Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

> > Usually there will be ~100mtres for this where you can
> > change at any time, but in OSM you have to decide on one merging
> > point.
>
>
I haven't seen a proposed scheme yet that really deals with this
> properly, they pretty much all pretend that a lane starts at a point.
> Personally I don't think that figuring this out is necessary for
> having multiple lanes, as we could continue the fiction that they
> start at a point. It'd be nice, but not necessary.


You can usually be accurate if you're not afraid of having overlapping ways.
 Lanes do start at a point, they just commonly overlap with another lane at
that point.

But yeah, even if you get down to the lane level, you're still mapping
logically and not physically.  If you want to map the physical, I think
you've gotta go with a boundary relation: landuse=right_of_way, surface=*.
 Hopefully you can keep the ways for routing purposes, though.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[James Livingston]

On 22/09/2009, at 10:57 AM, Martin Koppenhoefer wrote:
> 2009/9/22 Anthony :
>> It is possible to represent different surfaces and different  
>> maxspeeds
>> without using more than one way.  "maxspeed:lane=130;110";
>> "surface:lane=asphalt;concrete".  That's not necessarily the best  
>> solution,
>
> indeed, it won't be understood by none of the apps that are using our
> data and it doesn't say, which lane has which value...

The first bit is always going to be a problem, however we decide to  
map lanes the tools will need changing. Special tags or relations?  
They'll need to parse those to be useful. A new database construct?  
They'll need to understand that. There is no way around needing to  
modify the tools if we want more than a pile of tags that somehow  
relate to ways.

The second bit we can easily get around by picking either counting  
from the left or right. Maybe something like lane[n]:key=value where  
there is data, and on the node in a way where the lanes change or  
merge we have lane-mapping data that says lane A back along the way  
changes to lane B after the way.


> but that's exactly what I propose (map lanes explicitly) and it's
> against the separate-ways-only-when-physically-divided-paradigm
> (because an ambulance could change from one way to another)...

There's a fun question, what counts as "physically divided"? I'm seen  
many emergency vehicles cross "physically divided" roads. If we  
exclude those, then what counts as a vehicle for the purposes of  
dividing a way? Are bicycles includes? Four-wheel drives?

> Usually there will be ~100mtres for this where you can
> change at any time, but in OSM you have to decide on one merging
> point.

I haven't seen a proposed scheme yet that really deals with this  
properly, they pretty much all pretend that a lane starts at a point.  
Personally I don't think that figuring this out is necessary for  
having multiple lanes, as we could continue the fiction that they  
start at a point. It'd be nice, but not necessary.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 8:57 PM, Martin Koppenhoefer  wrote:

> 2009/9/22 Anthony :
> > It is possible to represent different surfaces and different maxspeeds
> > without using more than one way.  "maxspeed:lane=130;110";
> > "surface:lane=asphalt;concrete".  That's not necessarily the best
> solution,
>
> indeed, it won't be understood by none of the apps that are using our
> data and it doesn't say, which lane has which value...
>

There would obviously need to be a standard, formed through a proposal,
first.  :)

Left to right when the way is pointing up?

> Different turn-restrictions is already possible.  If you have a three lane
> > way with two lanes going straight and one turning right, you join three
> ways
> > at one node: one with three lanes, one with two lanes, one with one lane.
> > If you have a three lane road with two lanes going straight and one lane
> > going straight or turning right, ditto, except the way going straight has
> > three lanes instead of two.  Which lane is which is determined by the
> > geometry of the ways as they come out of the node.
>
> but that's exactly what I propose (map lanes explicitly) and it's
> against the separate-ways-only-when-physically-divided-paradigm
> (because an ambulance could change from one way to another)...
>

If you're going to go with
the separate-ways-only-when-physically-divided-paradigm, aren't you better
off using boundary relations rather than ways?

I meant the lanes for acceleration and breaking (when going on or from
> a highway). Usually there will be ~100mtres for this where you can
> change at any time, but in OSM you have to decide on one merging
> point.


True...  Perhaps it's a mistake to try to combine the physical and the
logical into one in the first place.

How about using ways to represent the suggested paths of travel, and
boundary relations to represent the physical road structure?

If you want to represent the physical road structure, mapping all the lanes
doesn't even cut it. When two lanes merge or diverge, they don't leave nice
little "lanes" of paved roadway in between them, they leave sections of
roadways that would be better described with polygons or boundary relations
rather than lanes.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Tue, Sep 22, 2009 at 5:22 PM, Lester Caine  wrote:
>>> Camp one is - single way with lanes=4 + bridge section
>>> Camp two would prefer all elements mapped in which case the bridge WOULD
>>> be a separate element.
>>
>> Camp three: multiple ways representing paths of travel, grouped with a
>> "bridge" relation to indicate they share a common bridge. This could
>> probably be seen as a compromise, and is (I think) a good interim
>> solution, if not a very-long-term solution.
>
> That is simply camp two ...
> The bridge relation would have to have a physical presence at some point!

In that case, just to clarify, I meant
Camp 2a) All elements mapped with the bridge as a *separate* element
(e.g. an area, not a relation)
Camp 2b) All elements mapped with the bridge as a grouping of elements
(a relation, including the ways on the bridge and optionally the area
indicating the bridge surface, etc.)

These are both covered by
http://wiki.openstreetmap.org/wiki/Relations/Proposed/Bridges_and_Tunnels

> The point I am trying to make is that the two camps HAVE to co-exist.
> What we are looking for is a way to 'rough in' the data, and provide a
> macro level view of things, and then add the NECESSARY detail below that
> so that micro mapping can simply be added to the macro model.

Agreed.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Peteris Krisjanis]

2009/9/22 Dave F. :
> Hi Peter
>> I can suggest to add bridge's physical form to this. Otherwise yeah,
>> why not. It would also help to indicate where bridge actually starts
>> and ends to help routing software logic.
>>
>>
> Why would routing software need to know the extent of a bridge?

For routing software for example to say: "Turn there, go across the
bridge and then after 100 meters go left?"

Bridge is kinda very big object in every town or city. Giving
perspective where it starts and where it ends you give additional
information for routing software to inform it's user how to find
things relatively to the bridge.

> Cheers
> Dave F.
>
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>



-- 
mortigi tempo
Pēteris Krišjānis

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Lester Caine]

Martin Koppenhoefer wrote:
> 2009/9/21 Anthony :
>> Osmarender does a pretty good job of those bridges:
>> http://www.openstreetmap.org/?lat=27.95907&lon=-82.53907&zoom=17&layers=0B00FTF
>> Notice how it combines the two ways heading south into one bridge even
>> without any additional information besides the fact that the two ways are
>> close together.
> 
> there are many cases of parallel bridges though, where you don't want
> them to render as one because they are two (or more), so being close
> cannot be the criteria.

And details like two or three deck bridges also require a little more 
information, although the layer model would at least keep cars off the 
train level ;)

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Lester Caine]

Roy Wallace wrote:
> On Tue, Sep 22, 2009 at 12:20 AM, Anthony  wrote:
>> A bridge should probably have its
>> own geometry.  And if a bridge has its own geometry (polygon or line and
>> width) and a layer tag you don't even need the relation, do you?  Anything
>> in the area of the bridge with the same layer is located on the bridge.
> 
> I like the idea.
> 
> BUT I shouldn't have to measure the bridge to be able to indicate that
> a section of way(s) goes over a bridge.
> 
> Using an area to mark a bridge in this way should be *optional*, in
> the same was as a POI can be marked as a node or an area. Therefore, a
> "bridge" relation would still be useful to indicate that multiple ways
> share a bridge, for when the bridge geometry is unknown.

Which is exactly the same rule as applies elsewhere. A car park can 
exist simply as a POI tag, or as fine detail, perhaps even with parking 
bays. The point is to make things transparently scalable. At the 
'county' level you don't need the shape of the car park, just a tag to 
it's presence so you can indicate where they are. Same with any object?

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Lester Caine]

Roy Wallace wrote:
> On Mon, Sep 21, 2009 at 11:15 PM, Lester Caine  wrote:
>>> Can I check? Are there people here who are suggesting that, in my case,
>>> they want to draw all four ways (yes, I know the footpath hasn't been
>>> mapped yet) as  a single way & specify the differences with lane tags?
>> Camp one is - single way with lanes=4 + bridge section
>> Camp two would prefer all elements mapped in which case the bridge WOULD
>> be a separate element.
> 
> Camp three: multiple ways representing paths of travel, grouped with a
> "bridge" relation to indicate they share a common bridge. This could
> probably be seen as a compromise, and is (I think) a good interim
> solution, if not a very-long-term solution.

That is simply camp two ...
The bridge relation would have to have a physical presence at some point!

The point I am trying to make is that the two camps HAVE to co-exist. 
What we are looking for is a way to 'rough in' the data, and provide a 
macro level view of things, and then add the NECESSARY detail below that 
so that micro mapping can simply be added to the macro model.

If that means a half way house of detailing the fine content of a way by 
splitting into multiple tracks, then the 'holder' for the bridge slots 
in to that model and can be expanded in detail later.

At some point it WILL become necessary to create elements like bridges 
as 'paintings' with accurate graphic detail, but the higher level view 
only needs the single way with a bridge flag - and there is no reason 
that could not be provided as a tag on a single macro level view of the 
way? At the macro level THAT makes sense, while once one breaks a way 
down into it's component tracks, a tag on that view may be of more 
advantage to fine detail routing, while a separate bridge element makes 
more sense for rendering in the smaller scales?

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/22 Martin Koppenhoefer :

> - is probably more work to map every lane

I disagree, you declare how many lanes and unless you need to tag
lanes independent of the way everything for the most part would be the
same as it is now.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Martin Koppenhoefer]

2009/9/22 Anthony :
> It is possible to represent different surfaces and different maxspeeds
> without using more than one way.  "maxspeed:lane=130;110";
> "surface:lane=asphalt;concrete".  That's not necessarily the best solution,

indeed, it won't be understood by none of the apps that are using our
data and it doesn't say, which lane has which value...

> Different turn-restrictions is already possible.  If you have a three lane
> way with two lanes going straight and one turning right, you join three ways
> at one node: one with three lanes, one with two lanes, one with one lane.
> If you have a three lane road with two lanes going straight and one lane
> going straight or turning right, ditto, except the way going straight has
> three lanes instead of two.  Which lane is which is determined by the
> geometry of the ways as they come out of the node.

but that's exactly what I propose (map lanes explicitly) and it's
against the separate-ways-only-when-physically-divided-paradigm
(because an ambulance could change from one way to another)...

>> but I also see the pros:
>> - is easier to understand and more intuitive to edit: better
>> maintenance: you see what is there
>
> That's an editor consideration.  You can make it *look* like you're editing
> three separate ways when you're really editing one way with three lanes.

I knew someone would raise this argument, but please explain how the
editor will know where the lanes exactly go.

>>
>> - is more precise in terms of positional accuracy
>
> I'm assuming the lanes are parallel.  If not, then yes, you need to use
> multiple ways.

they will never be parallel for the entire lenght, at least there will
be a start and an end, but often they will also turn at junctions...
>>
>> - can better represent stuff like ramps for motorways (parallel ways
>> that currently are mapped like intersections), cyclelanes, ...
>
> Why are they being mapped like intersections?  They shouldn't be.  You mean
> a gradual merge is being mapped at a right angle?  That's wrong, and there's
> no good reason for it (other than didn't have the time to get it right,
> anyway).

I meant the lanes for acceleration and breaking (when going on or from
a highway). Usually there will be ~100mtres for this where you can
change at any time, but in OSM you have to decide on one merging
point.

cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 6:31 PM, Martin Koppenhoefer  wrote:

> 2009/9/21 Anthony :
> > Right now, like it or not, ways are being used essentially in the manner
> I'm
> > suggesting we keep using them.  We don't combine everything that crosses
> a
> > bridge into a single way, and we don't break apart every lane into a
> > separate way.  Those saying that we should do one of these two things are
> > the ones trying to reinvent the wheel, and the burden of proof is on them
> to
> > prove that their new wheel is superior.
>
> It is needed for several situations that currently cannot be mapped
> completely (different surfaces, different maxspeeds, different
> turn-restrictions, etc.)


It is possible to represent different surfaces and different maxspeeds
without using more than one way.  "maxspeed:lane=130;110";
"surface:lane=asphalt;concrete".  That's not necessarily the best solution,
but it beats having two ways and some not-yet-proposed method of showing
that you can switch between them.

Different turn-restrictions is already possible.  If you have a three lane
way with two lanes going straight and one turning right, you join three ways
at one node: one with three lanes, one with two lanes, one with one lane.
If you have a three lane road with two lanes going straight and one lane
going straight or turning right, ditto, except the way going straight has
three lanes instead of two.  Which lane is which is determined by the
geometry of the ways as they come out of the node.

but I also see the pros:
> - is easier to understand and more intuitive to edit: better
> maintenance: you see what is there
>

That's an editor consideration.  You can make it *look* like you're editing
three separate ways when you're really editing one way with three lanes.


> - is more precise in terms of positional accuracy
>

I'm assuming the lanes are parallel.  If not, then yes, you need to use
multiple ways.


> - can better represent stuff like ramps for motorways (parallel ways
> that currently are mapped like intersections), cyclelanes, ...
>

Why are they being mapped like intersections?  They shouldn't be.  You mean
a gradual merge is being mapped at a right angle?  That's wrong, and there's
no good reason for it (other than didn't have the time to get it right,
anyway).
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Dave F.]

Anthony wrote:
>
> Also, shouldn't the S.R.589 be a bridge directly below it label in
> that image.
>
>
> I'm not sure what you're referring to there.  Can you give me a way ID 
> and a couple point IDs?
Sure:
http://www.openstreetmap.org/?mlat=27.95915&mlon=-82.53835&zoom=16&layers=0B00FTF
The ways appear to clash on the same layer
>  
>
> And the link to the North East, doesn't that pass over a river of
> some sorts?
>
>
> Looking at the Yahoo Map which we are allowed to derive from, which 
> may be out of date, it seems to be some sort of man-made canal.  I 
> don't remember it, I don't have any pictures of it, and I'm not 100% 
> sure it wasn't filled in, so I'm not going to add it unless I can 
> manage to confirm it somehow.  It's difficult mapping without the 
> benefit of the Yahoo aerials, and these Yahoo aerials are all out of date.
>
> It's awesome having more up-to-date info than Yahoo and Google, 
> especially knowing you helped make it.
Definitely



___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Liz]

On Tue, 22 Sep 2009, Dave F. wrote:
> Why would routing software need to know the extent of a bridge?
so pedestrians aren't encouraged to jump to the road below??


___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Dave F.]

Roy Wallace wrote:
> On Tue, Sep 22, 2009 at 7:20 AM, Dave F.  wrote:
>   
>> Hi Peter
>> 
>>> I can suggest to add bridge's physical form to this. Otherwise yeah,
>>> why not. It would also help to indicate where bridge actually starts
>>> and ends to help routing software logic.
>>>
>>>
>>>   
>> Why would routing software need to know the extent of a bridge?
>> 
>
> OSM is not just for routers. It should be a map of what exists in the world.
>
>
>   
True, but I was asking Peter why he thinks that routing software would 
need to know that info.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Martin Koppenhoefer]

2009/9/21 Anthony :
> Right now, like it or not, ways are being used essentially in the manner I'm
> suggesting we keep using them.  We don't combine everything that crosses a
> bridge into a single way, and we don't break apart every lane into a
> separate way.  Those saying that we should do one of these two things are
> the ones trying to reinvent the wheel, and the burden of proof is on them to
> prove that their new wheel is superior.

It is needed for several situations that currently cannot be mapped
completely (different surfaces, different maxspeeds, different
turn-restrictions, etc.) and I see some cons of mapping this
explicitly:
- is probably more work to map every lane
- will probably be a bigger file to handle in various apps

but I also see the pros:
- is easier to understand and more intuitive to edit: better
maintenance: you see what is there
- is more precise in terms of positional accuracy
- can better represent stuff like ramps for motorways (parallel ways
that currently are mapped like intersections), cyclelanes, ...

so I just wait for even bigger parts of Germany being finished, so
people have time for this ;-)

cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 5:29 PM, Martin Koppenhoefer  wrote:

> 2009/9/21 Anthony :
> > Osmarender does a pretty good job of those bridges:
> >
> http://www.openstreetmap.org/?lat=27.95907&lon=-82.53907&zoom=17&layers=0B00FTF
> > Notice how it combines the two ways heading south into one bridge even
> > without any additional information besides the fact that the two ways are
> > close together.
>
> there are many cases of parallel bridges though, where you don't want
> them to render as one because they are two (or more), so being close
> cannot be the criteria.
>

It can if you factor in the width of the ways.  If the distance between the
ways is equal to the average width of the ways (give or take half a meter),
you've got one bridge, not parallel bridges.

That wouldn't solve the problem if there's a gap between the ways but one
bridge, though.  I agree it's better to represent the bridge separately.

On Mon, Sep 21, 2009 at 5:31 PM, Roy Wallace  wrote:

> On Tue, Sep 22, 2009 at 7:17 AM, Dave F.  wrote:
> > 2. if the bridge is independent, the size (width) can be drawn to suit
>
> What do you mean? If I don't know the width of the bridge, I can't draw it.
>

If you know the distance between the centers of the ways, you can calculate
the minimum width the bridge could possibly be: distance between the centers
of the ways times two.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Tue, Sep 22, 2009 at 7:17 AM, Dave F.  wrote:
> Roy
>>  Therefore, a
>> "bridge" relation would still be useful to indicate that multiple ways
>> share a bridge, for when the bridge geometry is unknown.
>>
>>
> 1. Why do the ways need to indicate they share a bridge?

No particular reason other than that it is a fact that exists in the
world. In other words, we should be able to map whether or not the
ways do or do not share a bridge. As a consequence, users of OSM (e.g.
renderers) would be able to make better use of the data (e.g. to
render bridges).

> 2. if the bridge is independent, the size (width) can be drawn to suit

What do you mean? If I don't know the width of the bridge, I can't draw it.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 4:47 PM, Roy Wallace  wrote:

> On Mon, Sep 21, 2009 at 11:32 PM, Anthony  wrote:
> >
> > John seems to combine everything into a single way and treat the
> individual
> > "lanes" (some of the substructures aren't even really lanes) as
> > substructures.  Some people want to break every lane into a separate way,
> > and combine them into superstructures.  Frankly, these two plans are
> > essentially equivalent.
>
> Except that a superstructure of elements is already implemented (a
> relation).
>

True, true, I was thinking more in the "what if we were inventing everything
from scratch" context.


> > Personally, I don't want either of those.  I want the way to be whatever
> > logical unit is used for routing.
>
> This seems selfish (but thanks for being honest :)). Not everyone uses
> OSM for routing. The ultimate, all-inclusive goal should always be an
> accurate map of what exists in the world.
>

Might have to change the name of the project :).

Yes, not everyone uses OSM for routing.  But many people do, so the
information has to be maintained.  And if you have the routing information,
you can generate the map.

I want everyone to be happy, I really do.  I'm willing to listen to
proposals which have every lane as a way, and proposals to have every (I
don't even know how to define it) as a way.  But I think you've got an
uphill battle trying to implement it without losing information.

Right now, like it or not, ways are being used essentially in the manner I'm
suggesting we keep using them.  We don't combine everything that crosses a
bridge into a single way, and we don't break apart every lane into a
separate way.  Those saying that we should do one of these two things are
the ones trying to reinvent the wheel, and the burden of proof is on them to
prove that their new wheel is superior.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Tue, Sep 22, 2009 at 7:20 AM, Dave F.  wrote:
> Hi Peter
>> I can suggest to add bridge's physical form to this. Otherwise yeah,
>> why not. It would also help to indicate where bridge actually starts
>> and ends to help routing software logic.
>>
>>
> Why would routing software need to know the extent of a bridge?

OSM is not just for routers. It should be a map of what exists in the world.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Martin Koppenhoefer]

2009/9/21 Anthony :
> Osmarender does a pretty good job of those bridges:
> http://www.openstreetmap.org/?lat=27.95907&lon=-82.53907&zoom=17&layers=0B00FTF
> Notice how it combines the two ways heading south into one bridge even
> without any additional information besides the fact that the two ways are
> close together.

there are many cases of parallel bridges though, where you don't want
them to render as one because they are two (or more), so being close
cannot be the criteria.

cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Dave F.]

Hi Peter
> I can suggest to add bridge's physical form to this. Otherwise yeah,
> why not. It would also help to indicate where bridge actually starts
> and ends to help routing software logic.
>
>   
Why would routing software need to know the extent of a bridge?

Cheers
Dave F.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Dave F.]

Roy
>  Therefore, a
> "bridge" relation would still be useful to indicate that multiple ways
> share a bridge, for when the bridge geometry is unknown.
>
>   
1. Why do the ways need to indicate they share a bridge?
2. if the bridge is independent, the size (width) can be drawn to suit)

Cheers
Dave F.


___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Tue, Sep 22, 2009 at 6:55 AM, Peteris Krisjanis  wrote:
>> Camp three: multiple ways representing paths of travel, grouped with a
>> "bridge" relation to indicate they share a common bridge. This could
>> probably be seen as a compromise, and is (I think) a good interim
>> solution, if not a very-long-term solution.
>>
>
> I can suggest to add bridge's physical form to this. Otherwise yeah,
> why not. It would also help to indicate where bridge actually starts
> and ends to help routing software logic.

By the way, for those interested, the relation we are suggesting is
essentially already documented here:
http://wiki.openstreetmap.org/wiki/Relations/Proposed/Bridges_and_Tunnels

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Peteris Krisjanis]

> Camp three: multiple ways representing paths of travel, grouped with a
> "bridge" relation to indicate they share a common bridge. This could
> probably be seen as a compromise, and is (I think) a good interim
> solution, if not a very-long-term solution.
>

I can suggest to add bridge's physical form to this. Otherwise yeah,
why not. It would also help to indicate where bridge actually starts
and ends to help routing software logic.

Cheers,
Peter.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Tue, Sep 22, 2009 at 12:20 AM, Anthony  wrote:
>
> A bridge should probably have its
> own geometry.  And if a bridge has its own geometry (polygon or line and
> width) and a layer tag you don't even need the relation, do you?  Anything
> in the area of the bridge with the same layer is located on the bridge.

I like the idea.

BUT I shouldn't have to measure the bridge to be able to indicate that
a section of way(s) goes over a bridge.

Using an area to mark a bridge in this way should be *optional*, in
the same was as a POI can be marked as a node or an area. Therefore, a
"bridge" relation would still be useful to indicate that multiple ways
share a bridge, for when the bridge geometry is unknown.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Mon, Sep 21, 2009 at 11:32 PM, Anthony  wrote:
>
> John seems to combine everything into a single way and treat the individual
> "lanes" (some of the substructures aren't even really lanes) as
> substructures.  Some people want to break every lane into a separate way,
> and combine them into superstructures.  Frankly, these two plans are
> essentially equivalent.

Except that a superstructure of elements is already implemented (a relation).

> Personally, I don't want either of those.  I want the way to be whatever
> logical unit is used for routing.

This seems selfish (but thanks for being honest :)). Not everyone uses
OSM for routing. The ultimate, all-inclusive goal should always be an
accurate map of what exists in the world.

(By the way, don't forget that a "way" is just "a series of nodes". A
"way" itself has no meaning until given one with tags. For example, a
building outline is a way, but you certainly can't be routed around a
building outline.)

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Mon, Sep 21, 2009 at 11:15 PM, Lester Caine  wrote:
>
> > Can I check? Are there people here who are suggesting that, in my case,
> > they want to draw all four ways (yes, I know the footpath hasn't been
> > mapped yet) as  a single way & specify the differences with lane tags?
>
> Camp one is - single way with lanes=4 + bridge section
> Camp two would prefer all elements mapped in which case the bridge WOULD
> be a separate element.

Camp three: multiple ways representing paths of travel, grouped with a
"bridge" relation to indicate they share a common bridge. This could
probably be seen as a compromise, and is (I think) a good interim
solution, if not a very-long-term solution.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Dave F.]

Dave F. wrote:
> Anthony wrote:
>> On Mon, Sep 21, 2009 at 3:06 PM, Anthony > > wrote:
>>
>> On Mon, Sep 21, 2009 at 9:32 AM, Anthony > > wrote:
>>
>> On Mon, Sep 21, 2009 at 6:20 AM, Richard Mann
>> > > wrote:
>>
>> If we were just gathering data for rendering a
>> single-scale street map, we'd add tags to a single way,
>> and probably not bother with lane info.
>>
>>
>> Depends how micro you want to get.  With the detail people
>> seem to want, you'd probably be best off with a paint program 
>> :).
>>
>>
>> I'd love to one day see an open source map that looks like this:
>> 
>> http://www.mytbi.com/urs/content/design/linksstagei/images/aerials/v8-pr.jpg 
>>
>> (minus the part that's a photograph).
>>
>>
>> Osmarender does a pretty good job of those bridges: 
>> http://www.openstreetmap.org/?lat=27.95907&lon=-82.53907&zoom=17&layers=0B00FTF
>>  
>> 
>>   
>> Notice how it combines the two ways heading south into one bridge 
>> even without any additional information besides the fact that the two 
>> ways are close together.
> Not saying you mapped it but it appears to be missing a trunk_link to 
> the north-west.
> Also, shouldn't the S.R.589 be a bridge directly below it label in 
> that image.
> And the link to the North East, doesn't that pass over a river of some 
> sorts?
> Any idea what that aeroway-bridge is for?
>
>
Ah! just overlayed Yahoo. Is the interchange going or gone through a 
major redevelopment?


___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 3:06 PM, Anthony  wrote:

> On Mon, Sep 21, 2009 at 9:32 AM, Anthony  wrote:
>
>> On Mon, Sep 21, 2009 at 6:20 AM, Richard Mann <
>> richard.mann.westoxf...@googlemail.com> wrote:
>>
>>> If we were just gathering data for rendering a single-scale street map,
>>> we'd add tags to a single way, and probably not bother with lane info.
>>>
>>
>> Depends how micro you want to get.  With the detail people seem to want,
>> you'd probably be best off with a paint program :).
>>
>
> I'd love to one day see an open source map that looks like this:
> http://www.mytbi.com/urs/content/design/linksstagei/images/aerials/v8-pr.jpg(minus
>  the part that's a photograph).
>

Osmarender does a pretty good job of those bridges:
http://www.openstreetmap.org/?lat=27.95907&lon=-82.53907&zoom=17&layers=0B00FTF
Notice how it combines the two ways heading south into one bridge even
without any additional information besides the fact that the two ways are
close together.

Mapnik, not as much:
http://www.openstreetmap.org/?lat=27.95873&lon=-82.53908&zoom=17&layers=B000FTF

I don't believe either renderer factors in width or lane information, but I
haven't added that (unfortunately, I apparently can't use the image from
mytbi as a source for the number of lanes, because it isn't public domain).
Add in lane information, and width information (which is already supported
by the current api), and we're getting close to what I see as ideal.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 9:32 AM, Anthony  wrote:

> On Mon, Sep 21, 2009 at 6:20 AM, Richard Mann <
> richard.mann.westoxf...@googlemail.com> wrote:
>>
>> If we were just gathering data for rendering a single-scale street map,
>> we'd add tags to a single way, and probably not bother with lane info.
>>
>
> Depends how micro you want to get.  With the detail people seem to want,
> you'd probably be best off with a paint program :).
>

I'd love to one day see an open source map that looks like this:
http://www.mytbi.com/urs/content/design/linksstagei/images/aerials/v8-pr.jpg(minus
the part that's a photograph).

That's got lane information and much much more, and it's probably pretty
much doable in theory.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[d f]

From: Anthony 
To: d f 
Cc: OSM Talk 
Sent: Monday, 21 September, 2009 15:20:43
Subject: Re: [OSM-talk] Should Bridges be independent of their ways?


I think the important question is, does it add information?  Probably so.  A 
bridge really is more than just a collection of ways.  It might be 
significantly larger than the ways on it.  A bridge should probably have its 
own geometry.  And if a bridge has its own geometry (polygon or line and width) 
and a layer tag you don't even need the relation, do you?  Anything in the area 
of the bridge with the same layer is located on the bridge.

+1

The only issue I see is when when a bridge only consists of a single way, it'd 
be a pain to add *another* way, with the same geometry, to represent the 
bridge.  So the renderers would have to special case this.  Maybe

+1

Okay, I have a proposal.  I can bet some people are going to hate me for it, 
but I'm going to propose it anyway...

amenity=bridge (or would it be landuse=bridge?), to be attached to a way or 
polygon.  layer tag is used to indicate the layer.  If a bridge is equivalent 
to a single way, you can attach amenity/landuse=bridge to the way (after 
splitting) instead of creating a separate way.

bridge=yes could, and probably should, still be attached to the way.  It will 
indicate that the way is *on* (over?) a bridge, not that the way *is* a bridge.

No relations, unless you want to add them as redundant information to make it 
easier to calculate which ways are on which bridges (but this can be obtained 
from the geometry, the layer tag, and the bridge tag).


>Would it affect routers? Would a route be described as "cross this bridge, 
>then turn left in 200 metres"?
>

I doubt most routers are going to bother with information that isn't part of 
the way or the nodes directly on the way.

To be clearer I should have said "Turn left 200 metres after crossing this 
river"
To answer my own question, I think they would use such vernacular.  


>It would certain save time splitting the ways.
>

The way should probably still be split, at least to add the layer tag, and 
arguably to add the bridge=yes, which indicates that the way is indeed on a 
bridge.

I've spent so much time splitting ways for bridges so it's with regret that I 
agree they need to be split to define layers. 
+1



  ___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Alex Mauer]

On 09/21/2009 09:20 AM, Anthony wrote:
> On Mon, Sep 21, 2009 at 8:21 AM, d f  wrote:
> amenity=bridge (or would it be landuse=bridge?), to be attached to a way or
> polygon.  

manmade=bridge?



signature.asc
Description: OpenPGP digital signature
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 8:21 AM, d f  wrote:

> It was more an off the top of my head comment really.
> Would having it independent make it easier for the renderers?
>

I think the important question is, does it add information?  Probably so.  A
bridge really is more than just a collection of ways.  It might be
significantly larger than the ways on it.  A bridge should probably have its
own geometry.  And if a bridge has its own geometry (polygon or line and
width) and a layer tag you don't even need the relation, do you?  Anything
in the area of the bridge with the same layer is located on the bridge.

The only issue I see is when when a bridge only consists of a single way,
it'd be a pain to add *another* way, with the same geometry, to represent
the bridge.  So the renderers would have to special case this.  Maybe

Okay, I have a proposal.  I can bet some people are going to hate me for it,
but I'm going to propose it anyway...

amenity=bridge (or would it be landuse=bridge?), to be attached to a way or
polygon.  layer tag is used to indicate the layer.  If a bridge is
equivalent to a single way, you can attach amenity/landuse=bridge to the way
(after splitting) instead of creating a separate way.

bridge=yes could, and probably should, still be attached to the way.  It
will indicate that the way is *on* (over?) a bridge, not that the way *is* a
bridge.

No relations, unless you want to add them as redundant information to make
it easier to calculate which ways are on which bridges (but this can be
obtained from the geometry, the layer tag, and the bridge tag).

Would it affect routers? Would a route be described as "cross this bridge,
> then turn left in 200 metres"?
>

I doubt most routers are going to bother with information that isn't part of
the way or the nodes directly on the way.

It would certain save time splitting the ways.
>

The way should probably still be split, at least to add the layer tag, and
arguably to add the bridge=yes, which indicates that the way is indeed on a
bridge.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Mon, Sep 21, 2009 at 6:20 AM, Richard Mann <
richard.mann.westoxf...@googlemail.com> wrote:

> If we were just gathering data for routers, we would map every lane as a
> separate way, with relations for moving between each pair of adjacent lanes.
>

I disagree with that.  Dealing with relations when mapping a route is an
unnecessary burden.


> If we were just gathering data for rendering a single-scale street map,
> we'd add tags to a single way, and probably not bother with lane info.
>

Depends how micro you want to get.  With the detail people seem to want,
you'd probably be best off with a paint program :).


> I think routers ought to be able to cope with a single-way structure with
> extended tags. I think renderers would struggle (even more) to produce a
> good-looking multi-scale map if the data is held as multiple ways.
>

> I think the most generally-usable structure is a single way with lane info.
> This could be done with tags (and probably needs to be, to start with),
> though you might hope that editors would find a way of presenting it so that
> it looks like a set of parallel ways with tied nodes.
>

John seems to combine everything into a single way and treat the individual
"lanes" (some of the substructures aren't even really lanes) as
substructures.  Some people want to break every lane into a separate way,
and combine them into superstructures.  Frankly, these two plans are
essentially equivalent.

Personally, I don't want either of those.  I want the way to be whatever
logical unit is used for routing.  Then, you can take those ways and break
them into lanes (so long as the lanes do not have their own geometries
except maybe one node per lane at each endpoint).  *Or* you can combine
those ways into multiway structures like bridges.

As long as you keep all the detail provided by that, I don't really care how
you implement it, though.  If you implement it in a way which is burdensome
to routing software, the routing software can just reformat it before
processing, so long as you have all the data.  I think it's silly to combine
multiple ways into one and then create a structure to break them apart
again.  And I think it's silly to break ways apart and then create a
structure to put them back together again.  But if you want to go through
that process, so long as you keep the information that's currently there,
I'm not going to object.

I will object to any plan to combine multiple ways into one which *doesn't*
maintain the information provided by having multiple ways.  And the
pseudocode John last presented did that, plus my discussion with him
indicates that he doesn't understand the purpose of keeping ways separate in
the first place (the contradictory comments about a bridge being a single
way and a physically separated road being more than one way indicate this
clearly).



On Mon, Sep 21, 2009 at 8:21 AM, d f  wrote:

> Back to basics:
> Are we all agreed that, in principle, it would be better to be able to
> have a single bridge to carry multiple ways? 
>

I'm not totally convinced that this isn't just a software issue.  If the
renderers drew the ways as the proper width, they could avoid drawing bridge
indicators in places where there isn't any space between the roadways.

It wouldn't hurt to add this information even if it's redundant.  But I'm
not convinced it isn't redundant.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Lester Caine]

d f wrote:
> My Lord! What happened to my question?!?
> You lot don't half go on. :-)
It's the same problem ;)
micro mapping requires detail
macro model requires everything linked to the ways

> Back to basics:
> Are we all agreed that, in principle, it would be better to be able to 
> have a single bridge to carry multiple ways?
> 
> http://osm.org/go/eukOONRtk--
> 
> http://maps.google.co.uk/maps?ie=UTF8&ll=51.380995,-2.350388&spn=0.075105,0.16531&z=13
>  
> 
> 
> To answer Martin's query about a bridges independence:
> When editing, if you selected a bridge & moved it, it wouldn't move any 
> of the ways going over.
> 
> It was more an off the top of my head comment really.
> Would having it independent make it easier for the renderers?
> Would it affect routers? Would a route be described as "cross this 
> bridge, then turn left in 200 metres"?
> It would certain save time splitting the ways.
>  
> If there are reasons why the bridge needs to be tagged with the 
> different types of ways please let me know.
> 
> The bridge would have a width tag which the mapper would adjust to suit 
> all ways, thus saving the renderer the calculation.
THAT is the key point here
Width tags and other flags - against accurate mapping of details

> Can I check? Are there people here who are suggesting that, in my case, 
> they want to draw all four ways (yes, I know the footpath hasn't been 
> mapped yet) as  a single way & specify the differences with lane tags?
Camp one is - single way with lanes=4 + bridge section
Camp two would prefer all elements mapped in which case the bridge WOULD 
be a separate element.
At the moment we have a hoch podge somewhere in between, with people 
splitting ways and adding detail and others putting extra tags that 
identify the situation without providing the detail.

> Also I thought street was a "band" word & anybody quoting it was given a 
> slap on the wrist & told to go & stand on the naughty step :-)
Lots of streets around here, but the only things banned would be 
vehicles on the pedestrian only ones ;)

Anyway
Top down view - single way with lots of tags such as 'bridge'.
Bottom up view - every detail mapped and 'linked' in some way but each 
with it's own structural element.

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[d f]

Oh dear, I, of course, meant banned not band.  



  ___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[d f]

My Lord! What happened to my question?!?
You lot don't half go on. :-)

Back to basics:
Are we all agreed that, in principle, it would be better to be able to have a 
single bridge to carry multiple ways?

http://osm.org/go/eukOONRtk--

http://maps.google.co.uk/maps?ie=UTF8&ll=51.380995,-2.350388&spn=0.075105,0.16531&z=13

To answer Martin's query about a bridges independence:
When editing, if you selected a bridge & moved it, it wouldn't move any of the 
ways going over.

It was more an off the top of my head comment really. 
Would having it independent make it easier for the renderers?
Would it affect routers? Would a route be described as "cross this bridge, then 
turn left in 200 metres"?
It would certain save time splitting the ways.
 
If there are reasons why the bridge needs to be tagged with the different types 
of ways please let me know.

The bridge would have a width tag which the mapper would adjust to suit all 
ways, thus saving the renderer the calculation.

Can I check? Are there people here who are suggesting that, in my case, they 
want to draw all four ways (yes, I know the footpath hasn't been mapped yet) as 
 a single way & specify the differences with lane tags?

Also I thought street was a "band" word & anybody quoting it was given a slap 
on the wrist & told to go & stand on the naughty step :-)

Cheers
Dave F.



  ___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Peteris Krisjanis]

Problem is rather simple with this actually right on target aproach -
current practice and "it is too damn boring and difficult" or is also
known as "death by thousand paper cuts".

I like micromapping - I hate current quality of routing software and I
think there is whole posibility to have it more precise and "smarter".
For that we need precise maps (for all kind of traffic at least).
However let's admit - it is currently available only on selected
regions, cities or even districts of cities or towns. So move from
"highway" tag (who only last year have found solid ground to work on)
to seperate lines + relation ASAP wouldn't be very smart for now.

Maybe this need a vision bigger than one proposal? I would suggest a
plan with points to move slowly forward to posibility to have maximum
features to do micromapping.

I would vote for using highway as usual for places where is no
difference, and where it comes to split, split off seperate lines
(tagged as lines=1) with clear indication which is which.

Then we could go futher with discussing how to reform ways/lanes.

However, it doesn't change a original topic - I would like to see
bridges as seperated physical entities with different ways on the top
of it. It *is* wierd when two lines on the bridge are seperated.

Just my thoughts,
Cheers,
Peter

2009/9/21 Martin Koppenhoefer :
> 2009/9/21 Lester Caine :
>> Simple answer to that BULLSHIT.
>
> I agree even if maybe I would have put it in less harsh words ;-)
>
>> The only way to produce fine detail maps is with the correct
>> information. Adding a footpath at the side of a road which along the
>> length of the road meandering from one side to the other, is adjacent to
>> the road in places, or separated by a grass verge, or isolated by bushes
>> can not easily be managed by tags added to the road. Add the pull off
>> for a bus stop, with the changes in the path of the pavement for that.
>> The changes as a footpath merges with other obstructions at road
>> junctions ... and so on. Getting the renderer to guess what the real
>> situation is even with quite complex tags does not make sense.
>
> +1
>
>> A single way simply does not work at the micro level unless you make the
>> tagging so complex that only a computer could understand it? And it
>> would have to have distances as part of the tags so you know how far the
>> edge of the path is from the arbitrary center line of the base way.
>
> +1
>
> cheers,
> Martin
>
> ___
> talk mailing list
> talk@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>



-- 
mortigi tempo
Pēteris Krišjānis

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Martin Koppenhoefer]

2009/9/21 Lester Caine :
> Simple answer to that BULLSHIT.

I agree even if maybe I would have put it in less harsh words ;-)

> The only way to produce fine detail maps is with the correct
> information. Adding a footpath at the side of a road which along the
> length of the road meandering from one side to the other, is adjacent to
> the road in places, or separated by a grass verge, or isolated by bushes
> can not easily be managed by tags added to the road. Add the pull off
> for a bus stop, with the changes in the path of the pavement for that.
> The changes as a footpath merges with other obstructions at road
> junctions ... and so on. Getting the renderer to guess what the real
> situation is even with quite complex tags does not make sense.

+1

> A single way simply does not work at the micro level unless you make the
> tagging so complex that only a computer could understand it? And it
> would have to have distances as part of the tags so you know how far the
> edge of the path is from the arbitrary center line of the base way.

+1

cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Martin Koppenhoefer]

2009/9/20 John Smith :
> 2009/9/21 Anthony :
>
>> Irrelevant.  I never said you had to use relations.  In fact, I said you
>> don't.
>
> Others have suggested otherwise, to "group" ways that are on the same
> physical bridge.

why should that be "abusing" relations? There are several proposals
which suggest exactly that (bridge and tunnels, street, collected
way).

cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Lester Caine]

Richard Mann wrote:
> If we were just gathering data for routers, we would map every lane as a 
> separate way, with relations for moving between each pair of adjacent 
> lanes. If we were just gathering data for rendering a single-scale 
> street map, we'd add tags to a single way, and probably not bother with 
> lane info.
>  
> I think routers ought to be able to cope with a single-way structure 
> with extended tags. I think renderers would struggle (even more) to 
> produce a good-looking multi-scale map if the data is held as multiple ways.
>  
> I think the most generally-usable structure is a single way with lane 
> info. This could be done with tags (and probably needs to be, to start 
> with), though you might hope that editors would find a way of presenting 
> it so that it looks like a set of parallel ways with tied nodes.

Simple answer to that BULLSHIT.

The only way to produce fine detail maps is with the correct 
information. Adding a footpath at the side of a road which along the 
length of the road meandering from one side to the other, is adjacent to 
the road in places, or separated by a grass verge, or isolated by bushes 
can not easily be managed by tags added to the road. Add the pull off 
for a bus stop, with the changes in the path of the pavement for that. 
The changes as a footpath merges with other obstructions at road 
junctions ... and so on. Getting the renderer to guess what the real 
situation is even with quite complex tags does not make sense.

A single way simply does not work at the micro level unless you make the 
tagging so complex that only a computer could understand it? And it 
would have to have distances as part of the tags so you know how far the 
edge of the path is from the arbitrary center line of the base way.

I *AM* coming to the view that we perhaps need another layer of 
information? AT the macro level it is simply a way, but at the micro 
level a way would consist of all the detail necessary to accurately map 
the reality on the ground? But whether the macro view would need the 
'path=left', 'path=both' information as tags is questionable?

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Richard Mann :
> If we were just gathering data for routers, we would map every lane as a
> separate way, with relations for moving between each pair of adjacent lanes.
> If we were just gathering data for rendering a single-scale street map, we'd
> add tags to a single way, and probably not bother with lane info.

Not everything is tagged for the purpose of being rendered, nor should it.

> I think the most generally-usable structure is a single way with lane info.
> This could be done with tags (and probably needs to be, to start with),
> though you might hope that editors would find a way of presenting it so that
> it looks like a set of parallel ways with tied nodes.

We basically need to treat ways as a special type of relation, one
that has a number of lanes which can be tagged independent of the way
otherwise inherent the tags of the way.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Richard Mann]

If we were just gathering data for routers, we would map every lane as a
separate way, with relations for moving between each pair of adjacent lanes.
If we were just gathering data for rendering a single-scale street map, we'd
add tags to a single way, and probably not bother with lane info.

I think routers ought to be able to cope with a single-way structure with
extended tags. I think renderers would struggle (even more) to produce a
good-looking multi-scale map if the data is held as multiple ways.

I think the most generally-usable structure is a single way with lane info.
This could be done with tags (and probably needs to be, to start with),
though you might hope that editors would find a way of presenting it so that
it looks like a set of parallel ways with tied nodes.

Richard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Lester Caine]

Anthony wrote:
> On Sun, Sep 20, 2009 at 4:57 PM, Roy Wallace  > wrote:
> 
> On Mon, Sep 21, 2009 at 3:24 AM, Anthony  > wrote:
>  >
>  > Because that's the primary purpose for which maps are created. 
> To inform us
>  > how to get from place to place.
> 
> Be careful...big assumption.
> 
> You're right, I'm assuming too much.

In some parts of the world simply having a route from a to b would be 
nice to have, and I'm including areas of the world where a lot of detail 
has already been mapped.

BUT in a large area of the world we now have much of the coarse detail 
mapped, and it IS the fine detail which is now needed.

While a single way CAN be used in a lot of places, it has reached the 
point where expanding a way into multiple parallel elements is actually 
essential to add the fine detail. Not just for vehicle 'carriageways' on 
a road, but also to add sidewalk details for pedestrians and bikes.

In the case of bridges, where the main function is moving vehicles, 
there may be a pedestrian way as well, or this may take an alternative 
route. Trying to decipher what traffic is allowed down a way and what 
has to take an alternative track when producing a pedestrian or cycle 
route as opposed to a vehicle route ( and adding lorry restrictions to 
that ) could be handled by complex tagging of a single way. The fact 
that pedestrians can simply run across a motorway if they feel inclined 
does complicate the routing question, and on roads where simply crossing 
anywhere is practical, this needs to be catered for. Once a route 
becomes a complex set of linked ways, then tagging each element with 
'bridge' DOES become questionable, and a separate bridge element makes 
sense as well.

Should the tagging handle both situations - macro and micro mapping - YES
Does it currently - NO

We are now at a point where the data needs a number of levels of 
complexity, and ideas like having the editors display ways as multiple 
parallel tracks makes perfect sense - especially when you add sidewalk 
and other detail to that. We have perhaps reached the point where the 
level of complexity returned should depend on the zoom selected? Taking 
a large zoom returns single track ways, while small scale maps return 
all the fine detail of footpaths, road widths and location of road 
markings like visually impaired persons crossing markers 
AND if the fine detail has not been specifically mapped yet, it is 
inferred from the information available in the larger scale tags.

In my own case, mapping the footpaths down the side of the roads around 
here, the problem is showing which side of the road the walkway is, and 
where it takes an alternative route to the actual road route, perhaps 
even taking a completely different track. The tags to add these details 
to a single way are not currently documented, and neither are the 
guidelines for creating a multiple way route and flaging it as a single 
element :(

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> If it doesn't affect any routing information, then we shouldn't, because
> it's a waste of time, but I don't mind if you do.

It is a waste of time, I'm just trying to show that ways are treated
differently depending on the type of way, which doesn't reflect
reality and we should be mapping reality.

> That could be more easily represented by some sort of "uturn=no" though.

My point is, just because you can't u-turn doesn't mean it should be
dual carriage way

> You'll have to show me an intersection with such lines, or the law.  This
> looks to me like a passing restriction.

I did, one side of the centre line is solid, the other is broken, you
cannot cross from the solid side, you can from the broken side.

> Well, let's take this intersection:
> http://maps.google.com/maps?f=q&ie=UTF8&layer=c&cbll=-26.124795,152.574151&panoid=H08s6qv1gLXcd8hGtNhvwg&cbp=12,333.55,,0,2.6&ll=-26.124704,152.574123&spn=0,359.996175&z=18
>
> If you want to micromap that, I'd go with dual carriageways.

But it isn't dual carriage way, and it makes no sense to map it as
such, it would be nice to be able to map the lanes correctly however.

> Do we need a method to represent a single bridge with multiple ways?  Any
> suggestions?

Being able to tag lanes along a way maybe?

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> Okay, I looked up the law.  A double white continuous line is equivalent to
> our double double yellow line.  You may not pass across it, and you may not
> enter or leave the road across it (such as at a driveway).  So I would
> represent that as a dual carriageway, at least in any situation where
> routing software might otherwise direct someone to cross it (such as when
> making a right turn into a driveway).

I already told you this, however it doesn't make any sense to map
roads like that, so I don't know why other roads with physically
joined lanes should be treated any differently.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 4:57 PM, Roy Wallace  wrote:

> On Mon, Sep 21, 2009 at 3:24 AM, Anthony  wrote:
> >
> > Because that's the primary purpose for which maps are created.  To inform
> us
> > how to get from place to place.
>
> Be careful...big assumption.
>

You're right, I'm assuming too much.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Mon, Sep 21, 2009 at 4:12 AM, Anthony  wrote:
>
> On Sun, Sep 20, 2009 at 2:09 PM, John Smith 
> wrote:
>>
>> 2009/9/21 Anthony :
>> > Define "physically separated".
>>
>> The road base finishes or there is a concreate barrier or other form
>> of barrier that would prevent a car crossing, legally or otherwise.
>>
>> > Do you admit that a physical bridge can carry multiple ways, or not?  I
>>
>> A bridge is a single way
>
> Even if there is a concrete barrier or other form of barrier that would
> prevent a car crossing?  Why are bridges different?

Yep, John, you've contradicted yourself.

You said "A way is a physical feature physically seperated from other ways"

Then you said "physcially separated" means e.g. "a concrete barrier".

Therefore, a bridge with a concrete barrier separating lanes *carries
multiple ways* - which you then deny, saying "A bridge is a single
way".

Please clarify your interpretation of a *way*, or of *physical
separation*, so that you make sense.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Liz]

On Mon, 21 Sep 2009, Anthony wrote:
> Because that's the primary purpose for which maps are created.  To inform
> us how to get from place to place.


No, a map is a statement about the world, and a person's view of that world. 
That's why I mentally marked up 
> > However that doesn't reflect reality
>
> It does reflect reality.  It just reflects a different part of reality than
> you want it to reflect.

because there is no answer to some conflicts.

Maps are political statements too, which is why the Mercator projection is so 
popular in European classrooms and books.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Mon, Sep 21, 2009 at 4:09 AM, John Smith  wrote:
>
> If we are tagging lanes not ways then we need to do so en masse, not
> have 2 different systems that exists at present.

First things first - 1) propose a way to tag lanes, 2) start using it
in parallel to the current system, and see if it works. I doubt you
will be able to achieve anything more than gradual adoption, if any.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Mon, Sep 21, 2009 at 3:27 AM, John Smith  wrote:
>
> it simply isn't treated that way in all
> situation or each residential street would be 2 parallel ways, instead
> we use a single way to indication a pair of lanes, so which is it
> going to be do we need to split residential streets or do we treat
> things as a piece of asphalt when it suits us?

I think splitting ways "when it suits us" is actually a good pragmatic approach.

Having a clearly defined and documented standard for tagging
individual lanes is a good goal, but while we wait for that (John?),
using individual ways "when it suits us", and relating them with a
relation is at least probably a pretty good *interim* solution (or
long-term solution, if an individual lane-tagging proposal doesn't
gain traction...).

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Mon, Sep 21, 2009 at 3:24 AM, Anthony  wrote:
>
> Because that's the primary purpose for which maps are created.  To inform us
> how to get from place to place.

Be careful...big assumption.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Jason Gerecke]

On Sun, Sep 20, 2009 at 11:19 AM, John Smith wrote:

> 2009/9/21 Anthony :
>
>  > Even if there is a concrete barrier or other form of barrier that would
> > prevent a car crossing?  Why are bridges different?
>
> Usually because bridges narrow things to make it cheaper to cross a
> river etc, however my point still stands a way is a physical thing, if
> there is only one physical thing then that's all that should be mapped
> and lanes should be tagged independently of the way.
>
>
A way is not a physical thing. A way is just a group of nodes in the OSM
database. What that group of nodes represents is entirely up to the
community. We have ways that represent pieces of asphalt, but we also have
ways that represent completely non-physical city outlines. If I got people
to agree to the convention, ways could represent the airways that planes
often fly along.

I personally feel that when it comes to mapping roads, we should be mapping
the *network* (that is, the paths of travel), not the *asphalt*. If
individual paths of travel cannot be accurately represented with a single
way, they should be represented with multiple ways. Since a car cannot drive
on a bridge's bike path, there should be multiple ways comprising the single
bridge.

Jason G.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 3:23 PM, Anthony  wrote:

> On Sun, Sep 20, 2009 at 3:03 PM, John Smith wrote:
>
>> 2009/9/21 Anthony :
>>
> > I'm not talking about where passing is allowed, I'm talking about where
>> > turning is allowed.  In any case, once again there appear to be no
>> > intersections or places to turn, in which case it doesn't really matter.
>>
>> the link above you can only cross the road depending on the lane you are
>> in.
>>
>
> You'll have to show me an intersection with such lines, or the law.  This
> looks to me like a passing restriction.
>

Okay, I looked up the law.  A double white continuous line is equivalent to
our double double yellow line.  You may not pass across it, and you may not
enter or leave the road across it (such as at a driveway).  So I would
represent that as a dual carriageway, at least in any situation where
routing software might otherwise direct someone to cross it (such as when
making a right turn into a driveway).
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 3:03 PM, John Smith wrote:

> 2009/9/21 Anthony :
> > If you're allowed to cross it, for instance to make a turn, it should be
> > represented as one way.  If you aren't, it shouldn't.  In Florida and I
>
> As I point out below, you can't turn depending on the centre line not
> being solid. Should we create multiple ways for intermittent areas
> where you can't cross or not?
>

If it doesn't affect any routing information, then we shouldn't, because
it's a waste of time, but I don't mind if you do.

>>
> http://maps.google.com/maps?f=q&ie=UTF8&t=h&ll=-26.134596,152.582068&spn=0,359.990988&z=18&layer=c&cbll=-26.134719,152.583365&panoid=Zqx7qYT6v-zeBfuZnmywdA&cbp=12,270.23,,0,5.75
> >
> > In this case there appear to be no intersections or places to turn, in
> which
> > case it doesn't really matter.
>
> Irrelevent since you keep bringing up u-turns you can't do that either.
>

That could be more easily represented by some sort of "uturn=no" though.

> I'm not talking about where passing is allowed, I'm talking about where
> > turning is allowed.  In any case, once again there appear to be no
> > intersections or places to turn, in which case it doesn't really matter.
>
> the link above you can only cross the road depending on the lane you are
> in.
>

You'll have to show me an intersection with such lines, or the law.  This
looks to me like a passing restriction.

> If you can direct me to a site which explains these lines and what they
> > mean, I can give you a further response.
>
> It's really simple, solid line = you can't cross to turn in any
> respect, colour of the paint is irrelevent they used to use yellow
> they switched to white the law is still the same.
>

Well, let's take this intersection:
http://maps.google.com/maps?f=q&ie=UTF8&layer=c&cbll=-26.124795,152.574151&panoid=H08s6qv1gLXcd8hGtNhvwg&cbp=12,333.55,,0,2.6&ll=-26.124704,152.574123&spn=0,359.996175&z=18

If you want to micromap that, I'd go with dual carriageways.

>> You didn't express this opinion earlier, you were trying to show
> >> multiple ways where there is no physical barrier.
> >
> > Okay, fine, so we are in agreement?  All this stuff about mapping
> individual
> > lanes is off-topic?  We need a method to represent a single bridge with
> > multiple ways?  Any suggestions?
>
> How could we be in agreement you've completely ignored my last point.
>

Do we need a method to represent a single bridge with multiple ways?  Any
suggestions?
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :
> If you're allowed to cross it, for instance to make a turn, it should be
> represented as one way.  If you aren't, it shouldn't.  In Florida and I

As I point out below, you can't turn depending on the centre line not
being solid. Should we create multiple ways for intermittent areas
where you can't cross or not?

>> http://maps.google.com/maps?f=q&ie=UTF8&t=h&ll=-26.134596,152.582068&spn=0,359.990988&z=18&layer=c&cbll=-26.134719,152.583365&panoid=Zqx7qYT6v-zeBfuZnmywdA&cbp=12,270.23,,0,5.75
>
> In this case there appear to be no intersections or places to turn, in which
> case it doesn't really matter.

Irrelevent since you keep bringing up u-turns you can't do that either.

> I'm not talking about where passing is allowed, I'm talking about where
> turning is allowed.  In any case, once again there appear to be no
> intersections or places to turn, in which case it doesn't really matter.

the link above you can only cross the road depending on the lane you are in.

> If you can direct me to a site which explains these lines and what they
> mean, I can give you a further response.

It's really simple, solid line = you can't cross to turn in any
respect, colour of the paint is irrelevent they used to use yellow
they switched to white the law is still the same.

>> You didn't express this opinion earlier, you were trying to show
>> multiple ways where there is no physical barrier.
>
> Okay, fine, so we are in agreement?  All this stuff about mapping individual
> lanes is off-topic?  We need a method to represent a single bridge with
> multiple ways?  Any suggestions?

How could we be in agreement you've completely ignored my last point.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 2:39 PM, John Smith wrote:

> 2009/9/21 Anthony :
> > Mapping ways should follow the legal paths of travel, not the existence
> or
> > non-existence of concrete.  If concrete is the only form of legal
> barrier,
> > then fine, concrete can determine how we map.  But if a painted median is
>
> Where do you draw the line over painted median strips exactly,


If you're allowed to cross it, for instance to make a turn, it should be
represented as one way.  If you aren't, it shouldn't.  In Florida and I
believe most of the US, the legal distinction would be double yellow (
http://www.golocalnet.net/drive/dblyel.htm) means no passing, and should be
represented as one way; double double yellow (
http://www.golocalnet.net/drive/dblyel2.htm) means no crossing, and should
be represented as a dual carriageway.  But that obviously has to be adapted
for each jurisdiction.

the Bruce highway for example allows you to cross in some places and
> doesn't in others, but this is more a case of where it's safe to over
> take or not, should these sections where it's not safe to over take
> mean we draw 2 distinct ways as a result?
>
>
> http://maps.google.com/maps?f=q&ie=UTF8&t=h&ll=-26.134596,152.582068&spn=0,359.990988&z=18&layer=c&cbll=-26.134719,152.583365&panoid=Zqx7qYT6v-zeBfuZnmywdA&cbp=12,270.23,,0,5.75
>

In this case there appear to be no intersections or places to turn, in which
case it doesn't really matter.


> Also how would you do this:
>
>
> http://maps.google.com/maps?f=q&ie=UTF8&t=h&ll=-26.134789,152.590678&spn=0,359.986063&z=17&layer=c&cbll=-26.134677,152.590969&panoid=eVbTF6vxWoolDxDVBQpWNg&cbp=12,77.27,,0,5.07
>
> It's legal to cross the road from one side, but not the other, etc etc
> etc...
>

I'm not talking about where passing is allowed, I'm talking about where
turning is allowed.  In any case, once again there appear to be no
intersections or places to turn, in which case it doesn't really matter.

If you can direct me to a site which explains these lines and what they
mean, I can give you a further response.


> > But if there is a concrete barrier in place, you agree we might have two
> > ways going over one bridge.  So all this stuff about mapping individual
> > lanes is off-topic.
>
> You didn't express this opinion earlier, you were trying to show
> multiple ways where there is no physical barrier.
>

Okay, fine, so we are in agreement?  All this stuff about mapping individual
lanes is off-topic?  We need a method to represent a single bridge with
multiple ways?  Any suggestions?
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :
> Mapping ways should follow the legal paths of travel, not the existence or
> non-existence of concrete.  If concrete is the only form of legal barrier,
> then fine, concrete can determine how we map.  But if a painted median is

Where do you draw the line over painted median strips exactly, the
Bruce highway for example allows you to cross in some places and
doesn't in others, but this is more a case of where it's safe to over
take or not, should these sections where it's not safe to over take
mean we draw 2 distinct ways as a result?

http://maps.google.com/maps?f=q&ie=UTF8&t=h&ll=-26.134596,152.582068&spn=0,359.990988&z=18&layer=c&cbll=-26.134719,152.583365&panoid=Zqx7qYT6v-zeBfuZnmywdA&cbp=12,270.23,,0,5.75

Also how would you do this:

http://maps.google.com/maps?f=q&ie=UTF8&t=h&ll=-26.134789,152.590678&spn=0,359.986063&z=17&layer=c&cbll=-26.134677,152.590969&panoid=eVbTF6vxWoolDxDVBQpWNg&cbp=12,77.27,,0,5.07

It's legal to cross the road from one side, but not the other, etc etc etc...

> But if there is a concrete barrier in place, you agree we might have two
> ways going over one bridge.  So all this stuff about mapping individual
> lanes is off-topic.

You didn't express this opinion earlier, you were trying to show
multiple ways where there is no physical barrier.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 2:19 PM, John Smith wrote:

> 2009/9/21 Anthony :
> > In general, the more important question than whether or not you can
> U-turn
> > is whether or not you can cross the center line in order to make a left
> > (right if you drive on the other side) turn.  I believe the law in most
> US
> > jurisdictions is that this is allowed when physically possible unless
> there
> > is a painted median (double double yellow lines), traffic cones, etc.  In
> > the case of a painted median (double double yellow lines), if this is
> > represented as a single way, that is incorrect and should be fixed.  In
> the
> > US, there is probably a lot of this, because we imported so much from
> Tiger,
> > which represents pretty much everything except major highways as single
> ways
> > (and is a significant area needing cleanup).  Outside the US, I just
> don't
> > know the laws.
>
> Assumptions are dangerous things, I suppose this is why so many
> tourists have accidents because they assume the whole world is just
> like where they are from.
>

I am not assuming this, however.  In fact, I *specifically* pointed out the
fact that I was *not* assuming it.

Mapping ways should follow the legal paths of travel, not the existence or
non-existence of concrete.  If concrete is the only form of legal barrier,
then fine, concrete can determine how we map.  But if a painted median is
legally equivalent to a concrete barrier, we should map the two situations
equivalently.  Thus, at least in Florida, simply looking for concrete is not
sufficient.  (Yes, this is my opinion, but it is an opinion which I have
supported by many previous statements.)

> Even if there is a concrete barrier or other form of barrier that would
> > prevent a car crossing?  Why are bridges different?
>
> Usually because bridges narrow things to make it cheaper to cross a
> river etc, however my point still stands a way is a physical thing, if
> there is only one physical thing then that's all that should be mapped
> and lanes should be tagged independently of the way.
>

But if there is a concrete barrier in place, you agree we might have two
ways going over one bridge.  So all this stuff about mapping individual
lanes is off-topic.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :
> I reread the question and realize I didn't answer it.  U-turn laws vary
> greatly by jurisdiction.  Here in Florida you are allowed to U-turn on any
> road which isn't separated by a barrier or painted median, unless it is
> unsafe to do so or doing so obstructs other traffic.  But that is extremely
> jurisdiction specific.

U-turns are a turning issue, maybe they're legal in one place, maybe
they aren't in another these things should be mapped some how but this
is getting off track.

> In general, the more important question than whether or not you can U-turn
> is whether or not you can cross the center line in order to make a left
> (right if you drive on the other side) turn.  I believe the law in most US
> jurisdictions is that this is allowed when physically possible unless there
> is a painted median (double double yellow lines), traffic cones, etc.  In
> the case of a painted median (double double yellow lines), if this is
> represented as a single way, that is incorrect and should be fixed.  In the
> US, there is probably a lot of this, because we imported so much from Tiger,
> which represents pretty much everything except major highways as single ways
> (and is a significant area needing cleanup).  Outside the US, I just don't
> know the laws.

Assumptions are dangerous things, I suppose this is why so many
tourists have accidents because they assume the whole world is just
like where they are from.

> Even if there is a concrete barrier or other form of barrier that would
> prevent a car crossing?  Why are bridges different?

Usually because bridges narrow things to make it cheaper to cross a
river etc, however my point still stands a way is a physical thing, if
there is only one physical thing then that's all that should be mapped
and lanes should be tagged independently of the way.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 2:01 PM, Anthony  wrote:

> On Sun, Sep 20, 2009 at 1:52 PM, John Smith wrote:
>
>> 2009/9/21 Anthony :
>> > Not if you are free to cross the center line, for instance to make a
>> left
>> > turn across oncoming traffic to turn into a driveway.
>>
>> I didn't know you can u-turn on most trunk roads legally so why aren't
>> we showing those as 2 seperate ways?
>>
>
> Using 2 separate ways implies that you can't U-turn, except in places where
> the two ways are connected.
>

I reread the question and realize I didn't answer it.  U-turn laws vary
greatly by jurisdiction.  Here in Florida you are allowed to U-turn on any
road which isn't separated by a barrier or painted median, unless it is
unsafe to do so or doing so obstructs other traffic.  But that is extremely
jurisdiction specific.

In general, the more important question than whether or not you can U-turn
is whether or not you can cross the center line in order to make a left
(right if you drive on the other side) turn.  I believe the law in most US
jurisdictions is that this is allowed when physically possible unless there
is a painted median (double double yellow lines), traffic cones, etc.  In
the case of a painted median (double double yellow lines), if this is
represented as a single way, that is incorrect and should be fixed.  In the
US, there is probably a lot of this, because we imported so much from Tiger,
which represents pretty much everything except major highways as single ways
(and is a significant area needing cleanup).  Outside the US, I just don't
know the laws.

On Sun, Sep 20, 2009 at 2:09 PM, John Smith wrote:

> 2009/9/21 Anthony :
>  > Define "physically separated".
>
> The road base finishes or there is a concreate barrier or other form
> of barrier that would prevent a car crossing, legally or otherwise.
>
> > Do you admit that a physical bridge can carry multiple ways, or not?  I
>
> A bridge is a single way


Even if there is a concrete barrier or other form of barrier that would
prevent a car crossing?  Why are bridges different?
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> Using 2 separate ways implies that you can't U-turn, except in places where
> the two ways are connected.

Then why aren't we using multiple ways to indicate this for all
teritary, secondary, primary, trunk etc?

> Are you suggesting that we should never express our opinions on mailing
> lists, because expressing them doesn't make them true?

I'm asking you to qualify your opinion, you are making a statement
without displaying any valid reasoning on why it is the best solution.

> And in that situation, assuming you *are* allowed to cross the center line
> to make a left (right if you drive on the other side) turn, it would be
> easier to represent that as a single way with "uturn=no" than to create 2
> parallel ways and connect all the driveways, etc.  But either is acceptable
> to me.

However that's not the case, you are trying to tell me the majority of
lanes on the Sydney Harbour bridge should be tagged as multiple ways,
yet in other situations there is only a single way used this is
completely inconsistent.

If we are tagging lanes not ways then we need to do so en masse, not
have 2 different systems that exists at present.

> Define "physically separated".

The road base finishes or there is a concreate barrier or other form
of barrier that would prevent a car crossing, legally or otherwise.

> Do you admit that a physical bridge can carry multiple ways, or not?  I

A bridge is a single way, it usually has multiple lanes and we should
be able to tag these lanes rather than using ways and then coming up
with all sorts of silly hacks to then show those ways are related
because they are related they're all part of the same way.

> guess this will be determined when you define "physically separated".
> Hopefully you don't ignore this question like you did so many others.

 I haven't ignored any questions you seem to be ignoring my answers
however, how many times do you want me to state the same thing?

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 1:52 PM, John Smith wrote:

> 2009/9/21 Anthony :
> > Not if you are free to cross the center line, for instance to make a left
> > turn across oncoming traffic to turn into a driveway.
>
> I didn't know you can u-turn on most trunk roads legally so why aren't
> we showing those as 2 seperate ways?
>

Using 2 separate ways implies that you can't U-turn, except in places where
the two ways are connected.


> Just because someone expresses an opinion doesn't mean it is true or
> the best way to do something, merely an opinion someone has stated and
> others agree with.
>

Are you suggesting that we should never express our opinions on mailing
lists, because expressing them doesn't make them true?


> > If you want to treat a residential street with a clearly marked center
> line
> > and no allowance for U-turns as 2 parallel ways, be my guest.  You're
> > wasting your time, but I don't see how you're incorrect, as long as you
> > connect all the driveways and other locations where people are allowed to
> > cross the center line.
>
> In some places you aren't allowed to u-turn except where signed as a
> valid u-turn place.
>

And in that situation, assuming you *are* allowed to cross the center line
to make a left (right if you drive on the other side) turn, it would be
easier to represent that as a single way with "uturn=no" than to create 2
parallel ways and connect all the driveways, etc.  But either is acceptable
to me.

A way is a physical feature physically seperated from other ways,
>

Define "physically separated".


> anything joined is a single way but may have 1 or more lanes, if we
> could tag lanes this would be a moot argument and this is what we
> should be doing, then people wouldn't be trying to join ways with
> relations to describe a physical bridge which is a single physical
> object.
>

Do you admit that a physical bridge can carry multiple ways, or not?  I
guess this will be determined when you define "physically separated".
Hopefully you don't ignore this question like you did so many others.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :
> Not if you are free to cross the center line, for instance to make a left
> turn across oncoming traffic to turn into a driveway.

I didn't know you can u-turn on most trunk roads legally so why aren't
we showing those as 2 seperate ways?

> My justification was in the fifty million messages that came before this.

Lots of people believe in a god of one sort or another, does it make
it so just because billions of people say there really is a god?

If there is a god because billions of people say so, which one or ones
exist and which don't?

Just because someone expresses an opinion doesn't mean it is true or
the best way to do something, merely an opinion someone has stated and
others agree with.

> I believe we should define our terms based on the purpose we are trying to
> accomplish.  In the case of OSM, I think the purpose is to create maps,
> which are used to inform us how to get from place to place, so the proper
> definition of "way", at least in terms of ways carrying (vehicular,
> pedestrian, bicycle, train) traffic is essentially "A course that is or may
> be used in going from one place to another".

A way is a physical feature, lanes are features of ways, if all the
ways a physically connected to each other than it's a single way
otherwise residential streets are all tagged wrong and we need to redo
them as a way in each direction.

> If you want to treat a residential street with a clearly marked center line
> and no allowance for U-turns as 2 parallel ways, be my guest.  You're
> wasting your time, but I don't see how you're incorrect, as long as you
> connect all the driveways and other locations where people are allowed to
> cross the center line.

In some places you aren't allowed to u-turn except where signed as a
valid u-turn place.

A way is a physical feature physically seperated from other ways,
anything joined is a single way but may have 1 or more lanes, if we
could tag lanes this would be a moot argument and this is what we
should be doing, then people wouldn't be trying to join ways with
relations to describe a physical bridge which is a single physical
object.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 1:24 PM, John Smith wrote:

> 2009/9/21 Anthony :
>
> > But there are multiple ways in reality.  A "way" is a path of travel, not
> a
> > piece of asphalt.
>
> If that's the case why are most "ways" a lane in each direction?
>

I'm not sure that is even the case.


> Surely if what you are saying is true we should plot each direction as
> an individual way.
>

Not if you are free to cross the center line, for instance to make a left
turn across oncoming traffic to turn into a driveway.



On Sun, Sep 20, 2009 at 1:27 PM, John Smith wrote:

> 2009/9/21 Anthony :
> > On Sun, Sep 20, 2009 at 1:20 PM, John Smith 
> > wrote:
> >>
> >> 2009/9/21 Anthony :
> >>
> >> > True.  But it is the best way to do it.
> >>
> >> Says who?
> >
> > Says the person who made the statement.
>
> That doesn't justify anything, using that logic I could say anything
> and it would be true because I said it was.
>

My justification was in the fifty million messages that came before this.


> >> > But there are multiple ways in reality.  A "way" is a path of travel,
> >> > not a
> >> > piece of asphalt.
> >>
> >> Why is it?
> >
> > Because that's the primary purpose for which maps are created.  To inform
> us
> > how to get from place to place.
>
> I wasn't asking you about mapping in general I was asking you to
> clarify your statement,


I believe we should define our terms based on the purpose we are trying to
accomplish.  In the case of OSM, I think the purpose is to create maps,
which are used to inform us how to get from place to place, so the proper
definition of "way", at least in terms of ways carrying (vehicular,
pedestrian, bicycle, train) traffic is essentially "A course that is or may
be used in going from one place to another".


> it simply isn't treated that way in all
> situation or each residential street would be 2 parallel ways, instead
> we use a single way to indication a pair of lanes, so which is it
> going to be do we need to split residential streets or do we treat
> things as a piece of asphalt when it suits us?
>

If you want to treat a residential street with a clearly marked center line
and no allowance for U-turns as 2 parallel ways, be my guest.  You're
wasting your time, but I don't see how you're incorrect, as long as you
connect all the driveways and other locations where people are allowed to
cross the center line.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :
> On Sun, Sep 20, 2009 at 1:20 PM, John Smith 
> wrote:
>>
>> 2009/9/21 Anthony :
>>
>> > True.  But it is the best way to do it.
>>
>> Says who?
>
> Says the person who made the statement.

That doesn't justify anything, using that logic I could say anything
and it would be true because I said it was.

>> > But there are multiple ways in reality.  A "way" is a path of travel,
>> > not a
>> > piece of asphalt.
>>
>> Why is it?
>
> Because that's the primary purpose for which maps are created.  To inform us
> how to get from place to place.

I wasn't asking you about mapping in general I was asking you to
clarify your statement, it simply isn't treated that way in all
situation or each residential street would be 2 parallel ways, instead
we use a single way to indication a pair of lanes, so which is it
going to be do we need to split residential streets or do we treat
things as a piece of asphalt when it suits us?

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 1:20 PM, John Smith wrote:

> 2009/9/21 Anthony :
>
> > True.  But it is the best way to do it.
>
> Says who?
>

Says the person who made the statement.


> > But there are multiple ways in reality.  A "way" is a path of travel, not
> a
> > piece of asphalt.
>
> Why is it?
>

Because that's the primary purpose for which maps are created.  To inform us
how to get from place to place.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> But there are multiple ways in reality.  A "way" is a path of travel, not a
> piece of asphalt.

If that's the case why are most "ways" a lane in each direction?

Surely if what you are saying is true we should plot each direction as
an individual way.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> True.  But it is the best way to do it.

Says who?

> But there are multiple ways in reality.  A "way" is a path of travel, not a
> piece of asphalt.

Why is it?

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 1:13 PM, John Smith wrote:

> 2009/9/21 Anthony :
> > You clearly define "way" differently than I do, and differently than the
> > current definition.  The bridge most certainly has multiple ways in OSM
> > today.
>
> However that doesn't reflect reality


It does reflect reality.  It just reflects a different part of reality than
you want it to reflect.


> Just because it's tagged with multiple ways doesn't mean it's the best
> way to do it.
>

True.  But it is the best way to do it.

> I never claimed there are 2 bridges.  There is one bridge, with multiple
> > ways.  "bridge" does not equal "way".
>
> But there isn't multiple ways in reality, there is a single "way" or
> bridge with 4 lanes + any footway/cycleways.
>

But there are multiple ways in reality.  A "way" is a path of travel, not a
piece of asphalt.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :
> You clearly define "way" differently than I do, and differently than the
> current definition.  The bridge most certainly has multiple ways in OSM
> today.

However that doesn't reflect reality and we should just accept "that's
just the way it is" because that's how things work now, shouldn't we
tag reality no tag things because it fits in the DB?

> At least you admit there *is* physical separation between *some* lanes.  If

6 of them don't have any physical seperation.

> I never claimed there are 2 bridges.  There is one bridge, with multiple
> ways.  "bridge" does not equal "way".

But there isn't multiple ways in reality, there is a single "way" or
bridge with 4 lanes + any footway/cycleways.

> http://maps.google.com/maps?f=q&ie=UTF8&ll=40.727889,-74.100804&spn=0.002041,0.003825&t=h&z=18
> is one bridge, with more than one way.  You seem to be the only one claiming
> this isn't true.

Just because it's tagged with multiple ways doesn't mean it's the best
way to do it.

As most school teachers said to me as a kid "just because everyone is
jumping off a cliff are you going to jump off too?"

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 12:52 PM, John Smith wrote:

> That isn't tagging reality, the bridge doesn't have multiple ways
>

You clearly define "way" differently than I do, and differently than the
current definition.  The bridge most certainly has multiple ways in OSM
today.


> there isn't even physical seperation on the bridge itself between most
> of the lanes when the lanes shift, it's simply lights above the lanes
> indicating direction.
>

At least you admit there *is* physical separation between *some* lanes.  If
you propose treating the 8 lanes as two ways, one six lane way (with a
one-way restriction during some part of the day) and one two lane way (which
is always one way), I can accept that.  But I don't accept treating the two
roads that are clearly separated as the same way as the other roads which
are clearly not.

> Be specific about your proposed solution, though, and maybe I'll find that
> > I'm wrong.
>
> 1 bridge, multiple lanes. How much simplier do you want it?
>

That answers nothing.  Of course we have bridge and multiple lanes.


> There isn't 2 bridges please show me the second bridge because I'm
> sure it will be news to at least the 4.5million people in Sydney at
> the very least.
>

I never claimed there are 2 bridges.  There is one bridge, with multiple
ways.  "bridge" does not equal "way".

http://maps.google.com/maps?f=q&ie=UTF8&ll=40.727889,-74.100804&spn=0.002041,0.003825&t=h&z=18is
one bridge, with more than one way.  You seem to be the only one
claiming
this isn't true.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :
> I'm going to focus on the 8 traffic lanes.  The rest should be separate
> ways, if they are separate lanes (whether the 2 railway lines are 2 ways or
> 1 I really don't know or care), and if we care to bother mapping them
> correctly.  Currently, I'd represent the traffic lanes as three ways: four
> lanes, two lanes, and two lanes.  The middle way should be reversible based
> on time - I'm not sure if this is possible to do now or not.  It certainly
> doesn't require much work to add this feature.

That isn't tagging reality, the bridge doesn't have multiple ways
there isn't even physical seperation on the bridge itself between most
of the lanes when the lanes shift, it's simply lights above the lanes
indicating direction.

> Be specific about your proposed solution, though, and maybe I'll find that
> I'm wrong.

1 bridge, multiple lanes. How much simplier do you want it?

There isn't 2 bridges please show me the second bridge because I'm
sure it will be news to at least the 4.5million people in Sydney at
the very least.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 12:28 PM, Anthony  wrote:

> The middle way should be reversible based on time - I'm not sure if this is
> possible to do now or not.
>

Proposed solution is of the format "oneway:time{0:00-4:00;12:00-24:00} =
yes" and "oneway:time{4:00-12:00} = -1".
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 11:39 AM, John Smith wrote:

> 2009/9/21 Anthony :
>
> > I'd love for you to be able to do it.  Come up with a way to do it that
> > doesn't require rewriting all the editors, all the routing software, and
> > combining multiple ways into single ways, and we can both be happy.
>
> So you are telling me that the editors, the OSM protocol and OSM DB
> schema has always been static and never changed from the begining?
>

No.


> > Give me an exact situation, and give me your exact proposed solution.
>
> You'd probably need some tag similar to shop open hours, but in this
> case would determine the flow of traffic.
>

Thanks for the exact proposed solution.


> > If we're talking about the Walt Whitman Bridge, with its "zipper barrier"
> > (http://www.phillyroads.com/crossings/walt-whitman/img12.gif), it could
> be
> > represented presently as four ways, two in each direction, with time
> > restrictions so that at any time only two ways are open.
>
> I was talking about the Sydney Harbour bridge, it has 1 pedestrian
> lane, 1 cycle lane, 1 bus lane, 2 railway "lanes" and 8 traffic lanes,
> during peak hours there is 6 lanes one way and 2 lanes the other,
> outside of peak times it's 4 lanes each way.
>
> However it's only one physical bridge but with 13 lanes.
>

I'm going to focus on the 8 traffic lanes.  The rest should be separate
ways, if they are separate lanes (whether the 2 railway lines are 2 ways or
1 I really don't know or care), and if we care to bother mapping them
correctly.  Currently, I'd represent the traffic lanes as three ways: four
lanes, two lanes, and two lanes.  The middle way should be reversible based
on time - I'm not sure if this is possible to do now or not.  It certainly
doesn't require much work to add this feature.

Alternatively, if we don't mind giving the mapping software designers a
headache, we could use four ways: a four lane way with time restrictions, a
six lane way with time restrictions, a two lane way with time restrictions,
and a two lane way without time restrictions.  The six lane way would
overlap two of the others - but it would only be open at times that the
other two were not.  Long term, I'd prefer this solution.  If traffic can
change lanes, there should be one way.  If traffic cannot change lanes,
there should be more than one.

I am assuming, based on this picture (
http://maps.google.com/maps?t=h&q=-33.85,151.210556&ie=UTF8&ll=-33.857301,151.207609&spn=0.000559,0.000956&z=20&layer=c&cbll=-33.857467,151.207483&panoid=3v9_fz9vCEWtuw4aiCXEPg&cbp=12,25.49,,0,2.6),
that the right two lanes in the picture represent a separate set of traffic
which cannot change lanes into any of the other six lanes.

Be specific about your proposed solution, though, and maybe I'll find that
I'm wrong.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> I'd love for you to be able to do it.  Come up with a way to do it that
> doesn't require rewriting all the editors, all the routing software, and
> combining multiple ways into single ways, and we can both be happy.

So you are telling me that the editors, the OSM protocol and OSM DB
schema has always been static and never changed from the begining?

> Give me an exact situation, and give me your exact proposed solution.

You'd probably need some tag similar to shop open hours, but in this
case would determine the flow of traffic.

> If we're talking about the Walt Whitman Bridge, with its "zipper barrier"
> (http://www.phillyroads.com/crossings/walt-whitman/img12.gif), it could be
> represented presently as four ways, two in each direction, with time
> restrictions so that at any time only two ways are open.

I was talking about the Sydney Harbour bridge, it has 1 pedestrian
lane, 1 cycle lane, 1 bus lane, 2 railway "lanes" and 8 traffic lanes,
during peak hours there is 6 lanes one way and 2 lanes the other,
outside of peak times it's 4 lanes each way.

However it's only one physical bridge but with 13 lanes.

With some kind of spatial tagging you could tag a lane with a time and
which direction it was allowed and potentially a default if it doesn't
change or if it's outside peak hours.



I'm sorry, a single physical bridge, with multiple lanes not multiple ways.

> If you disagree with that last one, then you're proposing to undo a whole lot 
> of work that people have done.

Again, the OSM system was perfect from the start and nothing has ever
changed since because it was so perfect.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 10:57 AM, John Smith wrote:

> 2009/9/21 Anthony :
> > Fine.  You can add "maximum height" along with "maximum speed" to your
> list
> > of lane-specific data.
> >
> > I don't find it very important, but if you want to allow for this,
> without
> > messing up anyone else, feel free.
>
> I don't find people tagging botanical names on plants useful either,
> but they're doing it because they feel it is important.
>
> At present I am not easily able to tag individual lanes, but would
> love to be able to do it.
>

I'd love for you to be able to do it.  Come up with a way to do it that
doesn't require rewriting all the editors, all the routing software, and
combining multiple ways into single ways, and we can both be happy.

>> Ever been on a bridge with dynamic lane changes depending on time of day?
> >
> > Yes.
>
> At present how exactly would you tag it?
>

Give me an exact situation, and give me your exact proposed solution.

If we're talking about the Walt Whitman Bridge, with its "zipper barrier" (
http://www.phillyroads.com/crossings/walt-whitman/img12.gif), it could be
represented presently as four ways, two in each direction, with time
restrictions so that at any time only two ways are open.

That's not how I'd tag it, though.  I'd probably tag it as two ways with 3.5
lanes each.  Because I don't think the current mapping software would like
the four way solution, and I don't want to break anything.  Ultimately, if
we can get the mapping software to support it, I'd prefer the four way
solution, though.

That's an interesting scenario, though.  I'm interested in hearing your
solution.  You might even be able to convince me that you're right, if
you're specific enough.

> So you think those examples should all be represented as a single way?
> >
> > That's horrible.
>
> What exactly is so horrible about it exactly, they are all part of the
> same physical way, but there is various lanes and being able to tag
> them individually would make a lot more sense than trying to plot
> multiple parallel ways.
>

They're not part of the same physical way.

http://maps.google.com/maps?f=q&ie=UTF8&ll=40.727889,-74.100804&spn=0.002041,0.003825&t=h&z=18

Two ways.  One bridge.

>
> http://bikelaneblog.files.wordpress.com/2008/02/pulaski-bridge-walkway.jpg?w=324&h=241
> >
> > One bridge or two?
> >
> > If it's one, then you are off-topic.
>
> Please explain exactly how that is two bridges, it's physically one
> bridge with multiple lanes even if one of those lanes isn't for cars.
>

It's two ways and one bridge.

http://www.treehugger.com/bike-commuters-on-bridge-with-heavy-traffic.jpg

Two ways, one bridge.

http://maps.google.com/maps?f=q&ie=UTF8&ll=40.727889,-74.100804&spn=0.002041,0.003825&t=h&z=18

Two ways, one bridge.

If you disagree with that last one, then you're proposing to undo a whole
lot of work that people have done.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> Irrelevant.  I never said you had to use relations.  In fact, I said you
> don't.

Others have suggested otherwise, to "group" ways that are on the same
physical bridge.

> I wouldn't call the use of relations "nutty", though.

I was referring to a specific use case.

> Fine.  You can add "maximum height" along with "maximum speed" to your list
> of lane-specific data.
>
> I don't find it very important, but if you want to allow for this, without
> messing up anyone else, feel free.

I don't find people tagging botanical names on plants useful either,
but they're doing it because they feel it is important.

At present I am not easily able to tag individual lanes, but would
love to be able to do it.

>> Ever been on a bridge with dynamic lane changes depending on time of day?
>
> Yes.

At present how exactly would you tag it?

> So you think those examples should all be represented as a single way?
>
> That's horrible.

What exactly is so horrible about it exactly, they are all part of the
same physical way, but there is various lanes and being able to tag
them individually would make a lot more sense than trying to plot
multiple parallel ways.

> The maximum speed of a way is the maximum speed of that way.  If you want to
> *also* tag lane speeds, feel free.

But that isn't reflecting life, it's reducing it and in turn loosing
information.

> http://bikelaneblog.files.wordpress.com/2008/02/pulaski-bridge-walkway.jpg?w=324&h=241
>
> One bridge or two?
>
> If it's one, then you are off-topic.

Please explain exactly how that is two bridges, it's physically one
bridge with multiple lanes even if one of those lanes isn't for cars.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 10:27 AM, John Smith wrote:

> 2009/9/21 Anthony :
>
> > So this is a single way?
> >
> http://bikelaneblog.files.wordpress.com/2008/02/pulaski-bridge-walkway.jpg?w=324&h=241
> >
> > That's nutty.
>
> And abusing relations to do the same thing isn't?
>

Irrelevant.  I never said you had to use relations.  In fact, I said you
don't.

I wouldn't call the use of relations "nutty", though.

> As long as you are free to change lanes, I disagree.  The maxheight of a
> way
> > is the maximum height of the way.  Using a GPS doesn't permit you to
> ignore
> > signs which say to stay out of the right lane.
>
> You miss the point entirely, the GPS could indicate which lane you
> need to be in if you program it with your current height.
>

Fine.  You can add "maximum height" along with "maximum speed" to your list
of lane-specific data.

I don't find it very important, but if you want to allow for this, without
messing up anyone else, feel free.


> > Have you ever been on a bridge with a Jersey barrier?  Is that one
> bridge,
> > or two?
>
> Ever been on a bridge with dynamic lane changes depending on time of day?
>

Yes.

> Maybe the last one.  Anything else, is "physically separated", but could
> > easily be accomplished over a single bridge (one of them even *is* an
> > example of a single bridge).
>
> Do you even grasp the benefit of being able to tag lanes rather than ways?
>

Yes, I do.  Have you read what I said?


> Pretty much all the examples you posted are exactly why we need to be
> able to tag lanes, thanks for proving my point for me.
>

So you think those examples should all be represented as a single way?

That's horrible.


> You keep saying you can tag a way with the maximum maxspeed but that's
> nonsense, that isn't being able to tag the real world accurately
> that's tagging a subset of the real world.
>

The maximum speed of a way is the maximum speed of that way.  If you want to
*also* tag lane speeds, feel free.

http://bikelaneblog.files.wordpress.com/2008/02/pulaski-bridge-walkway.jpg?w=324&h=241

One bridge or two?

If it's one, then you are off-topic.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/21 Anthony :

> So this is a single way?
> http://bikelaneblog.files.wordpress.com/2008/02/pulaski-bridge-walkway.jpg?w=324&h=241
>
> That's nutty.

And abusing relations to do the same thing isn't?

> As long as you are free to change lanes, I disagree.  The maxheight of a way
> is the maximum height of the way.  Using a GPS doesn't permit you to ignore
> signs which say to stay out of the right lane.

You miss the point entirely, the GPS could indicate which lane you
need to be in if you program it with your current height.

> Now if there's routing information involved - if you can't switch lanes -
> then yes, this is important information.  But it's also the place for more
> than one way.

Ummm what are you talking about?

> Have you ever been on a bridge with a Jersey barrier?  Is that one bridge,
> or two?

Ever been on a bridge with dynamic lane changes depending on time of day?

> Maybe the last one.  Anything else, is "physically separated", but could
> easily be accomplished over a single bridge (one of them even *is* an
> example of a single bridge).

Do you even grasp the benefit of being able to tag lanes rather than ways?

Pretty much all the examples you posted are exactly why we need to be
able to tag lanes, thanks for proving my point for me.

You keep saying you can tag a way with the maximum maxspeed but that's
nonsense, that isn't being able to tag the real world accurately
that's tagging a subset of the real world.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 10:04 AM, John Smith wrote:

> 2009/9/20 Anthony :
>
> > Yes they are.  If they weren't physically separated, people would be
> driving
> > on top of each other.  If they weren't physically separated, they
> wouldn't
> > be called multiple lanes - they'd be called one lane.
>
> Pretty sure I left an "if" out, if the lanes are on top of the same
> physical thing, in this case a bridge it's a single way, just the
> lanes need tagging differently.
>

So this is a single way?
http://bikelaneblog.files.wordpress.com/2008/02/pulaski-bridge-walkway.jpg?w=324&h=241

That's nutty.


> > The maxspeed of a way is the maximum speed you can legally travel on that
> > way.  There's nothing inaccurate about tagging the way with the maximum
> > speed you can legally travel on it.
>
> That isn't accurate, the maximum speed varies by lane. Instead of
> maxspeed, what about maxheight, if several lanes have different
> maxheights it would be inaccurate and incorrect to mark either the
> highest or lowest maxheight since you might fit if you get in the
> right lane.
>

As long as you are free to change lanes, I disagree.  The maxheight of a way
is the maximum height of the way.  Using a GPS doesn't permit you to ignore
signs which say to stay out of the right lane.

Now if there's routing information involved - if you can't switch lanes -
then yes, this is important information.  But it's also the place for more
than one way.


> > Everything in the world is physically connected.  In this case, let's
> leave
> > out the cycle lane and footpath and just consider a dual carriage way
> (with
> > central reservation).  There are two roads and an island which go over
> one
> > bridge.  One way, two, or three?
>
> Is it one bridge or 2 bridges, if it's one bridge it's a single way
> with multiple lanes.
>

Have you ever been on a bridge with a Jersey barrier?  Is that one bridge,
or two?


> > One way is unacceptable.  If that's your proposal, it is to rewrite all
> the
> > editors and all the routing software, and then go through the database
> > combining all the dual carriageways which many of us worked hard to split
> up
> > in the first place.
>
> You really need to read my comments instead of making assumptions,
> dual carriage ways usually aren't phycially connected, not in the same
> respect as a bridge.
>

Searching Google for "physically separated", I have some examples.  Are any
of these examples of a single way?

http://www.flickr.com/photos/portlandtransport/3877499480/sizes/m/in/set-72157622203417368/
http://www.uppergreenside.org/wp-content/uploads/2007/03/physically-separated-bike-lanes-diagram.jpg
http://bikelaneblog.files.wordpress.com/2008/03/9th-ave-with-platic-barriers.jpg
http://bikelaneblog.files.wordpress.com/2008/02/pulaski-bridge-walkway.jpg?w=324&h=241
http://2.bp.blogspot.com/_kvgPyMB-iIs/SWkOfoVQVEI/DbA/3lCHczbQ700/s320/pinest_physical_bike-1.jpg
http://www.streetsblog.org/wp-content/uploads/moved/buffered_bike_lane4.jpg
http://www.transitmiami.com/wp-content/uploads/2009/08/Miami-Bicycle-Survey-384-1024x768.jpg
http://i.treehugger.com/images/2007/10/24/new%20york%20bike%20lane.jpg
http://farm3.static.flickr.com/2399/2367382978_f878d69045.jpg

Maybe the last one.  Anything else, is "physically separated", but could
easily be accomplished over a single bridge (one of them even *is* an
example of a single bridge).
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/20 Anthony :

> Yes they are.  If they weren't physically separated, people would be driving
> on top of each other.  If they weren't physically separated, they wouldn't
> be called multiple lanes - they'd be called one lane.

Pretty sure I left an "if" out, if the lanes are on top of the same
physical thing, in this case a bridge it's a single way, just the
lanes need tagging differently.

> The maxspeed of a way is the maximum speed you can legally travel on that
> way.  There's nothing inaccurate about tagging the way with the maximum
> speed you can legally travel on it.

That isn't accurate, the maximum speed varies by lane. Instead of
maxspeed, what about maxheight, if several lanes have different
maxheights it would be inaccurate and incorrect to mark either the
highest or lowest maxheight since you might fit if you get in the
right lane.

> Everything in the world is physically connected.  In this case, let's leave
> out the cycle lane and footpath and just consider a dual carriage way (with
> central reservation).  There are two roads and an island which go over one
> bridge.  One way, two, or three?

Is it one bridge or 2 bridges, if it's one bridge it's a single way
with multiple lanes.

> One way is unacceptable.  If that's your proposal, it is to rewrite all the
> editors and all the routing software, and then go through the database
> combining all the dual carriageways which many of us worked hard to split up
> in the first place.

You really need to read my comments instead of making assumptions,
dual carriage ways usually aren't phycially connected, not in the same
respect as a bridge.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 9:30 AM, Anthony  wrote:

> On Sun, Sep 20, 2009 at 9:10 AM, John Smith wrote:
>
>> If it's a single physical section, ie a bridge with all the lanes
>> physically connected then it should only be one way and we should be
>> able to tag the individual lanes.
>>
>
> Everything in the world is physically connected.  In this case, let's leave
> out the cycle lane and footpath and just consider a dual carriage way (with
> central reservation).  There are two roads and an island which go over one
> bridge.  One way, two, or three?
>

If you'd prefer a picture:
http://bikeportland.org/wp-content/bikelaneMontreal.jpg

One way, or more than one way?

If your answer is one way, I vehemently disagree.  If your answer is more
than one way, then tagging subsections of ways is off-topic.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 9:10 AM, John Smith wrote:

> 2009/9/20 Anthony :
>
> > In most cases I don't think a relation is the only solution either.  I
> don't
> > see it as an abuse, though.  It is clearly being used to show a relation
>
> Lanes aren't physically seperated
>

Yes they are.  If they weren't physically separated, people would be driving
on top of each other.  If they weren't physically separated, they wouldn't
be called multiple lanes - they'd be called one lane.

> As long as you only suggest tagging subsections of a way in those limited
> > cases where there is no routing information conveyed, I don't really care
> > how you implement it.
>
> > If there are 2 lanes going in the same direction, with maxspeeds 130 and
> > 110, the maxspeed in that direction is 130.  The fact that you have to be
> in
> > a particular lane to travel at that speed is irrelevant to routing
> software
> > so long as you can change lanes freely.
>
> I disagree, you should be able to tag each lane with the right speed,
> otherwise you aren't accurately tagging things.
>

The maxspeed of a way is the maximum speed you can legally travel on that
way.  There's nothing inaccurate about tagging the way with the maximum
speed you can legally travel on it.


> > To wit, this thread started with someone talking about "a cycle lane,
> dual
> > carriage way (with central reservtion) & footpath".  Are you suggesting
> that
> > this should be represented as *a single way*?  If so, I have to
> vehemently
> > disagree.  If not, then tagging subsections of a single way has nothing
> to
> > do with the thread, and is off-topic.
>
> If it's a single physical section, ie a bridge with all the lanes
> physically connected then it should only be one way and we should be
> able to tag the individual lanes.
>

Everything in the world is physically connected.  In this case, let's leave
out the cycle lane and footpath and just consider a dual carriage way (with
central reservation).  There are two roads and an island which go over one
bridge.  One way, two, or three?

One way is unacceptable.  If that's your proposal, it is to rewrite all the
editors and all the routing software, and then go through the database
combining all the dual carriageways which many of us worked hard to split up
in the first place.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/20 Anthony :

> In most cases I don't think a relation is the only solution either.  I don't
> see it as an abuse, though.  It is clearly being used to show a relation

Lanes aren't physically seperated so they shouldn't be split nor need
a relation to show they are physically joined, I call that abusing
relations.

> As long as you only suggest tagging subsections of a way in those limited
> cases where there is no routing information conveyed, I don't really care
> how you implement it.

> If there are 2 lanes going in the same direction, with maxspeeds 130 and
> 110, the maxspeed in that direction is 130.  The fact that you have to be in
> a particular lane to travel at that speed is irrelevant to routing software
> so long as you can change lanes freely.

I disagree, you should be able to tag each lane with the right speed,
otherwise you aren't accurately tagging things.

> Per lane speeds, when you are free to change lanes, are fairly irrelevant.
> I don't mind if you want to extend the system to handle them, but don't
> suggest this extension has anything to do with other problems where we
> already have multiple ways.

Are they validly tagged ways or is it simply a way to tag things to
deal with the DB as is?

> To wit, this thread started with someone talking about "a cycle lane, dual
> carriage way (with central reservtion) & footpath".  Are you suggesting that
> this should be represented as *a single way*?  If so, I have to vehemently
> disagree.  If not, then tagging subsections of a single way has nothing to
> do with the thread, and is off-topic.

If it's a single physical section, ie a bridge with all the lanes
physically connected then it should only be one way and we should be
able to tag the individual lanes.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sun, Sep 20, 2009 at 12:49 AM, John Smith wrote:

> 2009/9/20 Anthony :
>
> > That's an editor issue.  If the editor wants to display lanes in a single
> > way as parallel ways, and let you edit them if need be, it can do that.
>
> It's also a DB/framework issue, I don't think relations should be
> abused for this purpose, instead the DB needs to be extended to cope
> with lanes being individually tagged.
>

In most cases I don't think a relation is the only solution either.  I don't
see it as an abuse, though.  It is clearly being used to show a relation
among multiple ways.  In some cases I can see a reason to tag subsections of
a way.  Different lane speeds in a way in which traffic is free to change
lanes would be one example (although I'd caution you to ensure that maxspeed
remains on the way as the maximum speed you can travel in *any* lane,
because that is how you calculate the cost that needs to be input into any
shortest path algorthim).  In many cases I see a reason to create multiple
ways, and in fact in many cases (including the example at the top of this
thread) that is what is currently being done.  If there is separate routing
information, you need separate ways.

As long as you only suggest tagging subsections of a way in those limited
cases where there is no routing information conveyed, I don't really care
how you implement it.

For example, you may have different maxpseed=* depending on the lane
> you are in, for example near here there is 2 lanes, one in each
> direction, each lane for a short section has a different maxspeed,
> I've taged it as maxspeed:forward=* but this won't scale to say 5
> lanes in each direction and 4 different maximum speeds depending on
> which lane you are in, and I know this situation exists as I've seen
> it. 130, 110, 100, 80, 80
>

If there are 2 lanes going in the same direction, with maxspeeds 130 and
110, the maxspeed in that direction is 130.  The fact that you have to be in
a particular lane to travel at that speed is irrelevant to routing software
so long as you can change lanes freely.

On the other hand, if you have to pick which lane to be in, and stay in that
lane for a certain significant period of time, then there should be a split
into two ways at the decision point, because routing software needs to know
about that - to say "get in the left lane if you want to go faster" or
whatever.  Do you see how if you are free to change lanes there's no need
for the routing software to tell you anything (and in fact, it *should not*
tell you anything)?  You just look up at the signs and make your own
decision, which you can change back at any time.

Per lane speeds, when you are free to change lanes, are fairly irrelevant.
I don't mind if you want to extend the system to handle them, but don't
suggest this extension has anything to do with other problems where we
already have multiple ways.

To wit, this thread started with someone talking about "a cycle lane, dual
carriage way (with central reservtion) & footpath".  Are you suggesting that
this should be represented as *a single way*?  If so, I have to vehemently
disagree.  If not, then tagging subsections of a single way has nothing to
do with the thread, and is off-topic.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/20 Anthony :

> That's an editor issue.  If the editor wants to display lanes in a single
> way as parallel ways, and let you edit them if need be, it can do that.

It's also a DB/framework issue, I don't think relations should be
abused for this purpose, instead the DB needs to be extended to cope
with lanes being individually tagged.

My suggestion about the editor is that's how ways could be represented
in editors so lanes could be individually tagged if needed, if there
is no need for lanes to be tagged differently to ways then they
shouldn't be, it's just that in some cases they need to be.

For example, you may have different maxpseed=* depending on the lane
you are in, for example near here there is 2 lanes, one in each
direction, each lane for a short section has a different maxspeed,
I've taged it as maxspeed:forward=* but this won't scale to say 5
lanes in each direction and 4 different maximum speeds depending on
which lane you are in, and I know this situation exists as I've seen
it. 130, 110, 100, 80, 80

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sat, Sep 19, 2009 at 11:37 PM, John Smith wrote:

> 2009/9/20 Anthony :
> > This can be done without resorting to mapping each lane separately.  If
> you
> > have a three lane road with no lane change restrictions or physical
> > barriers, you map it as one way, with three lanes, with the position as
> the
> > center of the three lanes.  When the road goes to two lanes, you map it
> as
> > one way, with two lanes, with the position as the center of the two
> lanes.
>
> I wasn't suggesting to map each lane separately, however an editor
> could display lanes and it would be so much better to display them as
> parallel ways which could be edited if they needed to be.
>

That's an editor issue.  If the editor wants to display lanes in a single
way as parallel ways, and let you edit them if need be, it can do that.

All that's needed is an unambiguous way to represent all the various
scenarios.  There are some issues where I think your method of using a
single way is appropriate (per lane speed when traffic is free to change
lanes freely is the only one I can think of at the moment though), and some
issues where I think using a single way is completely inappropriate (per
lane access restrictions, per lane geometries, per lane turn restrictions).
I should be able to run a shortest path algorithm on a single set of nodes
and edges.  Add whatever features you want to that, but don't take that
away.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Sun, Sep 20, 2009 at 1:37 PM, John Smith  wrote:
>
> I wasn't suggesting to map each lane separately, however an editor
> could display lanes and it would be so much better to display them as
> parallel ways which could be edited if they needed to be.

John, do you concede that there are some situations where lanes are
not parallel or are separated? How would you propose to model those
situations? Do you concede that sometimes there is a need to map an
individual lane as a series of nodes (way), and that in those
circumstances there is a need to group lanes (ways) with a relation?

Also, how would you propose to model transitions between sections of
ways with different lane configurations, if lanes are not individually
modeled as a series of nodes (way)?

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/20 Anthony :
> This can be done without resorting to mapping each lane separately.  If you
> have a three lane road with no lane change restrictions or physical
> barriers, you map it as one way, with three lanes, with the position as the
> center of the three lanes.  When the road goes to two lanes, you map it as
> one way, with two lanes, with the position as the center of the two lanes.


I wasn't suggesting to map each lane separately, however an editor
could display lanes and it would be so much better to display them as
parallel ways which could be edited if they needed to be.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sat, Sep 19, 2009 at 8:39 AM, Martin Koppenhoefer  wrote:
>
> There is indeed a problem with bridges (in cases like yours it looks
> like several bridges where in reality there is just one, then there
> are bridge-names that can differ from the streetname, etc.)
>

I wonder if this perhaps isn't a problem with the data after all.  As long
as the way widths are given, it's not ambiguous.  If the distance between
two ways is less than or (approximately) equal to the average width of the
two ways, you've got one bridge.  If the distance is greater, you've got two
bridges.

I guess you might have a separator on a single bridge, but that can always
be represented as another way.

And it might be difficult with the current editors to get the separation
exactly correct, but there's no reason the editor can't allow you to select
two ways, set the distance in meters between them, and automagically make
one parallel to the other separated by the distance you choose.

This is only really a problem if the renderer ignores width.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Roy Wallace]

On Sun, Sep 20, 2009 at 1:43 AM, Martin Koppenhoefer
 wrote:
>
> yes, but that's not the problem: straight parallel ways. The problem
> arises when they change (become one more or less), on intersections,
> etc. Try to imagine a situation like the one I posted above in a
> geometrically reduced system: it will get way too confusing. If we map
> lanes where they are, there is another benefit: positional correctness
> and ease of topological structure: you see what you do.

There seems to me to be good arguments for both of the suggestions
being discussed. For clarity, here is a summary of the two main
options:

1) modeling a *lane* as a series of nodes (as suggested by Martin)
2) modeling a *set of parallel lanes* as a series of nodes (as
suggested by John)

In 1), tags referring to the lane are applied to the way that
explicitly represents the lane. Lanes of the same street are grouped
with a relation.

In 2), tags referring to the lane are (somehow) applied to the way
that explicitly represents the entire street (group of parallel
lanes).

I would like to think that a structural database change is not
necessary to implement 2). All that would be required is a way to
denote which tags refer to which individual lanes (e.g. lane:*=*). I
know, John, that you prefer a change to database structure instead.
Note first, however, that regardless of whether lane:*=* or a new
database table is used, editors could be changed to allow for the kind
of editing you describe
(http://wiki.openstreetmap.org/images/4/46/Lane_group_example1_screen_2.png).
It's still 2).

Personally, I think both 1) and 2) could be used, as applicable.
Specifically, 1) where lanes are not parallel, and 2) otherwise.

I prefer 1), though, as it is more extensible (e.g. to use of further
relations indicating ability to change lanes - with 2), it's likely to
get messy). I also don't agree with John that 1) is an "abuse" of
relations.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sat, Sep 19, 2009 at 4:43 PM, Anthony  wrote:

> Perhaps there could be some sort of special designation for a way with 3
> lanes at the beginning and 2 lanes at the end, which designates whether the
> right or left lane ends, if you really want to get into the fine detail.
>

To clarify, at the point where the third lane starts to end, you split the
way.  Before the point you have lanes=3, after the point you have some sort
of "lanes=3;2", "laneends=right".  Then, at the point where the third lane
is completely gone, you split the way again.  Before the point you still
have "lanes=3;2","laneends=right".  After the point you have lanes=2.  In
all cases you map the physical center of all the lanes.

No need for relations for something simple like that.  Relations would be
nice for divided highways (a separate way could even go in the middle to
represent the island or whatever divider is present), for bridges, and stuff
like that, though.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sat, Sep 19, 2009 at 11:43 AM, Martin Koppenhoefer <
dieterdre...@gmail.com> wrote:

> yes, but that's not the problem: straight parallel ways. The problem
> arises when they change (become one more or less), on intersections,
> etc. Try to imagine a situation like the one I posted above in a
> geometrically reduced system: it will get way too confusing. If we map
> lanes where they are, there is another benefit: positional correctness
> and ease of topological structure: you see what you do.
>

This can be done without resorting to mapping each lane separately.  If you
have a three lane road with no lane change restrictions or physical
barriers, you map it as one way, with three lanes, with the position as the
center of the three lanes.  When the road goes to two lanes, you map it as
one way, with two lanes, with the position as the center of the two lanes.

Perhaps there could be some sort of special designation for a way with 3
lanes at the beginning and 2 lanes at the end, which designates whether the
right or left lane ends, if you really want to get into the fine detail.

When a lane doesn't end, but becomes an exit only, it's even simpler.  You
pick a decision point where the (let's say right) lane becomes exit only.
You split the way at that point.  Before the decision point you have one
way, with three lanes, with the position as the center of the three lanes.
After the decision point you have two ways - the left way has two lanes with
the position as the center of the two lanes; the right way has one lane with
the position as the center of the lane.  Then when the right lane goes off
to exit, you just follow that geometry.
http://www.openstreetmap.org/?lat=27.944845&lon=-82.538208&zoom=18&layers=B000FTF
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Anthony]

On Sat, Sep 19, 2009 at 9:24 AM, John Smith wrote:

> 2009/9/19 Martin Koppenhoefer :
> > don't get you. Isn't "mapping lanes" just the same like what I
> > suggested? I'm in favour of mapping all lanes and ways as well, but
> > you DO need relations to combine them into streets (indicating kind of
> > separation and / or possibility to change lanes). I was in this case
> > just talking about the bridge, but for streets I can imagine the same
> > procedure (and add green, dividers, walls, curbs, etc. as well)
>
> Why do we need relations to combine "lanes" into "ways".
>

The suggestion was to use relations to combine "ways" into "streets".


> Wouldn't it make more sense to tag lanes of ways?
>

Only if those lanes have identical geometries and traffic is free to change
lanes.  The use of a single way implies that traffic is free to travel
between any parts of a way.

What if you have a single bridge with two lanes traveling in the same
direction which are separated by a Jersey barrier (
http://en.wikipedia.org/wiki/Jersey_barrier)?  It would be incorrect to use
a single way to represent that bridge.  There are two ways and one bridge.

A relation seems like the only appropriate way to represent this, and I
don't see how it's a hack.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Martin Koppenhoefer]

this is a lane issue and needs to
> be solved on a lane basis, not on a way basis.

+1, that's what I say.

 cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Iván Sánchez Ortega]

El Sábado, 19 de Septiembre de 2009, John Smith escribió:
> I don't see the problem, you just need to be able to tag which lanes
> merge into which, or which diverge, this is a lane issue and needs to
> be solved on a lane basis, not on a way basis.

Per-lane speed limits, per-lane traffic access restrictions, proper topology 
of complex crossings...

And don't forget relating lanes to lanes, to say if you can change lanes or 
not (overpassing/incorporating to traffic/whatever).

-- 
--
Iván Sánchez Ortega 

http://ivan.sanchezortega.es
MSN:i_eat_s_p_a_m_for_breakf...@hotmail.com
Jabber:ivansanc...@jabber.org ; ivansanc...@kdetalk.net
IRC: ivansanchez @ OFTC & freenode

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/20 Martin Koppenhoefer :

> yes, but that's not the problem: straight parallel ways. The problem
> arises when they change (become one more or less), on intersections,
> etc. Try to imagine a situation like the one I posted above in a
> geometrically reduced system: it will get way too confusing. If we map
> lanes where they are, there is another benefit: positional correctness
> and ease of topological structure: you see what you do.

I don't see the problem, you just need to be able to tag which lanes
merge into which, or which diverge, this is a lane issue and needs to
be solved on a lane basis, not on a way basis.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[Martin Koppenhoefer]

2009/9/19 John Smith :
> 2009/9/19 Martin Koppenhoefer :
>
>> what do you mean? We are already doing this: lanes=3
>
> That only says how many lanes, it doesn't describe restrictions or
> properties of individual lanes.
>
>> In simple cases you don't need it, and when it get's complex, IMHO
>> explicit mapping is the only transparent and easy way to solve the
>> issue.
>> One of my favourite example is this situation:
>> http://maps.google.it/maps?hl=de&ie=UTF8&ll=41.866627,12.49679&spn=0.000684,0.001206&t=h&z=20
>
> No, more like this:
>
> http://wiki.openstreetmap.org/images/4/46/Lane_group_example1_screen_2.png
>

yes, but that's not the problem: straight parallel ways. The problem
arises when they change (become one more or less), on intersections,
etc. Try to imagine a situation like the one I posted above in a
geometrically reduced system: it will get way too confusing. If we map
lanes where they are, there is another benefit: positional correctness
and ease of topological structure: you see what you do.

cheers,
Martin

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Should Bridges be independent of their ways?

[John Smith]

2009/9/20 Mike Harris :
> Claudius - I think you may have answered the question I just asked - thanks
> - I must admit that I hadn't seen this proposal before. Once again,
> relations prove powerful!

Yes except they get abused when we should be looking towards
micromapping techniques, not hacks to get round the current DB/frame
work limitations.

___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk