Re: [Tagging] Tram tracks running in a road

[Jo]

Hi Luca,

Thank you for letting me know. That was indeed an oversight. I fixed it by
removing the tags from the little highway part, then combining that way
with the railway=tram. There are no bus lines that are passing that part,
so it's enough to have it as railway tram. It took many hours to
incorporate those tracks, even though we have them as open data from UrbIS.
They have the rails though and we want to combine those rails into tracks.

If you want to see high resolution imagery for Brussels, the
AGIV(laanderen) is available in JOSM's list of WMS sources.

Jo


2015-02-09 8:57 GMT+01:00 Luca Sigfrido Percich :

> Sorry I meant the railway tag. just added a note on the map
>
> 2015-02-09 8:54 GMT+01:00 Luca Sigfrido Percich :
>
>> Hi Jo,
>>
>> I was looking closely at your example, and noticed that maybe the highway
>> tag is missing from this way: http://www.openstreetmap.org/way/40283536
>> because the tram line looks interrupted there.
>>
>> Sig
>>
>> 2015-02-07 1:12 GMT+01:00 Jo :
>>
>>> The reason to use separate ways for trams can be seen in the other tram
>>> tracks I mapped:
>>>
>>> http://www.openstreetmap.org/#map=18/50.83181/4.33280
>>>
>>> You can clearly see that very often the rails don't follow the asphalt
>>> where the cars drive. Cars can make 90 degree turns, the tram rails need to
>>> follow smoother curves.
>>>
>>> So it's only in situations where buses follow the tram bedding:
>>>
>>> http://www.openstreetmap.org/#map=19/50.82432/4.33561
>>>
>>> that one can have common ways for the tram and that service road.
>>>
>>> * For normal streets we draw 2 ways for the tracks and 1 for the asphalt
>>> road.
>>> * For dual carriageways we draw 2 ways for the tracks and 2 for each
>>> side + sometimes a service way between the tracks, when the buses use it
>>> too.
>>>
>>> It's rather exceptional that the service road and the tram rails can use
>>> the same OSM way.
>>>
>>> Keep in mind it's only a model to represent reality. A model which uses
>>> lines for what in reality are areas, so whatever we do, it will never be a
>>> perfect fit.
>>>
>>>
>>> I'm sure André won't agree with me, but to implement the solution he
>>> proposes, we'd have to restart OSM from scratch. And even though it may
>>> simplify and solve some things, it would make other stuff a lot harder.
>>>
>>> Jo
>>>
>>> 2015-02-07 0:31 GMT+01:00 Janko Mihelić :
>>>
 2015-02-06 17:29 GMT+01:00 Luca Sigfrido Percich <
 luca.perc...@gmail.com>:

>
> We could also user a lanes modifier:
> lanes=3
> lanes:backward=2
> tram:lanes:backward=yes|no
> tram:forward=yes
>
>
 I think this is the best way to tag this. There's a great map paint
 style for seeing roads in towns in JOSM, and it helps a lot with tagging
 lanes. It's called Lanes and road attributes. Unfortunately, it doesn't
 show trams, but if we start tagging them, it will probably start rendering
 them. Right now, I use psv:lanes:forward=designated|no, because psv means
 all public service vehicles.

 http://wiki.openstreetmap.org/wiki/Key:lanes:psv

 And in my town those lanes are reserved for trams, buses and taxis.

 Janko Mihelić

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


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

Re: [Tagging] Tram tracks running in a road

[Jo]

I like to have those tram tracks as long as possible. So I tend to only
split them for the switches, where the route relations need it, or where
the combination with highway tags 'needs' it.

https://www.openstreetmap.org/way/40283768/history

2015-02-09 9:13 GMT+01:00 Jo :

> Hi Luca,
>
> Thank you for letting me know. That was indeed an oversight. I fixed it by
> removing the tags from the little highway part, then combining that way
> with the railway=tram. There are no bus lines that are passing that part,
> so it's enough to have it as railway tram. It took many hours to
> incorporate those tracks, even though we have them as open data from UrbIS.
> They have the rails though and we want to combine those rails into tracks.
>
> If you want to see high resolution imagery for Brussels, the
> AGIV(laanderen) is available in JOSM's list of WMS sources.
>
> Jo
>
>
> 2015-02-09 8:57 GMT+01:00 Luca Sigfrido Percich :
>
>> Sorry I meant the railway tag. just added a note on the map
>>
>> 2015-02-09 8:54 GMT+01:00 Luca Sigfrido Percich :
>>
>>> Hi Jo,
>>>
>>> I was looking closely at your example, and noticed that maybe the
>>> highway tag is missing from this way:
>>> http://www.openstreetmap.org/way/40283536 because the tram line looks
>>> interrupted there.
>>>
>>> Sig
>>>
>>> 2015-02-07 1:12 GMT+01:00 Jo :
>>>
 The reason to use separate ways for trams can be seen in the other tram
 tracks I mapped:

 http://www.openstreetmap.org/#map=18/50.83181/4.33280

 You can clearly see that very often the rails don't follow the asphalt
 where the cars drive. Cars can make 90 degree turns, the tram rails need to
 follow smoother curves.

 So it's only in situations where buses follow the tram bedding:

 http://www.openstreetmap.org/#map=19/50.82432/4.33561

 that one can have common ways for the tram and that service road.

 * For normal streets we draw 2 ways for the tracks and 1 for the
 asphalt road.
 * For dual carriageways we draw 2 ways for the tracks and 2 for each
 side + sometimes a service way between the tracks, when the buses use it
 too.

 It's rather exceptional that the service road and the tram rails can
 use the same OSM way.

 Keep in mind it's only a model to represent reality. A model which uses
 lines for what in reality are areas, so whatever we do, it will never be a
 perfect fit.


 I'm sure André won't agree with me, but to implement the solution he
 proposes, we'd have to restart OSM from scratch. And even though it may
 simplify and solve some things, it would make other stuff a lot harder.

 Jo

 2015-02-07 0:31 GMT+01:00 Janko Mihelić :

> 2015-02-06 17:29 GMT+01:00 Luca Sigfrido Percich <
> luca.perc...@gmail.com>:
>
>>
>> We could also user a lanes modifier:
>> lanes=3
>> lanes:backward=2
>> tram:lanes:backward=yes|no
>> tram:forward=yes
>>
>>
> I think this is the best way to tag this. There's a great map paint
> style for seeing roads in towns in JOSM, and it helps a lot with tagging
> lanes. It's called Lanes and road attributes. Unfortunately, it doesn't
> show trams, but if we start tagging them, it will probably start rendering
> them. Right now, I use psv:lanes:forward=designated|no, because psv means
> all public service vehicles.
>
> http://wiki.openstreetmap.org/wiki/Key:lanes:psv
>
> And in my town those lanes are reserved for trams, buses and taxis.
>
> Janko Mihelić
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
>

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


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

Re: [Tagging] Access restrictions for shoulder lanes?

[Paul Johnson]

On Fri, Feb 6, 2015 at 7:00 AM, Bryce Nesbitt  wrote:

> In the USA occasional sections of even Interstate highways are open to
> bicycles,
> where no equivalent route exists. There's some argument to tag these as
> bike paths to avoid the tag soup of lanes,
> and ensure the (unusual) situation is perfectly clear.
>

This is not a sensible assumption and I'm frankly getting a little sick of
having to mythbust this every few weeks just because 98% of America happens
to live next to the few, largely urban, exceptions to the norm.  It's not
anywhere as rare as you make it out to be.  The Federal Highway
Administration indicates that the default for any way in the US unless
otherwise locally defined, even freeways, is bicycle=yes
.
Bicycles even qualify for the HOV lane unless it presents a hazard.

Even in car-centric regions like California, the number of freeway miles
that ban bikes is greatly overshadowed by the vast majority of miles that
go by the default of allowing bicycles.  In my experience, with a few
goofball exceptions largely in the midwest (such as, say, various sections
of US 75, US 412 and (to a far lesser extent) I 40 in Oklahoma that have a
minimum speed limit, yet is the only sensible route and in some cases the
only physically possible route, and thus the ban is both routinely ignored
and rarely enforced for the same reason it isn't enforced on farm equipment
(which poses a far greater hazard as this equipment often spans multiple
lanes) either; however I do try to tag anything that isn't a bike route and
has a minimum speed limit as bicycle=no per Oklahoma's legal idiosyncrasy),
there's very few segments except for the most urban settings where a ban is
even a sensible suggestion in the first place.  Wyoming could be retagged
right now if it isn't already:  There's not one spot in the freeway system
in that state that bans bicycles
.

Can we finally bury this myth that bicycle=no is somehow even remotely the
norm for American freeways?  That said, regardless of the restriction, it's
a good idea to tag bicycle=* and foot=* explicitly on trunk and motorway
routes as there still seems to be widespread misconception on this and
could interfere with ideal routing if excluded.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Access restrictions for shoulder lanes?

[Paul Johnson]

On Fri, Feb 6, 2015 at 7:21 AM, fly  wrote:

> Am 06.02.2015 um 14:00 schrieb Bryce Nesbitt:
> > In the USA occasional sections of even Interstate highways are open to
> > bicycles,
> > where no equivalent route exists. There's some argument to tag these as
> > bike paths to avoid the tag soup of lanes,
> > and ensure the (unusual) situation is perfectly clear.
>
> Why is bicycle=yes and motorroad=no not enough ?
>


Well, the concept of a motorroad, to the best of my knowledge, isn't a
concept that exists legally anywhere in the US and I doubt very many people
in the US (save for people on this list or who have some windshield time in
the EU, and possibly not even then) even know what a motorroad is; this
would be a misapplication of the tag.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Access restrictions for shoulder lanes?

[Martin Koppenhoefer]

2015-02-02 16:03 GMT+01:00 AYTOUN RALPH :

> The highway=motorway already implies that there is two or more lanes plus
> an emergency hard shoulder



no, highway=motorway implies that it is a motorway. These can also have
just one lane (exceptional case) and do not need to have a hard shoulder
(also more or less exceptional case).

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

Re: [Tagging] Access restrictions for shoulder lanes?

[Martin Koppenhoefer]

2015-02-02 18:06 GMT+01:00 Paul Johnson :

> parking:lane=emergency seems like a good value.




emergency stops are typically not considered "parking". I don't think its a
good idea to use something with "parking" for emergency lanes tagging.

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

Re: [Tagging] Tram tracks running in a road

[Martin Koppenhoefer]

2015-02-07 1:12 GMT+01:00 Jo :

> Keep in mind it's only a model to represent reality. A model which uses
> lines for what in reality are areas, so whatever we do, it will never be a
> perfect fit.
>


+1, why not draw area:highway=* for the road areas to model that the trams
run on the road and not separately?

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

Re: [Tagging] Tram tracks running in a road

[Martin Koppenhoefer]

2015-02-08 22:32 GMT+01:00 Jo :

> is it one asphalt way with one track? Then I agree. Or is it one asphalt
> way with two tracks, one for each direction of the tram lines? Then I'd
> draw 3 ways, 2 for the tracks, and 1 for the highway.
>


+1
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Tram tracks running in a road

[Martin Koppenhoefer]

2015-02-09 8:29 GMT+01:00 Markus Lindholm :

> The road isn't between the tracks.
>


you could understand this by looking at the width of the road.



> In my opinion it's better
> to have two ways, one in each direction with highway and railway tags
> on both.
>


-1
if the highway isn't a dual carriageway, you shouldn't have 2 ways to
represent it. Having highway and railway tags on the same object might also
lead to other problems: you won't always know which tag / attribute belongs
to which entity, or in other words it will only work if all properties of
the railway and the highway are the same, e.g. oneway, width, surface.

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

Re: [Tagging] Tram tracks running in a road

[Martin Vonwald]

Just one quick question: are there any tags describing the track bed and
ties?

2015-02-07 11:19 GMT+01:00 Martin Vonwald :

> 2015-02-07 0:31 GMT+01:00 Janko Mihelić :
>
>> I think this is the best way to tag this. There's a great map paint style
>> for seeing roads in towns in JOSM, and it helps a lot with tagging lanes.
>> It's called Lanes and road attributes. Unfortunately, it doesn't show
>> trams, but if we start tagging them, it will probably start rendering them.
>>
>
> Let me know if there's a place with a lot of such tags and I try to update
> the style.
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Tram tracks running in a road

[Janko Mihelić]

2015-02-09 13:30 GMT+01:00 Martin Vonwald :

> Just one quick question: are there any tags describing the track bed and
> ties?
>

Here is a huge list of railway tags:

http://wiki.openstreetmap.org/wiki/OpenRailwayMap/Tagging

There is embedded=yes/no/pavement/wood/metal/plastic which tells if the
rails are embedded in the street. I would say that a railroad cannot be a
part of the street if it has embedded=no.

I couldn't find a tag for describing railroad ties, which is strange.

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

Re: [Tagging] Access restrictions for shoulder lanes?

[John F. Eldredge]

On February 9, 2015 4:32:36 AM CST, Paul Johnson  wrote:
> On Fri, Feb 6, 2015 at 7:00 AM, Bryce Nesbitt 
> wrote:
> 
> > In the USA occasional sections of even Interstate highways are open
> to
> > bicycles,
> > where no equivalent route exists. There's some argument to tag these
> as
> > bike paths to avoid the tag soup of lanes,
> > and ensure the (unusual) situation is perfectly clear.
> >
> 
> This is not a sensible assumption and I'm frankly getting a little
> sick of
> having to mythbust this every few weeks just because 98% of America
> happens
> to live next to the few, largely urban, exceptions to the norm.  It's
> not
> anywhere as rare as you make it out to be.  The Federal Highway
> Administration indicates that the default for any way in the US unless
> otherwise locally defined, even freeways, is bicycle=yes
> .
> Bicycles even qualify for the HOV lane unless it presents a hazard.
> 
> Even in car-centric regions like California, the number of freeway
> miles
> that ban bikes is greatly overshadowed by the vast majority of miles
> that
> go by the default of allowing bicycles.  In my experience, with a few
> goofball exceptions largely in the midwest (such as, say, various
> sections
> of US 75, US 412 and (to a far lesser extent) I 40 in Oklahoma that
> have a
> minimum speed limit, yet is the only sensible route and in some cases
> the
> only physically possible route, and thus the ban is both routinely
> ignored
> and rarely enforced for the same reason it isn't enforced on farm
> equipment
> (which poses a far greater hazard as this equipment often spans
> multiple
> lanes) either; however I do try to tag anything that isn't a bike
> route and
> has a minimum speed limit as bicycle=no per Oklahoma's legal
> idiosyncrasy),
> there's very few segments except for the most urban settings where a
> ban is
> even a sensible suggestion in the first place.  Wyoming could be
> retagged
> right now if it isn't already:  There's not one spot in the freeway
> system
> in that state that bans bicycles
> .
> 
> Can we finally bury this myth that bicycle=no is somehow even remotely
> the
> norm for American freeways?  That said, regardless of the restriction,
> it's
> a good idea to tag bicycle=* and foot=* explicitly on trunk and
> motorway
> routes as there still seems to be widespread misconception on this and
> could interfere with ideal routing if excluded.
> 
> 
> 
> 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

According to , 
14 US states out of 50 allow bicycle use on Interstate highways. Wyoming is the 
only state that allows it state-wide.  So, 72% of US states don't allow 
bicycles on Interstates. I don't have the percentages by highway mile 
available, but I would call a 72% majority the norm.

-- 
John F. Eldredge -- j...@jfeldredge.com
"Darkness cannot drive out darkness: only light can do that. Hate cannot drive 
out hate: only love can do that." -- Martin Luther King, Jr.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Access restrictions for shoulder lanes?

[Paul Johnson]

Wikipedia seems to be incomplete on this; I'm presently unaware of any
state that has a statewide prohibition.

On Mon, Feb 9, 2015 at 7:22 AM, John F. Eldredge 
wrote:

> On February 9, 2015 4:32:36 AM CST, Paul Johnson 
> wrote:
>
>> On Fri, Feb 6, 2015 at 7:00 AM, Bryce Nesbitt 
>> wrote:
>>
>>> In the USA occasional sections of even Interstate highways are open to
>>> bicycles,
>>> where no equivalent route exists. There's some argument to tag these as
>>> bike paths to avoid the tag soup of lanes,
>>> and ensure the (unusual) situation is perfectly clear.
>>>
>>
>> This is not a sensible assumption and I'm frankly getting a little sick
>> of having to mythbust this every few weeks just because 98% of America
>> happens to live next to the few, largely urban, exceptions to the norm.
>> It's not anywhere as rare as you make it out to be.  The Federal Highway
>> Administration indicates that the default for any way in the US unless
>> otherwise locally defined, even freeways, is bicycle=yes
>> .
>> Bicycles even qualify for the HOV lane unless it presents a hazard.
>>
>> Even in car-centric regions like California, the number of freeway miles
>> that ban bikes is greatly overshadowed by the vast majority of miles that
>> go by the default of allowing bicycles.  In my experience, with a few
>> goofball exceptions largely in the midwest (such as, say, various sections
>> of US 75, US 412 and (to a far lesser extent) I 40 in Oklahoma that have a
>> minimum speed limit, yet is the only sensible route and in some cases the
>> only physically possible route, and thus the ban is both routinely ignored
>> and rarely enforced for the same reason it isn't enforced on farm equipment
>> (which poses a far greater hazard as this equipment often spans multiple
>> lanes) either; however I do try to tag anything that isn't a bike route and
>> has a minimum speed limit as bicycle=no per Oklahoma's legal idiosyncrasy),
>> there's very few segments except for the most urban settings where a ban is
>> even a sensible suggestion in the first place.  Wyoming could be retagged
>> right now if it isn't already:  There's not one spot in the freeway
>> system in that state that bans bicycles
>> .
>>
>> Can we finally bury this myth that bicycle=no is somehow even remotely
>> the norm for American freeways?  That said, regardless of the restriction,
>> it's a good idea to tag bicycle=* and foot=* explicitly on trunk and
>> motorway routes as there still seems to be widespread misconception on this
>> and could interfere with ideal routing if excluded.
>>
>> --
>>
>> Tagging mailing list
>> Tagging@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/tagging
>>
>>
> According to <
> http://en.wikipedia.org/wiki/Non-motorized_access_on_freeways>;, 14 US
> states out of 50 allow bicycle use on Interstate highways. Wyoming is the
> only state that allows it state-wide. So, 72% of US states don't allow
> bicycles on Interstates. I don't have the percentages by highway mile
> available, but I would call a 72% majority the norm.
>
> --
> John F. Eldredge -- j...@jfeldredge.com
> "Darkness cannot drive out darkness: only light can do that. Hate cannot
> drive out hate: only love can do that." -- Martin Luther King, Jr.
>
> ___
> 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] Tram tracks running in a road

[fly]

Am 08.02.2015 um 19:45 schrieb Janko Mihelić:
> 2015-02-08 17:48 GMT+01:00 fly :
>> Am 07.02.2015 um 11:19 schrieb Martin Vonwald:
>>> 2015-02-07 0:31 GMT+01:00 Janko Mihelić :
 2015-02-06 17:29 GMT+01:00 Luca Sigfrido Percich :
>
> We could also user a lanes modifier:
> lanes=3
> lanes:backward=2
> tram:lanes:backward=yes|no
> tram:forward=yes

To get it clear, I only use these tags as addition on highway=* tagged
with lanes:-Taggingsystem. There is always a separate way with railway=*
+ track=1 for each track.

>> Actually, I use an even more general approach:
>> railway:forward=tram
>> railway:lanes:backward=tram|no
>>
>> together with access I also use train
>> access:lanes:backward=no|yes
>> train:lanes:backward=designated|no
>>
> 
> I don't understand why you would use railway:forward=* and
> railway:lanes:forward=*. Aren't those two redundant?

Yes, the first is used for lanes:forward=1 while the second one for
values greater than 1.

Please read carefully, I used forward and backward and followed the
quoted example.

> Also, why train and not tram?

tram would work in my case but I know also rails used by freight train
which are embedded into the road.

I see train as a more general access tag compared to tram, cable_car,
light_train etc.

> I like railway:lanes:forward/backward=* because sometimes you can have
> rails on the street which are not used by anything, so using tram/train
> doesn't make sense. But if you say tram:lanes:forward/backward=* than that
> implies that there are rails there, so no other tags are needed anymore.

I would always use railway:forward/backward= for
lanes:forward/backward=1 but won't insist on that one.

For more than one lane in one directions I use
railway:lanes:forward/backward=* as I already mentioned above this is
not only tram specific.

train or tram are access tags which I only use if I need them usually
only in cases of access=no or access:lanes:forward/backward=no|no|yes or
similar.

Forgot to send some links the other day:

[1] https://www.openstreetmap.org/#map=18/47.99472/7.85038
[2] https://www.openstreetmap.org/#map=19/47.98524/7.83145

cu fly

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

Re: [Tagging] Tram tracks running in a road

[Daniel Koć]

W dniu 2015-02-09 o 12:50, Martin Koppenhoefer pisze:


2015-02-07 1:12 GMT+01:00 Jo >:


Keep in mind it's only a model to represent reality. A model which
uses lines for what in reality are areas, so whatever we do, it
will never be a perfect fit.



+1, why not draw area:highway=* for the road areas to model that the 
trams run on the road and not separately?




BTW - I currently try to resurrect the idea of rendering street areas, 
feel free to contribute here:


https://github.com/gravitystorm/openstreetmap-carto/issues/180

We have much more examples where it would help us stay closer to the 
reality.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] courtyards

[John Sturdy]

Here's another variation: the courtyard of Limerick's Milk Market:
http://www.openstreetmap.org/way/115572313


It was originally open at the top, but now has a canopy that covers
most of it; and it's not a leisure facility.


__John

On Mon, Feb 9, 2015 at 7:42 AM, Friedrich Volkmann  wrote:
> On 08.02.2015 22:17, Warin wrote:
>
>>> >From a technical point of view they are typically associated with fire 
>>> >protection (way to leave the building, access for firefighters),
>>
>> If the courtyard is fully enclosed by buildings or by one building .. they
>> are not part of a fire escape (protection), those require exit to an open
>> area - not one that is fully enclosed. So the use as fire protection will
>> depend on  the courtyard. And my thinking is that a true 'courtyard' is
>> fully enclosed?
>
> We need to be able to map partially enclosed courtyards as well, e.g.:
> https://www.openstreetmap.org/#map=19/48.17839/16.34189
> (The courtyards are named Hof 1 ... Hof 7.)
>
> But I agree that a courtyard *typically* is fully enclosed by buildings,
> thus not an emergency feature. There's an approved tag entrance=emergency
> for emergency exits, and I'd suggest a tag like emergency=access for spots
> and alleys designed to be accessible for fire fighters.
>
> I think that, from a technical point view, the main function of a courtyard
> is to yield sunlight to building rooms that are not adjacent to the
> building's outer margin. All other uses, such as recreation, parking or
> emergency access, are subsequent.
>
> --
> Friedrich K. Volkmann   http://www.volki.at/
> Adr.: Davidgasse 76-80/14/10, 1100 Wien, Austria
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

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

[Tagging] Feature Proposal - Voting - traffic_signals (Lukas Schaus)

[Lukas Schaus]

Hi,

My proposal concerning the modelling of traffic signals is now open  
for voting.


Thank you


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

[Tagging] Feature Proposal - Voting - traffic_signals (Lukas Schaus)

[Lukas Schaus]

Hi,

My proposal concerning the modelling of traffic signals is now open
for voting.

http://wiki.openstreetmap.org/wiki/Proposed_features/Traffic_Signals#Voting

Thank you




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

Re: [Tagging] courtyards

[Stephen Gower]

On Fri, Feb 06, 2015 at 11:14:17PM +0100, Friedrich Volkmann wrote:
> Courtyards use to be mapped as "inner" members of building multipolygons. We
> can also use the multipolygon relation to assign a name to the bullding. If
> we want to assign a name to the courtyard, we must assign it to the way. But
> then we need some kind of physical tag in addition. Applications won't know
> what do do with the name when there aren't any other tags.

Here in Oxford (where we have many examples of named quadrangles/courtyards)
I see examples where they are tagged as highway=footway areas (e.g.
http://www.openstreetmap.org/way/301895528 ) but more often the central
section of lawn has been named (e.g. http://www.openstreetmap.org/way/228244550 
)

In reality, it is neither the paving or the lawn that is the named feature,
it's the architectual feature containing these and itself bounded by the
buildings (although in the case of cloistered courtyards, the covered
arcades around the edge are arguably both part of the building and the
courtyard). I support creation of a tag for more consistantly marking these
named features, but I have no idea where in the tagging structure it is
best placed (building/landuse/amenity/etc all have their problems).

S

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

Re: [Tagging] Feature Proposal - Voting - traffic_signals (Lukas Schaus)

[fly]

Am 09.02.2015 um 15:35 schrieb Lukas Schaus:
> Hi,
> 
> My proposal concerning the modelling of traffic signals is now open
> for voting.
> 
> http://wiki.openstreetmap.org/wiki/Proposed_features/Traffic_Signals#Voting

You did not comment on my question about micromapping a junction and
adding a highway=traffic_signal at the pedestrian crossing or the stop
line for each direction separately. Have a look at the examples below
[1],[2].

For complete direction separated junctions like [3] we probable will not
even need any relation.

Please, consider this tagging style and show me how this will work
together with your proposal.

Altogether, I am not sure if this relation is needed at all but for sure
not at the current base.

Still would prefer simple tags on the nodes if possible.

cu fly


[1] https://www.openstreetmap.org/#map=19/48.10739/7.85080
[2] https://www.openstreetmap.org/#map=18/48.06123/7.81258
[3] https://www.openstreetmap.org/#map=19/47.98530/7.82814

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

Re: [Tagging] Feature Proposal - Voting - traffic_signals (Lukas Schaus)

[Lukas Sommer]

I would strongly recommend to not use type=traffic_signals because we
have already a tag highway=traffic_signals. This would cause
confusion. Furthermore, type=traffic_signals does not describe well
what your prososal wants to do. With your relation, you don’t want to
represent traffic signals but you want to represent traffic signal
phases. Something like type=traffic_signals_phase would IMHO be
appropriate.

2015-02-09 15:47 GMT, fly :
> Am 09.02.2015 um 15:35 schrieb Lukas Schaus:
>> Hi,
>>
>> My proposal concerning the modelling of traffic signals is now open
>> for voting.
>>
>> http://wiki.openstreetmap.org/wiki/Proposed_features/Traffic_Signals#Voting
>
> You did not comment on my question about micromapping a junction and
> adding a highway=traffic_signal at the pedestrian crossing or the stop
> line for each direction separately. Have a look at the examples below
> [1],[2].
>
> For complete direction separated junctions like [3] we probable will not
> even need any relation.
>
> Please, consider this tagging style and show me how this will work
> together with your proposal.
>
> Altogether, I am not sure if this relation is needed at all but for sure
> not at the current base.
>
> Still would prefer simple tags on the nodes if possible.
>
> cu fly
>
>
> [1] https://www.openstreetmap.org/#map=19/48.10739/7.85080
> [2] https://www.openstreetmap.org/#map=18/48.06123/7.81258
> [3] https://www.openstreetmap.org/#map=19/47.98530/7.82814
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>


-- 
Lukas Sommer

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

Re: [Tagging] Tram tracks running in a road

[Markus Lindholm]

On 9 February 2015 at 12:58, Martin Koppenhoefer  wrote:
> 2015-02-09 8:29 GMT+01:00 Markus Lindholm :
>> The road isn't between the tracks.
> you could understand this by looking at the width of the road.

Doesn't seem to be an ideal solution to draw the objects in a way that
differs from ground truth and trying to use tags to convey that they
actually on top of each other.

> Having highway and railway tags on the same object might also
> lead to other problems: you won't always know which tag / attribute belongs
> to which entity, or in other words it will only work if all properties of
> the railway and the highway are the same, e.g. oneway, width, surface.

Not really a problem as you could always qualify the tag if there's a
risk of ambiguity.

/Markus

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

Re: [Tagging] Tram tracks running in a road

[Jo]

As I said before, it's simply a model. We have always drawn 1 way for each
stretch of asphalt. We also draw 1 way per track. They sometimes overlap,
but it turns out that this doesn't happen all that often.

Maybe this is the reason why they decided to draw rails instead of tracks
in the UrbIS dataset. But that's a different model. Probably makes more
sense in a GIS system.

Jo

2015-02-09 18:56 GMT+01:00 Markus Lindholm :

> On 9 February 2015 at 12:58, Martin Koppenhoefer 
> wrote:
> > 2015-02-09 8:29 GMT+01:00 Markus Lindholm :
> >> The road isn't between the tracks.
> > you could understand this by looking at the width of the road.
>
> Doesn't seem to be an ideal solution to draw the objects in a way that
> differs from ground truth and trying to use tags to convey that they
> actually on top of each other.
>
> > Having highway and railway tags on the same object might also
> > lead to other problems: you won't always know which tag / attribute
> belongs
> > to which entity, or in other words it will only work if all properties of
> > the railway and the highway are the same, e.g. oneway, width, surface.
>
> Not really a problem as you could always qualify the tag if there's a
> risk of ambiguity.
>
> /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] Feature Proposal - Voting - traffic_signals (Lukas Schaus)

[Lukas Sommer]

I know that I’m a little late with this comment – I missed this while
reading the proposal. Sorry.

Maybe that’s something that can be changed in the prososal – if
current voters agree?

2015-02-09 17:29 GMT, Lukas Sommer :
> I would strongly recommend to not use type=traffic_signals because we

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