[Indoor] is indoor=level walled ?

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

[Indoor] is indoor=level walled ?

PanierAvide-2
Hello,

I'm currently working on a project involving editing of indoor features.
Regarding Simple Indoor Tagging scheme :

"indoor=level is optional and not intended for rendering purposes, but
for having a place to add additional information like a floor name"

So according to wiki, indoor=level doesn't involve having a implicit
wall following the geometry. Is it something we agree on ?

In that case, if I need to create an explicit wall following the contour
of the indoor=level feature, I should create another way with
indoor=wall tag. According to wiki, indoor=wall can only be used on
lines, so if I create a closed line, it should not render as an area
(same behaviour as footways). Is it also something we agree on ?

Last question, if I want to create a wall as an area (imagine for
example a wide pillar where no one can enter), should I just use
indoor=wall + area=yes ?

Best regards,

--
Adrien P.


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

Re: [Indoor] is indoor=level walled ?

Warin
On 23/07/19 21:42, PanierAvide wrote:

> Hello,
>
> I'm currently working on a project involving editing of indoor
> features. Regarding Simple Indoor Tagging scheme :
>
> "indoor=level is optional and not intended for rendering purposes, but
> for having a place to add additional information like a floor name"
>
> So according to wiki, indoor=level doesn't involve having a implicit
> wall following the geometry. Is it something we agree on ?
>
> In that case, if I need to create an explicit wall following the
> contour of the indoor=level feature, I should create another way with
> indoor=wall tag. According to wiki, indoor=wall can only be used on
> lines, so if I create a closed line, it should not render as an area
> (same behaviour as footways). Is it also something we agree on ?
>
> Last question, if I want to create a wall as an area (imagine for
> example a wide pillar where no one can enter), should I just use
> indoor=wall + area=yes ?

I'd use indoor=room for an area that is enclosed by a wall, after all
that is a room.

indoor=wall to me is a free standing barrier within a room.

A pillar? Humm you could model it as a wall.

If the building has multiple levels then using the level=* tag says what
level the thing is on. A wall could be only on one level, or a few ..
I'd want to tag that . A pillar could flow from bottom level to top level.

I don't use indoor=level, I do use level=* and level:ref=*.

See https://wiki.openstreetmap.org/wiki/Simple_Indoor_Tagging

And I'd look at  some tagging examples on level up
https://wiki.openstreetmap.org/wiki/Simple_Indoor_Tagging#Tagging_Examples_.28alphabetical.29

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

Re: [Indoor] is indoor=level walled ?

PanierAvide-2
Hello,

Thanks for your answer. The use case I have in mind is a parking
building, where ground floor hasn't really any wall around it. As I use
indoor=level to add information about the floor (name in particular), I
was asking this to make sure this indoor=level object doesn't create any
implicit wall. If so, I can model rest of this floor using
indoor=area/room without having a wall closing the whole level.

I also use level=* as defined in Simple Indoor Tagging, and I'm aware of
level:ref=*. But level:ref=* is made for saying "this feature belongs
the level named XYZ", which is another use case.

Best regards,

Adrien P.

Le 24/07/2019 à 03:18, Warin a écrit :

> I'd use indoor=room for an area that is enclosed by a wall, after all
> that is a room.
>
> indoor=wall to me is a free standing barrier within a room.
>
> A pillar? Humm you could model it as a wall.
>
> If the building has multiple levels then using the level=* tag says
> what level the thing is on. A wall could be only on one level, or a
> few .. I'd want to tag that . A pillar could flow from bottom level to
> top level.
>
> I don't use indoor=level, I do use level=* and level:ref=*.
>
> See https://wiki.openstreetmap.org/wiki/Simple_Indoor_Tagging
>
> And I'd look at  some tagging examples on level up
> https://wiki.openstreetmap.org/wiki/Simple_Indoor_Tagging#Tagging_Examples_.28alphabetical.29 
>
>
> _______________________________________________
> 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: [Indoor] is indoor=level walled ?

Tordanik
In reply to this post by PanierAvide-2
On 23.07.19 13:42, PanierAvide wrote:
> So according to wiki, indoor=level doesn't involve having a implicit
> wall following the geometry. Is it something we agree on ?

indoor=level does not add an implicit wall.

However, building and building:part probably do have outer walls by
default, regardless of indoor=level or any other indoor mapping. There's
no solution for switching these walls off yet. We can't really assume
that the absence of an indoor=wall or indoor=room element means there's
no outer wall – fully indoor-mapped buildings are the exception, not the
rule.

Therefore, my suggestion would be to include a tag like wall=no on your
indoor=level polygon for these unwalled levels.

> According to wiki, indoor=wall can only be used on
> lines, so if I create a closed line, it should not render as an area
> (same behaviour as footways). Is it also something we agree on ?

Yes, a closed way with indoor=wall should be considered a line, not an area.

> Last question, if I want to create a wall as an area (imagine for
> example a wide pillar where no one can enter), should I just use
> indoor=wall + area=yes ?

Using area=yes makes sense here, yes.

Conceptually, I'm a bit uncomfortable with wall polygons because they
break the logic for doors and windows (which are nodes on the wall
ways). But for something completely solid like a pillar, this is
probably not going to cause problems.

Tobias

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

Re: [Indoor] is indoor=level walled ?

dieterdreist


sent from a phone

> On 26. Jul 2019, at 12:14, Tobias Knerr <[hidden email]> wrote:
>
> Therefore, my suggestion would be to include a tag like wall=no on your
> indoor=level polygon for these unwalled levels.


are they “indoor” if there is no wall? What makes them be inside rather than outside?

Are covered open spaces inside or outside? I would tend to “outside”
as the climate will be similar to the climate around the structure, rather than comparable to the room climate of a completely enclosed space.

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

Re: [Indoor] is indoor=level walled ?

PanierAvide-2
Thanks for your two answers.

Le 26/07/2019 à 13:02, Martin Koppenhoefer a écrit :
> are they “indoor” if there is no wall? What makes them be inside rather than outside?
>
> Are covered open spaces inside or outside? I would tend to “outside”
> as the climate will be similar to the climate around the structure, rather than comparable to the room climate of a completely enclosed space.

They are still indoors (at least conceptually) because the building
footprint is still there. Having ground floor unwalled doesn't mean that
upper levels can't have wall. So, in that case, you suggest that
building for example start at level 1 (with min_level=1) and everything
on ground floor is outside ? Then, how can I map for example a very
small room on this area ? Still with indoor=room, even if technically
not inside a building ?

I like the suggestion of Tobias to add wall=no on indoor=level, it seems
to break less thing and make explicit the exception of one particular
level being unwalled.

Best regards,

Adrien.


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

Re: [Indoor] is indoor=level walled ?

Warin
In reply to this post by dieterdreist
On 26/07/19 21:02, Martin Koppenhoefer wrote:

>
> sent from a phone
>
>> On 26. Jul 2019, at 12:14, Tobias Knerr <[hidden email]> wrote:
>>
>> Therefore, my suggestion would be to include a tag like wall=no on your
>> indoor=level polygon for these unwalled levels.
>
> are they “indoor” if there is no wall? What makes them be inside rather than outside?
>
> Are covered open spaces inside or outside? I would tend to “outside”
> as the climate will be similar to the climate around the structure, rather than comparable to the room climate of a completely enclosed space.

I thinking of a building where the lower levels are used for parking vehicles while the upper levels have shops and/or residences. .
The parking levels usually have no walls to provide ventilation (and it is cheaper), they usually have a barrier to stop people and cars going over the edge but not a full heigh wall. So they are partially enclosed and partially open.

They provide economic shelter from the weather (rain/sun).

I would not describe them as being 'walled' nor 'outside'.


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

Re: [Indoor] is indoor=level walled ?

dieterdreist
In reply to this post by PanierAvide-2
Am Fr., 26. Juli 2019 um 13:13 Uhr schrieb PanierAvide <[hidden email]>:
Le 26/07/2019 à 13:02, Martin Koppenhoefer a écrit :
> are they “indoor” if there is no wall? What makes them be inside rather than outside?
>
> Are covered open spaces inside or outside? I would tend to “outside”
> as the climate will be similar to the climate around the structure, rather than comparable to the room climate of a completely enclosed space.

They are still indoors (at least conceptually) because the building
footprint is still there. Having ground floor unwalled doesn't mean that
upper levels can't have wall.



sure, but if other levels have walls it doesn't make those levels without walls "inside", does it?


 
So, in that case, you suggest that
building for example start at level 1 (with min_level=1) and everything
on ground floor is outside ?


no, I would put it like this: the ground floor is still part of the building, but it is outside. Like a balcony for example. Would you say a balcony is "inside"?

 
Then, how can I map for example a very
small room on this area ? Still with indoor=room, even if technically
not inside a building ?


is this a room with some kind of delimiting feature? I would not say it is not part of the building, I would only say the space that is not enclosed by walls (including windows, glass facades, etc.) is not "indoor".

Cheers
Martin

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

Re: [Indoor] is indoor=level walled ?

dieterdreist
Am Fr., 26. Juli 2019 um 13:18 Uhr schrieb Martin Koppenhoefer <[hidden email]>:
no, I would put it like this: the ground floor is still part of the building, but it is outside. Like a balcony for example. Would you say a balcony is "inside"?


I guess this was too short, here's a more exhaustive take on the typical situations:

1. iconic building by le Corbu: https://upload.wikimedia.org/wikipedia/de/a/af/Villa_Savoye_2015.jpg This is a typical example for a raised modernist building.
the space where you can see chairs is IMHO clearly not "indoor", I would tend to accept it is part of the building (because it is "created"/delimited by the building and intended as usuable space), but you could also argue it is part of the garden, the architect even emphasizes this by using the same pavement as for the driveway (at least it looks like this on the picture).

These are typically cases where the building is raised above the ground in order to make use of a covered outdoor space, e.g. to use it as part of the garden, or to park a car, or as common space for the residents.


I would tend to count the outdoor space below the "house" as not being part of the building (conceptually, the building is standing on legs, and while the legs are part of it, the area where they stand could be considered as not part of it). The area not being usable/accessible contributes to this judgement.
These are generally cases where the building is raised above a "hostile" environment, e.g. to protect it from water, wild animals, enemies, or to create a level surface in an inclined surrounding. Typically the space below is not used in these cases. I would not consider the (unmodified / unaltered) ground below the building to be part of the building.


In all cases, I would not consider these indoor spaces, because they can not be heated or cooled, while you may be protected from the sun and precipitation you will still feel more outside than inside, typically.

I acknowledge there are many different situations and you will have to assess these individually, there will surely be a lot of edge cases. How you see them may also depend on the climate in the area in general, e.g. there are also lots of houses that are neither cooled nor heated, and some may have openings that cannot be closed rather than windows.

Cheers,
Martin




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

Re: [Indoor] is indoor=level walled ?

PanierAvide-2

Thanks for this feedback. In these examples, I would say that there is still a clear delimitation of what outside and what is inside, so can be addressed with Simple 3D buildings modelling. My question is oriented in a particular case where you don't have a very precise delimitation of inside/outside, like this parking lot :

https://commons.wikimedia.org/wiki/File:Parking_Building_(41640900211).jpg

As level 0 doesn't have wall, if you are near the building "limit" you can consider being outside, but at the center of this level you are clearly inside (covered, maybe warmer). So how can we represent this lack of walls, but looking more like something inside ?

Best regards,

Adrien P.
Le 26/07/2019 à 13:39, Martin Koppenhoefer a écrit :
Am Fr., 26. Juli 2019 um 13:18 Uhr schrieb Martin Koppenhoefer <[hidden email]>:
no, I would put it like this: the ground floor is still part of the building, but it is outside. Like a balcony for example. Would you say a balcony is "inside"?


I guess this was too short, here's a more exhaustive take on the typical situations:

1. iconic building by le Corbu: https://upload.wikimedia.org/wikipedia/de/a/af/Villa_Savoye_2015.jpg This is a typical example for a raised modernist building.
the space where you can see chairs is IMHO clearly not "indoor", I would tend to accept it is part of the building (because it is "created"/delimited by the building and intended as usuable space), but you could also argue it is part of the garden, the architect even emphasizes this by using the same pavement as for the driveway (at least it looks like this on the picture).

These are typically cases where the building is raised above the ground in order to make use of a covered outdoor space, e.g. to use it as part of the garden, or to park a car, or as common space for the residents.


I would tend to count the outdoor space below the "house" as not being part of the building (conceptually, the building is standing on legs, and while the legs are part of it, the area where they stand could be considered as not part of it). The area not being usable/accessible contributes to this judgement.
These are generally cases where the building is raised above a "hostile" environment, e.g. to protect it from water, wild animals, enemies, or to create a level surface in an inclined surrounding. Typically the space below is not used in these cases. I would not consider the (unmodified / unaltered) ground below the building to be part of the building.


In all cases, I would not consider these indoor spaces, because they can not be heated or cooled, while you may be protected from the sun and precipitation you will still feel more outside than inside, typically.

I acknowledge there are many different situations and you will have to assess these individually, there will surely be a lot of edge cases. How you see them may also depend on the climate in the area in general, e.g. there are also lots of houses that are neither cooled nor heated, and some may have openings that cannot be closed rather than windows.

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: [Indoor] is indoor=level walled ?

dieterdreist


Am Fr., 26. Juli 2019 um 13:58 Uhr schrieb PanierAvide <[hidden email]>:

Thanks for this feedback. In these examples, I would say that there is still a clear delimitation of what outside and what is inside, so can be addressed with Simple 3D buildings modelling. My question is oriented in a particular case where you don't have a very precise delimitation of inside/outside, like this parking lot :

https://commons.wikimedia.org/wiki/File:Parking_Building_(41640900211).jpg

As level 0 doesn't have wall, if you are near the building "limit" you can consider being outside, but at the center of this level you are clearly inside (covered, maybe warmer). So how can we represent this lack of walls, but looking more like something inside ?



frankly I would not consider this parking on any level visible in the picture as "indoor". An indoor parking would be something like this: https://www.parkrideflyusa.com/facility-photos/33/indoor-parking.jpg 
Parking spaces are generally edge cases because of the ventilation requirements, but in the picture you showed there is hardly anything that can be considered "walls", I would call them "fences" or maybe "grates".

Cheers,
Martin

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

Re: [Indoor] is indoor=level walled ?

Joseph Eisenberg
Well a parking garage may not really be "indoor", but it's certainly
"inside of a building" according to the database model, since the
garage will be mapped as a building=*

So what's important is how we may the lack of walls.

I agree with the earlier suggestion by Tobias to add wall=no for
clarity when using indoor=level, in this case.

Joseph

On 7/26/19, Martin Koppenhoefer <[hidden email]> wrote:

> Am Fr., 26. Juli 2019 um 13:58 Uhr schrieb PanierAvide <
> [hidden email]>:
>
>> Thanks for this feedback. In these examples, I would say that there is
>> still a clear delimitation of what outside and what is inside, so can be
>> addressed with Simple 3D buildings modelling. My question is oriented in
>> a
>> particular case where you don't have a very precise delimitation of
>> inside/outside, like this parking lot :
>>
>> https://commons.wikimedia.org/wiki/File:Parking_Building_(41640900211).jpg
>>
>> As level 0 doesn't have wall, if you are near the building "limit" you
>> can
>> consider being outside, but at the center of this level you are clearly
>> inside (covered, maybe warmer). So how can we represent this lack of
>> walls,
>> but looking more like something inside ?
>>
>
>
> frankly I would not consider this parking on any level visible in the
> picture as "indoor". An indoor parking would be something like this:
> https://www.parkrideflyusa.com/facility-photos/33/indoor-parking.jpg
> Parking spaces are generally edge cases because of the ventilation
> requirements, but in the picture you showed there is hardly anything that
> can be considered "walls", I would call them "fences" or maybe "grates".
>
> Cheers,
> Martin
>

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