOSM
validator.

I have opened more than 200 issues on
various bug trackers of  OSM software that I can
and want to fix.

There are many OSM notes on my city.
One of them is about an incorrectly
split road.
Etc etc (and all that is just OSM activities
that are mostly entertainment/hobby in my case)
Presence of such ways is indicator,
but just because N mappers are
active nearby does not mean that they
all N agree with that tagging

And for basically any kind of mistake
I can find many cases where it happens.
Again, have you tried comparing
popularity of both methods in 
region where you are not editing?___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 13:04 by snusmumriken.map...@runbox.com:

> On Fri, 2019-10-11 at 12:48 +0200, Martin Koppenhoefer wrote:
>
>> Am Fr., 11. Okt. 2019 um 12:39 Uhr schrieb Snusmumriken <
>> snusmumriken.map...@runbox.com>:
>> > Just to be clear, I'm not advocating that legal separation MUST
>> > lead to
>> > way separation. Just that a rule that wouldn't allow it would be a
>> > very
>> > bad rule. What makes most sense based upon the ground truth should
>> > be
>> > followed.
>>
>>
>>
>> generally, in OSM the ways highway=* represent the carriageway.
>> "legal separations" often are just lane markings, i.e. they do not
>> constitute a carriageway, hence are not to be mapped individually
>> each with their way. We have generally followed this definition,
>>
>
> Who is this 'we' you're speaking in behalf of?
>
>From context it is fairly obvious that 
Martin is describing typical mapping
methods encountered during his OSM
activities.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Florian Lohoff]

Hi

On Thu, Oct 10, 2019 at 10:59:53PM +0200, Mateusz Konieczny wrote:
> > I had a quick 10 Minute Look at Mapillary and i have found 10s of
> > examples of separate way although no physical barrier. 
> >
> It can be easily done for any kind of mistake.
> 
> Have you tried comparing it to split way tagging?
> 
> It may be a good argument is that tagging
> is appearing often. Just because it is appearing at 
> all is not very interesting, we know this.

I would call them mistaked. By intuition i would map
these areas exactly as they are right now.

And these stuff is not some ice road at the north pole. These
are streets which have been touched by 1000s of mappers and
you call all of them beeing inexperienced noobs making mistakes?

Flo
-- 
Florian Lohoff f...@zz.de
UTF-8 Test: The  ran after a , but the  ran away


signature.asc
Description: PGP signature
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Cycling relation misuse

[Dave F via Tagging]

Are you able to properly verify these are all "Random road your cycling 
club likes to ride on the weekend" & not designated/signed routes?


ATM it appears you're vetting them purely on the class of highway used.

Designated cycle routes can go along "just regular roads, with no 
designation for cyclists."


DaveF





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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Fri, 2019-10-11 at 12:48 +0200, Martin Koppenhoefer wrote:
> Am Fr., 11. Okt. 2019 um 12:39 Uhr schrieb Snusmumriken <
> snusmumriken.map...@runbox.com>:
> > Just to be clear, I'm not advocating that legal separation MUST
> > lead to
> > way separation. Just that a rule that wouldn't allow it would be a
> > very
> > bad rule. What makes most sense based upon the ground truth should
> > be
> > followed.
> 
> 
> 
> generally, in OSM the ways highway=* represent the carriageway.
> "legal separations" often are just lane markings, i.e. they do not
> constitute a carriageway, hence are not to be mapped individually
> each with their way. We have generally followed this definition,

Who is this 'we' you're speaking in behalf of?


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

Re: [Tagging] Cycling relation misuse

[Richard Fairhurst]

Martin Koppenhoefer wrote:
> wouldn't it be better to delete them from OSM if they are made up?

It would, but I have limited hours in the day to police every single cycle
route relation in OSM.

I lose track of the amount of time I spent on user messages and changeset
comments trying to get the Great Divide Mountain Bike Route properly tagged
as route=mtb... it even says Mountain Bike in its name, for crying out loud.

Richard



--
Sent from: http://gis.19327.n8.nabble.com/Tagging-f5258744.html

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

Re: [Tagging] Cycling relation misuse

[Martin Koppenhoefer]

Am Fr., 11. Okt. 2019 um 12:38 Uhr schrieb Richard Fairhurst <
rich...@systemed.net>:

> (I have a fair few lines of code in cycle.travel's rendering and routing
> codes to blacklist certain routes in OSM which are made up or otherwise
> unsuitable.)



wouldn't it be better to delete them from OSM if they are made up?

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Martin Koppenhoefer]

Am Fr., 11. Okt. 2019 um 12:39 Uhr schrieb Snusmumriken <
snusmumriken.map...@runbox.com>:

> Just to be clear, I'm not advocating that legal separation MUST lead to
> way separation. Just that a rule that wouldn't allow it would be a very
> bad rule. What makes most sense based upon the ground truth should be
> followed.




generally, in OSM the ways highway=* represent the carriageway. "legal
separations" often are just lane markings, i.e. they do not constitute a
carriageway, hence are not to be mapped individually each with their way.
We have generally followed this definition, and if we were to change it, we
should have good reasons. If the only reason is that we need fewer turn
restrictions, then I believe we should find a better workaround than giving
up the carriageway=highway definition.

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Fri, 2019-10-11 at 11:21 +0200, Martin Koppenhoefer wrote:
> 
> 
> Am Fr., 11. Okt. 2019 um 11:10 Uhr schrieb Snusmumriken <
> snusmumriken.map...@runbox.com>:
> > It is up to the driver. I think he can ignore most of the traffic
> > laws
> > in the cause of getting as fast and as safe to where he needs to
> > go. So
> > he would use his own judgment and not so much what a routing engine
> > tells him what he can do.
> 
> 
> 
> you are missing the point: when the emergency vehicle gets the call,
> the routing engine will suggest a route to approach the place of
> action from where it is now, and depending on the osm data (and other
> data like traffic congestion, unaccessible roads, etc.) it may
> suggest different routes. Of course you can dismiss this in general
> and say: "the driver will know where to go" or "will use his own
> judgement", i.e. would not use OSM data at all, but this is not the
> reality, in reality, OSM is used more and more in emergency
> scenarios. There are companies dedicated to provide OSM-data-based
> infrastructure for use by emergency services. I have seen it.


Thanks for clearing that out. I still think it is better to map for the
99.99% of drivers who need to follow the law strictly. Special tagging
for different emergency vehicles could be applied.

Just to be clear, I'm not advocating that legal separation MUST lead to
way separation. Just that a rule that wouldn't allow it would be a very
bad rule. What makes most sense based upon the ground truth should be
followed.


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

Re: [Tagging] Cycling relation misuse

[Richard Fairhurst]

John Willis wrote:
> I want to delete these fake “mountain workout” relations that 
> should be mapped in strava or a similar workout app. 

Fully agree. Go for it.

OSM is for verifiable, signposted cycle routes and verifiable, real cycling
infrastructure. If a route is on the way to being signposted then it can be
mapped with state=proposed.

There are literally millions of personal favourite rides in guidebooks and
on third-party websites but with no supporting evidence on the ground. There
is no place for these in OSM.

(I have a fair few lines of code in cycle.travel's rendering and routing
codes to blacklist certain routes in OSM which are made up or otherwise
unsuitable.)

Richard



--
Sent from: http://gis.19327.n8.nabble.com/Tagging-f5258744.html

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Fri, 2019-10-11 at 11:32 +0200, Mateusz Konieczny wrote:
> 
> 
> 11 Oct 2019, 11:19 by snusmumriken.map...@runbox.com:
> > My assumption is that pedestrian routing engine would stick to
> > sidewalks and crossings and not to tell the pedestrian to cross a
> > street where there is no crossing. The individual pedestrian can of
> > course make up his own mind what legal/physical risks are
> > acceptable to
> > save a bit of time
> 
> It again depends on a country.
> And in same countries depends on additional details.
> 
> In many places crossing road may be as
> legal and/or as safe as crossing on a
> designated crossing.

I would think that whether it is safe or not would mostly depend on the
time of day and how much traffic there happens to be and thus best left
to the individual to judge.


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

Re: [Tagging] Cycling relation misuse

[John Willis via Tagging]

Am I misunderstanding something fundamental?  Mapping cycle route relations 
Sounds a lot like mapping bus routes: mapping the designated routes of existing 
public/private routes seems to be useful - mapping where you like to drive your 
RV around With a bus route relation and inter-mixing that into official bus 
route relations sounds like a disaster. 

I was under the impression cycle route relations (especially with a network=* 
designation) were for mapping designated cycleway routes - not mapping wherever 
bicycle=yes is implicit or implied, or whatever route I happen to enjoy riding 
on weekends.

Of course The the relation can include any way - it might include cycle Lanes 
in large roads or segments of roads used to link cycling roads together - but 
just any random road your cycling club likes to ride on the weekend? A route 
that is 100% trunk road from end-to-end with 0% cycling lanes or paths and no 
official designation as a route for cyclists? Is that part of a "cycling route 
network?" Is my favorite Canoeing path around a lake ferry route relation? 

It reeks of polluting the actual designated cycling routes (which are not even 
half-finished in my area, relation-wise) with relations of random roads which 
are just regular roads, with no designation for cyclists. It's like if I 
designated my daily commute a "cycle route relation, network=local" just so I 
can get a bright blue line in OpenCycleMap, rather than creating/downloading a 
route in my cycling app on my phone for my own private use - mapping for the 
renderer IMO. 

Is there something Im not understanding? Can anyone make a route relation for 
any Way regardless if it is actually a designated oute by a city, signed, or 
publically documented?

Javbw

> On Oct 11, 2019, at 5:58 PM, Warin <61sundow...@gmail.com> wrote:
> 
> On 11/10/19 18:04, John Willis via Tagging wrote:
>> Questions about using cycle relations properly:
>> 
>> I am mapping and repairing cycle roads in the Kanto/Tokyo area. There are a 
>> lot of designated cycling roads that follow a long rivers and other water 
>> features out into the countryside, making up a regional system, and a lot of 
>> smaller local cycling roads (also along small rivers) that connect 
>> neighborhoods and towns together.
>> example: https://www.openstreetmap.org/relation/3218181
>> 
>> I’m working to get all the individual ways of the cycle roads into relations 
>> and to properly classify these (local/regional, etc).
>> 
>> But on the cycling layer of OSM, I find regular roads labeled as cycle 
>> routes: mountain roads where professional cyclists like to exercise labeled 
>> as a “cycling route”, which seems like “mapping for the renderer”.
>> 
>> example: https://www.openstreetmap.org/relation/8066243
>> 
>> - They don’t seem to be cycling roads - all the relation members are trunk 
>> roads or similar - no cycleways whatsoever.
> 
> There is no requirement for a cycle route to use cycleways, even in part.
>> 
>> -they are dangerous routes with no side-paths, sidewalks, or dedicated cycle 
>> lanes - just regular roads.
>> 
>> - they are exercise loops or hill climbs for pro cyclistsand serve no 
>> purpose for travelers or commuters.
> 
> Never the less they could be seen as cycle routes - frequently used by 
> cyclists?
> 
>> 
>> - they are not, AFAIK, part of an official “cycling network”. The 
>> Super-relation someone has added all cycle routes to ( 関東地方サイクリングロード・ネットワーク 
>> ). https://www.openstreetmap.org/relation/8051094 also seems to be made-up 
>> and not official either - the name only returns one result (the OSM data 
>> page) when searched.
>> 
>> 
>> 
>> To me, these non-cycle routes are just garbage relations meant to have the 
>> route show up on the cycling view of OSM for people doing workouts.
> 
> I have had a commuting cyclist map into OSM cycling lanes .. that are not 
> there, shared paths that are not shared.. I would much rather that were 
> mapped as routes showing the actual infrastructure that is there.
> 
>> 
>> I want to delete these fake “mountain workout” relations that should be 
>> mapped in strava or a similar workout app.
> 
> If the route shows that regular roads are used .. possibly use the 
> description key to state the nature of the route?
> 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging


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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 11:22 by colin.sm...@xs4all.nl:

>
> On 2019-10-11 11:09, Snusmumriken wrote:
>
>
>> It is up to the driver. I think he can ignore most of the traffic laws
>>  in the cause of getting as fast and as safe to where he needs to go. So
>>  he would use his own judgment and not so much what a routing engine
>>  tells him what he can do.
>>
> That may be the case in some countries, but in the UK there are limitations 
> on what laws emergency vehicles can and cannot break.
>  
> Countries also have differing definitions of what constitutes an "emergency 
> vehicle" for these purposes. A regular doctor on his way to an emergency, 
> organs for transplant for example... in the UK they cannot use blue lights, 
> and "people die" because these vehicles get stuck in traffic.
>  
> See this website for all the complexities:
> http://www.ukemergency.co.uk/blue-light-use/ 
> 
>  
> The question is, how to model this in OSM? Or do we just model for normal 
> cars? Just like for trucks, routing for emergency vehicles needs 
> parameterisation for the vehicle characteristics and the specific use to 
> which it is being put *at that moment*.
>
In this case I would map as done so far.

Split on physical splits etc.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 11:09 by snusmumriken.map...@runbox.com:

> On Fri, 2019-10-11 at 10:57 +0200, Mateusz Konieczny wrote:
>
>>
>>
>>
>> 11 Oct 2019, 10:50 by snusmumriken.map...@runbox.com:
>> > On Fri, 2019-10-11 at 10:31 +0200, Martin Koppenhoefer wrote:
>> > > Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <
>> > > snusmumriken.map...@runbox.com>:
>> > > > A level strip of grass can be crossed by any car. With a big
>> > > SUV
>> > > > you
>> > > > can cross curbs and so on. It's just a questions about how big
>> > > your
>> > > > car
>> > > > is and the nature of the physical separation. But I don't think
>> > > > that
>> > > > OSM should be about that, but rather to be a map database to be
>> > > > able to
>> > > > provide a _legal_ route from A to B.
>> > > 
>> > > 
>> > > 
>> > > what is not legal for you may be legal for someone else, for
>> > > example
>> > > an emergency vehicle...
>> > 
>> > Yes, exactly, but as I wrote "You have to remember that some
>> > physical
>> > separation are just as easy to cross as a painted line."
>> > 
>> > So a level strip of grass would be just as problematic for the
>> > emergency vehicle routing engine as a painted line.
>>
>> Maybe it depends on location but in
>> Poland emergency vehicles routinely
>> ignore road paintings, one-way restrictions,
>> traffic lights, turn restrictions etc.
>>
>> And I have never seen an emergency vehicle
>> crossing a grass median.
>>
>> And it seem obvious that crossing a grass median
>> is trickier than crossing just a painted line.
>>
>
> It is up to the driver. I think he can ignore most of the traffic laws
> in the cause of getting as fast and as safe to where he needs to go. So
> he would use his own judgment and not so much what a routing engine
> tells him what he can do.
>
The problem is the initial routing -
selecting nearest emergency vehicle(s),
initial proposed route etc.

And such software is also used for planning purposes - 
coverage of area by available firefighters etc.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 11:19 by snusmumriken.map...@runbox.com:
> My assumption is that pedestrian routing engine would stick to
> sidewalks and crossings and not to tell the pedestrian to cross a
> street where there is no crossing. The individual pedestrian can of
> course make up his own mind what legal/physical risks are acceptable to
> save a bit of time
>
It again depends on a country.
And in same countries depends on additional details.
In many places crossing road may be as
legal and/or as safe as crossing on a
designated crossing.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Colin Smale]

On 2019-10-11 11:09, Snusmumriken wrote:

> It is up to the driver. I think he can ignore most of the traffic laws
> in the cause of getting as fast and as safe to where he needs to go. So
> he would use his own judgment and not so much what a routing engine
> tells him what he can do.

That may be the case in some countries, but in the UK there are
limitations on what laws emergency vehicles can and cannot break. 

Countries also have differing definitions of what constitutes an
"emergency vehicle" for these purposes. A regular doctor on his way to
an emergency, organs for transplant for example... in the UK they cannot
use blue lights, and "people die" because these vehicles get stuck in
traffic. 

See this website for all the complexities: 
http://www.ukemergency.co.uk/blue-light-use/ 

The question is, how to model this in OSM? Or do we just model for
normal cars? Just like for trucks, routing for emergency vehicles needs
parameterisation for the vehicle characteristics and the specific use to
which it is being put *at that moment*.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Philip Barnes]

Not just the driver. Routing software can be used to determine which vehicle 
can give the quickest response.

Phil (trigpoint)

On Friday, 11 October 2019, Snusmumriken wrote:
> On Fri, 2019-10-11 at 10:57 +0200, Mateusz Konieczny wrote:
> > 
> > 
> > 
> > 11 Oct 2019, 10:50 by snusmumriken.map...@runbox.com:
> > > On Fri, 2019-10-11 at 10:31 +0200, Martin Koppenhoefer wrote:
> > > > Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <
> > > > snusmumriken.map...@runbox.com>:
> > > > > A level strip of grass can be crossed by any car. With a big
> > > > SUV
> > > > > you
> > > > > can cross curbs and so on. It's just a questions about how big
> > > > your
> > > > > car
> > > > > is and the nature of the physical separation. But I don't think
> > > > > that
> > > > > OSM should be about that, but rather to be a map database to be
> > > > > able to
> > > > > provide a _legal_ route from A to B.
> > > > 
> > > > 
> > > > 
> > > > what is not legal for you may be legal for someone else, for
> > > > example
> > > > an emergency vehicle...
> > > 
> > > Yes, exactly, but as I wrote "You have to remember that some
> > > physical
> > > separation are just as easy to cross as a painted line."
> > > 
> > > So a level strip of grass would be just as problematic for the
> > > emergency vehicle routing engine as a painted line.
> > 
> > Maybe it depends on location but in
> > Poland emergency vehicles routinely
> > ignore road paintings, one-way restrictions,
> > traffic lights, turn restrictions etc.
> > 
> > And I have never seen an emergency vehicle
> > crossing a grass median.
> > 
> > And it seem obvious that crossing a grass median
> > is trickier than crossing just a painted line.
> 
> It is up to the driver. I think he can ignore most of the traffic laws
> in the cause of getting as fast and as safe to where he needs to go. So
> he would use his own judgment and not so much what a routing engine
> tells him what he can do.
> 
> 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>

-- 
Sent from my Sailfish device
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Martin Koppenhoefer]

Am Fr., 11. Okt. 2019 um 11:10 Uhr schrieb Snusmumriken <
snusmumriken.map...@runbox.com>:

> It is up to the driver. I think he can ignore most of the traffic laws
> in the cause of getting as fast and as safe to where he needs to go. So
> he would use his own judgment and not so much what a routing engine
> tells him what he can do.




you are missing the point: when the emergency vehicle gets the call, the
routing engine will suggest a route to approach the place of action from
where it is now, and depending on the osm data (and other data like traffic
congestion, unaccessible roads, etc.) it may suggest different routes. Of
course you can dismiss this in general and say: "the driver will know where
to go" or "will use his own judgement", i.e. would not use OSM data at all,
but this is not the reality, in reality, OSM is used more and more in
emergency scenarios. There are companies dedicated to provide
OSM-data-based infrastructure for use by emergency services. I have seen it.

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Thu, 2019-10-10 at 19:53 +0200, Markus wrote:
> On Thu, 10 Oct 2019 at 16:10, Snusmumriken
>  wrote:
> > For example if you try to create a routing advice for a car
> > journey.
> > Let's say that the journey starts at Main street number 10 and that
> > Main street is a two way street where the two directions are
> > legally
> > separated. Let's say that number 10 is on the right-hand side of
> > the
> > road and we are in a country that drives on the right side. Let's
> > further say that the shortest way to the destination would be to
> > cross
> > the legal separation and take left. But that would be illegal. But
> > there is no way the routing engine could know that. Unless the two
> > directions are separated.
> 
> That's not true. There's another way to tell routers that it is
> illegal to change lanes: by adding that information to the highway=*
> way. There's already a tag for this: change:langes [1] (> 90 000
> uses).

That tag is about lane changing, I don't see how it could be applied to
my example

> 
> While mapping separate ways where there is no physical barrier works
> for car routing, it breaks pedestrian routing and there's likely no
> way to fix this. Pedestrians usually are allowed to cross a painted
> line that cars aren't allowed to cross (at least in Europe).
> Therefore, if the road in your example is mapped with two separate
> ways, a routing engine would make pedestrians do a detour (possibly a
> long detour), even though they could just cross the street.

My assumption is that pedestrian routing engine would stick to
sidewalks and crossings and not to tell the pedestrian to cross a
street where there is no crossing. The individual pedestrian can of
course make up his own mind what legal/physical risks are acceptable to
save a bit of time


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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Fri, 2019-10-11 at 10:57 +0200, Mateusz Konieczny wrote:
> 
> 
> 
> 11 Oct 2019, 10:50 by snusmumriken.map...@runbox.com:
> > On Fri, 2019-10-11 at 10:31 +0200, Martin Koppenhoefer wrote:
> > > Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <
> > > snusmumriken.map...@runbox.com>:
> > > > A level strip of grass can be crossed by any car. With a big
> > > SUV
> > > > you
> > > > can cross curbs and so on. It's just a questions about how big
> > > your
> > > > car
> > > > is and the nature of the physical separation. But I don't think
> > > > that
> > > > OSM should be about that, but rather to be a map database to be
> > > > able to
> > > > provide a _legal_ route from A to B.
> > > 
> > > 
> > > 
> > > what is not legal for you may be legal for someone else, for
> > > example
> > > an emergency vehicle...
> > 
> > Yes, exactly, but as I wrote "You have to remember that some
> > physical
> > separation are just as easy to cross as a painted line."
> > 
> > So a level strip of grass would be just as problematic for the
> > emergency vehicle routing engine as a painted line.
> 
> Maybe it depends on location but in
> Poland emergency vehicles routinely
> ignore road paintings, one-way restrictions,
> traffic lights, turn restrictions etc.
> 
> And I have never seen an emergency vehicle
> crossing a grass median.
> 
> And it seem obvious that crossing a grass median
> is trickier than crossing just a painted line.

It is up to the driver. I think he can ignore most of the traffic laws
in the cause of getting as fast and as safe to where he needs to go. So
he would use his own judgment and not so much what a routing engine
tells him what he can do.


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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 10:50 by snusmumriken.map...@runbox.com:

> On Fri, 2019-10-11 at 10:31 +0200, Martin Koppenhoefer wrote:
>
>> Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <
>> snusmumriken.map...@runbox.com>:
>> > A level strip of grass can be crossed by any car. With a big SUV
>> > you
>> > can cross curbs and so on. It's just a questions about how big your
>> > car
>> > is and the nature of the physical separation. But I don't think
>> > that
>> > OSM should be about that, but rather to be a map database to be
>> > able to
>> > provide a _legal_ route from A to B.
>>
>>
>>
>> what is not legal for you may be legal for someone else, for example
>> an emergency vehicle...
>>
>
> Yes, exactly, but as I wrote "You have to remember that some physical
> separation are just as easy to cross as a painted line."
>
> So a level strip of grass would be just as problematic for the
> emergency vehicle routing engine as a painted line.
>
Maybe it depends on location but in
Poland emergency vehicles routinely
ignore road paintings, one-way restrictions,
traffic lights, turn restrictions etc.
And I have never seen an emergency vehicle
crossing a grass median.

And it seem obvious that crossing a grass median
is trickier than crossing just a painted line.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Cycling relation misuse

[Warin]

On 11/10/19 18:04, John Willis via Tagging wrote:

Questions about using cycle relations properly:

I am mapping and repairing cycle roads in the Kanto/Tokyo area. There 
are a lot of designated cycling roads that follow a long rivers and 
other water features out into the countryside, making up a regional 
system, and a lot of smaller local cycling roads (also along small 
rivers) that connect neighborhoods and towns together.

example: https://www.openstreetmap.org/relation/3218181

I’m working to get all the individual ways of the cycle roads into 
relations and to properly classify these (local/regional, etc).


But on the cycling layer of OSM, I find regular roads labeled as cycle 
routes: mountain roads where professional cyclists like to exercise 
labeled as a “cycling route”, which seems like “mapping for the renderer”.


example: https://www.openstreetmap.org/relation/8066243

- They don’t seem to be cycling roads - all the relation members are 
trunk roads or similar - no cycleways whatsoever.


There is no requirement for a cycle route to use cycleways, even in part.


-they are dangerous routes with no side-paths, sidewalks, or dedicated 
cycle lanes - just regular roads.


- they are exercise loops or hill climbs for pro cyclistsand serve no 
purpose for travelers or commuters.


Never the less they could be seen as cycle routes - frequently used by 
cyclists?




- they are not, AFAIK, part of an official “cycling network”. The 
Super-relation someone has added all cycle routes to ( 
関東地方サイクリングロード・ネットワーク ). 
https://www.openstreetmap.org/relation/8051094 also seems to be 
made-up and not official either - the name only returns one result 
(the OSM data page) when searched.




To me, these non-cycle routes are just garbage relations meant to have 
the route show up on the cycling view of OSM for people doing workouts.


I have had a commuting cyclist map into OSM cycling lanes .. that are 
not there, shared paths that are not shared.. I would much rather that 
were mapped as routes showing the actual infrastructure that is there.




I want to delete these fake “mountain workout” relations that should 
be mapped in strava or a similar workout app.


If the route shows that regular roads are used .. possibly use the 
description key to state the nature of the route?


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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Mateusz Konieczny]

11 Oct 2019, 10:31 by dieterdre...@gmail.com:

> Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <> 
> snusmumriken.map...@runbox.com > >:
>
>>
>> A level strip of grass can be crossed by any car. With a big SUV you
>>  can cross curbs and so on. It's just a questions about how big your car
>>  is and the nature of the physical separation. But I don't think that
>>  OSM should be about that, but rather to be a map database to be able to
>>  provide a _legal_ route from A to B.
>>
>
>
>
> what is not legal for you may be legal for someone else, for example an 
> emergency vehicle...
>
And this is not just a theory, in Poland
OSM is actually used for routing of at
least some emergency vehicles.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Fri, 2019-10-11 at 10:31 +0200, Martin Koppenhoefer wrote:
> Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <
> snusmumriken.map...@runbox.com>:
> > A level strip of grass can be crossed by any car. With a big SUV
> > you
> > can cross curbs and so on. It's just a questions about how big your
> > car
> > is and the nature of the physical separation. But I don't think
> > that
> > OSM should be about that, but rather to be a map database to be
> > able to
> > provide a _legal_ route from A to B.
> 
> 
> 
> what is not legal for you may be legal for someone else, for example
> an emergency vehicle...

Yes, exactly, but as I wrote "You have to remember that some physical
separation are just as easy to cross as a painted line."

So a level strip of grass would be just as problematic for the
emergency vehicle routing engine as a painted line.



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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Philip Barnes]

On Friday, 11 October 2019, Martin Koppenhoefer wrote:
> Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <
> snusmumriken.map...@runbox.com>:
> 
> >
> > A level strip of grass can be crossed by any car. With a big SUV you
> > can cross curbs and so on. It's just a questions about how big your car
> > is and the nature of the physical separation. But I don't think that
> > OSM should be about that, but rather to be a map database to be able to
> > provide a _legal_ route from A to B.
> >
> 
> 
> 
> what is not legal for you may be legal for someone else, for example an
> emergency vehicle...
>
+100
 
Also a cyclist can dismount and become a pedestrian to cross the road.

Phil (trigpoint)

-- 
Sent from my Sailfish device
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Cycling relation misuse

[Joseph Eisenberg]

Have you discussed this with the individual mappers via changeset messages
or on a Japanese forum/mailing list?

Joseph

On Fri, Oct 11, 2019 at 4:12 PM John Willis via Tagging <
tagging@openstreetmap.org> wrote:

> Questions about using cycle relations properly:
>
> I am mapping and repairing cycle roads in the Kanto/Tokyo area. There are
> a lot of designated cycling roads that follow a long rivers and other water
> features out into the countryside, making up a regional system, and a lot
> of smaller local cycling roads (also along small rivers) that connect
> neighborhoods and towns together.
> example: https://www.openstreetmap.org/relation/3218181
>
> I’m working to get all the individual ways of the cycle roads into
> relations and to properly classify these (local/regional, etc).
>
> But on the cycling layer of OSM, I find regular roads labeled as cycle
> routes: mountain roads where professional cyclists like to exercise labeled
> as a “cycling route”, which seems like “mapping for the renderer”.
>
> example: https://www.openstreetmap.org/relation/8066243
>
> - They don’t seem to be cycling roads - all the relation members are trunk
> roads or similar - no cycleways whatsoever.
>
> -they are dangerous routes with no side-paths, sidewalks, or dedicated
> cycle lanes - just regular roads.
>
> - they are exercise loops or hill climbs for pro cyclistsand serve no
> purpose for travelers or commuters.
>
> - they are not, AFAIK, part of an official “cycling network”. The
> Super-relation someone has added all cycle routes to ( 関東地方サイクリングロード・ネットワーク
> ). https://www.openstreetmap.org/relation/8051094  also seems to be
> made-up and not official either - the name only returns one result (the OSM
> data page) when searched.
>
>
>
> To me, these non-cycle routes are just garbage relations meant to have the
> route show up on the cycling view of OSM for people doing workouts.
>
> I want to delete these fake “mountain workout” relations that should be
> mapped in strava or a similar workout app.
>
> Javbw
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Martin Koppenhoefer]

Am Fr., 11. Okt. 2019 um 10:26 Uhr schrieb Snusmumriken <
snusmumriken.map...@runbox.com>:

>
> A level strip of grass can be crossed by any car. With a big SUV you
> can cross curbs and so on. It's just a questions about how big your car
> is and the nature of the physical separation. But I don't think that
> OSM should be about that, but rather to be a map database to be able to
> provide a _legal_ route from A to B.
>



what is not legal for you may be legal for someone else, for example an
emergency vehicle...

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Snusmumriken]

On Thu, 2019-10-10 at 17:57 +0200, Martin Koppenhoefer wrote:
> 
> 
> Am Do., 10. Okt. 2019 um 16:10 Uhr schrieb Snusmumriken <
> snusmumriken.map...@runbox.com>:
> > For example if you try to create a routing advice for a car
> > journey.
> > Let's say that the journey starts at Main street number 10 and that
> > Main street is a two way street where the two directions are
> > legally
> > separated. Let's say that number 10 is on the right-hand side of
> > the
> > road and we are in a country that drives on the right side. Let's
> > further say that the shortest way to the destination would be to
> > cross
> > the legal separation and take left. But that would be illegal. But
> > there is no way the routing engine could know that. Unless the two
> > directions are separated.
> 
> 
> or the kind of legal separation is mapped so that the software could
> know. 
> Or you park your car on the opposite side of the road and cross it as
> a pedestrian. Or maybe you'll finding a free parking spot much
> farther away and have to walk quite a bit. Or maybe they drive on the
> left, you're the prime minister, and your driver will park the car...
> 
> Of course it does not matter for those cases where you may not cross
> the divider legally and you do not plan to do so, and it is mapped as
> if you could not even physically, but there are usecases where you
> might want to either cross illegally, or you have the special right
> to do so, and then it should be possible to determine whether there
> is a physical possibility or not.

You have to remember that some physical separation are just as easy to
cross as a painted line. 

A level strip of grass can be crossed by any car. With a big SUV you
can cross curbs and so on. It's just a questions about how big your car
is and the nature of the physical separation. But I don't think that
OSM should be about that, but rather to be a map database to be able to
provide a _legal_ route from A to B.





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

Re: [Tagging] railway crossings with cycleways

[John Willis via Tagging]

I think it depends on the country - most cycling roads in Japan “end” and dump 
you into a sidewalk to cross rail crossings - otherwise they either avoid the 
grade crossing or end and put you into regular road traffic to use the road’s 
grade crossing. they are basically 1-2km long sections of cycleway that 
constantly “end" and you either join the road traffic (as a car) or use 
pedestrian routes (crosswalks, sidewalks) to get around obstacles (like 
railways or crossing a road with an island or similar). 

The official documentation for this cycling road details the “detour” on the 
regular roads to use the car’s grade crossing because they don’t want a cycle 
grade crossing. 

https://www.openstreetmap.org/#map=18/36.29001/139.36018=C

you can see the detour in blue on the map
http://www.kendoseibi.pref.gunma.jp/section/dourokikaku/hp/download/05hebi1.pdf 



I’m sure there is a dedicated “cycleway grade crossing” somewhere in Japan, but 
in the few hundred KM of cycleways I have ridden, I have never seen one. most 
non-motorcar grade crossings are for peds or for tractors. 

But I’m also sure there are countries with hundreds of dedicated cycleway 
crossings.  

Perhaps a separate “cycleway grade crossing” is appropriate  - iD recently 
started supporting cycleway crossings. Perhaps this tag is also needed.

But please remember that the basic premise that a cyclist is a variant of 
pedestrian is the idea for a lot of cycling infrastructure around the world, 
and making a cycleway crossing a “road” crossing rather than a “path” crossing 
would be considered bad mapping in a lot of places. 

Please consider this when coming up with a solution to this problem. 

Javbw


> On Oct 9, 2019, at 11:07 PM, Vɑdɪm  wrote:
> 
> On the other hand the Vienna Convention on Road Traffic mentions crossings
> for cyclists separately
> (https://www.unece.org/fileadmin/DAM/trans/conventn/Conv_road_traffic_EN.pdf):
> 
>> 3. (a) The standing or parking of a vehicle on the carriageway shall be
>> prohibited:
>> (i) On pedestrian crossings, on *crossings for cyclists*, and on
>> level-crossings;
> 
> 
> 
> 
> 
> --
> Sent from: http://gis.19327.n8.nabble.com/Tagging-f5258744.html
> 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

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

Re: [Tagging] Divided highways, and not so divided highways, one way or two

[Alan Mackie]

On Thu, 10 Oct 2019 at 15:50, Vɑdɪm  wrote:

> Florian Lohoff-2 wrote
> > On Thu, Oct 10, 2019 at 08:38:28AM +0200, Frederik Ramm wrote:
> > Mapping large, multi-lane roads with a "do not cross line" in the
> > middle as single line requires 4-5 times the number of turn
> > restrictions. These are number i am estimating from my own experience
> > mapping it one or the other way.
> > At every way junction one has to model every disallowed way/turn.
> > From my experience this is very error prone.
>
> +1
>
> Also there are some arrangements which probably do not have a simple
> solution even with centre=* tag suggested here.
>
> For example a street with a tramway track in the middle separated from the
> rest of the roadbed by dividing lines at each side which vehicles cannot
> cross: https://goo.gl/maps/VHKbwjMoCVwawHxU9. By the way tramway tracks
> are
> drawn with two separate ways, so a single way line in the middle would make
> you think that the tramway tracks are not in the middle of the roadbed but
> at its sides.
>
> Another example is a bus lane in the middle of the road: 2 lanes of forward
> traffic, a forward bus lane, a backward bus lane, backward traffic
> https://goo.gl/maps/FubkLdHRP6DHLkv86.
>
> Yes another one is a bus lane on the right side, but it turns on the left
> through 4 lanes of normal forward traffic which not allowed to turn left
> https://goo.gl/maps/QFcfDW9h7cQ3UMJaA.
>
> I think for some of the more complex streets some grouping concept is
needed whether via areas or relations. The latter is probably a bit
fragile, the former is at least somewhat consistent with man_made=bridge
and previous 'junction area' proposals.

There was a talk along these lines at SOTM recently, although I'm not so
keen on the areas overlapping:
https://media.ccc.de/v/sotm2019-1038-is-the-osm-data-model-creaking-#t=1263

For the primary way mapping I don't think we should be splitting ways
according to direction unless there is a barrier or median.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

[Tagging] Cycling relation misuse

[John Willis via Tagging]

Questions about using cycle relations properly:

I am mapping and repairing cycle roads in the Kanto/Tokyo area. There are a lot 
of designated cycling roads that follow a long rivers and other water features 
out into the countryside, making up a regional system, and a lot of smaller 
local cycling roads (also along small rivers) that connect neighborhoods and 
towns together. 
example: https://www.openstreetmap.org/relation/3218181 


I’m working to get all the individual ways of the cycle roads into relations 
and to properly classify these (local/regional, etc). 

But on the cycling layer of OSM, I find regular roads labeled as cycle routes: 
mountain roads where professional cyclists like to exercise labeled as a 
“cycling route”, which seems like “mapping for the renderer”.

example: https://www.openstreetmap.org/relation/8066243 


- They don’t seem to be cycling roads - all the relation members are trunk 
roads or similar - no cycleways whatsoever. 

-they are dangerous routes with no side-paths, sidewalks, or dedicated cycle 
lanes - just regular roads.

- they are exercise loops or hill climbs for pro cyclistsand serve no purpose 
for travelers or commuters.

- they are not, AFAIK, part of an official “cycling network”. The 
Super-relation someone has added all cycle routes to ( 関東地方サイクリングロード・ネットワーク ). 
https://www.openstreetmap.org/relation/8051094 
  also seems to be made-up and 
not official either - the name only returns one result (the OSM data page) when 
searched.  



To me, these non-cycle routes are just garbage relations meant to have the 
route show up on the cycling view of OSM for people doing workouts. 

I want to delete these fake “mountain workout” relations that should be mapped 
in strava or a similar workout app. 

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