lanes = 0

classic Classic list List threaded Threaded
63 messages Options
1234
Reply | Threaded
Open this post in threaded view
|

Re: lanes = 0

yo paseopor
In Spain is easy: when there is no marks =  lanes=1
The lack of the mark lanes is the reason why when a Spaniard drives by Rome thinks Italian people are crazy, because they overtake you in the same big lane, but ONLY one lane (lane without marks). One lane= one car. 
lane with no marks is =1 (driver's school book says that) so lane=0 would be impossible and ununderstable for a Spaniard wherever in the World.

Cheers 
Yopaseopor



On Sat, Jun 15, 2019 at 6:59 PM Martin Koppenhoefer <[hidden email]> wrote:


sent from a phone

> On 15. Jun 2019, at 01:10, Joseph Eisenberg <[hidden email]> wrote:
>
> This requirement is fine for Europe, but the presence of lane markings
> is not reliable in all of the world.
>
> In developing countries, such as here in Indonesia, the presence of
> painted lane markings is inconsistent. Often cheap pain is used
> instead of more durable thermoplastic, so the markings only last a
> year. After that the road still functions the same, even though the
> markings are no longer visible.
>
> There are also sections of primary or trunk road that are at least 6
> or 7 meters wide and freshly painted, but have not yet been marked and
> may not be for a number of years. I tag these as lanes=2 because the
> road is clearly wide enough for two lanes.
>
> And here in town the main road was recently marked with 2 lanes in
> each direction, but before it already functioned as 4 lanes because
> the width was sufficient.
>
> While tagging the width is useful, I believe tagging the presence of
> "de facto" lanes is reasonable in developing countries and places
> where painted lane markings are not frequently used.



This description is a perfect fit for the situation in central Italy as well, not having marked lanes can happen on 2+2 roads for years and for many kilometers. Often there are lane markings for some part of the road while they are missing on others. Generally they are aiming at having lanes, but it isn’t pursued with high priority ;-)
I can understand the argument that lanes have to be painted in order to be there, but it isn’t the reality I am observing.

We shouldn’t dismiss lane_markings=no as it can solve both cases: no lanes marked but lanes=n is set, and no lanes tag set (confirmation the tag wasn’t forgotten).

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

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

Re: lanes = 0

Warin
In reply to this post by Tobias Zwick
On 16/06/19 22:53, Tobias Zwick wrote:

> Okay, to wrap this up, I added this title in the wiki and referenced back to this discussion, advising to not use lanes=0/1.5/none to signify no lane markings but instead use something like lane_markings=no.
>
> https://wiki.openstreetmap.org/wiki/Key:lanes#No_lane_markings
>
> ---
>
> Additionally, I noted that after a similar discussion about lanes=1.5 in the German forum in 2017, the wiki page was changed to stress that the lanes-key is for *marked* traffic lanes. The change was announced on the Talk page and the German forum discussion linked there.
>
> I did not change the formulation back but only added the outcome of this discussion to that topic on the Talk page because I do not feel legitimated to do that as the 2017 wiki change was also done only after discussion in the community, same as now:
> https://wiki.openstreetmap.org/wiki/Talk:Key:lanes#No_centerline_-_one_or_two_lanes.3F

The community that decided that lanes must be marked in geography small and probably failed to consider the rest of the world.
Fine for them to set 'rules' locally but that looks to be causing problems in other parts of the world.
I have made comment on the discussion page.
https://wiki.openstreetmap.org/wiki/Talk:Key:lanes#Marked_or_unmarked_lanes

>
> [1] https://forum.openstreetmap.org/viewtopic.php?pid=627975#p627975
>
> On 15/06/2019 18:55, Martin Koppenhoefer wrote:
>>
>> sent from a phone
>>
>>> On 15. Jun 2019, at 01:10, Joseph Eisenberg <[hidden email]> wrote:
>>>
>>> This requirement is fine for Europe, but the presence of lane markings
>>> is not reliable in all of the world.
>>>
>>> In developing countries, such as here in Indonesia, the presence of
>>> painted lane markings is inconsistent. Often cheap pain is used
>>> instead of more durable thermoplastic, so the markings only last a
>>> year. After that the road still functions the same, even though the
>>> markings are no longer visible.
>>>
>>> There are also sections of primary or trunk road that are at least 6
>>> or 7 meters wide and freshly painted, but have not yet been marked and
>>> may not be for a number of years. I tag these as lanes=2 because the
>>> road is clearly wide enough for two lanes.
>>>
>>> And here in town the main road was recently marked with 2 lanes in
>>> each direction, but before it already functioned as 4 lanes because
>>> the width was sufficient.
>>>
>>> While tagging the width is useful, I believe tagging the presence of
>>> "de facto" lanes is reasonable in developing countries and places
>>> where painted lane markings are not frequently used.
>>
>>
>> This description is a perfect fit for the situation in central Italy as well, not having marked lanes can happen on 2+2 roads for years and for many kilometers. Often there are lane markings for some part of the road while they are missing on others. Generally they are aiming at having lanes, but it isn’t pursued with high priority ;-)
>> I can understand the argument that lanes have to be painted in order to be there, but it isn’t the reality I am observing.
>>
>> We shouldn’t dismiss lane_markings=no as it can solve both cases: no lanes marked but lanes=n is set, and no lanes tag set (confirmation the tag wasn’t forgotten).
>>
>> Cheers, Martin
>> _______________________________________________
>>


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

Re: lanes = 0

Greg Troxel-2
In reply to this post by Joseph Eisenberg
Joseph Eisenberg <[hidden email]> writes:

> This requirement is fine for Europe, but the presence of lane markings
> is not reliable in all of the world.
>
> In developing countries, such as here in Indonesia, the presence of
> painted lane markings is inconsistent. Often cheap pain is used
> instead of more durable thermoplastic, so the markings only last a
> year. After that the road still functions the same, even though the
> markings are no longer visible.

It is not just about developing countries.  In my part of the US, there
are many roads whicha have either no paint at all, or have white lines
at the edges (so you can see where the edges are at night).   Almost all
of these roads are wide enough for two cars to pass comfortably, but not
really wider than that.  This seems really obviously one lane in each
direction, and everybody who drives here gets that.  There is a legal
requirement to stay on the right of the imaginary center lane (absent a
reason such as passing a pedestrian); you can be cited for "operating
left of center" entire reasonably on a two-cars-wide road with no
markings -- but that will only happen if you are left of center
egregiously or on a blind curve or rise.


So that's a long way of saying that "lane markings" should not be
required for lanes=N; it is enough to observe the local conventions.

I agree that a finer-grained tag that says if there are markings or not
is sensible.  But the most important thing is to describe how traffic
actually behaves.

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

Re: lanes = 0

Paul Johnson-3


On Fri, Jun 21, 2019 at 5:49 PM Greg Troxel <[hidden email]> wrote:
Joseph Eisenberg <[hidden email]> writes:

> This requirement is fine for Europe, but the presence of lane markings
> is not reliable in all of the world.
>
> In developing countries, such as here in Indonesia, the presence of
> painted lane markings is inconsistent. Often cheap pain is used
> instead of more durable thermoplastic, so the markings only last a
> year. After that the road still functions the same, even though the
> markings are no longer visible.

It is not just about developing countries.  In my part of the US, there
are many roads whicha have either no paint at all, or have white lines
at the edges (so you can see where the edges are at night).   Almost all
of these roads are wide enough for two cars to pass comfortably, but not
really wider than that.  This seems really obviously one lane in each
direction, and everybody who drives here gets that.  There is a legal
requirement to stay on the right of the imaginary center lane (absent a
reason such as passing a pedestrian); you can be cited for "operating
left of center" entire reasonably on a two-cars-wide road with no
markings -- but that will only happen if you are left of center
egregiously or on a blind curve or rise.


So that's a long way of saying that "lane markings" should not be
required for lanes=N; it is enough to observe the local conventions

 In that example, I think it'd be better to just tag width=* instead of lanes=*.

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

Re: lanes = 0

Nita S.
Without digging deeply into width, presumably it measured in meters ? I can see value there, but the width has to be an average, and the cleared area for passage of vehicles.

On Sun, Jun 23, 2019 at 3:12 PM Paul Johnson <[hidden email]> wrote:


On Fri, Jun 21, 2019 at 5:49 PM Greg Troxel <[hidden email]> wrote:
Joseph Eisenberg <[hidden email]> writes:

> This requirement is fine for Europe, but the presence of lane markings
> is not reliable in all of the world.
>
> In developing countries, such as here in Indonesia, the presence of
> painted lane markings is inconsistent. Often cheap pain is used
> instead of more durable thermoplastic, so the markings only last a
> year. After that the road still functions the same, even though the
> markings are no longer visible.

It is not just about developing countries.  In my part of the US, there
are many roads whicha have either no paint at all, or have white lines
at the edges (so you can see where the edges are at night).   Almost all
of these roads are wide enough for two cars to pass comfortably, but not
really wider than that.  This seems really obviously one lane in each
direction, and everybody who drives here gets that.  There is a legal
requirement to stay on the right of the imaginary center lane (absent a
reason such as passing a pedestrian); you can be cited for "operating
left of center" entire reasonably on a two-cars-wide road with no
markings -- but that will only happen if you are left of center
egregiously or on a blind curve or rise.


So that's a long way of saying that "lane markings" should not be
required for lanes=N; it is enough to observe the local conventions

 In that example, I think it'd be better to just tag width=* instead of lanes=*.
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

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

Re: lanes = 0

Warin
The width default unit is metres, see https://wiki.openstreetmap.org/wiki/Map_Features/Units for other units and how to state them.

For a road I would tag the minimum width - so any vehicle entering can pass along the entire length and not get stuck.



On 24/06/19 07:56, Nita S. wrote:
Without digging deeply into width, presumably it measured in meters ? I can see value there, but the width has to be an average, and the cleared area for passage of vehicles.

On Sun, Jun 23, 2019 at 3:12 PM Paul Johnson <[hidden email]> wrote:


On Fri, Jun 21, 2019 at 5:49 PM Greg Troxel <[hidden email]> wrote:
Joseph Eisenberg <[hidden email]> writes:

> This requirement is fine for Europe, but the presence of lane markings
> is not reliable in all of the world.
>
> In developing countries, such as here in Indonesia, the presence of
> painted lane markings is inconsistent. Often cheap pain is used
> instead of more durable thermoplastic, so the markings only last a
> year. After that the road still functions the same, even though the
> markings are no longer visible.

It is not just about developing countries.  In my part of the US, there
are many roads whicha have either no paint at all, or have white lines
at the edges (so you can see where the edges are at night).   Almost all
of these roads are wide enough for two cars to pass comfortably, but not
really wider than that.  This seems really obviously one lane in each
direction, and everybody who drives here gets that.  There is a legal
requirement to stay on the right of the imaginary center lane (absent a
reason such as passing a pedestrian); you can be cited for "operating
left of center" entire reasonably on a two-cars-wide road with no
markings -- but that will only happen if you are left of center
egregiously or on a blind curve or rise.


So that's a long way of saying that "lane markings" should not be
required for lanes=N; it is enough to observe the local conventions

 In that example, I think it'd be better to just tag width=* instead of lanes=*.
_______________________________________________



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

Re: lanes = 0

Greg Troxel-2
In reply to this post by Paul Johnson-3
Paul Johnson <[hidden email]> writes:

>  In that example, I think it'd be better to just tag width=* instead of
> lanes=*.

Perhaps, but then data consumers have to figure how how many cars are
supposed to be side by side.  That number really is local convention;
one road I use is really not wide enough for 2, but people always do it.
So I favor using the lanes tag to specify how many lanes are actually
typically in use.  Then with width one can get the average width of
those.

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

Re: lanes = 0

Paul Johnson-3
That so doesn't work in the real world, though.  Residential streets are typically about 35 feet wide curb to curb.  Is this lanes=4 or lanes=2?  Trick question, it's neither because parking is allowed on the curb on both sides and the middle portion isn't wide enough to allow to cars to pass side to side in the middle.

On Sun, Jun 23, 2019 at 8:02 PM Greg Troxel <[hidden email]> wrote:
Paul Johnson <[hidden email]> writes:

>  In that example, I think it'd be better to just tag width=* instead of
> lanes=*.

Perhaps, but then data consumers have to figure how how many cars are
supposed to be side by side.  That number really is local convention;
one road I use is really not wide enough for 2, but people always do it.
So I favor using the lanes tag to specify how many lanes are actually
typically in use.  Then with width one can get the average width of
those.

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

Re: lanes = 0

Allroads
So there are lanes and virtual lanes.
 
We must make a good distinction, I must be able to see immediately, whether I am dealing with a marked lane or a virtual lane, that has no marking.
 
Do not expect from a mapper, at a marked lane, also to set  marked = yes. (or else) to make the distinction.
See wikipedia and else, there all marked.
 
A two-way road without a marking in our country, does not have lanes!  (law). Although, you can pass each other. There, we could have also a new tagcombination! But not lanes=* , these are marked! (law)
To make a good distinction, it must be immediately clear.
 
What do you think of:
 
 
 
lanes: virtual = (number),   lanes that have no markings. Not a second tag needed.
 
The same method as there is used highway: virtual = pedestrian, to make a route line over a pedestrian area. Or over a field, a beach.
 
You could say, lanes are created in the UK, lanes are created in OSM, these lanes where written down as marked lanes, to use lanes=* for virtual lanes was a abuse of the tag lanes=* , if you do use it, you make the definition unclear and that should be avoided, there is a new tag needed.
Problem solved.
 
 
 
 
 
Quote: yo_paseopor
In Spain is easy: when there is no marks =  lanes=1
 
Also when they are a passable, two way road?
 
When there are no marking there are no lanes.
lanes=1, like on a highway link, is indicating one way, one direction.
 
A lot of lanes=1 are deleted in our country, because they are not a lane (rijstrook)(law).
 
 
Allroads.
 

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

Re: lanes = 0

Nita S.
Let's try looking at this from a different direction. All roads must have the capacity to convey one vehicle width. Some roads have formal lane markings ( in which case the number of lanes will be obvious) and other road types have informal (i.e. none) lane markings. On the latter, the number of lanes is somewhat variable, and may depend upon the road surface type, the types of vehicles approaching, and recent precipitation ( which could fill edges, and dynamically contract the width). 

So maybe trying to put a finite number of lanes on an informal road situation may be a long process that satisfies no one. On informal roads , width may be the only viable metric. 

On Mon, Jun 24, 2019 at 7:41 AM Allroads <[hidden email]> wrote:
So there are lanes and virtual lanes.
 
We must make a good distinction, I must be able to see immediately, whether I am dealing with a marked lane or a virtual lane, that has no marking.
 
Do not expect from a mapper, at a marked lane, also to set  marked = yes. (or else) to make the distinction.
See wikipedia and else, there all marked.
 
A two-way road without a marking in our country, does not have lanes!  (law). Although, you can pass each other. There, we could have also a new tagcombination! But not lanes=* , these are marked! (law)
To make a good distinction, it must be immediately clear.
 
What do you think of:
 
 
 
lanes: virtual = (number),   lanes that have no markings. Not a second tag needed.
 
The same method as there is used highway: virtual = pedestrian, to make a route line over a pedestrian area. Or over a field, a beach.
 
You could say, lanes are created in the UK, lanes are created in OSM, these lanes where written down as marked lanes, to use lanes=* for virtual lanes was a abuse of the tag lanes=* , if you do use it, you make the definition unclear and that should be avoided, there is a new tag needed.
Problem solved.
 
 
 
 
 
Quote: yo_paseopor
In Spain is easy: when there is no marks =  lanes=1
 
Also when they are a passable, two way road?
 
When there are no marking there are no lanes.
lanes=1, like on a highway link, is indicating one way, one direction.
 
A lot of lanes=1 are deleted in our country, because they are not a lane (rijstrook)(law).
 
 
Allroads.
 
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

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

Re: lanes = 0

yo paseopor
In reply to this post by Allroads
 Also when they are a passable, two way road?
BOE-020_Codigo_de_Trafico_y_Seguridad_Vial
Page 50
 Carril. Banda longitudinal en que puede estar subdividida la calzada, delimitada o no por marcas viales longitudinales, siempre que tenga una anchura suficiente para permitir la circulación de una fila de automóviles que no sean motocicletas.
 
 It is the place of the road , limited or not by marks , that has sufficient width to fit a row of motor vehicles (not motorcycles).
In a road, if you can fit a row of motor vehicles, then , in Spain,  it is a lane.

So...
lane=1
oneway=no
could be possible

Also if you have the width to fit a motor vehicle it is a lane so...

In a road lane=0 is not possible in Spain

Salut i carrils (Health and lanes)
yopaseopor

On Mon, Jun 24, 2019 at 1:41 PM Allroads <[hidden email]> wrote:
So there are lanes and virtual lanes.
 
We must make a good distinction, I must be able to see immediately, whether I am dealing with a marked lane or a virtual lane, that has no marking.
 
Do not expect from a mapper, at a marked lane, also to set  marked = yes. (or else) to make the distinction.
See wikipedia and else, there all marked.
 
A two-way road without a marking in our country, does not have lanes!  (law). Although, you can pass each other. There, we could have also a new tagcombination! But not lanes=* , these are marked! (law)
To make a good distinction, it must be immediately clear.
 
What do you think of:
 
 
 
lanes: virtual = (number),   lanes that have no markings. Not a second tag needed.
 
The same method as there is used highway: virtual = pedestrian, to make a route line over a pedestrian area. Or over a field, a beach.
 
You could say, lanes are created in the UK, lanes are created in OSM, these lanes where written down as marked lanes, to use lanes=* for virtual lanes was a abuse of the tag lanes=* , if you do use it, you make the definition unclear and that should be avoided, there is a new tag needed.
Problem solved.
 
 
 
 
 
Quote: yo_paseopor
In Spain is easy: when there is no marks =  lanes=1
 
Also when they are a passable, two way road?
 
When there are no marking there are no lanes.
lanes=1, like on a highway link, is indicating one way, one direction.
 
A lot of lanes=1 are deleted in our country, because they are not a lane (rijstrook)(law).
 
 
Allroads.
 
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

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

Re: lanes = 0

Joseph Eisenberg
On 6/24/19, yo paseopor <[hidden email]> wrote:

> BOE-020_Codigo_de_Trafico_y_Seguridad_Vial
> Page 50
>  Carril. Banda longitudinal en que puede estar subdividida la calzada,
> delimitada o no por marcas viales longitudinales, siempre que tenga una
> anchura suficiente para permitir la circulación de una fila de automóviles
> que no sean motocicletas.

Here's my translation, FWIW

"Lane. Lengthwise strip into which a roadway may be subdivided,
whether or not delineated by longitudinal road markings, as long as it
is wide enough to allow the flow of a queue of automobiles, other than
motorcycles."

So in Spain a lane is a concept of a linear part of a road which is
wide enough to be used by a line of automobiles, whether or not it is
marked by painted lines. I believe this is a common definition in many
other countries as well.

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

Re: lanes = 0

dieterdreist
In reply to this post by Nita S.


sent from a phone

> Am 24.06.2019 um 13:57 schrieb Nita S. <[hidden email]>:
>
> Some roads have formal lane markings ( in which case the number of lanes will be obvious) and other road types have informal (i.e. none) lane markings.


even with markings it is not always clear, I know places with unidirectional traffic and one marked lane but effectively space for 2 small cars (plus a shoulder which is sometimes used despite it is not legally possible) , so that there is frequent overtaking. Not comparable to one true lane.


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

Re: lanes = 0

Paul Johnson-3
In reply to this post by yo paseopor


On Mon, Jun 24, 2019 at 7:09 AM yo paseopor <[hidden email]> wrote:
 Also when they are a passable, two way road?
BOE-020_Codigo_de_Trafico_y_Seguridad_Vial
Page 50
 Carril. Banda longitudinal en que puede estar subdividida la calzada, delimitada o no por marcas viales longitudinales, siempre que tenga una anchura suficiente para permitir la circulación de una fila de automóviles que no sean motocicletas.
 
 It is the place of the road , limited or not by marks , that has sufficient width to fit a row of motor vehicles (not motorcycles).
In a road, if you can fit a row of motor vehicles, then , in Spain,  it is a lane.

So...
lane=1
oneway=no
could be possible

Yes, and that's actually already a thing.  It looks like this: https://en.wikipedia.org/wiki/Single-track_road  Extremely common in national forests in the US.  Somewhat unusual but not unheard of on particularly rural highways in the US and Canada (BC 17 between Victoria and Port Renfrew comes to mind offhand as having quite a few sections where it goes down to a single lane for both directions).

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

Re: lanes = 0

Richard Z.
In reply to this post by Warin
On Thu, Jun 13, 2019 at 12:59:27PM +1000, Warin wrote:

> Hi,
>
>
> There are a few uses of lanes=0... I would think these are errors. Even if
> unmarked a road would have at least one lane otherwise it is not really a
> road.
>
>
> But looking at tag info there are a fair few uses fo it in various
> locations. So ... what is it used for?

this might also be something like an attempt at oneway=reversible

Richard

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

extend unsigned to describe "no info on the ground" for a key

marc marc
In reply to this post by Tobias Zwick
Le 13.06.19 à 15:15, Tobias Zwick a écrit :
> it semantically refers to the <...> key

it is a problem common to many keys but there is no overall coherence.

extend unsigned used for name with unsigned=<name of the key>
is easy to make a link betweeen the key and "no info on the ground".

it avoids:
- using funny no*=yes
- namespaces without any coherence between usecase
- to be generalizable : any key can be filled in this way without having
to discuss the value of the key or without having to code a difference
rule in the tools that can benefit from it (I think for example of
streetcomplete)

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

Re: extend unsigned to describe "no info on the ground" for a key

dieterdreist


Am Mi., 3. Juli 2019 um 15:48 Uhr schrieb marc marc <[hidden email]>:
Le 13.06.19 à 15:15, Tobias Zwick a écrit :
> it semantically refers to the <...> key

it is a problem common to many keys but there is no overall coherence.

extend unsigned used for name with unsigned=<name of the key>
is easy to make a link betweeen the key and "no info on the ground".

it avoids:
- using funny no*=yes
- namespaces without any coherence between usecase
- to be generalizable : any key can be filled in this way without having
to discuss the value of the key or without having to code a difference
rule in the tools that can benefit from it (I think for example of
streetcomplete)



"unsigned" means there is no sign on the ground, this would not avoid noname=yes or nohousenumber=yes because they state there is no name or housenumber, not that it isn't signed.

Cheers
Martin


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

Re: extend unsigned to describe "no info on the ground" for a key

Paul Allen
On Wed, 3 Jul 2019 at 14:56, Martin Koppenhoefer <[hidden email]> wrote:

"unsigned" means there is no sign on the ground, this would not avoid noname=yes or nohousenumber=yes because they state there is no name or housenumber, not that it isn't signed.

Surely "unsigned" means that the weights can only be positive.  Trucks with negative weights are
not allowed.  Therefore the data schema can accommodate weights with an unsigned int.

It also means that the sign, stating that only positive weights are permitted, does not have a
signature.

What "unsigned" doesn't do is identify how the mapper came to any conclusion about the weight
limit or how other mappers may verify it.

--
Paul


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

Re: extend unsigned to describe "no info on the ground" for a key

marc marc
Le 03.07.19 à 16:55, Paul Allen a écrit :
> What "unsigned" doesn't do is identify how the mapper came to any
> conclusion about the weight
> limit or how other mappers may verify it.

unsigned just said "no info on the ground"
the panel may have fallen, stolen or not yet installed,
the unsigned key says nothing else.
Moreover, when you notice that a bridge has no sign, you generally
cannot deduce anything more than "there is no sign". Similarly, when I
do a survey and I notice that a house does not have the usual sign
indicating its house number, I can said that the sign is not there.

another time, with another source (official or local knowledge),
this information could be completed (no limit, no number, no name,...)
with another key. the unsigned key doesn't try to solve others issues.
one type of information -> one key

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

Re: extend unsigned to describe "no info on the ground" for a key

Paul Allen
On Wed, 3 Jul 2019 at 16:16, marc marc <[hidden email]> wrote:
Le 03.07.19 à 16:55, Paul Allen a écrit :
> What "unsigned" doesn't do is identify how the mapper came to any
> conclusion about the weight
> limit or how other mappers may verify it.

unsigned just said "no info on the ground"

The same can be achieved by omitting the tag.  Possibly supplemented with a fixme.  A fixme
is better because quality tools can help mappers see where more information is needed.

the panel may have fallen, stolen or not yet installed,
the unsigned key says nothing else.

It does when, as seems to be implied by you,  an object can end up with
maxweight=3.5 + unsigned=maxweight without telling anyone how or why
the mapper decided that the maxweight is, in fact, 3.5.

Moreover, when you notice that a bridge has no sign, you generally
cannot deduce anything more than "there is no sign".

Actually, it's worse than that.  Because you cannot deduce what the absent sign is
about.  Is it an absent sign for a speed limit, or a weight restriction, or something else?
I'd expect, for modern bridges on modern roads, the speed limit of the road as appears
on signs somewhere (possibly a long way) before and after the bridge applies to the
bridge itself.  I'd only expect speed limit signs on the bridge if the limit is lower than for
the rest of the road.  By your logic, though, I should tag modern bridges on modern roads
with unsigned=maxspeed.

Similarly, when I  do a survey and I notice that a house does not have the usual
sign indicating its house number, I can said that the sign is not there.

Which would be a little annoying around here, because maybe a tenth of the houses do
not have numbers, only names.  Most of those name-only houses have never had numbers.
You think it sensible to tag unsigned=addr:housenumber for those?  Many houses around
here have numbers without names.  You think it sensible to tag unsigned=addr:housename
for those?
 
another time, with another source (official or local knowledge),
this information could be completed (no limit, no number, no name,...)
with another key. the unsigned key doesn't try to solve others issues.
one type of information -> one key

I don't think it actually solves any issues that are not better handled with a fixme or a
source tag, or simply omitting the tag.

--
Paul


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
1234