subject:"Re\: \[Tagging\] layer=\-1, rivers, bridges and tunnels"

On Wed, 2 Apr 2014, Martin Koppenhoefer wrote:

 2014-04-02 13:51 GMT+02:00 Richard Z. ricoz@gmail.com:
   It is not wrong by itself but there are many circumstances
   where it
   is plain wrong
 
 +1, it is not wrong as long as there aren't any crossing / overlapping
 objects that have a different layer in OSM (e.g. no layer tag in OSM =
 implicit layer=0) but are physically or conceptually on the same layer in
 reality. Basically you'd have to put all these other objects as well to
 layer=-1, which in turn would most probably require you to put even more
 objects on layer=-1 and so on.

So how would bridge+layer=1 = bridge+layer=2 transition happen, given 
such a non-sense rule. Definately a road continuing is supposedly on the 
same layer as the previous part? Does it imply layer is cast to stone for 
the whole road network because all connections to a node are definately 
sharing the level at the node. Or what did you mean?!?


-- 
 i.

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Wed, 2 Apr 2014, Dave Swarthout wrote:

 C'mon guys. Tagging an entire river at layer=-1 is simply not the way to do
 things, unless it is a covered river or one that runs underground. What
 other possible justification is there other than not wanting to do the work
 of tagging bridges with a layer=1? If you argue it's okay if there aren't
 any crossing/overlapping objects do you mean on the entire planet?

C'mon guys. It should be done with sensible defaults rather than forcing 
mappers to tell such plain obvious things. Rivers tend to pretty 
universally go below bridges, don't you agree?

-- 
 i.

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-04-02 Thread Nelson A. de Oliveira

On Wed, Apr 2, 2014 at 9:53 AM, Ilpo Järvinen ilpo.jarvi...@helsinki.fi wrote:
 C'mon guys. It should be done with sensible defaults rather than forcing
 mappers to tell such plain obvious things. Rivers tend to pretty
 universally go below bridges, don't you agree?

Or bridges go above rivers? :-)

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Wed, 2 Apr 2014, Nelson A. de Oliveira wrote:

 On Wed, Apr 2, 2014 at 9:53 AM, Ilpo Järvinen ilpo.jarvi...@helsinki.fi 
 wrote:
  C'mon guys. It should be done with sensible defaults rather than forcing
  mappers to tell such plain obvious things. Rivers tend to pretty
  universally go below bridges, don't you agree?
 
 Or bridges go above rivers? :-)

Definately, both :), which was my point. Adding layer in this case is just 
to satisfy some lawyer rather than useful work.

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-04-02 Thread Martin Koppenhoefer

2014-04-02 15:10 GMT+02:00 Ilpo Järvinen ilpo.jarvi...@helsinki.fi:

 Definately, both :), which was my point. Adding layer in this case is just
 to satisfy some lawyer rather than useful work.





not adding layer tags to objects crossing on different layers is incomplete
data. It can still be rendered correctly in most cases, i.e. the simple
ones, but this doesn't make the data complete. I agree it is often more a
formal requirement then a practical necessity to tag the bridge with
layer=1 (can be guessed correctly according to common sense for 99.9% of
the cases, agreed), while tagging the river for longer parts with layer=-1
will most probably lead to inconsistencies sooner or later, so I see it as
bad style (though not necessarily wrong).


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-04-02 Thread Pieren

On Wed, Apr 2, 2014 at 3:10 PM, Ilpo Järvinen ilpo.jarvi...@helsinki.fi wrote:

  Rivers tend to pretty universally go below bridges, don't you agree?

 Or bridges go above rivers? :-)

^^
Sounds like the glass being half empty or half full. We will never
reconcile the two points of view ;-)

Pieren

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-04-02 Thread Richard Z.

On Wed, Apr 02, 2014 at 03:53:25PM +0300, Ilpo Järvinen wrote:
 On Wed, 2 Apr 2014, Dave Swarthout wrote:
 
  C'mon guys. Tagging an entire river at layer=-1 is simply not the way to do
  things, unless it is a covered river or one that runs underground. What
  other possible justification is there other than not wanting to do the work
  of tagging bridges with a layer=1? If you argue it's okay if there aren't
  any crossing/overlapping objects do you mean on the entire planet?
 
 C'mon guys. It should be done with sensible defaults rather than forcing 
 mappers to tell such plain obvious things. Rivers tend to pretty 
 universally go below bridges, don't you agree?

yes. It should be done with sensible presets, we can not change the defaults
for features that are already in use for ages.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-04-02 Thread Richard Z.

On Tue, Apr 01, 2014 at 11:21:51PM -0700, Bryce Nesbitt wrote:

 April 1st aside: the number of important implicit assumptions is relatively
 small.  Rivers under, power lines over, closed ways under except if they're
 tagged building, etc.  Currently this type of layering is implicit in
 various bits rendering software, but
 it could be formalized at the tag definition level to help meet certain
 mapper expectations.

there are some more assumptions, rivers are visible and not under forests, 
residential
areas maybe others.
The way mapnik handles this is black magic.

Also, when formalizing this we should make a distinction between rendering
order and vertical object relations.

 In the case of the river/highway layer warning: if the warning had never
 existed, chances are the various workaround schemes would never have come
 up.  Rivers would run under roadways, and tagging would be needed only in
 the rare case of a ford or an arroyo with no culvert.

That warning is indeed the cause of lots of hassle. It is not wrong but
until it is balanced by more complete validation of layers, rivers etc
it seems to cause more damage than good.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Wed, 2 Apr 2014, Richard Z. wrote:

 On Wed, Apr 02, 2014 at 03:53:25PM +0300, Ilpo Järvinen wrote:
  On Wed, 2 Apr 2014, Dave Swarthout wrote:
  
   C'mon guys. Tagging an entire river at layer=-1 is simply not the way to 
   do
   things, unless it is a covered river or one that runs underground. What
   other possible justification is there other than not wanting to do the 
   work
   of tagging bridges with a layer=1? If you argue it's okay if there aren't
   any crossing/overlapping objects do you mean on the entire planet?
  
  C'mon guys. It should be done with sensible defaults rather than forcing 
  mappers to tell such plain obvious things. Rivers tend to pretty 
  universally go below bridges, don't you agree?
 
 yes. It should be done with sensible presets, we can not change the defaults
 for features that are already in use for ages.

This is just misinformation. The default is possible to add even though 
many keep repeating it's impossible. We just need to ensure that the 
layer=* overrides and weights more than the default. Together it would 
ensure that _only_ those cases which do not have layers defined yet get 
affected.

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-25 Thread Dave F.


On 14/03/2014 15:57, Pieren wrote:


You don't see the point where adding one layer=-1  is easier than
adding 10 layer=1 ?


Not when you could have other entities passing under the bridges.
I see it as lazy  less accurate. Making OSM more accurate is a primary 
consideration when editing



I see the layer tag in tunnel/bridge in simple
cases but you should not follow all recommendations as fixed in stone.


True, but you should follow the correct ones.

---
This email is free from viruses and malware because avast! Antivirus protection 
is active.
http://www.avast.com


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-25 Thread Dave Swarthout

+1 for

Not when you could have other entities passing under the bridges.
I see it as lazy  less accurate. Making OSM more accurate is a primary
consideration when editing


On Tue, Mar 25, 2014 at 7:16 PM, Dave F. dave...@madasafish.com wrote:

 On 14/03/2014 15:57, Pieren wrote:


 You don't see the point where adding one layer=-1  is easier than
 adding 10 layer=1 ?


 Not when you could have other entities passing under the bridges.
 I see it as lazy  less accurate. Making OSM more accurate is a primary
 consideration when editing

  I see the layer tag in tunnel/bridge in simple
 cases but you should not follow all recommendations as fixed in stone.


 True, but you should follow the correct ones.

 ---
 This email is free from viruses and malware because avast! Antivirus
 protection is active.
 http://www.avast.com


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




-- 
Dave Swarthout
Homer, Alaska
Chiang Mai, Thailand
Travel Blog at http://dswarthout.blogspot.com
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-25 Thread Richard Z.

On Mon, Mar 24, 2014 at 10:38:08PM +0100, fly wrote:
 On 24.03.2014 20:45, Richard Z. wrote:
  On Mon, Mar 24, 2014 at 11:02:35AM -0300, Fernando Trebien wrote:
  On Mon, Mar 24, 2014 at 10:27 AM, Friedrich Volkmann b...@volki.at wrote:
 
 As it might be even hard to define the ground level (we just have a
 discussion on talk-de@ about houses built on slops), I would never say
 that an negative layer value is an indicator/synonym for underground.
 
  Again, no mention in the wiki
  (http://wiki.openstreetmap.org/wiki/Key:layer) to negative layer
  values being used to represent the idea of underground.
  
  not an explicit mention, but if there is an object X1 with implicit layer=0,
  with no level or location tags, and another object X2 with a layer=-1 than
  there are not too many possibilities where to find X2. It could be 
  underground
  or it could be under a large overhanging rock. Both should have explicit
  tags to clarify the situation.
  
  
  No, except for underground rivers. They do exist in karst regions...
  
  we need a way to tag underground rivers and lakes. layer=-1 itself is not
  sufficient, we need additional tags. Perhaps tunnel=cave but this would
  only describe part of those phenomena.
 
 How about location=underground ?

seems good to me.

 covered=yes is another useful tag (eg your overhanging rock)

covered is already complicated enough and this might be too much of
stretching the original idea.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-24 Thread Friedrich Volkmann

On 14.03.2014 15:51, Fernando Trebien wrote:
 This is a small issue that came up recently in Brazil. In my
 understanding, the layer tag has no specific meaning other than to
 specify a rendering order.

That is a common misconception by people who worked with graphics editing
software such as Photoshop and Corel Draw. In these applications, layers are
only used for ordering and grouping of objects. As opposed to that abstract
layer model, we use a pysical layer model in OSM where layer=0 means
ground-level, layer0 means underground, and layer0 means above ground-level.

Renderers typically use a different rendering style for underground
features, e.g. dashed  or greyed-out lines, or these features may even be
omitted. The only way how renderers can determine whether a feature is
underground is by looking at the layer tag. tunnel=* and bridge=* do not
necessarily mean that a feature is underground / above ground. There was
even a negative voting on this (a proposal for implying layer=-1 or 1
respectively). And the other way round, not all underground objects are
tunnels or culverts. A tunnel is a way with an entrance on each end. It is
not a tunnel if it has a dead end. And what about POIs and areas? It would
be stupid to tag them as tunnels. The tunnel=* and bridge=* tags do help the
renderers when it comes to the curved bridge and tunnel signatures, but the
dashing can only depend on the layer tag.

 The wiki, however, states that it is wrong
 to tag a whole river with layer=-1. The reason for that, as far as I
 could figure, is because current validators (such as JOSM's or
 KeepRight's) will not issue a warning on a waterway x highway crossing
 when their layers are different, leading some users into tagging the
 river with layer=-1 in order to get rid of warnings about missing
 bridges and tunnels.

The warning is ok, but the big problem with validators is that mappers mess
around with the data without local knowledge, just to shut up the validator.
They connect ways which are not really connected, they insert bridges and
culverts that do not really exist, and so on. Among these so-called fixes,
the layer=-1 on waterways are relatively harmless, because it is so obvious
that they are wrong. A culvert that has been made up by a sofa mapper is
much more difficult to correct, because you only know when you go there.

 Do you agree that the river can be tagged with layer=-1 as long as
 this value is correct in relation to the layer of other
 nearby/crossing ways?

No, except for underground rivers. They do exist in karst regions...

-- 
Friedrich K. Volkmann   http://www.volki.at/
Adr.: Davidgasse 76-80/14/10, 1100 Wien, Austria

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-24 Thread Martin Koppenhoefer



 Am 24/mar/2014 um 14:27 schrieb Friedrich Volkmann b...@volki.at:
 
 As opposed to that abstract
 layer model, we use a pysical layer model in OSM where layer=0 means
 ground-level, layer0 means underground, and layer0 means above ground-level.


AFAIK it is not defined like this, rather it is meant to describe real world 
stacking (as opposed to rendering order) at a local point: when 2 objects cross 
the one with the higher layer is above, if both layers are equal they are on 
the same level

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-24 Thread Fernando Trebien

On Mon, Mar 24, 2014 at 10:27 AM, Friedrich Volkmann b...@volki.at wrote:
 On 14.03.2014 15:51, Fernando Trebien wrote:
 This is a small issue that came up recently in Brazil. In my
 understanding, the layer tag has no specific meaning other than to
 specify a rendering order.

 That is a common misconception by people who worked with graphics editing
 software such as Photoshop and Corel Draw. In these applications, layers are
 only used for ordering and grouping of objects. As opposed to that abstract
 layer model, we use a pysical layer model in OSM where layer=0 means
 ground-level, layer0 means underground, and layer0 means above ground-level.

This is not written anywhere in the wiki:
http://wiki.openstreetmap.org/wiki/Key:layer

 Renderers typically use a different rendering style for underground
 features, e.g. dashed  or greyed-out lines, or these features may even be
 omitted. The only way how renderers can determine whether a feature is
 underground is by looking at the layer tag. tunnel=* and bridge=* do not
 necessarily mean that a feature is underground / above ground. There was
 even a negative voting on this (a proposal for implying layer=-1 or 1
 respectively). And the other way round, not all underground objects are
 tunnels or culverts. A tunnel is a way with an entrance on each end. It is
 not a tunnel if it has a dead end. And what about POIs and areas? It would
 be stupid to tag them as tunnels. The tunnel=* and bridge=* tags do help the
 renderers when it comes to the curved bridge and tunnel signatures, but the
 dashing can only depend on the layer tag.

The concept of underground seems more closely related to the tags
level and location:
http://wiki.openstreetmap.org/wiki/Key:level
http://wiki.openstreetmap.org/wiki/Key:location

Again, no mention in the wiki
(http://wiki.openstreetmap.org/wiki/Key:layer) to negative layer
values being used to represent the idea of underground.

 The wiki, however, states that it is wrong
 to tag a whole river with layer=-1. The reason for that, as far as I
 could figure, is because current validators (such as JOSM's or
 KeepRight's) will not issue a warning on a waterway x highway crossing
 when their layers are different, leading some users into tagging the
 river with layer=-1 in order to get rid of warnings about missing
 bridges and tunnels.

 The warning is ok, but the big problem with validators is that mappers mess
 around with the data without local knowledge, just to shut up the validator.
 They connect ways which are not really connected, they insert bridges and
 culverts that do not really exist, and so on. Among these so-called fixes,
 the layer=-1 on waterways are relatively harmless, because it is so obvious
 that they are wrong. A culvert that has been made up by a sofa mapper is
 much more difficult to correct, because you only know when you go there.

I understand this issue, but don't you think validators could be a
little bit more clever, so that users couldn't trick them? In some
previous message I described some very simple logic that would make
these tricks (in this situation) impossible to work around, and this
would discourage people from using layer=-1 on the rivers.

 Do you agree that the river can be tagged with layer=-1 as long as
 this value is correct in relation to the layer of other
 nearby/crossing ways?

 No, except for underground rivers. They do exist in karst regions...
 --
 Friedrich K. Volkmann   http://www.volki.at/
 Adr.: Davidgasse 76-80/14/10, 1100 Wien, Austria

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



-- 
Fernando Trebien
+55 (51) 9962-5409

The speed of computer chips doubles every 18 months. (Moore's law)
The speed of software halves every 18 months. (Gates' law)

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-24 Thread Richard Z.

On Mon, Mar 24, 2014 at 11:02:35AM -0300, Fernando Trebien wrote:
 On Mon, Mar 24, 2014 at 10:27 AM, Friedrich Volkmann b...@volki.at wrote:
  On 14.03.2014 15:51, Fernando Trebien wrote:
  This is a small issue that came up recently in Brazil. In my
  understanding, the layer tag has no specific meaning other than to
  specify a rendering order.
 
  That is a common misconception by people who worked with graphics editing
  software such as Photoshop and Corel Draw. In these applications, layers are
  only used for ordering and grouping of objects. As opposed to that abstract
  layer model, we use a pysical layer model in OSM where layer=0 means
  ground-level, layer0 means underground, and layer0 means above 
  ground-level.
 
 This is not written anywhere in the wiki:
 http://wiki.openstreetmap.org/wiki/Key:layer

to clear up this misunderstnding, this has been a relatively recent change 
by me. If you look at it the wording has changed but the practical application
has not.

Earlier there was a longstanding mention of layer=0 as ground-level
along with a complicated definition of what ground-level is supposed to be.

I have changed that mainly because the way ground-level was defined was
prone to interpretation problems and difficult to define better. 

The new formulation of the kind higher value means higher above does 
not need the explicit definition of ground-level.
Also the definition of layer=0 as ground-level confused too many people to 
tag elevated roads and similar with layer=1.

However, the new formulation is so that in practice layer=0 will always be 
ground-level except perhaps in very complicated urban areas:


Ways passing above other ways on a bridge will have a higher layer value, ways 
passing in tunnels bellow other ways will have lower (negative) values. All 
ways without an explicit value are assumed to have layer 0.


 
 Again, no mention in the wiki
 (http://wiki.openstreetmap.org/wiki/Key:layer) to negative layer
 values being used to represent the idea of underground.

not an explicit mention, but if there is an object X1 with implicit layer=0,
with no level or location tags, and another object X2 with a layer=-1 than
there are not too many possibilities where to find X2. It could be underground
or it could be under a large overhanging rock. Both should have explicit
tags to clarify the situation.


  No, except for underground rivers. They do exist in karst regions...

we need a way to tag underground rivers and lakes. layer=-1 itself is not
sufficient, we need additional tags. Perhaps tunnel=cave but this would
only describe part of those phenomena.

I am compiling a list of landforms and natural objects that would be worthwile
to map.

Richard


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-24 Thread fly

On 24.03.2014 20:45, Richard Z. wrote:
 On Mon, Mar 24, 2014 at 11:02:35AM -0300, Fernando Trebien wrote:
 On Mon, Mar 24, 2014 at 10:27 AM, Friedrich Volkmann b...@volki.at wrote:

As it might be even hard to define the ground level (we just have a
discussion on talk-de@ about houses built on slops), I would never say
that an negative layer value is an indicator/synonym for underground.

 Again, no mention in the wiki
 (http://wiki.openstreetmap.org/wiki/Key:layer) to negative layer
 values being used to represent the idea of underground.
 
 not an explicit mention, but if there is an object X1 with implicit layer=0,
 with no level or location tags, and another object X2 with a layer=-1 than
 there are not too many possibilities where to find X2. It could be underground
 or it could be under a large overhanging rock. Both should have explicit
 tags to clarify the situation.
 
 
 No, except for underground rivers. They do exist in karst regions...
 
 we need a way to tag underground rivers and lakes. layer=-1 itself is not
 sufficient, we need additional tags. Perhaps tunnel=cave but this would
 only describe part of those phenomena.

How about location=underground ?
covered=yes is another useful tag (eg your overhanging rock)

fly

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-16 Thread Martin Koppenhoefer



 Am 16/mar/2014 um 01:42 schrieb Richard Z. ricoz@gmail.com:
 
 Also building=bridge is the wrong tag for this bridge



why? Let's be cautious with judgements like wrong tag and even more in 
situations where there is not clearly a generally adopted tagging scheme (like 
for bridges).

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Sat, Mar 15, 2014 at 01:24:07AM +0200, Ilpo Järvinen wrote:
 On Sat, 15 Mar 2014, Richard Z. wrote:

   Therefore, everyone needs now to handle those hardly useful layer 
   warnings about trivial cases (and waste their time on correcting them). 
  
  even worse, people just apply layer=-1 to thousands of miles of rivers and
  similar tricks to hide those warnings.
 
 Which proves my point. The mappers didn't like the unnecessary burden
 nor the warnings which do in no way improve quality but only reduce 
 signal-to-noise of the validator.

I am in favor of having the warnings fully configurable. Obviously if you
do not know an area well you can't decide whether a waterway crossing should 
be a bridge, culvert or a ford and should not be bothered with such warnings.
And I am thinking this warning should be off by default because it is one
of the most frequently useless warnings that I know.

File tickets for the JOSM validator whenever you think the validator could 
be improved or otherwise fine tuned.

Other warnings otoh should be added. There are many instances of tunnel=culvert
without a layer and almost all of them were accidental errors - someone added
the culvert to the wrong segment of the way.

As of the bridges, the editing software could make it a lot easier to create 
them. Currently its quite many manual steps to insert a bridge properly and 
I think there would be a demand to have a plugin or whatever doing it easier.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-15 Thread André Pirard

Hi,

I wonder why we make bridges split and split and split the roads.
In reality, bridges are pieces of concrete or stonework at level -1
under an uninterrupted foil of tarmac at level 0.
Or at level 0 if it's understood that the renderer knows what's a bridge.
And the renderer knows, as it draws two thin stripes beside the road.
So, a bridge can be a little way segment overlaying the road.
This lets the routing software ignore the unnecessary complication of
having to account for bridges as part of the route.
This lets the bridge having its own attributes, unrelated to the road,
for example a different name.
This makes obsolete discussions wondering if the bridge must be split in
two because the road changes in the middle.
Etc. etc., all pieces clutch in very neatly.
And BTW, this is similar to tunnel=culvert which is an optional feature
of a bridge and that surprises no one at layer -1.
And now, if we put bridges and culverts at -1, the rivers or streams are
normally at -2.
Tunnels (inside which the road runs) should be segments too, at level +1
or 0.

I have tagged a number of streams and rivers at -2 -1 0 and I find it
appreciable to have an instant view of where the complete main stream
is, if not exaggeratedly long, as well as less prone to errors.

Cheers,

André.



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

Re: [Tagging] layer=-1, rivers, bridges and tunnels



 Am 14/mar/2014 um 15:51 schrieb Fernando Trebien fernando.treb...@gmail.com:
 
 Do you agree that the river can be tagged with layer=-1 as long as
 this value is correct in relation to the layer of other
 nearby/crossing ways?


I would discourage you to do so. Layer tags should only be applied to ways that 
actually cross other objects on different layers (ie without intersecting them).

To me the Josm way of warning seems correct, as a crossing of objects on 
different layers should issue a warning.

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels



 Am 14/mar/2014 um 16:36 schrieb Pieren pier...@gmail.com:
 
 Real case from real world : a deep ditch where the stream is not
 underground but below the ground level, is crossing a village
 where we have 10 bridges. Either you add 10 times layer=1 on the
 bridges or you add 1 time layer=-1 on the stream.


in a deep ditch the waterway is still on ground level, just that the ground is 
lower at this point than it is in the surroundings. 

I would advocate for adding layer=1 to the bridges and leave the waterway 
without layer tag

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels



 Am 14/mar/2014 um 16:35 schrieb Fernando Trebien fernando.treb...@gmail.com:
 
 From this logic, layer=-1 means the object is rendered beneath
 anything that has layer=0 (or, conversely, that anything with layer=0
 is rendered on top of anything with layer=-1). It does not mean that
 it is in fact below it (though it almost always is).


no, the opposite is true, you will render stuff as you like (when using dashed 
lines for underground features you might opt to render them above other lines 
for instance), but the real world stacking order is given by the layer tag

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels



 Am 14/mar/2014 um 19:55 schrieb Fernando Trebien fernando.treb...@gmail.com:
 
 I don't think you should be required to check the river's layer tag.
 Validators should do this job for you, it's quite easy to write a rule
 for that.


first you'll have to download all data along this river in order to make this 
work,seems easier to download all data along a bridge when you add it then to 
put layer tags on long ways which tend to extend into not downloaded areas when 
you edit

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-15 Thread Peter Wendorff

Hi,

I agree partially with you here.
Yes, adding bridges in addition to the road is possible and may be a
good idea.
What we currently map as being a bridge in fact is the property of the
road is on a bridge instead.
Changing the current tagging scheme to duplicate the corresponding
segment of the way and tag the bridge as a separate, but again linear
object is worse in all but one point.
The only point this is better in is that a street with a continuous name
may not have to be splitted because of the bridge; but on the other hand
we do so for anything else, too: speed restrictions, footway or not,
highway type, surface and anything else; so it doesn't solve an issue
dedicated to bridges.

On the other hand it doesn't solve the issue with multiple parallel ways
on the same bridge, e.g. considering a dual carriage way on one bridge
construction we currently map the property road is on a bridge again
on both parts of the dual carriage way independently, but it's
impossible to decide from the data (usually) if it's one bridge or two
bridges.
Your proposal to duplicate the way does not solve this issue either, as
you would still need two separate ways here.

regards
Peter


Am 15.03.2014 13:25, schrieb André Pirard:
 Hi,
 
 I wonder why we make bridges split and split and split the roads.
 In reality, bridges are pieces of concrete or stonework at level -1
 under an uninterrupted foil of tarmac at level 0.
 Or at level 0 if it's understood that the renderer knows what's a bridge.
 And the renderer knows, as it draws two thin stripes beside the road.
 So, a bridge can be a little way segment overlaying the road.
 This lets the routing software ignore the unnecessary complication of
 having to account for bridges as part of the route.
 This lets the bridge having its own attributes, unrelated to the road,
 for example a different name.
 This makes obsolete discussions wondering if the bridge must be split in
 two because the road changes in the middle.
 Etc. etc., all pieces clutch in very neatly.
 And BTW, this is similar to tunnel=culvert which is an optional feature
 of a bridge and that surprises no one at layer -1.
 And now, if we put bridges and culverts at -1, the rivers or streams are
 normally at -2.
 Tunnels (inside which the road runs) should be segments too, at level +1
 or 0.
 
 I have tagged a number of streams and rivers at -2 -1 0 and I find it
 appreciable to have an instant view of where the complete main stream
 is, if not exaggeratedly long, as well as less prone to errors.
 
 Cheers,
 
 André.
 
 
 
 
 
 
 ___
 Tagging mailing list
 Tagging@openstreetmap.org
 https://lists.openstreetmap.org/listinfo/tagging
 


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-15 Thread John Packer

I believe there was a proposal for tagging a bridge separately:
man_made=bridge. I think it would be really nice to have the actual outline
of the bridge rendered
Em 15/03/2014 10:02, Peter Wendorff wendo...@uni-paderborn.de escreveu:

 Hi,

 I agree partially with you here.
 Yes, adding bridges in addition to the road is possible and may be a
 good idea.
 What we currently map as being a bridge in fact is the property of the
 road is on a bridge instead.
 Changing the current tagging scheme to duplicate the corresponding
 segment of the way and tag the bridge as a separate, but again linear
 object is worse in all but one point.
 The only point this is better in is that a street with a continuous name
 may not have to be splitted because of the bridge; but on the other hand
 we do so for anything else, too: speed restrictions, footway or not,
 highway type, surface and anything else; so it doesn't solve an issue
 dedicated to bridges.

 On the other hand it doesn't solve the issue with multiple parallel ways
 on the same bridge, e.g. considering a dual carriage way on one bridge
 construction we currently map the property road is on a bridge again
 on both parts of the dual carriage way independently, but it's
 impossible to decide from the data (usually) if it's one bridge or two
 bridges.
 Your proposal to duplicate the way does not solve this issue either, as
 you would still need two separate ways here.

 regards
 Peter


 Am 15.03.2014 13:25, schrieb André Pirard:
  Hi,
 
  I wonder why we make bridges split and split and split the roads.
  In reality, bridges are pieces of concrete or stonework at level -1
  under an uninterrupted foil of tarmac at level 0.
  Or at level 0 if it's understood that the renderer knows what's a bridge.
  And the renderer knows, as it draws two thin stripes beside the road.
  So, a bridge can be a little way segment overlaying the road.
  This lets the routing software ignore the unnecessary complication of
  having to account for bridges as part of the route.
  This lets the bridge having its own attributes, unrelated to the road,
  for example a different name.
  This makes obsolete discussions wondering if the bridge must be split in
  two because the road changes in the middle.
  Etc. etc., all pieces clutch in very neatly.
  And BTW, this is similar to tunnel=culvert which is an optional feature
  of a bridge and that surprises no one at layer -1.
  And now, if we put bridges and culverts at -1, the rivers or streams are
  normally at -2.
  Tunnels (inside which the road runs) should be segments too, at level +1
  or 0.
 
  I have tagged a number of streams and rivers at -2 -1 0 and I find it
  appreciable to have an instant view of where the complete main stream
  is, if not exaggeratedly long, as well as less prone to errors.
 
  Cheers,
 
  André.
 
 
 
 
 
 
  ___
  Tagging mailing list
  Tagging@openstreetmap.org
  https://lists.openstreetmap.org/listinfo/tagging
 


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

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Sat, Mar 15, 2014 at 01:25:16PM +0100, André Pirard wrote:
 Hi,
 
 I wonder why we make bridges split and split and split the roads.

do not like that too much either.

 In reality, bridges are pieces of concrete or stonework at level -1
 under an uninterrupted foil of tarmac at level 0.

but in our model we don't map the concret, nor do we map cellars instead
of houses. We map bridges as a property of the road.

There is also the possibility to use man_made=bridge instead.

 Or at level 0 if it's understood that the renderer knows what's a bridge.
 And the renderer knows, as it draws two thin stripes beside the road.
 So, a bridge can be a little way segment overlaying the road.

it must be somehow connected to the road though. Do you advocate overlapping
ways?


 And BTW, this is similar to tunnel=culvert which is an optional feature
 of a bridge and that surprises no one at layer -1.
 And now, if we put bridges and culverts at -1, the rivers or streams are
 normally at -2.
 Tunnels (inside which the road runs) should be segments too, at level +1
 or 0.
 
 I have tagged a number of streams and rivers at -2 -1 0 and I find it
 appreciable to have an instant view of where the complete main stream
 is, if not exaggeratedly long, as well as less prone to errors.

I think everyone else who will come across this tagging will remove
your layer tags as incorrect.

Seems to me the wiki should be kept in sync better.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-15 Thread Peter Wendorff

Hi John,
yes, that's one possibility; knew that already, but thanks for pointing
the list to the link.

regards
Peter

Am 15.03.2014 14:16, schrieb John Packer:
 I believe there was a proposal for tagging a bridge separately:
 man_made=bridge. I think it would be really nice to have the actual outline
 of the bridge rendered
 Em 15/03/2014 10:02, Peter Wendorff wendo...@uni-paderborn.de escreveu:
 
 Hi,

 I agree partially with you here.
 Yes, adding bridges in addition to the road is possible and may be a
 good idea.
 What we currently map as being a bridge in fact is the property of the
 road is on a bridge instead.
 Changing the current tagging scheme to duplicate the corresponding
 segment of the way and tag the bridge as a separate, but again linear
 object is worse in all but one point.
 The only point this is better in is that a street with a continuous name
 may not have to be splitted because of the bridge; but on the other hand
 we do so for anything else, too: speed restrictions, footway or not,
 highway type, surface and anything else; so it doesn't solve an issue
 dedicated to bridges.

 On the other hand it doesn't solve the issue with multiple parallel ways
 on the same bridge, e.g. considering a dual carriage way on one bridge
 construction we currently map the property road is on a bridge again
 on both parts of the dual carriage way independently, but it's
 impossible to decide from the data (usually) if it's one bridge or two
 bridges.
 Your proposal to duplicate the way does not solve this issue either, as
 you would still need two separate ways here.

 regards
 Peter


 Am 15.03.2014 13:25, schrieb André Pirard:
 Hi,

 I wonder why we make bridges split and split and split the roads.
 In reality, bridges are pieces of concrete or stonework at level -1
 under an uninterrupted foil of tarmac at level 0.
 Or at level 0 if it's understood that the renderer knows what's a bridge.
 And the renderer knows, as it draws two thin stripes beside the road.
 So, a bridge can be a little way segment overlaying the road.
 This lets the routing software ignore the unnecessary complication of
 having to account for bridges as part of the route.
 This lets the bridge having its own attributes, unrelated to the road,
 for example a different name.
 This makes obsolete discussions wondering if the bridge must be split in
 two because the road changes in the middle.
 Etc. etc., all pieces clutch in very neatly.
 And BTW, this is similar to tunnel=culvert which is an optional feature
 of a bridge and that surprises no one at layer -1.
 And now, if we put bridges and culverts at -1, the rivers or streams are
 normally at -2.
 Tunnels (inside which the road runs) should be segments too, at level +1
 or 0.

 I have tagged a number of streams and rivers at -2 -1 0 and I find it
 appreciable to have an instant view of where the complete main stream
 is, if not exaggeratedly long, as well as less prone to errors.

 Cheers,

 André.






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



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

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


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-15 Thread Frank Little


Martin Koppenhoefer wrote:
Am 14/mar/2014 um 15:51 schrieb Fernando Trebien 
fernando.treb...@gmail.com:


Do you agree that the river can be tagged with layer=-1 as long as
this value is correct in relation to the layer of other
nearby/crossing ways?


I would discourage you to do so. Layer tags should only be applied to ways 
that actually cross other objects on different layers (ie without 
intersecting them).


I agree totally with: Layer tags should only be applied to ways that actually 
cross other objects.


At its simplest, a layer tag is a hint to a renderer which of  two crossing 
ways should be rendered later (i.e. on top). If a renderer does not apply the 
real world knowledge that a bridge (by its definition) crosses over a way 
(road, water, whatever) underneath, then it can still take the hint to render 
it correctly. The renderers have no problem interpreting the situation 
correctly, with or without the layer tag, afaik.


A layer tag is not a way to define the relative height of different objects. 
Some of the discussion on the proposal's talk page is confused about that.


I would tag the structure (bridge or tunnel) with a layer tag*.
I would not tag a river or stream along its entire length.

Rivers, streams, canals, etc. are surface features (in most cases). The mere 
fact that the bed of a waterway is often  at a lower level than the 
surrounding ground level is not relevant for the layer tag since hinting for 
correct rendering is not necessary. (In the Netherlands and other polder 
areas, waterways are often above the surrounding area.)


*Actually, as I made clear on talk when we had this discussion very recently, 
I would prefer not to use the layer tag at all in most of these cases. The 
fact that somewhere between one quarter (taginfo) and one third (overpass 
turbo samples in the Netherlands) do not use a layer tag with bridges 
indicates to me that it is not as clear cut as people are suggesting. (Note: I 
realise that there are specific cases where explicit tagging for layer hinting 
is necessary (e.g. bridges or viaducts layered vertically). These are 
relatively rare.) 



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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

Alright. I see that applying layer to long ways is bad for several
reasons. Surely this could be turned into a validation warning.

But what's the difference between tagging the bridge with layer=1 and
tagging the river underneath with layer=-1? Some people seem to think
that both are necessary, many think it's best to use layer=1 on the
bridge, I'm saying that layer=-1 on the river (let's say a short
section, not the entire length) is equivalent. Is it not equivalent?
Is it wrong? If it is wrong, why is it wrong?

On Sat, Mar 15, 2014 at 12:53 PM, Frank Little frank...@xs4all.nl wrote:
 Martin Koppenhoefer wrote:

 Am 14/mar/2014 um 15:51 schrieb Fernando Trebien
 fernando.treb...@gmail.com:

 Do you agree that the river can be tagged with layer=-1 as long as
 this value is correct in relation to the layer of other
 nearby/crossing ways?


 I would discourage you to do so. Layer tags should only be applied to ways
 that actually cross other objects on different layers (ie without
 intersecting them).

 I agree totally with: Layer tags should only be applied to ways that
 actually cross other objects.

 At its simplest, a layer tag is a hint to a renderer which of  two crossing
 ways should be rendered later (i.e. on top). If a renderer does not apply
 the real world knowledge that a bridge (by its definition) crosses over a
 way (road, water, whatever) underneath, then it can still take the hint to
 render it correctly. The renderers have no problem interpreting the
 situation correctly, with or without the layer tag, afaik.

 A layer tag is not a way to define the relative height of different objects.
 Some of the discussion on the proposal's talk page is confused about that.

 I would tag the structure (bridge or tunnel) with a layer tag*.
 I would not tag a river or stream along its entire length.

 Rivers, streams, canals, etc. are surface features (in most cases). The mere
 fact that the bed of a waterway is often  at a lower level than the
 surrounding ground level is not relevant for the layer tag since hinting for
 correct rendering is not necessary. (In the Netherlands and other polder
 areas, waterways are often above the surrounding area.)

 *Actually, as I made clear on talk when we had this discussion very
 recently, I would prefer not to use the layer tag at all in most of these
 cases. The fact that somewhere between one quarter (taginfo) and one third
 (overpass turbo samples in the Netherlands) do not use a layer tag with
 bridges indicates to me that it is not as clear cut as people are
 suggesting. (Note: I realise that there are specific cases where explicit
 tagging for layer hinting is necessary (e.g. bridges or viaducts layered
 vertically). These are relatively rare.)

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



-- 
Fernando Trebien
+55 (51) 9962-5409

The speed of computer chips doubles every 18 months. (Moore's law)
The speed of software halves every 18 months. (Gates' law)

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-15 Thread Frank Little


Fernando Trebien wrote:

Alright. I see that applying layer to long ways is bad for several
reasons. Surely this could be turned into a validation warning.

But what's the difference between tagging the bridge with layer=1 and
tagging the river underneath with layer=-1? Some people seem to think
that both are necessary, many think it's best to use layer=1 on the
bridge, I'm saying that layer=-1 on the river (let's say a short
section, not the entire length) is equivalent. Is it not equivalent?
Is it wrong? If it is wrong, why is it wrong?

I don't think 'wrong' is the way to approach this; afaik, they are indeed 
equivalent.
There are four alternatives which mappers follow, none of which are 'wrong': 
tag the bridge segment, tag the water segment under the bridge, tag both, tag 
neither.


I've run waterway=stream or =canal (or =ditch, I think) through a few of the 
small rivers and streams here in the Netherlands. Roads need splitting to make 
bridges, so it makes sense to do all the relevant tagging on the road segment 
with the bridge tag when you are working on it.


I don't have any reason to split the waterway=*, so I just draw on without 
stopping. Since I put the name on the waterway and not on the riverbank, it 
leaves it to the renderer to find a good place to fit in the 
river/stream/canal name. In principle, that should mean a cleaner map if the 
renderer can work out the proper placement (difficult job, though).


So I would be against splitting the waterway at a bridge and tagging it 
layer=-1 on practical grounds. And you can be sure that it would cause 
confusion with other mappers who would imagine that you are trying to model an 
inverted siphon with the piece of waterway tagged layer=-1, or that you had 
simply made a mistake. (The 'level' confusion again.) 



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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

the validator will only prevent the most obvious errors but will give
you no clue how to fix them correctly

I know. But two or three rounds of trial and error with the validator
should be enough to bring a new user to an acceptable representation.

there is no difference between connections in endpoints or in a
crossing point as far as I can tell.

You're right. I should have used overlap not cross. That's what I
meant really. If a waterway and a highway cross, that means they share
a common node, so it represents a ford and maybe a dam. If they
overlap but do not cross, they should, in principle, be in different
vertical positions, so they should have a different value in the
layer tag.

or one of covered,location,indoor,steps,lift or level, maybe more.

I have to read more about them an check their usage, but they could
all be incorporated in the same rule in the validator. It's just a few
extra values in a set referred to in a rule. Surely this set can grow
over time.

except for indoor mapping and maybe other weird cases.

I'm not so much involved with indoor mapping yet, but I think the
rules would still apply. What I know about indoor mapping (possibly
too little): it is being done in such a way that people will first
filter by level and then render. So layer probably applies within
level. If you have two ways that overlap at the same level but do
share a node, they must sit at different layers, right? They must be
vertically displaced, so one of them should be an indoor tunnel and
the other should be an indoor bridge (or something alike), right?

also railways?

As far as I can imagine, it should apply to railways too. Also to
combinations of railways with highways, and railways with waterways.
(If it does not, please show me an example.)

more general: not connected, different layer values and not one of
bridge,tunnel,covered,location,indoor,steps,lift, no level tag and
a few more things to take into account.

I believe you mean that these propositions are joined with AND
logic: not connected AND different layer values AND not of
{bridge,tunnel,...} AND no level tag, right?

Different layer values AND not one of {bridge,tunnel,...} will issue a
warning for the way without bridge=* or layer=* that goes underneath
bridge=yes+layer=1, right?

identical to d?

Exactly, that's the point I'm trying to state. Layer is a relative
value, its actual values should not be assigned any special meaning.
Layer=0 does not mean ground level, and layer=-1 does not mean
underground nor should be forbidden for rivers (as long as it obeys
the other thumb rule: use layer only in short ways or short spans of
a way).

unless indoor or other strange cases

Correct, let's add within the same level to all of those rules, and
assume level=0 when level is not specified in a tag. Then they all
work also for indoor mapping.

It is a lot easier saying that every bridge and tunnel must have a
layer tag and enforce that than catching all the situations mentioned
in situation d.

Yes but it is also much harder to get everyone in the world to follow
it. Even a person that knows that rule may forget to apply it
sometimes. I know that this same reason does not apply to many other
similar situations in which a validation rule would be much more
complex than that.

With some luck, you can restrict d to waterways and it becomes easy.

Fair enough, since the current problem mostly concerns waterways. I
just tried to arrive at a more generic rule, which I believed would be
more useful in the long term.

On Fri, Mar 14, 2014 at 5:12 PM, Richard Z. ricoz@gmail.com wrote:
 On Fri, Mar 14, 2014 at 03:55:39PM -0300, Fernando Trebien wrote:
 I don't think you should be required to check the river's layer tag.
 Validators should do this job for you, it's quite easy to write a rule
 for that.

 validators can check for many errors but if you want to change
 anything you have to understand the whole situation.
 Imagine you want to add a new bridge to a complex freeway intersection
 with junctions and overpasses.. the validator will only prevent the
 most obvious errors but will give you no clue how to fix them
 correctly.

 Given two ways that cross internally (excluding connections at
 endpoints), and considering the layer value defined explicitly in a
 tag or implicitly 0 when the tag is missing, have the validator issue
 a warning in the following situations:

 there is no difference between connections in endpoints or in a crossing
 point as far as I can tell.

 1. The ways have the same layer value and are unconnected. (They
 should be connected, or else something is surely missing. This could
 actually be considered an error.)

 except for aerial ways and similar exceptions

 1.1. Also warn if if one way is a waterway and the other is a highway
 and the connection is not explicitly a ford. (It should be, for
 clarity. If it's not, it's also possibly not a ford, therefore the
 connection is wrong.)

 there is also the odd case of highways

Re: [Tagging] layer=-1, rivers, bridges and tunnels

I thought a bit more and this statement I said is incorrect:

Correct, let's add within the same level to all of those rules, and
assume level=0 when level is not specified in a tag. Then they all
work also for indoor mapping.

The correct wording of those warning rules, taking indoor mapping into
account, should be:

---
If level is missing, use the value of location instead for the
following rules. Otherwise, consider it equal to 0.

When layer is equal between both ways, and so is level, warn when:
1. The ways are unconnected (ie. they just overlap without sharing a node).
2. The ways are connected, one way is a waterway and the other is a
highway and the connection is not explicitly a ford or a dam.

When only level is equal between both ways, warn when:
3. The ways have different layer value and both are missing a tunnel
or a bridge tag.
4. The layer value of a bridge is inferior to that of a way that is
not a bridge.
5. The layer value of a tunnel is superior to that of a way that is
not a tunnel.

In addition, also warn when location=underground/underwater and level
 0, and when location=overground and level  0.
---

I thought about the meaning of covered and steps, but they don't say
anything about the vertical order of the overlapping elements, so it
may be interesting to get users to declare that order when ways
overlap within the same level.

Please see if you can find a situation that would not be identified by
those rules, or one that would be identified incorrectly.

On Sat, Mar 15, 2014 at 2:06 PM, Fernando Trebien
fernando.treb...@gmail.com wrote:
 the validator will only prevent the most obvious errors but will give
 you no clue how to fix them correctly

 I know. But two or three rounds of trial and error with the validator
 should be enough to bring a new user to an acceptable representation.

 there is no difference between connections in endpoints or in a
 crossing point as far as I can tell.

 You're right. I should have used overlap not cross. That's what I
 meant really. If a waterway and a highway cross, that means they share
 a common node, so it represents a ford and maybe a dam. If they
 overlap but do not cross, they should, in principle, be in different
 vertical positions, so they should have a different value in the
 layer tag.

 or one of covered,location,indoor,steps,lift or level, maybe more.

 I have to read more about them an check their usage, but they could
 all be incorporated in the same rule in the validator. It's just a few
 extra values in a set referred to in a rule. Surely this set can grow
 over time.

 except for indoor mapping and maybe other weird cases.

 I'm not so much involved with indoor mapping yet, but I think the
 rules would still apply. What I know about indoor mapping (possibly
 too little): it is being done in such a way that people will first
 filter by level and then render. So layer probably applies within
 level. If you have two ways that overlap at the same level but do
 share a node, they must sit at different layers, right? They must be
 vertically displaced, so one of them should be an indoor tunnel and
 the other should be an indoor bridge (or something alike), right?

 also railways?

 As far as I can imagine, it should apply to railways too. Also to
 combinations of railways with highways, and railways with waterways.
 (If it does not, please show me an example.)

 more general: not connected, different layer values and not one of
 bridge,tunnel,covered,location,indoor,steps,lift, no level tag and
 a few more things to take into account.

 I believe you mean that these propositions are joined with AND
 logic: not connected AND different layer values AND not of
 {bridge,tunnel,...} AND no level tag, right?

 Different layer values AND not one of {bridge,tunnel,...} will issue a
 warning for the way without bridge=* or layer=* that goes underneath
 bridge=yes+layer=1, right?

 identical to d?

 Exactly, that's the point I'm trying to state. Layer is a relative
 value, its actual values should not be assigned any special meaning.
 Layer=0 does not mean ground level, and layer=-1 does not mean
 underground nor should be forbidden for rivers (as long as it obeys
 the other thumb rule: use layer only in short ways or short spans of
 a way).

 unless indoor or other strange cases

 Correct, let's add within the same level to all of those rules, and
 assume level=0 when level is not specified in a tag. Then they all
 work also for indoor mapping.

 It is a lot easier saying that every bridge and tunnel must have a
 layer tag and enforce that than catching all the situations mentioned
 in situation d.

 Yes but it is also much harder to get everyone in the world to follow
 it. Even a person that knows that rule may forget to apply it
 sometimes. I know that this same reason does not apply to many other
 similar situations in which a validation rule would be much more
 complex than that.

 With some luck, you can restrict d to waterways and it

Re: [Tagging] layer=-1, rivers, bridges and tunnels

Here are a few arguable reasons to split the waterway and tag it with layer=-1:
1. Bridges may come in pairs for dual carriageways. In this case, it's
a single layer tag for the waterway versus 2 layer tags for the
bridges. This may happen many times in a row. In this case, it makes
sense to split the waterway at 1 point (dividing into urban and not
urban parts) and tag the whole urban part with layer=-1. That's the
case in my hometown (54 tags, one for every bridge vs 1 tag only + 1
split waterway), see here towards the East:
http://www.openstreetmap.org/#map=17/-30.04781/-51.22689
2. If you split only near the bridges, the name of the waterway will
be rendered between the bridges, which is the optimal position. (But
this could be considered mapping for the renderer.)

Situation 1 happens in many other cities across the world, and if you
tag the bridge as layer=1, you may end up inverting the rendering
order of highways, leading to this:
http://www.openstreetmap.org/way/138032009

(The 'level' confusion again.)

If this is a common mistake, let's write the distinction at the very
top of the respective article in the wiki.

On Sat, Mar 15, 2014 at 1:39 PM, Frank Little frank...@xs4all.nl wrote:
 Fernando Trebien wrote:

 Alright. I see that applying layer to long ways is bad for several
 reasons. Surely this could be turned into a validation warning.

 But what's the difference between tagging the bridge with layer=1 and
 tagging the river underneath with layer=-1? Some people seem to think
 that both are necessary, many think it's best to use layer=1 on the
 bridge, I'm saying that layer=-1 on the river (let's say a short
 section, not the entire length) is equivalent. Is it not equivalent?
 Is it wrong? If it is wrong, why is it wrong?

 I don't think 'wrong' is the way to approach this; afaik, they are indeed
 equivalent.
 There are four alternatives which mappers follow, none of which are 'wrong':
 tag the bridge segment, tag the water segment under the bridge, tag both,
 tag neither.

 I've run waterway=stream or =canal (or =ditch, I think) through a few of the
 small rivers and streams here in the Netherlands. Roads need splitting to
 make bridges, so it makes sense to do all the relevant tagging on the road
 segment with the bridge tag when you are working on it.

 I don't have any reason to split the waterway=*, so I just draw on without
 stopping. Since I put the name on the waterway and not on the riverbank, it
 leaves it to the renderer to find a good place to fit in the
 river/stream/canal name. In principle, that should mean a cleaner map if the
 renderer can work out the proper placement (difficult job, though).

 So I would be against splitting the waterway at a bridge and tagging it
 layer=-1 on practical grounds. And you can be sure that it would cause
 confusion with other mappers who would imagine that you are trying to model
 an inverted siphon with the piece of waterway tagged layer=-1, or that you
 had simply made a mistake. (The 'level' confusion again.)

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



-- 
Fernando Trebien
+55 (51) 9962-5409

The speed of computer chips doubles every 18 months. (Moore's law)
The speed of software halves every 18 months. (Gates' law)

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Sat, Mar 15, 2014 at 02:06:13PM -0300, Fernando Trebien wrote:
 the validator will only prevent the most obvious errors but will give
 you no clue how to fix them correctly
 
 I know. But two or three rounds of trial and error with the validator
 should be enough to bring a new user to an acceptable representation.

the validator has no idea how the crossing is supposed to look like, this
is a very optimistic assumption optimistic for anything but the simplest 
cases.
 

 or one of covered,location,indoor,steps,lift or level, maybe more.
 
 I have to read more about them an check their usage, but they could
 all be incorporated in the same rule in the validator. It's just a few
 extra values in a set referred to in a rule. Surely this set can grow
 over time.

I have a bunch of search strings for JOSM on my user page,
  https://wiki.openstreetmap.org/wiki/User:RicoZ
use those as tests for the assumptions. In most part of the world what 
they match are obvious errors - with the exception of some parts of 
Chicago City which I do not know well enough to judge what is going on
there.

Reply to level related things in next email.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-15 Thread Peter Wendorff

Am 15.03.2014 19:19, schrieb Fernando Trebien:
 Here are a few arguable reasons to split the waterway and tag it with 
 layer=-1:
 1. Bridges may come in pairs for dual carriageways. In this case, it's
 a single layer tag for the waterway versus 2 layer tags for the
 bridges. This may happen many times in a row. In this case, it makes
 sense to split the waterway at 1 point (dividing into urban and not
 urban parts) and tag the whole urban part with layer=-1. That's the
 case in my hometown (54 tags, one for every bridge vs 1 tag only + 1
 split waterway), see here towards the East:
 http://www.openstreetmap.org/#map=17/-30.04781/-51.22689
On the other hand you don't have to split anything if you put the layer
tag on the bridges because the bridge is already a separate object.

And even bridges span several parallel osm-ways often: cyclepaths
footpaths, streets, railway lines... along the waterway, one sided or
both-sided.

 2. If you split only near the bridges, the name of the waterway will
 be rendered between the bridges, which is the optimal position. (But
 this could be considered mapping for the renderer.)
This is heavily mapping for the renderer. A good and powerful renderer would
a) join ways with the same name but different detail tags for layer
positioning
b) place the label where no bridge or way above the tunnel is in
conflict regarding the space on the canvas.

This may not be the case out of performance reasons for an online
rendering system like the mapnik stylesheet on osm.org, but that's
another issue.

 Situation 1 happens in many other cities across the world, and if you
 tag the bridge as layer=1, you may end up inverting the rendering
 order of highways, leading to this:
 http://www.openstreetmap.org/way/138032009
good point, but I would consider this a bug independent of rendering as
the same may occur on the way below the bridge as well, if there's a
join of that way with another one without a layer tag.x

anyone going to report this as a bug in the stylesheet?

regards
Peter


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Sat, Mar 15, 2014 at 03:19:36PM -0300, Fernando Trebien wrote:

 Situation 1 happens in many other cities across the world, and if you
 tag the bridge as layer=1, you may end up inverting the rendering
 order of highways, leading to this:
 http://www.openstreetmap.org/way/138032009

what exactly is the problem here? Colour artefacts on the Quai des Gevres?
If it is affected by the layer than it is a bug in Mapnik - and no, we 
should not use layer to fix bugs in Mapnik.

Looking at the data in JOSM I see a few problems - for example there 
is no reason why this particular way should have any layer tag at all.
It seems completely useless and one of the reasons I want to enforce
the no layer tag without a bridge/tunnel rule which would catch
similar accidents.

While not wrong in this case it is also not needed to have the bridge 
at layer=2.

Also building=bridge is the wrong tag for this bridge (tagging for 
Mapnik again ?) but if it is used at all the ways entering the bridge 
ought at least share a node with the bridge where they are entering it.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread Fernando Trebien

Layer= tag clearly (logically) implies that some data is above or
below some other data. At least to my logic.

From this logic, layer=-1 means the object is rendered beneath
anything that has layer=0 (or, conversely, that anything with layer=0
is rendered on top of anything with layer=-1). It does not mean that
it is in fact below it (though it almost always is). That's the
situation of a river passing under a bridge. Likewise, layer=0 means
the object is rendered below anything that has layer=1. Also the same
situation for the situation of a bridge over a river (or of a river
under a bridge). Both approaches should be possible.

At least OsmAnd renders all waterways with layer=-1 with dashed
casing, as if they were underground, which to me makes sense

This makes absolutely no sense to me. OsmAnd has confused the concept
of rendering layer order with the concept of level (underground,
on the ground, above the ground). It may also have confused it with
the concept of a tunnel, which is rendered using dashed casing in
OSM-Carto (the default map style of the main website).

I don't understand why anyone would do this. That's it. Why?

Reason: to avoid having to add a layer tag to every bridge. If a
single river is crossed by 100 bridges, it is easier to add a layer=-1
tag to the river than to add a layer=1 to each of the 100 bridges.

Another reason: to work around the limitations of validators in order
to avoid warnings about missing bridges. But IMHO, this is the wrong
way to handle the situation, since a workaround is easy AND is
exploited in practice. See here:
http://wiki.openstreetmap.org/wiki/Talk:Key:layer#layer.3D-1_erroneously_used_for_many_italian_rivers

So, again : why tag things with with layer= tag when they are
essentially on ground level = not above or below other data (or at
least natural objects that might be mapped one day *) ? Am I missing
something here?

I believe you are missing this: layer=0 does not mean ground level.
The closest to that idea is level=0, even though it is not always
ground level but nearly so almost always. Layer=0 means: render
this to layer 0. Layer=-1 means: render this to layer -1, which is
rendered before layer 0, therefore, anything on layer 0 is drawn on
top of things on layer -1.

On Fri, Mar 14, 2014 at 12:15 PM, Jaakko Helleranta.com
jaa...@helleranta.com wrote:
Re; tagging a (complete or longer segment of a) river with layer=-1

I don't understand why anyone would do this. That's it. Why?

Layer= tag clearly (logically) implies that some data is above or below some
other data. At least to my logic. And I don't seem to be the only one with
this thinking. At least OsmAnd renders all waterways with layer=-1 with
dashed casing, as if they were underground, which to me makes sense, but
what makes all rivers with this unnecessary tag render oddly. There are
quite a number of such problematic waterways (also) in Nicaragua and Haiti.

So, again : why tag things with with layer= tag when they are essentially on
ground level = not above or below other data (or at least natural objects
that might be mapped one day *) ? Am I missing something here?

Cheers,
-Jaakko

*) I do add layer=1 to all bridges I map even if there wouldn't be any data
under then at the moment and think that this kind of layer tag use makes
sense

--
Sent from my Android device. * +505-8845-3391 * http://about.me/jaakkoh

El mar 14, 2014 8:52 a.m., Fernando Trebien fernando.treb...@gmail.com
escribió:

Hello everyone,

This is a small issue that came up recently in Brazil. In my
understanding, the layer tag has no specific meaning other than to
specify a rendering order. The wiki, however, states that it is wrong
to tag a whole river with layer=-1. The reason for that, as far as I
could figure, is because current validators (such as JOSM's or
KeepRight's) will not issue a warning on a waterway x highway crossing
when their layers are different, leading some users into tagging the
river with layer=-1 in order to get rid of warnings about missing
bridges and tunnels. So, I think that the validation rule is
inadequate: it should warn when a waterway crosses a highway that has
no bridge or tunnel tag, regardless of the value in the layer tag.
(Fords are the only exception, AFAIK.)

Do you agree that the river can be tagged with layer=-1 as long as
this value is correct in relation to the layer of other
nearby/crossing ways?

--
Fernando Trebien
+55 (51) 9962-5409

The speed of computer chips doubles every 18 months. (Moore's law)
The speed of software halves every 18 months. (Gates' law)

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

--
Fernando Trebien
+55 (51) 9962-5409

The speed of computer chips

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 4:15 PM, Jaakko Helleranta.com jaa...@helleranta.com

 At least OsmAnd renders all waterways with layer=-1 with
 dashed casing, as if they were underground, which to me makes sense

That's clearly a bug. Waterways underground is specified by tunnel=*

 So, again : why tag things with with layer= tag when they are essentially on
 ground level = not above or below other data (or at least natural objects
 that might be mapped one day *) ? Am I missing something here?

Real case from real world : a deep ditch where the stream is not
underground but below the ground level, is crossing a village
where we have 10 bridges. Either you add 10 times layer=1 on the
bridges or you add 1 time layer=-1 on the stream.

Pieren

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 11:51:47AM -0300, Fernando Trebien wrote:
 Hello everyone,
 
 This is a small issue that came up recently in Brazil. In my
 understanding, the layer tag has no specific meaning other than to
 specify a rendering order. The wiki, however, states that it is wrong
 to tag a whole river with layer=-1. The reason for that, as far as I
 could figure, is because current validators (such as JOSM's or
 KeepRight's) will not issue a warning on a waterway x highway crossing
 when their layers are different, leading some users into tagging the
 river with layer=-1 in order to get rid of warnings about missing
 bridges and tunnels. So, I think that the validation rule is
 inadequate: it should warn when a waterway crosses a highway that has
 no bridge or tunnel tag, regardless of the value in the layer tag.
 (Fords are the only exception, AFAIK.)

agree that the error check should be stronger, location, pipeline and covered 
being other exceptions.

 Do you agree that the river can be tagged with layer=-1 as long as
 this value is correct in relation to the layer of other
 nearby/crossing ways?

in theory yes. However nearby is a problem as rivers can be very long.

Many people simply tag rivers with layer=-1 without even thinking about
the fact that the rivers may now collide with tunnels some hundreds of miles
away.

Furthermore there is a convention that if there is a crossing with a bridge, 
the bridge should have the layer tag and not the way bellow it. Similar tunnel.

For both tunnels and bridges layer is now considered mandatory thus it is
totally useless to put rivers at layer=-1 unless they are in a tunnel.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread fly

On 14.03.2014 16:36, Pieren wrote:
 On Fri, Mar 14, 2014 at 4:15 PM, Jaakko Helleranta.com jaa...@helleranta.com
 
 At least OsmAnd renders all waterways with layer=-1 with
 dashed casing, as if they were underground, which to me makes sense
 
 That's clearly a bug. Waterways underground is specified by tunnel=*
 
 So, again : why tag things with with layer= tag when they are essentially on
 ground level = not above or below other data (or at least natural objects
 that might be mapped one day *) ? Am I missing something here?
 
 Real case from real world : a deep ditch where the stream is not
 underground but below the ground level, is crossing a village
 where we have 10 bridges. Either you add 10 times layer=1 on the
 bridges or you add 1 time layer=-1 on the stream.

Well, I do not get your problem, as bridge/tunnel always need a layer
tag and you already have to cut the ways to tag the bridge/tunnel, why
not simply add the layer to the bridge/tunnel and leave everything else
untouched ?

Using layer tags on long ways is discouraged as you often only need it
on a small part and it makes it more difficult as you always have to
check the layer tag of these long ways before adding a new bridge/tunnel
with appropriate layer value.

cu
fly


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 04:36:26PM +0100, Pieren wrote:
 On Fri, Mar 14, 2014 at 4:15 PM, Jaakko Helleranta.com jaa...@helleranta.com
 
  At least OsmAnd renders all waterways with layer=-1 with
  dashed casing, as if they were underground, which to me makes sense
 
 That's clearly a bug. Waterways underground is specified by tunnel=*
 
  So, again : why tag things with with layer= tag when they are essentially on
  ground level = not above or below other data (or at least natural objects
  that might be mapped one day *) ? Am I missing something here?
 
 Real case from real world : a deep ditch where the stream is not
 underground but below the ground level, is crossing a village
 where we have 10 bridges. Either you add 10 times layer=1 on the
 bridges or you add 1 time layer=-1 on the stream.

wiki says that every bridge should have a layer tag. If you are lazy
you can as well omit the layer altogether, it will be still rendered 
correctly.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 4:45 PM, Richard Z. ricoz@gmail.com wrote:

 in theory yes. However nearby is a problem as rivers can be very long.
 Many people simply tag rivers with layer=-1 without even thinking about
 the fact that the rivers may now collide with tunnels some hundreds of miles
 away.

In general, we should avoid single ways running over hundreds of
miles. And we have QA tools to detect such issues.

 Furthermore there is a convention that if there is a crossing with a bridge,
 the bridge should have the layer tag and not the way bellow it. Similar 
 tunnel.
 For both tunnels and bridges layer is now considered mandatory thus it is
 totally useless to put rivers at layer=-1 unless they are in a tunnel.

Who decided this and where ? The wiki about bridge says Bridges
should have a layer=*, . should is not must.

Pieren

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 4:48 PM, fly lowfligh...@googlemail.com wrote:

 Well, I do not get your problem, as bridge/tunnel always need a layer
 tag and you already have to cut the ways to tag the bridge/tunnel, why
 not simply add the layer to the bridge/tunnel and leave everything else
 untouched ?

You don't see the point where adding one layer=-1  is easier than
adding 10 layer=1 ? I see the layer tag in tunnel/bridge in simple
cases but you should not follow all recommendations as fixed in stone.

Pieren

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 4:48 PM, Richard Z. ricoz@gmail.com wrote:

 wiki says that every bridge should have a layer tag. If you are lazy
 you can as well omit the layer altogether, it will be still rendered
 correctly.

It's not a question of laziness. Setting layer=-1 to the waterway
instead of 10 bridges just demonstrates that you understood the
original concept of the tag layer.

Pieren

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread fly

On 14.03.2014 16:57, Pieren wrote:
 On Fri, Mar 14, 2014 at 4:48 PM, fly lowfligh...@googlemail.com wrote:
 
 Well, I do not get your problem, as bridge/tunnel always need a layer
 tag and you already have to cut the ways to tag the bridge/tunnel, why
 not simply add the layer to the bridge/tunnel and leave everything else
 untouched ?
 
 You don't see the point where adding one layer=-1  is easier than
 adding 10 layer=1 ? I see the layer tag in tunnel/bridge in simple
 cases but you should not follow all recommendations as fixed in stone.

No, as you need to split the river leading to two more objects.

Or do you simply tag the whole way with layer=-1 which is not a good
practices as the next mapper who adds a tunnel has to check the river to
find out that she/he needs to use -2 instead of -1 as value.

cu fly

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread Florian Lohoff

On Fri, Mar 14, 2014 at 05:01:10PM +0100, Pieren wrote:
 On Fri, Mar 14, 2014 at 4:48 PM, Richard Z. ricoz@gmail.com wrote:
 
  wiki says that every bridge should have a layer tag. If you are lazy
  you can as well omit the layer altogether, it will be still rendered
  correctly.
 
 It's not a question of laziness. Setting layer=-1 to the waterway
 instead of 10 bridges just demonstrates that you understood the
 original concept of the tag layer.

Still i feel setting a layer -1 on a waterway and NOT tagging any
bridges is lazyness. I remove the layer=-1 in those cases and wait
for keepright to come up with layer violations.

Flo
-- 
Florian Lohoff f...@zz.de


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread Richard Mann

Setting the river to layer=-1, and the bridge to layer=0 (or 1) avoids a
range of rendering artefacts when roads have casings (which they usually
do). Good practice is only applying that to a shortish section of river,
obviously.

I don't know why the wiki has a statement against it - it always seemed
like a unilateral I don't like it from Nathan Edgar the second.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 04:51:18PM +0100, Pieren wrote:
 On Fri, Mar 14, 2014 at 4:45 PM, Richard Z. ricoz@gmail.com wrote:
 
  in theory yes. However nearby is a problem as rivers can be very long.
  Many people simply tag rivers with layer=-1 without even thinking about
  the fact that the rivers may now collide with tunnels some hundreds of miles
  away.
 
 In general, we should avoid single ways running over hundreds of
 miles. And we have QA tools to detect such issues.

well rivers are thousands of miles long. They don't care about QA tools.

  Furthermore there is a convention that if there is a crossing with a bridge,
  the bridge should have the layer tag and not the way bellow it. Similar 
  tunnel.
  For both tunnels and bridges layer is now considered mandatory thus it is
  totally useless to put rivers at layer=-1 unless they are in a tunnel.
 
 Who decided this and where ? The wiki about bridge says Bridges
 should have a layer=*, . should is not must.

it is a polite must.

There has been a proposal long ago for bridges to have implicit an layer
and it was not accepted.

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread Fernando Trebien

I don't think you should be required to check the river's layer tag.
Validators should do this job for you, it's quite easy to write a rule
for that. Here's an example:

Given two ways that cross internally (excluding connections at
endpoints), and considering the layer value defined explicitly in a
tag or implicitly 0 when the tag is missing, have the validator issue
a warning in the following situations:

1. The ways have the same layer value and are unconnected. (They
should be connected, or else something is surely missing. This could
actually be considered an error.)
1.1. Also warn if if one way is a waterway and the other is a highway
and the connection is not explicitly a ford. (It should be, for
clarity. If it's not, it's also possibly not a ford, therefore the
connection is wrong.)
2. The ways have different layer values and both are missing a tunnel
or a bridge tag. (One of them must be either a bridge or a tunnel.
They can both be tunnels or bridges, but they can't be none of those
two simultaneously in the real world.)
2.1. Additionally, if one of them is a bridge and the other is a
tunnel or is neither a tunnel nor a bridge: the bridge should have a
greater layer value.
2.2. Similarly, if one is a tunnel, its layer value should be lower if
the other is a bridge or has neither tag.

These rules apply to any arbitrary combination of stacked waterways
and highways that I can think of right now. A few examples using two
overlapping ways:

a. The ways are connected and do not have a layer tag: everything is
ok, no rules issue a warning.
b. The ways are not connected and do not have a layer tag: rule 1
issues a warning. They must either be connected or lie at different
layer levels.
c. The ways are not connected, both have the same layer (say layer=3
or layer=-4), and have no other tags: rule 1 issues a warning. Similar
to situation b.
d. The ways are not connected and one of them has a layer=-1 tag and
no other tags: rule 2 issues a warning.
e. The ways are not connected and one of them has a layer=1 tag and no
other tags: rule 2 issues a warning too.
f. The ways are not connected, one of them is a bridge with layer=2
and the other is a tunnel with layer=5: rule 2.1 issues a warning.
g. The ways are not connected, one of them is a tunnel with layer=1
and the other is neither a bridge nor has a layer tag (layer=0 is
assumed): rule 2.2 issues a warning.

Actually, situation d is what would discourage people from using
layer=-1 to work around today's validator warnings. With this ruleset,
it's impossible to eliminate the warning without actually taking
action on bridges.

On Fri, Mar 14, 2014 at 12:48 PM, fly lowfligh...@googlemail.com wrote:
 On 14.03.2014 16:36, Pieren wrote:
 On Fri, Mar 14, 2014 at 4:15 PM, Jaakko Helleranta.com 
 jaa...@helleranta.com

 At least OsmAnd renders all waterways with layer=-1 with
 dashed casing, as if they were underground, which to me makes sense

 That's clearly a bug. Waterways underground is specified by tunnel=*

 So, again : why tag things with with layer= tag when they are essentially on
 ground level = not above or below other data (or at least natural objects
 that might be mapped one day *) ? Am I missing something here?

 Real case from real world : a deep ditch where the stream is not
 underground but below the ground level, is crossing a village
 where we have 10 bridges. Either you add 10 times layer=1 on the
 bridges or you add 1 time layer=-1 on the stream.

 Well, I do not get your problem, as bridge/tunnel always need a layer
 tag and you already have to cut the ways to tag the bridge/tunnel, why
 not simply add the layer to the bridge/tunnel and leave everything else
 untouched ?

 Using layer tags on long ways is discouraged as you often only need it
 on a small part and it makes it more difficult as you always have to
 check the layer tag of these long ways before adding a new bridge/tunnel
 with appropriate layer value.

 cu
 fly


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



-- 
Fernando Trebien
+55 (51) 9962-5409

The speed of computer chips doubles every 18 months. (Moore's law)
The speed of software halves every 18 months. (Gates' law)

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 04:36:26PM +0100, Pieren wrote:
 On Fri, Mar 14, 2014 at 4:15 PM, Jaakko Helleranta.com jaa...@helleranta.com
 
  At least OsmAnd renders all waterways with layer=-1 with
  dashed casing, as if they were underground, which to me makes sense
 
 That's clearly a bug. Waterways underground is specified by tunnel=*

this is not a bug. Rivers do not flow through thin air but through areas
which are tagged as landuse=forest, landuse=meadow or some other.

So what happens if you have a river with a layer=-1 going through
a landuse=forest with a layer=0? Quite clearly the river should be
hidden under the forest and my guess is 5 out of 10 renderers do 
exactly that.

Mapnik is way too fault tolerant and works around this obvious breakage.
Maybe it should be changed to be less fault tolerant.

  So, again : why tag things with with layer= tag when they are essentially on
  ground level = not above or below other data (or at least natural objects
  that might be mapped one day *) ? Am I missing something here?
 
 Real case from real world : a deep ditch where the stream is not
 underground but below the ground level, is crossing a village
 where we have 10 bridges. Either you add 10 times layer=1 on the
 bridges or you add 1 time layer=-1 on the stream.

This is one rare example where it may at first sight appear to be of 
some use. 
However as it is a rare example it will cause more work in the end than 
if you do it the normal way. Nobody expects it being done like this so 
it will cause more thinking for anybody trying to change something 
there.
Everyone who tries to edit it must download the complete data along the 
way that is tagged layer=-1 and understand all crossings there. 
What happens if it is a bigger village or a small town? This does not 
scale well.

This one rare case where it might be of theoretical use is in my opinion
far outweighed by the disadvantages of allowing it:

* prevents validators from doing some very easy and useful checks
* very many people use layer=-1 for waterways incorrectly and taking your
  idea into account would make it much harder to properly detect and fix 
  such errors

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 03:55:39PM -0300, Fernando Trebien wrote:
 I don't think you should be required to check the river's layer tag.
 Validators should do this job for you, it's quite easy to write a rule
 for that.

validators can check for many errors but if you want to change
anything you have to understand the whole situation. 
Imagine you want to add a new bridge to a complex freeway intersection
with junctions and overpasses.. the validator will only prevent the
most obvious errors but will give you no clue how to fix them
correctly.
 
 Given two ways that cross internally (excluding connections at
 endpoints), and considering the layer value defined explicitly in a
 tag or implicitly 0 when the tag is missing, have the validator issue
 a warning in the following situations:

there is no difference between connections in endpoints or in a crossing 
point as far as I can tell.

 1. The ways have the same layer value and are unconnected. (They
 should be connected, or else something is surely missing. This could
 actually be considered an error.)

except for aerial ways and similar exceptions

 1.1. Also warn if if one way is a waterway and the other is a highway
 and the connection is not explicitly a ford. (It should be, for
 clarity. If it's not, it's also possibly not a ford, therefore the
 connection is wrong.)

there is also the odd case of highways across dams, those are connected
with the waterway

 2. The ways have different layer values and both are missing a tunnel
 or a bridge tag. (One of them must be either a bridge or a tunnel.
 They can both be tunnels or bridges, but they can't be none of those
 two simultaneously in the real world.)

or one of covered,location,indoor,steps,lift or level, maybe more.

 2.1. Additionally, if one of them is a bridge and the other is a
 tunnel or is neither a tunnel nor a bridge: the bridge should have a
 greater layer value.
 2.2. Similarly, if one is a tunnel, its layer value should be lower if
 the other is a bridge or has neither tag.

except for indoor mapping and maybe other weird cases.

 These rules apply to any arbitrary combination of stacked waterways
 and highways that I can think of right now. 

also railways?

  A few examples using two
 overlapping ways:
 
 a. The ways are connected and do not have a layer tag: everything is
 ok, no rules issue a warning.
 b. The ways are not connected and do not have a layer tag: rule 1
 issues a warning. They must either be connected or lie at different
 layer levels.
 c. The ways are not connected, both have the same layer (say layer=3
 or layer=-4), and have no other tags: rule 1 issues a warning. Similar
 to situation b.
 d. The ways are not connected and one of them has a layer=-1 tag and
 no other tags: rule 2 issues a warning.

more general: not connected, different layer values and not one of
bridge,tunnel,covered,location,indoor,steps,lift, no level tag and 
a few more things to take into account.

I am not sure it is so easy to catch all that.

 e. The ways are not connected and one of them has a layer=1 tag and no
 other tags: rule 2 issues a warning too.

identical to d?

 f. The ways are not connected, one of them is a bridge with layer=2
 and the other is a tunnel with layer=5: rule 2.1 issues a warning.

unless indoor or other strange cases

 g. The ways are not connected, one of them is a tunnel with layer=1
 and the other is neither a bridge nor has a layer tag (layer=0 is
 assumed): rule 2.2 issues a warning.

 
 Actually, situation d is what would discourage people from using
 layer=-1 to work around today's validator warnings. With this ruleset,
 it's impossible to eliminate the warning without actually taking
 action on bridges.

It is a lot easier saying that every bridge and tunnel must have a layer 
tag and enforce that than catching all the situations mentioned in 
situation d.

With some luck, you can restrict d to waterways and it becomes easy

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread Ilpo Järvinen

On Fri, 14 Mar 2014, Richard Z. wrote:

 On Fri, Mar 14, 2014 at 04:51:18PM +0100, Pieren wrote:
  On Fri, Mar 14, 2014 at 4:45 PM, Richard Z. ricoz@gmail.com wrote:
 
 There has been a proposal long ago for bridges to have implicit an layer
 and it was not accepted.

Was that for bridges being equal to layer=1 (which would obviously be bad 
assumption) or for less than what layer tag can specify (e.g. +/-0.1 for 
bridges/tunnels or whatever  1)?

-- 
 i.

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

On Fri, Mar 14, 2014 at 10:34:41PM +0200, Ilpo Järvinen wrote:
 On Fri, 14 Mar 2014, Richard Z. wrote:
 
  On Fri, Mar 14, 2014 at 04:51:18PM +0100, Pieren wrote:
   On Fri, Mar 14, 2014 at 4:45 PM, Richard Z. ricoz@gmail.com wrote:
  
  There has been a proposal long ago for bridges to have implicit an layer
  and it was not accepted.
 
 Was that for bridges being equal to layer=1 (which would obviously be bad 
 assumption) or for less than what layer tag can specify (e.g. +/-0.1 for 
 bridges/tunnels or whatever  1)?


https://wiki.openstreetmap.org/wiki/Proposed_features/default_layer_for_bridge_and_tunnel

Richard

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

Re: [Tagging] layer=-1, rivers, bridges and tunnels

2014-03-14 Thread Ilpo Järvinen

On Fri, 14 Mar 2014, Richard Z. wrote:

 On Fri, Mar 14, 2014 at 10:34:41PM +0200, Ilpo Järvinen wrote:
  On Fri, 14 Mar 2014, Richard Z. wrote:
  
   On Fri, Mar 14, 2014 at 04:51:18PM +0100, Pieren wrote:
On Fri, Mar 14, 2014 at 4:45 PM, Richard Z. ricoz@gmail.com wrote:
   
   There has been a proposal long ago for bridges to have implicit an layer
   and it was not accepted.
  
  Was that for bridges being equal to layer=1 (which would obviously be bad 
  assumption) or for less than what layer tag can specify (e.g. +/-0.1 for 
  bridges/tunnels or whatever  1)?
 
 https://wiki.openstreetmap.org/wiki/Proposed_features/default_layer_for_bridge_and_tunnel

Sadly the mapper point of view is almost totally missing from that 
discussion. I've understood that is to most valuable resourse we've and
we should try to avoid putting any unnecessary burden on them.

Therefore, everyone needs now to handle those hardly useful layer 
warnings about trivial cases (and waste their time on correcting them). 
And in fact, I've wasted some time just on that today while what I'm 
really after is real geometry errors whose fixing would be much much more 
benefial but JOSM validator did not differentiate these two cases for me 
but follows such a bad spec.


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

Re: [Tagging] layer=-1, rivers, bridges and tunnels