Re: [Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[me]

On 10/06/16 at 10:46am, Greg Troxel wrote:
> 
> Martin Koppenhoefer  writes:
> 
> (This is just a longer note about non-admin-boundary settlements and why
> they are particularly tricky in a lot of New England, sort of separate
> From the node/line/way discussion.)
> 
> >> Il giorno 09 giu 2016, alle ore 18:50, Christoph Hormann 
> >>  ha scritto:
> >> 
> >> If you can verifiably map a settlement as a linear way you can also map 
> >> it as an area.  Usually neither is the case so most populated places 
> >> are mapped as nodes.
> >
> > I think that most settlements could be mapped as areas, there's often
> > a border where many people would agree that inside is the settlement
> > and outside is not
> 
> In rural areas, I think that's true.  But around me (mostly built up
> because of proximity to Boston), it isn't true and the borders are very
> hard to know.  Even the people that live there will find it hard and
> disagree and tell you that the question doesn't quite make sense.
> 
> As a concrete example, consider the "hamlet" of South Acton:
> 
>   https://www.openstreetmap.org/node/158813473
> 
> This is not an administative boundary (which would be "Acton",
> admin_level=8), but a name for a populated place.
> 
> This place name is longstanding, probably dating to before the American
> Revolution.  It appears on USGS topo maps.  It has a railway station of
> the same name.  It is known to to the locals.  I believe that the
> railway went there (~1846) and has a stop because the village was
> important.
> 
> To those who think of South Acton as the historical village, it's clear
> that the buildings on either side of the railroad bridge are in the
> village.  But as you go away from the center, it's very hard to draw a
> line.  (There's a further complication that the intersection of 27/111,
> traditionally "Kelly's Corner", is often called South Acton, as it is
> more significant commerce-wise today.)
> 
> There are many other examples, where what used to be a village with very
> little on the road to the next village is now a place where there are a
> clump of older houses among a sea of houses covering the whole town.
> 
> Still, it's entirely reasonable to try to draw a polygon, as long as its
> done by the locals over beer.  And also to have a node, which is far
> easier to place uncontroversially, as there is usually an obvious
> cluster of houses much older than the rest, and useful even if there is
> a polygon.

Agree - for me the big problem with placenames as just nodes is that it
then assumes that all areas are circular when doing reverse geocoding to
get the name of an area. But where I live this simply isn't the case -
For example Edinburgh Old Town looks like this:

http://nominatim.openstreetmap.org/search.php?q=edinburgh+old+town=1=

So for the various services that do the reverse lookups to work properly
we ended up dividing the city up into suburbs, all as areas. As has been
said, I agree this is really hard, and has involved discussions over
beer. But it does mean that at that level the classifications are much
better than with just nodes. 

At that point, I discovered that some renderers like the cycle map layer
with render names for nodes and ways so we end up with duplicates :(

I did submit a pull request:
https://github.com/gravitystorm/openstreetmap-carto/pull/427

But it all got very complicated, but these changes never went very far,
see
https://github.com/gravitystorm/openstreetmap-carto/pull/546

Cheers
Chris


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

Re: [Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[Greg Troxel]

Martin Koppenhoefer  writes:

(This is just a longer note about non-admin-boundary settlements and why
they are particularly tricky in a lot of New England, sort of separate
From the node/line/way discussion.)

>> Il giorno 09 giu 2016, alle ore 18:50, Christoph Hormann 
>>  ha scritto:
>> 
>> If you can verifiably map a settlement as a linear way you can also map 
>> it as an area.  Usually neither is the case so most populated places 
>> are mapped as nodes.
>
> I think that most settlements could be mapped as areas, there's often
> a border where many people would agree that inside is the settlement
> and outside is not

In rural areas, I think that's true.  But around me (mostly built up
because of proximity to Boston), it isn't true and the borders are very
hard to know.  Even the people that live there will find it hard and
disagree and tell you that the question doesn't quite make sense.

As a concrete example, consider the "hamlet" of South Acton:

  https://www.openstreetmap.org/node/158813473

This is not an administative boundary (which would be "Acton",
admin_level=8), but a name for a populated place.

This place name is longstanding, probably dating to before the American
Revolution.  It appears on USGS topo maps.  It has a railway station of
the same name.  It is known to to the locals.  I believe that the
railway went there (~1846) and has a stop because the village was
important.

To those who think of South Acton as the historical village, it's clear
that the buildings on either side of the railroad bridge are in the
village.  But as you go away from the center, it's very hard to draw a
line.  (There's a further complication that the intersection of 27/111,
traditionally "Kelly's Corner", is often called South Acton, as it is
more significant commerce-wise today.)

There are many other examples, where what used to be a village with very
little on the road to the next village is now a place where there are a
clump of older houses among a sea of houses covering the whole town.

Still, it's entirely reasonable to try to draw a polygon, as long as its
done by the locals over beer.  And also to have a node, which is far
easier to place uncontroversially, as there is usually an obvious
cluster of houses much older than the rest, and useful even if there is
a polygon.


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

Re: [Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[Martin Koppenhoefer]

sent from a phone

> Il giorno 09 giu 2016, alle ore 18:50, Christoph Hormann 
>  ha scritto:
> 
> If you can verifiably map a settlement as a linear way you can also map 
> it as an area.  Usually neither is the case so most populated places 
> are mapped as nodes.



I think that most settlements could be mapped as areas, there's often a border 
where many people would agree that inside is the settlement and outside is not



>  In case of a settlement consisting exclusively of 
> buildings densely located along a road at both sides tagging that 
> stretch of road as place=hamlet or similar might be a good compact way 
> to map it but i have not yet seen a case like this in reality.


I would prefer an area for this case as well, only disadvantages with a line, 
besides maybe you can draw a line a bit faster than a polygon, but it would 
still not convey any notion of spatial extension, while a polygon can do 
exactly this.

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

Re: [Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[Martin Koppenhoefer]

sent from a phone

> Il giorno 09 giu 2016, alle ore 18:41, Greg Troxel  ha 
> scritto:
> 
> Perhaps there should be some more explicit tagging to denote
> uncertainty.  But drawing a polygon around the buildings that are part
> of the settlement, and including areas that if a new building were built
> there, it would be considered part of the settlement seems sensible.
> This is logically well-defined, even if it's hard to answer in detail.


+1, I don't know why the OP thinks that a polygon is not suitable for locality 
and hamlet, I believe it definitely is, and if the wiki says the opposite it 
should be corrected 



> 
> I can see an argument that a point is wrong, but really a point is a
> different representation, showing not really the centroid but the
> logical center of the settlement.


+1


> 
> A line would be a polygon with no width but significant length, which
> omits covering houses and has some new notion of near.   So it's sort of
> a blend of an area and the center, and that seems to raise more issues
> and complexity than it helps.


+1, to me a line makes no sense for representing an area (which any kind of 
settlement clearly is)


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

Re: [Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[Warin]

On 6/10/2016 12:56 AM, Amacri wrote:


Hi,

Tag place=isolated_dwelling, which should be used to name an almost 
unpopulated place (often in the mountains), can be currently defined 
as node (point) or polygon (area), but not as a way (line). Same 
consideration for place=hamlet or place=locality: using way is not 
allowed for them.


It often happens that, when a place cannot be clearly delimited in the 
mountains, traditional maps typically draw its text, where the text 
itself outlines and shapes the covered region.




If these are 'places' then they should be tagged as 'places'.

If the 'place' can be verified as a line then tag it as a line.

BUT the old map may simply be using the text placement as an indication 
of an area rather than a line.


I would not rely on the map/s alone .. needs a bit more investigation?


place=locality has some 7,800 uses as a line presently.

So I would use it as a line until more information is available. Add a 
note to that effect?



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

Re: [Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[Christoph Hormann]

On Thursday 09 June 2016, Amacri wrote:
>
> Provided that marking such areas with polylines can lead to
> inappropriate tagging as the boundaries are not well definable, they
> do not deserve just a point, as they may represent an extended area,
> that can be shaped through its text (often the original shape can be
> found in historic maps too, e.g., available in municipal archives.
>
> For these reasons, I would consider appropriate representing such
> places as areas when boundaries are well definable, or through lines
> when there is no deterministic way to define an actual boundary. I
> anyway would limit visibility of points (nodes) to very high zoom
> levels (e.g., zoom level

The thing is if something is not verifiably defined in its basic 
localization - like if you have a certain name known to refer to a 
place of some sort but if you ask a number of knowledgeable locals 
where it is they all point vaguely at some place but each of them a 
distinctly different one - that place is not mappable in OSM at all - 
neither as a node, a linear way or an area.

If on the other hand there is a verifiable localization of the place - 
which does not necessarily have to be in the form of either of these 
three well established feature types - it can be mapped in that form.

However if you think the place is not well suited to be mapped as a node 
and neither as an area this does dot necessarily make it a good 
candidate to be mapped as a linear way.  So far you have not given a 
reason why for the type of place you are thinking of a way as a good 
representation (except for the fact that creating a label for it in a 
map in a basic form is fairly easy).

When thinking about mapping and tagging it is usually best to completely 
ignore the question of how something can be possibly rendered in a map 
at first and simply consider how it can be represented in the database 
in a way that allows following mappers visiting the area to objectively 
verify if the mapping is accurate or not.

> In case a new feature would be needed, which would be the most
> appropriate name? natural=upland? Please, suggest.

Actually you have not really said how you want to define natural=upland.  
If it is as unspecific as place=locality there is no point in creating 
a new tag.  If it is more specific you need to tell how a mapper can 
decide if something is to be tagged natural=upland.  Note 
the 'natural'-key generally implies something to be a verifiable 
feature even without a name.

>
> Otherwise, would you consider appropriate requesting to enable lines
> for place=isolated_dwelling, place=hamlet or place=locality?

I cannot at the moment think of a situation where mapping 
place=isolated_dwelling or place=hamlet as a linear way makes sense.  
If you can verifiably map a settlement as a linear way you can also map 
it as an area.  Usually neither is the case so most populated places 
are mapped as nodes.  In case of a settlement consisting exclusively of 
buildings densely located along a road at both sides tagging that 
stretch of road as place=hamlet or similar might be a good compact way 
to map it but i have not yet seen a case like this in reality.

place=locality is simply too vague to give a definite answer.  You could 
for example think of a certain stretch of coast with a certain name.  
But in this case place=locality would be wrong anyway because 
natural=coastline + name=foo would be perfectly sufficient.

On a general note - when things are mapped as nodes this is frequently 
done with the implicit notion that this is a location with a certain 
tolerance margin.  You might think of mapping something with a linear 
way as a method to specify an anisotropic tolerance, well localized in 
one direction but poorly in another.  However that is not what you 
actually do when you map it as a way - on the contrary you much more 
specifically localize it.

-- 
Christoph Hormann
http://www.imagico.de/

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

Re: [Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[Greg Troxel]

"Amacri"  writes:

> Tag place=isolated_dwelling, which should be used to name an almost
> unpopulated place (often in the mountains), can be currently defined as node
> (point) or polygon (area), but not as a way (line). Same consideration for
> place=hamlet or place=locality: using way is not allowed for them.
>
> It often happens that, when a place cannot be clearly delimited in the
> mountains, traditional maps typically draw its text, where the text itself
> outlines and shapes the covered region.

I think you are blurring rendering and denoting the data.

Points make sense for settlements, as there is often a single place
(which is really not quite a place) that people consider the center of
town.  Areas make sense too, although as you say there is a fuzzy edge
to what is part of the settlement and what is not.

It seems like the motivation for line is to aid in rendering with text.
Is that true?  Or are you trying  to define historical things when the
only source data is text on a map, with no more information?

Perhaps there should be some more explicit tagging to denote
uncertainty.  But drawing a polygon around the buildings that are part
of the settlement, and including areas that if a new building were built
there, it would be considered part of the settlement seems sensible.
This is logically well-defined, even if it's hard to answer in detail.

I can see an argument that a point is wrong, but really a point is a
different representation, showing not really the centroid but the
logical center of the settlement.

A line would be a polygon with no width but significant length, which
omits covering houses and has some new notion of near.   So it's sort of
a blend of an area and the center, and that seems to raise more issues
and complexity than it helps.

I run into some of this around me.  There are town names, which have
actual boundaries with monuments, and there is little uncertainty.
There are also place names that are not towns, and are usually areas of
towns.  Many had an original meaning where there was a cluster of
houses, and sometimes there is now an intersection of modern roads with
shops that people thing of as that place name.  I tend to put the node
on the traditional place, as that seems the usage favored by those who
understand the most.  But these are all point type nodes, not area; in
modern times there is continuous building and no defined edges to the
historical settlements.



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

[Tagging] Request for new tag "natural=upland" (as way) or enabling "way" for "place" tags

[Amacri]

Hi,

 

Tag place=isolated_dwelling, which should be used to name an almost
unpopulated place (often in the mountains), can be currently defined as node
(point) or polygon (area), but not as a way (line). Same consideration for
place=hamlet or place=locality: using way is not allowed for them.

 

It often happens that, when a place cannot be clearly delimited in the
mountains, traditional maps typically draw its text, where the text itself
outlines and shapes the covered region.

 

Check for instance this map found in Intenet:

http://4.bp.blogspot.com/-mQlKv2pP5Qg/UWMx9WybzPI/Ay8/QC6o0CsvyBQ/s1
600/M.+Erio+cartina+traccia.JPG

 

You can see that there are places, named with capital letters in this case,
where the text itself shapes the related region.

 

Examples in that map: PLATA, COSTA DEL CIVELLO, HORNA HUTA, POZZETTA SECCA,
SPIAZZI DEI MERCANTI, CHEMPLE, BOSCO SCURO, BOSCO PERGOLA, VERENETTA, etc.

 

These labels represent named sectors of specific mountains, without clear
boundaries.

 

Notice that they do not mark ridges, aretes, valleys, so we cannot use these
already existing tags to name them.

 

Sometimes they are sub-ridges, sometimes are partial woods (but the zone can
be part of a wood and sometimes includes pastures); often (or maybe usually)
they are uplands (which do not have a tag in OSM currently).

 

They are not place=region & region:type=mountain_area, which should be
dedicated to tagging consistent group of mountains and not a place of
limited size within a single mountain.

 

For place=region & region:type=mountain_area there are many examples in OSM
showing that specific boundaries of mountain group are definable and
effective and DEM can be exploited to discover the most appropriate
boundaries. Conversely, DEM might not of help at all for local names
attributed to small areas within a single mountain (e.g., a single mountain
might have a number of uplands, let's say 5 or 10).

Provided that there could be no relation between DEM and local places within
a single mountain, sometimes, like in the case here below, boundaries could
be appropriately imagined basing on the current vegetation:

 

 
http://wiki.openstreetmap.org/wiki/Talk:Proposed_features/Mountains (ref.
"sidearm, subpeak, i-dont-know" and related image

http://wiki.openstreetmap.org/wiki/File:2014-05-18_11-30-25.jpg)

 

Anyway, in most cases, names refer to ancient usage of territories. Ancient
woodlands might have different boundaries than modern woods. Wood-pastures
can be human creation induced in past ages by grazing animals and
selectively clearing trees on the land. Some of them remain stable over
centuries, some don't. For instance, when in past centuries the production
of wool was a main human activity on a specific land, grass pastures were
prominent to provide natural food for sheep; now, if one tries to name a
medieval place using the boundaries of last century woods (naturally
regenerated or possibly consolidated in recent years through artificial
planting) he will improperly tag a place by forcing a border redefinition of
a different original perimeter; also tagging a wood area hypothetically
reflecting its ancient name inside a modern wood (or even intersecting an
existing wood representing current land use - e.g. a wood inside a wood or a
wood/pasture intersecting a wood) might not be correct (other than forcing
the mapper to artificially define a border to a region named in past ages
which might not be scientifically identified nowadays).

 

Provided that marking such areas with polylines can lead to inappropriate
tagging as the boundaries are not well definable, they do not deserve just a
point, as they may represent an extended area, that can be shaped through
its text (often the original shape can be found in historic maps too, e.g.,
available in municipal archives.

For these reasons, I would consider appropriate representing such places as
areas when boundaries are well definable, or through lines when there is no
deterministic way to define an actual boundary. I anyway would limit
visibility of points (nodes) to very high zoom levels (e.g., zoom level
>=15) to incentivize mappers to be as accurate as possible on tagging (lines
and areas shall be strongly privileged to simple nodes).

 

Lines (way) shall be dragged defining the appropriate curve, where each node
of the line should be placed with reference to the most common attribution
to the referenced named place (e.g., basing decision on local knowledge, on
old/authoritative documents e.g. found in local municipalities, on public
maps with right to be used in OSM, etc.).

 

Maybe a hypothetic natural=upland (to be represented as node, way or
polygon) could address them. (Or please let me know alternative proposals.).
Perhaps, natural=mountain_upland could avoid abuse of this tag.

 

They