Re: [Tagging] 'ongoing' U-turn restriction

[Martin Koppenhoefer]

sent from a phone

> On 8 Dec 2016, at 19:13, Colin Smale  wrote:
> 
> "no overtaking" is too broad IMHO.


maybe we'd have to define what "overtaking" means, in contrast to e.g. "passing 
an obstacle" (e.g. slow moving tractor). If overtaking is broad or not is not 
completely clear currently, because there's no definition for it...


Generally, while the idea seems appealing also to me to model the world and the 
traffic rules, and have the computer find out the rest, 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] 'ongoing' U-turn restriction, plus side turns

[André Pirard]

On 2016-12-08 09:36, Volker Schmidt wrote:
> On 2016-12-08 03:09, André Pirard wrote:
>> On 2016-12-07 22:53, Martijn van Exel wrote:
>>> Hey all, 
>>>
>>> Apparently there are roads that have an 'ongoing' U-turn
>>> restriction, see the sign here for
>>> example: http://openstreetcam.org/details/10053/304 
>>>
>>> I know how to tag this for an intersection but not how to do it for
>>> an entire way where U-turns are prohibited.
>> The concept of a node where one cannot U-turn and where one could
>> U-turn 1 meter before it is a strange one.
>> In Belgium, signal C33
>> 
>> indicates that one cannot U-turn from the position of the signal up
>> to and including the next crossing.  It's obviously always in force
>> over a span of the highway, it indicates a one-way section and the
>> simplest way to indicate that is oneway=yes.
>
> This is interesting and closely related to the question how we tag, on
> two-way roads, continuous middle lines (or double continuous middle
> lines) that imply continuous stretches of road with "no-u-turn", no
> passing/overtaking, and "no-left-turn" (in countries with right-hand
> traffic).
> In addition there seems to be countries (I am not sure) where u-turns
> are generally forbidden on certain classes of roads.
> Why do we not use u_turn=no on highways, similar to the overtaking
> tag? I am also wondering why "overtaking" in reality is not frequently
> use either, worldwide only 40,000 tags.

I have taken permission to reintroduce the quote showing what's interesting.
And thank you for recalling a subject I once raised.
Those Belgian white lines preventing U-turn (without oneway) are
jurisdiction-dependent indeed as well said further.
I had abandoned the discussion, taking it as the usual "no" the answers
that it's not so everywhere.
But as said further too, what we map is not the color or form of the
lines but their meaning.

And what I mostly mentioned before, is that those Belgian lines not only
forbid U-turn,
but also turning left (in right-hand drive) when coming from a side road.
I find this important because it makes a lot of missing, difficult to
map turn relations unnecessary.
But I suppose it would have, because of jurisdiction-dependency, mapped
differently than U-turn.

Interesting discussion.

Cheers

André.



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

Re: [Tagging] 'ongoing' U-turn restriction

[Colin Smale]

On 2016-12-08 15:12, Martin Koppenhoefer wrote:

> On 8 Dec 2016, at 12:25, Colin Smale  wrote: 
> 
>> Martin, your interpretation is jurisdiction-dependant. There is sometimes a 
>> distinction between single and double lines as well (Queensland is an 
>> example I just found). The objectively verifiable fact is the road marking, 
>> thereafter you have to interpret the local laws to understand its effect.
> 
> Yes, if you map with the divider tag you don't have to understand the 
> meaning, your direct uneducated observation is sufficient for mapping, still 
> I'd consider the resulting effect from interpreting the local law something 
> "objectively verifiable" as well - and more interesting for OSM if there are 
> significant differences in meaning. Usually we do add our interpretation 
> (e.g. the access rights of a road, put on the highway) rather than limiting 
> us to pure physical facts (e.g. an access restriction sign), and we clearly 
> prioritize the former.

Absolutely! But we need more subtle nuances in our tagging - "no
overtaking" is too broad IMHO. As has been mentioned many times before,
this is a case where we could benefit from a curated database (in
whatever form!) of semantic implications based on geographic areas, so
we could suffice with tagging the divider and any other road signs (as a
physical fact) and have a clear set of rules which these facts imply.
For example if we have divider=single_solid_line the database may
contain rules for all the different states to state what that actually
means. In one country it may be "no overtaking", in another it might be
"no crossing the line except to take a turn". If we develop an explicit
tagging scheme for all these nuances, this database could be considered
"defaults in the absence of explicit tagging to the contrary". But now I
am being rather heretical and at risk of being burnt at the stake for
mentioning documented defaults 

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

Re: [Tagging] 'ongoing' U-turn restriction

[Richard Welty]

On 12/8/16 9:34 AM, Greg Troxel wrote:
> Also, for what it's worth, in massachusetts.us, we have two kinds of
> solid yellow (center) lines.  One can pass with care give a single line,
> but may not cross the line or pass if double.
many states post explicit "no passing" signs at the beginning of sections
of road with solid yellow lines.

richard

-- 
rwe...@averillpark.net
 Averill Park Networking - GIS & IT Consulting
 OpenStreetMap - PostgreSQL - Linux
 Java - Web Applications - Search




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

Re: [Tagging] 'ongoing' U-turn restriction

[Greg Troxel]

Martin Koppenhoefer  writes:

> sent from a phone
>
>> On 8 Dec 2016, at 12:25, Colin Smale  wrote:
>> 
>> Martin, your interpretation is jurisdiction-dependant. There is
>> sometimes a distinction between single and double lines as well
>> (Queensland is an example I just found). The objectively verifiable
>> fact is the road marking, thereafter you have to interpret the local
>> laws to understand its effect.
>
> Yes, if you map with the divider tag you don't have to understand the
> meaning, your direct uneducated observation is sufficient for mapping,
> still I'd consider the resulting effect from interpreting the local
> law something "objectively verifiable" as well - and more interesting
> for OSM if there are significant differences in meaning. Usually we do
> add our interpretation (e.g. the access rights of a road, put on the
> highway) rather than limiting us to pure physical facts (e.g. an
> access restriction sign), and we clearly prioritize the former.

I agree with Martin.  In addition, a program computing routes has to
work on semantics, not signs, unless the jurisdiction semantics are
encoded in a separate database.

Also, for routing, what's needed is to compute a route which can be more
or less followed legally by a driver who understands the rules.

Also, for what it's worth, in massachusetts.us, we have two kinds of
solid yellow (center) lines.  One can pass with care give a single line,
but may not cross the line or pass if double.  Except that in practice,
one can pass in a double yellow line zone if the vehicle being passed is
really slow (agriculture or bicycle) or there is some other unusual
situation.   Also, most people who drive here have no idea what the
U-turn rules are :-)


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

Re: [Tagging] 'ongoing' U-turn restriction

[Martin Koppenhoefer]

sent from a phone

> On 8 Dec 2016, at 12:25, Colin Smale  wrote:
> 
> Martin, your interpretation is jurisdiction-dependant. There is sometimes a 
> distinction between single and double lines as well (Queensland is an example 
> I just found). The objectively verifiable fact is the road marking, 
> thereafter you have to interpret the local laws to understand its effect.
> 


Yes, if you map with the divider tag you don't have to understand the meaning, 
your direct uneducated observation is sufficient for mapping, still I'd 
consider the resulting effect from interpreting the local law something 
"objectively verifiable" as well - and more interesting for OSM if there are 
significant differences in meaning. Usually we do add our interpretation (e.g. 
the access rights of a road, put on the highway) rather than limiting us to 
pure physical facts (e.g. an access restriction sign), and we clearly 
prioritize the former.

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

Re: [Tagging] shop=estate_agent and office=estate_agent

[Andy Townsend]

On 07/12/2016 22:48, LeTopographeFou wrote:


Hi,

I will probably reopen an explosive case but I would like to know 
where we are reguarding shop=estate_agent vs office=estate_agent.




Personally I use both - "shop" for places that are primarily B2C and 
fitted out as shops, and "office" for places that are more B2B, 
back-office, or service-led, or "feel more like offices".


Cheers,

Andy


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

Re: [Tagging] shop=estate_agent and office=estate_agent

[Éric Gillet]

2016-12-08 11:25 GMT+01:00 Martin Koppenhoefer :

> > On 8 Dec 2016, at 01:49, Mike N  wrote:
> >
> > Like all things in OSM, I'm not sure how to migrate to a new consensus
> of office=
>
>
> just use office, and add a hint to the wiki. As was pointed out above,
> actual migration has already taken place, given the usage numbers and
> support by the 2 mostly used editors.
>

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

Re: [Tagging] 'ongoing' U-turn restriction

[Colin Smale]

Martin, your interpretation is jurisdiction-dependant. There is
sometimes a distinction between single and double lines as well
(Queensland is an example I just found). The objectively verifiable fact
is the road marking, thereafter you have to interpret the local laws to
understand its effect. A no-overtaking sign is just as "variable" as a
continuous line - it can mean different things in different
jurisdictions.

If we map the road markings and signs, we cannot be wrong. If we try to
map their effects, and do it badly, we will miss a lot of information
which may be relevant to e.g. motorcyclists or tractor drivers. Mapping
the effects properly is a lot of work "no overtaking except to pass
agricultural vehicles or one motorcycle without sidecar overtaking
another, PLUS no turning to cross the line except to access a property
for vehicles < 3.5t" (random invention but perfectly existable) 

//colin 

On 2016-12-08 11:35, Martin Koppenhoefer wrote:

> sent from a phone 
> 
> On 8 Dec 2016, at 09:36, Volker Schmidt  wrote:
> 
>> how we tag, on two-way roads, continuous middle lines (or double continuous 
>> middle lines) that imply continuous stretches of road with "no-u-turn", no 
>> passing/overtaking, and "no-left-turn" (in countries with right-hand 
>> traffic).
> 
> it is a common misreading of continuous lines to believe they mean no 
> passing/overtaking, but they don't. You may not cross the line, but if there 
> is sufficient space you can still overtake/pass by other vehicles (i.e. in 
> many situations the line does indeed prevent from overtaking (legally) 
> because of lack of such space). 
> 
> One osm tag for this is "divider" but it is neither promoted nor widespread: 
> http://taginfo.osm.org/keys/divider#values 
> 
> cheers, 
> Martin  
> ___
> 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] 'ongoing' U-turn restriction

[Martin Koppenhoefer]

sent from a phone

> On 8 Dec 2016, at 09:36, Volker Schmidt  wrote:
> 
> how we tag, on two-way roads, continuous middle lines (or double continuous 
> middle lines) that imply continuous stretches of road with "no-u-turn", no 
> passing/overtaking, and "no-left-turn" (in countries with right-hand traffic).



it is a common misreading of continuous lines to believe they mean no 
passing/overtaking, but they don't. You may not cross the line, but if there is 
sufficient space you can still overtake/pass by other vehicles (i.e. in many 
situations the line does indeed prevent from overtaking (legally) because of 
lack of such space).

One osm tag for this is "divider" but it is neither promoted nor widespread: 
http://taginfo.osm.org/keys/divider#values

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

Re: [Tagging] shop=estate_agent and office=estate_agent

[Martin Koppenhoefer]

sent from a phone

> On 8 Dec 2016, at 01:49, Mike N  wrote:
> 
> Like all things in OSM, I'm not sure how to migrate to a new consensus of 
> office=


just use office, and add a hint to the wiki. As was pointed out above, actual 
migration has already taken place, given the usage numbers and support by the 2 
mostly used editors.

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

Re: [Tagging] Mediterranean "bar" aka pub

[Martin Koppenhoefer]

sent from a phone

On 8 Dec 2016, at 02:15, Alejandro S.  wrote:

>> Some are indeed better tagged with amenity=cafe
> 
> 
> 
> 
> And maybe some of them as amenity=pub?


maybe in very few exceptions, usually not. There are pubs in Italy and they are 
completely different from bars. Pubs also have necessarily tables and chairs.

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

Re: [Tagging] 'ongoing' U-turn restriction

[Volker Schmidt]

This is interesting and closely related to the question how we tag, on
two-way roads, continuous middle lines (or double continuous middle lines)
that imply continuous stretches of road with "no-u-turn", no
passing/overtaking, and "no-left-turn" (in countries with right-hand
traffic).
In addition there seems to be countries (I am not sure) where u-turns are
generally forbidden on certain classes of roads.
Why do we not use u_turn=no on highways, similar to the overtaking tag? I
am also wondering why "overtaking" in reality is not frequently use either,
worldwide only 40,000 tags.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging