Re: [Tagging] Tag for a plateau or tableland?

[Andrew Harvey]

Plateau, table, tablelands all used in Australia.

landform=plateau has 5 usages
https://taginfo.openstreetmap.org/tags/landform=plateau

natural=plateau seems fine to me. Could be called walls sometimes, but to
me it's unclear if that's the term for the plateau or the cliffs that
surround the plateau.

On Wed, 17 Apr 2019 at 14:56, Joseph Eisenberg 
wrote:

> I'm surprised that I can't find an established tag or wiki page for a
> plateau, mesa, or tableland; an area of raised land that is flat on
> top:
>
> See https://en.wikipedia.org/wiki/Plateau and
> https://en.wikipedia.org/wiki/Mesa and
> https://en.wikipedia.org/wiki/Table_(landform)
>
> Small plateaus or mesas may be called a "Butte" in the USA, or a
> tableland in some English-speaking regions. See
> https://en.wikipedia.org/wiki/Butte
>
> They are known as "Mesas" in Spanish-speaking areas and related
> languages. I believe "plateau" is from French?
>
> These features are common in arid areas of the southwestern United
> States and in many other places that have the appropriate geology:
> usually there is a hard layer of rock on top of softer layers. This
> often forms a flat area with steep slopes or even cliffs at the sides.
> They can also from from volcanic activity, or from glaciation.
>
> I searched taginfo for "tableland", "table_land", "table-land",
> "plateau" and "mesa".
>
> There are 94 natural=plateau and 3 natural=mesa.
> I found no uses of natural=table or table_land or tableland or tableland
>
> Is natural=plateau the best option? This sounds fine to me, as an
> American English speaker, but I'd like to know if it's the best
> British English option.
>
> ___
> 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] Avoid using place=locality - find more specific tags instead

[Graeme Fitzpatrick]

On Wed, 17 Apr 2019 at 12:21, Joseph Eisenberg 
wrote:

> I have reviewed all the features tagged as place=locality in 2 places
> in the USA and 2 in Europe, and found that 3 out of 4, place=locality
> is usually used for features that could be tagged with a more specific
> tag.
>

Thanks for your extremely detailed research Joseph!

Have just done some checking around our area & found that a lot of the
place=locality listings should probably be =suburb or =neighbourhood (& a
lot of them were done by me, so I'll have to go in & fix them all!).

Couple to get an opinion on though, please.

https://www.openstreetmap.org/node/4313816641#map=14/-28.1502/153.3205 is
Austinville, which is the area located along the length of Austinville
Road. Austinville is a valid locality name, but there is no actual spot you
can point to & say that's Austinville - there are no shops, petrol station
etc, only houses / farms & a Community Hall.

What would it be if it's not a =locality? =village / hamlet / suburb /
neighbourhood?

Same thing for Springbrook:
https://www.openstreetmap.org/node/4316622215#map=14/-28.1872/153.2702,
which is the area on top of Springbrook Mountain. Once again, fairly
sparsely populated - maybe 500?, but no distinct centre of town, except one
general store, a Primary School & a handful of cafe's.

There are a few others around that are similar - they have a population,
but there are only 100 - 200 people living in an area of 200+ sq k - seems
way to sparse to be labelled as village or anything similar?

Thanks

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

[Tagging] Tag for a plateau or tableland?

[Joseph Eisenberg]

I'm surprised that I can't find an established tag or wiki page for a
plateau, mesa, or tableland; an area of raised land that is flat on
top:

See https://en.wikipedia.org/wiki/Plateau and
https://en.wikipedia.org/wiki/Mesa and
https://en.wikipedia.org/wiki/Table_(landform)

Small plateaus or mesas may be called a "Butte" in the USA, or a
tableland in some English-speaking regions. See
https://en.wikipedia.org/wiki/Butte

They are known as "Mesas" in Spanish-speaking areas and related
languages. I believe "plateau" is from French?

These features are common in arid areas of the southwestern United
States and in many other places that have the appropriate geology:
usually there is a hard layer of rock on top of softer layers. This
often forms a flat area with steep slopes or even cliffs at the sides.
They can also from from volcanic activity, or from glaciation.

I searched taginfo for "tableland", "table_land", "table-land",
"plateau" and "mesa".

There are 94 natural=plateau and 3 natural=mesa.
I found no uses of natural=table or table_land or tableland or tableland

Is natural=plateau the best option? This sounds fine to me, as an
American English speaker, but I'd like to know if it's the best
British English option.

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

Re: [Tagging] documenting cycleway=crossing

[John Willis via Tagging]

> highway=path



This is all a result on an incomplete tagging set, and using highway=path as a 
catch-all tag to avoid creation and documentation of missing tags. Any and all 
of the footway=* values and sub-tags should also be available for cycleway=* 
path=* and bridleway=* as well.

Also, there is "pedestrian street”  but no "cycling street” -  another missing 
highway=* value. these force unnecessary disparities in tagging method. all 
those videos I see of people in Amsterdam cycling on wide dedicated cycling 
roads with signals and stop lights and corssing and whatnot seems like a road 
to me. 

but I am not here for that today, just cycleway=crossing. 

When mapping the cycling “roads” I encounter here in Japan, it is a cycleway. 
it may have foot=yes, but it is a cycleway. it is built and graded and signed 
and has curves and access ramps to be a cycleway. it is not a footway or a 
path. it is a cycleway. It quacks like a cycleway, so it is one. 

In many instances, cycleways are interrupted by large trunk roads, forcing 
cyclists onto a footway=sidewalk and to use regular pedestrian infrastructure 
at lights and signals. this is especially true where a cycling road follows a 
river, and a trunk road crossing the river via a bridge forces cyclists to use 
footways and a nearby intersection crosswalk to get to the other side of the 
bridge to continue on the cycling road. in rural areas with very long cycling 
roads, this is common. I know how to tag all that. That is not for 
cycleway=crossing.

But there are also many dedicated cycleways marked with their own crossings 
when they have to cross a smaller road where a bypass tunnel is not practical. 


Here is a location where I have cycleway and footway bridges, unmarked 
crossings, and marked crossings. There are some cuttings and tunnels too. 

https://www.openstreetmap.org/#map=18/36.28425/137.90828 



Similar to how we have bridges, tunnels, cuttings, and other features of the 
way, “crossing” is not any different. the handwringing over having to to use 
“path” is taking it way too far. 

We wouldn't be doing such mental gymnastics for a bridge - it too is a property 
of the way, and tags the method of crossing another way (the river). 

"It is a cycleway. it is a bridge.” is no different than “it is a cycleway. "It 
is a road crossing” , beyond the intersecting node. 

Regardless if it is appropriate for use in your country,  think we can be 
flexible enough to use cycleway=crossing in situations in countries where it is 
appropriate. 

Javbw


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

Re: [Tagging] Avoid using place=locality - find more specific tags instead

[Joseph Eisenberg]

I have reviewed all the features tagged as place=locality in 2 places
in the USA and 2 in Europe, and found that 3 out of 4, place=locality
is usually used for features that could be tagged with a more specific
tag.

Summary:

1) Hawaii: less than 12 out of 143 place=locality features in Hawaii
are correct uses; most are also type=boundary relations, and most of
the nodes could be tagged as another feature like natural=beach,
natural=cape

2) Delaware: all 15 closed ways and relations in Delaware are
boundary=*, and 2 out of the 3 nodes can be tagged with more specific
feature tags.

3) Liechtenstein: 39 nodes. 8 have a word or suffix that defines a
specific feature like "wald" = wood, "berg" = hill/mountain. I don't
know German well enough to guess any of the others, or they appear to
be just a name.

4) Andorra: 2/3rds (39 of 58 features) can be tagged with a more
specific tag by translating the name. The other 19 features cannot be
clearly translated. All are nodes except one small square.

Details:
I selected small states or countries so that I could open the whole
data file in JOSM. The data was downloaded between October 2018 and
January 2019.

Hawaii:
- 143 place=locality
- 72 nodes
- 26  closed ways (all tagged boundary=census) - which are also
usually members of:
- 70 relations (all are tagged with boundary=census or boundary=administrative)

All the closed ways and relations are also tagged “type=boundary”

Of the 72 nodes:
- 22 are also tagged natural=cape
- 3 are tagged tourism=attraction (one of these is also historic=yes)

Out of the remaining 47 nodes, several have names that suggest they
should have other tags:
- 8 are named “* Beach” (=> natural=beach)
- 2 are named Business Park or Business Center (landuse=commercial)
- 2 “* Point” (natural=cape or natural=peninsula)
- 2 road junctions: “Four Corners” and “Old Saddle Road Junction”
(highway=junction)
- 1 may be a lake (“Green Lake”) (natural=water water=lake)
- 1  Paiko Peninsula (natural=peninsula)
- 1  Haleakalā Observatories (man_made=observatory)
- 1  Honokanai'a Base Camp (tourism=camp_site)
- 1  Kauhakō Crater (geological=caldera or natural=caldera and natural=cliff)
- 1  “Keālia Boardwalk Entrance” (tourism=attraction and highway=footpath)
- 1  Hapuna Beach Prince Hotel Resort (tourism=hotel or =resort)
- 1  Maui Swap Meet (=> landuse=retail or amenity=marketplace)
- 1 Ka‘ū Desert (natural=desert or natural=grassland/heath/scrub etc)
- 1 Pepeekeo Mill (abandoned:man_made=works?)
- 1 “Split Rock” (natural=rock or =cliff or =peak)
- 1 Lehua Landing
- 1 likely beach or bay - Punalu’u is also a bay and a beach (mapped separately)

This leaves 20 objects, all nodes with a Hawaiian-based name which I
cannot interpret. 15 are on the Big Island, mostly the south-east
quadrant. Half (12)  have GNIS tags from an import, 13 have ele=*
(most low, but 3 are above 600 meters and might be peaks or ridges?).
8 of these are on the coast, and likely represent beaches, bays or
capes. Actually 2 of these have a bay with is similar name right next
to them.
So  less than 12 out of 143 place=locality features in Hawaii are
correct uses of the tag for an “unpopulated places that is not another
more specific feature”; less than 9%, and it’s possible that even the
features might have better tags if reviewed by a local mapper.


Delaware:

There are 18 features tagged place=locality
4 are relations tagged type=boundary
11 are closed ways tagged boundary=administrative (9) or =census (2)
3 are nodes:
- “Ragan” - next to a the Northeast corridor, a busy railway, and an
unnamed industrial area. This may be the name of the industrial area,
or may be the name of a historic railway station now closed.
- “Russell Complex” - on the University of Delaware campus, next to
“Russell Hall A”, B, C, D, and E - probably landuse=residential
- “Tower Circle” - in a cemetery, next to a building with description
“this is a block tower burial monument with a capacity of at least 80
people” and a circular highway=service, and a highway=path that
circles the tower. I suspect the name should either be on the service
highway or the footway.

It appears that all of the closed ways and relations in Delaware are
incorrectly tagged place=locality, and 2 out of the 3 nodes can be
tagged with more specific feature tags.


Liechtenstein:
- Use of place=locality is better here.

There are 39 nodes tagged place=locality; there are no ways or relations
- 1 is tagged natural=saddle
- 1 is tagged natural=cave_entrance
- 2 are “* Wiesen” which probably should be natural=grassland or
landuse=meadow if I’ve translated correctly
- 1 is
- 1 is Lawena Obersäss - not sure how to translate this?
- 2 have ele=1200 and ele=1273; one is “Mitätsch” and the other
“Gerawald” => natural=wood or landuse=forest?
- 1 is *berg => hill, peak or ridge?
- The other 31  have a name in German without a suffix that I
recognize, so would need a local mapper to review.

Andorra:
58 features are tagged place=locality
1 

Re: [Tagging] Subtag for place=locality?

[Joseph Eisenberg]

> I might just as well use a multipolygon relation for them as I have in
the past.

Yes, that’s my recommendation. You just need a tag that is for groups of
lakes.

Archipelagos are mapped as multipolygon relations tagged with
place=archipelago, name= and type=multipolygon. This makes it easy to
search for, does not duplicate the place=island tags on each island, and
can be rendered with existing tools.

A named group of lakes is similar to a water equivalent of an archipelago
(especially if they are not connected by rivers)
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] documenting cycleway=crossing

[Mateusz Konieczny]

Apr 16, 2019, 3:47 PM by marc_marc_...@hotmail.com:

> Le 16.04.19 à 15:25, Mateusz Konieczny a écrit :
>
>> Apr 16, 2019, 2:50 PM by >> marc_marc_...@hotmail.com 
>> >> :
>>
>>  > highway=path/highway=footway that has bicycle=yes/bicycle=designated.
>>
>>  highway=path + path=crossing + bicycle=yes|designate
>>  or highway=footway + footway=crossing + bicycle=yes|designate
>>  is not enought ?
>>
>> It is enough, I rephrased it. Hopefully it is now more redable
>> direct link: >> https://wiki.openstreetmap.org/wiki/Tag:cycleway=crossing 
>> 
>> edit: 
>> https://wiki.openstreetmap.org/w/index.php?title=Tag:cycleway=crossing=revision=1843156=1843110
>>  
>> 
>>
>
> yes it's more redable, but i still find it strange.
> I think this tag only makes sense when one of these
> two ways crosses a road :
> highway=cycleway
> or
> highway=path|footway + cycleway=*
>
> if there is no cycleway=* before the crossing,
> it is normal that the crossing way does not have a cycleway
> e.g. a highway=path + cycleway=no + bicycle=yes crossing a road
> must not have a cycleway=crossing. no cycleway exist there.
>
I never used this tagging but it happens. For example 
http://mpi.krakow.pl/pliki/293415/3 
(image from 
http://krakow.pl/aktualnosci/226594,1912,komunikat,awaria_wodociagu_na_skrzyzowaniu_ul__opolskiej_i_al__29_listopada.html
 

 )

https://www.openstreetmap.org/?mlat=50.08591=19.95453#map=19/50.08591/19.95453
 

has dedicated cycleway crossing, without cycleway on either side
(depending on how pedantic you are, cycling on surrounding footways may be
even considered illegal).


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

Re: [Tagging] Avoid using place=locality - find more specific tags instead

[Kevin Kenny]

On Tue, Apr 16, 2019 at 10:26 AM Joseph Eisenberg
 wrote:
> I apologize for being unnecessarily polemic by mentioning deprecation.
> I have no intention of investing time in such a proposal. I mainly
> wanted to suggest that the mappers on this mailing list think about
> using more specific tags, and check the features in their local area.

I agree that the tag is undesirable.

Nevertheless, there are named places that don't really have any
current feature to tag; it's a name that's persisted after the
corresponding feature no longer existed.

Another example I can think of is "Shattuck Clearing".  There was once
a ranger station there, but it was shut down in the 1960s and
demolished in the 1970s. There's really nothing remaining at its site.
The area, as the name suggests, had once been cleared, but it's grown
to trees. More than one trail meets there, so highway=junction is a
possibility, but it's a stretch. Equally a stretch is that there's a
guidepost and trail register (one of the mandatory ones - you're
required to sign the book when you pass one) at the site. Admittedly,
few non-hikers (and non-riders - it's accessible by horse) know the
name - it's about a 25-km walk to the nearest drivable road. The truck
trail has been disused since the ranger station was abandoned, and is
entirely impassable to anything on wheels. Nevertheless, the hikers,
skiers and riders who visit it do know the name, and it would raise no
eyebrows if one told another, "I'll leave a note in the register box
at Shattuck Clearing", or "we'll make camp somewhere near Shattuck
Clearing." Even though there's no clearing.

All of that history is really material for OHM. What it is *today* is
a locality.

It's an exception. There aren't many localities that wouldn't be
better tagged as something else. But there are some.

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

Re: [Tagging] documenting cycleway=crossing

[Hubert87]

Hi,

it feels a bit wrong to me to use cycleway=crossing on a highway=footway
or highway=path, as described on the wiki page
.
I might have been doing it wrong all the time but I always use
hw=footway + footway=crossing
hw=cycleway + cycleway=crossing
hw=path + path=crossing

Accessibility should imo be tagged using access-tags
bicycle=yes/designated/etc, or implied using default hw=*-values.



Am 16.04.2019 um 15:47 schrieb marc marc:

Le 16.04.19 à 15:25, Mateusz Konieczny a écrit :

Apr 16, 2019, 2:50 PM by marc_marc_...@hotmail.com:

  > highway=path/highway=footway that has bicycle=yes/bicycle=designated.

 highway=path + path=crossing + bicycle=yes|designate
 or highway=footway + footway=crossing + bicycle=yes|designate
 is not enought ?

It is enough, I rephrased it. Hopefully it is now more redable
direct link: https://wiki.openstreetmap.org/wiki/Tag:cycleway%3Dcrossing
edit:
https://wiki.openstreetmap.org/w/index.php?title=Tag%3Acycleway%3Dcrossing=revision=1843156=1843110

yes it's more redable, but i still find it strange.
I think this tag only makes sense when one of these
two ways crosses a road :
highway=cycleway
or
highway=path|footway + cycleway=*

if there is no cycleway=* before the crossing,
it is normal that the crossing way does not have a cycleway
e.g. a highway=path + cycleway=no + bicycle=yes crossing a road
must not have a cycleway=crossing. no cycleway exist there.
___
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] Avoid using place=locality - find more specific tags instead

[Joseph Eisenberg]

On 4/16/19, Dave Swarthout  wrote:
> Joseph wrote: We recently discussed place=locality, and I now believe this
> tag should be avoided, and perhaps deprecated."
>
> I cannot agree.

I apologize for being unnecessarily polemic by mentioning deprecation.
I have no intention of investing time in such a proposal. I mainly
wanted to suggest that the mappers on this mailing list think about
using more specific tags, and check the features in their local area.

> In the case of Alaska, these named places are so remote that there is no
> chance of me ever doing a survey

Certainly, I would not expect you to edit features outside of your
local area. Here in Indonesia there are thousands of place=village
nodes that were imported from bad government data, and many are
kilometers away from the correct location, but I'm not going to delete
them or move them unless I have good local knowledge about the area.

> Warin's question is also relevant: what about place=island or place=islet?

These are perfectly good tags, I'm sorry if I said anything that
implied I disapproved of them. There's no problem with having a place
tag for a feature with no population. Place=island is very well
defined by the coastline (or by being surrounded by natural=water /
waterway=riverbank)

The problem with locality is that it is not specific enough.The
requirement that a locality no population is a good detail.

> @MarKus: Regarding the tagging of islands or lake groups (clusters), I've
> already begun to use the type=group tag and hope that someone will push
> OSM-Carto to render such relations in the future.

It will be very difficult to handle such relations in osm2pgsql, the
tool that is used to import the database for rendering, as long as the
group relation can include other relations, ways, and nodes in one
object.

Is there any reason that lake groups cannot be tagged as multipolygon
relations? These are already handled by most database users, including
Openstreetmap-Carto.

-Joseph

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

Re: [Tagging] Subtag for place=locality?

[Eugene Alvin Villar]

On Tue, Apr 16, 2019 at 10:00 PM Joseph Eisenberg <
joseph.eisenb...@gmail.com> wrote:

> I added some comments to the talk page of your "type=group" relation
> proposal.
>
> I would recommend simplifying the proposal to just be for groups of
> nodes, because there are already relations for multipolygons (areas)
>

I disagree. I think a group of areas is semantically different from a
multipolygon. If I will relate this to GeoJSON concepts, the first is like
a GeoJSON FeatureCollection where individual Features are
Polygons/MultiPolygons, while the second is just a GeoJSON MultiPolygon.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Subtag for place=locality?

[Joseph Eisenberg]

I added some comments to the talk page of your "type=group" relation proposal.

I would recommend simplifying the proposal to just be for groups of
nodes, because there are already relations for multipolygons (areas)
and linear ways (waterway, route, etc), and it will be very difficulty
for database users to handle multiple types of objects in one
relation.

Also, I believe every database object needs a feature tag.
Multipolygon relations work in this way too. This may mean creating
new tags specific to groups of objects in some cases, but it's not
good to depend on taking tags off of the members of the relation,
since this will often lead to invalid results and is not currently
supported by editors like ID and JOSM, or most database users.

On 4/16/19, Martin Koppenhoefer  wrote:
>
>
> sent from a phone
>
>> On 16. Apr 2019, at 09:13, Joseph Eisenberg 
>> wrote:
>>
>> A "group" usually has more than 2 members, but I can't think of an
>> objective cut-off point above 2 or 3. If "three's a crowd" it's also a
>> group, no?
>>
>> So I think it's reasonable for mappers using place=archipelago to
>> describe a group of as few as 2 or 3 islands.
>>
>> Similarly, if you use a tag like "natural=lake_group" to describe
>> named groups of interconnected lakes, this could be used for as few as
>> 2 or 3 lakes, or as many as hundreds.
>
>
> the group relation requires at least 2 objects:
> https://wiki.openstreetmap.org/wiki/Proposed_features/Group_Relation
>
> It may not be absolutely typical natural language to speak of a group of
> two, but proposing also a pair relation seemed overkill ;-) and a group of
> three is fine.
>
> You would not need to specify whether it is a group of trees, lakes or
> islands, because it should be implicit through the members, but of course
> you could add tags like natural=archipelago if it makes sense to you.
>
>
> Cheers, Martin
>
>

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

Re: [Tagging] Ireland, imports and maybe locality is correct  | Re: Subtag for place=locality?

[Joseph Eisenberg]

I'm sorry for assuming these features were imported. That's a lot of
work! Thanks for getting those administrative boundaries into the
database.

A place=locality is defined as "an unpopulated location for which
there is no extant feature to which the tag could be associated;" the
original proposal said it was for a "named place that has no
population."

It seems like the majority townlands as in Ireland are populated,
however: https://en.wikipedia.org/wiki/Townland - and they are
administrative divisions, so it's not necessary for them to be tagged
with "place=*" in addition to the administrative boundary tags.

On 4/16/19, Rory McCann  wrote:
> On 15/04/2019 03:55, Joseph Eisenberg wrote:
>> https://taginfo.openstreetmap.org/keys/locality#values
>>
>> This tag is already used 65,000 times, but actually on boundaries; it
>> was used for an import in Ireland with the values locality=townland
>> and locality=subtownland. (These seem to be incorrect usages, because
>> townlands seem to be populated places)
>
> The townlands (& subtownlands) in Ireland were not imported. The were
> hand mapped. Yes, it's possible for a small number of mappers to map
> ~65,000 things (over several years).
>
> I gave a talk about the project at SotM 2016:
> https://2016.stateofthemap.org/2016/mapping-irelands-61000-administrative-boundaries/
>
> I believe the `locality=townland` was added for something for nominatim
> or geocoding, but it depends on what JOSM preset one was using at the
> same time .
>
> Given that 99.5%+ of the usage of the `locality` key was intentionally
> added by Irish mappers (and not a bad import), how can you say it's
> wrong? 
>
>
> --
> Rory
>
>
> ___
> 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] documenting cycleway=crossing

[marc marc]

Le 16.04.19 à 15:25, Mateusz Konieczny a écrit :
> Apr 16, 2019, 2:50 PM by marc_marc_...@hotmail.com:
> 
>  > highway=path/highway=footway that has bicycle=yes/bicycle=designated.
> 
> highway=path + path=crossing + bicycle=yes|designate
> or highway=footway + footway=crossing + bicycle=yes|designate
> is not enought ?
> 
> It is enough, I rephrased it. Hopefully it is now more redable
> direct link: https://wiki.openstreetmap.org/wiki/Tag:cycleway%3Dcrossing
> edit: 
> https://wiki.openstreetmap.org/w/index.php?title=Tag%3Acycleway%3Dcrossing=revision=1843156=1843110

yes it's more redable, but i still find it strange.
I think this tag only makes sense when one of these
two ways crosses a road :
highway=cycleway
or
highway=path|footway + cycleway=*

if there is no cycleway=* before the crossing,
it is normal that the crossing way does not have a cycleway
e.g. a highway=path + cycleway=no + bicycle=yes crossing a road
must not have a cycleway=crossing. no cycleway exist there.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] documenting cycleway=crossing

[Mateusz Konieczny]

Apr 16, 2019, 2:50 PM by marc_marc_...@hotmail.com:

>  > highway=path/highway=footway that has bicycle=yes/bicycle=designated.
>
> highway=path + path=crossing + bicycle=yes|designate
> or highway=footway + footway=crossing  + bicycle=yes|designate
> is not enought ?
>
It is enough, I rephrased it. Hopefully it is now more redable
direct link: https://wiki.openstreetmap.org/wiki/Tag:cycleway%3Dcrossing 

edit: 
https://wiki.openstreetmap.org/w/index.php?title=Tag%3Acycleway%3Dcrossing=revision=1843156=1843110
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Subtag for place=locality?

[Martin Koppenhoefer]

sent from a phone

> On 16. Apr 2019, at 11:30, Christoph Hormann  wrote:
> 
> and not just a name you have heard from your 
> grandfather to apply to a place around here somewhere but you can't 
> really specifiy what it refers to now.


if you (or your grandfather) can specify _where_ (around where) it is, then I 
would see it as a case for place=locality, you do not need to know where the 
name comes from.

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

[Tagging] Ireland, imports and maybe locality is correct  | Re: Subtag for place=locality?

[Rory McCann]

On 15/04/2019 03:55, Joseph Eisenberg wrote:

https://taginfo.openstreetmap.org/keys/locality#values

This tag is already used 65,000 times, but actually on boundaries; it
was used for an import in Ireland with the values locality=townland
and locality=subtownland. (These seem to be incorrect usages, because
townlands seem to be populated places)


The townlands (& subtownlands) in Ireland were not imported. The were 
hand mapped. Yes, it's possible for a small number of mappers to map 
~65,000 things (over several years).


I gave a talk about the project at SotM 2016: 
https://2016.stateofthemap.org/2016/mapping-irelands-61000-administrative-boundaries/


I believe the `locality=townland` was added for something for nominatim 
or geocoding, but it depends on what JOSM preset one was using at the 
same time .


Given that 99.5%+ of the usage of the `locality` key was intentionally 
added by Irish mappers (and not a bad import), how can you say it's 
wrong? 



--
Rory


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

Re: [Tagging] Subtag for place=locality?

[Martin Koppenhoefer]

sent from a phone

> On 16. Apr 2019, at 09:13, Joseph Eisenberg  
> wrote:
> 
> A "group" usually has more than 2 members, but I can't think of an
> objective cut-off point above 2 or 3. If "three's a crowd" it's also a
> group, no?
> 
> So I think it's reasonable for mappers using place=archipelago to
> describe a group of as few as 2 or 3 islands.
> 
> Similarly, if you use a tag like "natural=lake_group" to describe
> named groups of interconnected lakes, this could be used for as few as
> 2 or 3 lakes, or as many as hundreds.


the group relation requires at least 2 objects:
https://wiki.openstreetmap.org/wiki/Proposed_features/Group_Relation

It may not be absolutely typical natural language to speak of a group of two, 
but proposing also a pair relation seemed overkill ;-) and a group of three is 
fine.

You would not need to specify whether it is a group of trees, lakes or islands, 
because it should be implicit through the members, but of course you could add 
tags like natural=archipelago if it makes sense to you.


Cheers, Martin 

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

Re: [Tagging] Subtag for place=locality?

[Martin Koppenhoefer]

sent from a phone

> On 15. Apr 2019, at 22:55, Dave Swarthout  wrote:
> 
> (BTW, I would recommend tagging archipelagos as simple nodes or as
> multipolygon relations that include all of the islands. The wiki pages
> suggests using a "type=cluster" relation, but this would be hard to
> use)


the usecase for cluster or group relations would be islands mapped only as 
nodes, which you can not put into polygon relations. 

Cheers, Martin 


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

Re: [Tagging] documenting cycleway=crossing

[marc marc]

Le 16.04.19 à 12:35, Mateusz Konieczny a écrit :
> I turned
> https://wiki.openstreetmap.org/w/index.php?title=Tag:cycleway%3Dcrossing
> from redirect into a separate page.

great.

 > cycleway=crossing is additional tag <...> with highway=cycleway

ok

 > highway=path/highway=footway that has bicycle=yes/bicycle=designated.

highway=path + path=crossing + bicycle=yes|designate
or highway=footway + footway=crossing  + bicycle=yes|designate
is not enought ?
I find it strange to strange to add a cycleway if the way before
the crossing doesn't have a cycleway=* tag
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Subtag for place=locality?

[Paul Allen]

On Tue, 16 Apr 2019 at 07:52, Joseph Eisenberg 
wrote:

> That's a challenging one, but it's possible to use a lifecycle prefix
> like proposed:aeroway=aerodrome or abandoned=yes?
>
> If 2 prefixes can be added, you could use
> abandoned:proposed:aeroway=aerodrome
>

To my mind there is a problem with applying lifecycle prefixes to physical
objects: they vanish
from the map (standard carto).  It's great for cases like a pub that has
closed but there's a
possibility of it being bought and re-opened: disused:amenity=pub.  Not so
good for
abandoned buildings: abandoned:building=yes because something that is
clearly present
on the ground (with broken windows and other damage) vanishes from the
map.  Which is
why, for physical objects, I end up using abandoned=yes and disused=yes.

Maybe the renderer should treat lifecycle prefixes on physical objects
differently.  Or the wiki
should be amended to state that disused=yes is valid and preferred to a
lifecycle prefix on
physical objects.  Or something.

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

Re: [Tagging] Avoid using place=locality - find more specific tags instead

[Christoph Hormann]

On Tuesday 16 April 2019, Dave Swarthout wrote:
>
> @MarKus: Regarding the tagging of islands or lake groups (clusters),
> I've already begun to use the type=group tag and hope that someone
> will push OSM-Carto to render such relations in the future.

On a general note:  It is very unlikely that software developers are 
going to open the can of worms of interpreting relations that have 
other relations as members.  Especially for tools that need to deal 
with differential data updates like osm2pgsql.

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

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

Re: [Tagging] Avoid using place=locality - find more specific tags instead

[Dave Swarthout]

Joseph wrote: We recently discussed place=locality, and I now believe this
tag
should be avoided, and perhaps deprecated."

I cannot agree.

In the case of Alaska, these named places are so remote that there is no
chance of me (or any other OSM mapper for that matter), ever doing a survey
to determine if those place names are in use by locals (trappers hunters,
canoeists) or not. I'm willing to change my tagging practices and ADD a new
and better designed tag reflecting the status of such places as can best be
determined from DigitalGlobe imagery but I am certainly not going to remove
the place=locality tag from them.

Warin's question is also relevant: what about place=island or place=islet?
FYI, Alaska currently has more than 500K nodes, 5646 ways and 186 relations
that represent either a place=island or place=islet and I'm still adding
more of them daily.

@MarKus: Regarding the tagging of islands or lake groups (clusters), I've
already begun to use the type=group tag and hope that someone will push
OSM-Carto to render such relations in the future.

On Tue, Apr 16, 2019 at 5:26 AM Markus  wrote:

> Hi
>
> On Tue, 16 Apr 2019 at 09:40, Joseph Eisenberg
>  wrote:
> >
> > Two of the examples need new tags created:
> > 3 lakes with a name: needs a new tag, perhaps natural=lake_group as a
> > multipolygon relation?
>
> There is already a proposed and used type=group relation for all kind
> of named groups:
>
> https://wiki.openstreetmap.org/wiki/Proposed_features/Group_Relation
>
> Regards
>
> Markus
>
> ___
> 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] documenting cycleway=crossing

[Mateusz Konieczny]

Apr 16, 2019, 8:21 AM by tagging@openstreetmap.org:

> there are ~12,000 uses of cycleway=crossing  on ways according to taginfo, 
> but no documentation for the tag. 
>
To reduce confusion in the future I turned 
https://wiki.openstreetmap.org/w/index.php?title=Tag:cycleway%3Dcrossing 

from redirect into a separate page.

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

Re: [Tagging] documenting cycleway=crossing

[Mateusz Konieczny]

Apr 16, 2019, 12:22 PM by p...@trigpoint.me.uk:

>
>
> On Tuesday, 16 April 2019, Mateusz Konieczny wrote:
>
>>
>> Apr 16, 2019, 8:21 AM by >> tagging@openstreetmap.org 
>> >> :
>>
>> >> highway=cycleway cycleway=crossing
>> >>
>> >
>> > I have been mapping more and more cycleways here in Japan, and there is a 
>> > true need for cycleway=crossing. 
>> >
>> > Unlike some other countries where the cycleway is more akin to a road, 
>> > cycleways here in Japan are often treated like footpaths, and dead-end 
>> > into sidewalks and use pedestrian crosswalks for crossing roads. I have no 
>> > problem using relation links to incluse non-cycleway ways into a larger 
>> > cyclepath replation, but often times a cyclepath crosses a road and there 
>> > is heavily painted markings, such as a zebra crossing. 
>> >
>> It is the same situation as in Poland. I am using highway=crossing 
>> bicycle=yes
>> (and highway=crossing bicycle=no for cases where cycleway crosses road and 
>> cyclist is legally
>> obligated to dismount and cross as a pedestrian)
>>
>
> I think that should be bicycle=dismount rather than no.
>
> Thats what I use in such circumstances, and matches associated signage.
>
> Routers do support this tag.
>
In Poland separate bicycle=dismount and bicycle=no are misleading as in traffic 
law these two
are equivalent. Also, in Poland it matches signage better - traffic signs in 
such situation are
not depicting dismounted cyclists but simply no cyclists or even "no bicycles" 
sign.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] tags for a live stock sale yard

[bkil]

Thank you for the links. Note that the page about the selling options
raises more questions than it answers (how common are the listed
options?). While it is incidental that auction houses may also be
equipped with a stockyard if they also hold livestock auctions, this
is not a necessity for some of the options. Although, I acknowledge
that the few USA livestock "market" POI I've checked in OSM were
placed near barns by farm owners, so calling them "auction houses"
seem a bit stretch.

It looks like we are going to be having separate tagging conventions
for livestock auctions compared to livestock markets, as it would
probably not be comfortable for mappers in the USA and Australia to
add auction=yes to every livestock auctioning marketplace without
presets.

The working of a common market is different from an auction house. In
the former, the seller bring a few of their own home produce, remains
at their table all day long, waiting for potential customers and
working with average prices. This method can work even in case of very
low attendance, as unsold items could be brought to market on the next
day as well. While auctioning of items can be a much quicker
undertaking, especially for big producers having a large number of
items for sale.

From the viewpoint of a buyer, you can come to the marketplace at any
time when it is open to purchase a given good, while you will need to
attend an auction according to a fixed schedule if you want to get
something, although at the end of the day, you may leave empty handed.

I don't feel strongly about the exact tags we will use for this
purpose, although putting livestock auctions below the general concept
describing auction houses that can auction various things can have its
advantages. For example:
amenity=auction_house + auction=livestock;grain
amenity=auction + auction=livestock;grain
I'm not sure what other kind of merchandise you sell through auctions,
the most common ones around here deal with vehicles, antiques, art,
jewelry or real estate.

Here's the alternative:
amenity=livestock_auction
The problem with this one is that I have a feeling that mappers would
try to abuse it by typing livestock_market by choice or by accident,
however that is a separate concept in other parts of the world.

As hinted earlier, most towns around here hold frequent livestock
markets ("állatpiac") at a fixed place similar to how common markets
("piac") are being held on certain days. For smaller communities and
less common livestock, livestock trade shows (fairs,
"állatbörze"/"állatvásár") are being organized in various regions,
sometimes on areas dedicated for this purpose, while at other places
on common grounds. On both of these, haggling is the common purchase
method, not auctioning.

According to our tax office, we also have a legal category for
livestock auctions ("állatárverés"), although they seem to be rarely
held.

This is how a livestock market/fair looks like:
https://www.programturizmus.hu/ajanlat-dunafoldvari-orszagos-allat-es-kirakodovasar.html
https://www.szoljon.hu/cimke/allatvasar/

Here's how common they are, although many do not possess online presence:
https://www.programturizmus.hu/ajanlat-allatvasar.html
http://www.vasarokfesztivalok.eoldal.hu/cikkek/orszagos-vasarok--a---k-/

According to Wikipedia and personal experience, It is customary to
classify marketplaces by function:
marketplace=livestock/antique/farmers/fish/flea/flower/craft

I feel that farmers' market, fish market and crafts market are related
and may perhaps be combined into a single category of "producers'
market", but do share what you think about this. This term is mostly
used in connection with farmer's (and artisanal handicraft) markets in
Hungary, as fish markets are rare, but we have a lot from every other
category.

On Tue, Apr 16, 2019 at 1:47 AM Graeme Fitzpatrick
 wrote:
>
>
> On Tue, 16 Apr 2019 at 06:52, bkil  wrote:
>>
>> Warin did not mention how they are purchased, i.e., whether an auction
>> is being held
>
>
> Yep, as mentioned previously, livestock are (almost ?) always sold via 
> auction. Have a look at https://www.facebook.com/watch/?v=2082183102010269 
> which shows a little bit of the auction process in the background.
>
> Another description of it is: 
> http://www.mareebasaleyards.com.au/index.php?option=com_content=article=53:purchase=31:general=46
>
>>
>> This sounds like a different feature:
>> https://en.wiktionary.org/wiki/stockyard
>> "(US) An enclosed yard, with pens, sheds etc. or stables, where
>> livestock is kept temporarily before being slaughtered, treated, sold,
>> or shipped etc."
>
>
> Yes, that's a yard where stock is held. A sales yard is like this 
> http://www.dalbyregionalsaleyards.com.au/about/ which is broadly similar, but 
> not quite the same.
>
>> I'm still looking for the answer regarding the ratio of auction vs.
>> haggling in USA livestock markets/auctions,
>
>
> Not easy to find, but here's some info: 
> 

Re: [Tagging] documenting cycleway=crossing

[Philip Barnes]

On Tuesday, 16 April 2019, Mateusz Konieczny wrote:
> 
> Apr 16, 2019, 8:21 AM by tagging@openstreetmap.org:
> 
> >> highway=cycleway cycleway=crossing
> >>
> >
> > I have been mapping more and more cycleways here in Japan, and there is a 
> > true need for cycleway=crossing. 
> >
> > Unlike some other countries where the cycleway is more akin to a road, 
> > cycleways here in Japan are often treated like footpaths, and dead-end into 
> > sidewalks and use pedestrian crosswalks for crossing roads. I have no 
> > problem using relation links to incluse non-cycleway ways into a larger 
> > cyclepath replation, but often times a cyclepath crosses a road and there 
> > is heavily painted markings, such as a zebra crossing. 
> >
> It is the same situation as in Poland. I am using highway=crossing bicycle=yes
> (and highway=crossing bicycle=no for cases where cycleway crosses road and 
> cyclist is legally
> obligated to dismount and cross as a pedestrian)

I think that should be bicycle=dismount rather than no.

Thats what I use in such circumstances, and matches associated signage.

Routers do support this tag.

Phil (trigpoint)






> 
> >
> > there are ~12,000 uses of cycleway=crossing  on ways according to taginfo, 
> > but no documentation for the tag. 
> >
> 
> https://wiki.openstreetmap.org/w/index.php?title=Tag:cycleway%3Dcrossing=no
>  
> 
>  
> redirects to https://wiki.openstreetmap.org/wiki/Tag:footway%3Dcrossing 
>  that has some
> documentation.
> 
> See also "footway=crossing and cycleway=crossing are sometimes used on ways 
> which lead 
> from a sidewalk to the crossing node (The node which has this 
> highway=crossing tag) "
> at https://wiki.openstreetmap.org/wiki/Tag:highway%3Dcrossing 
>  - see
> https://wiki.openstreetmap.org/w/index.php?search=cycleway%3Dcrossing
> I found it by searching cycleway=crossing at 
> https://wiki.openstreetmap.org/w/index.php?search 
> 
> 
> 
> 
>

-- 
Sent from my Sailfish device
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] documenting cycleway=crossing

[Mateusz Konieczny]

Apr 16, 2019, 8:21 AM by tagging@openstreetmap.org:

>> highway=cycleway cycleway=crossing
>>
>
> I have been mapping more and more cycleways here in Japan, and there is a 
> true need for cycleway=crossing. 
>
> Unlike some other countries where the cycleway is more akin to a road, 
> cycleways here in Japan are often treated like footpaths, and dead-end into 
> sidewalks and use pedestrian crosswalks for crossing roads. I have no problem 
> using relation links to incluse non-cycleway ways into a larger cyclepath 
> replation, but often times a cyclepath crosses a road and there is heavily 
> painted markings, such as a zebra crossing. 
>
It is the same situation as in Poland. I am using highway=crossing bicycle=yes
(and highway=crossing bicycle=no for cases where cycleway crosses road and 
cyclist is legally
obligated to dismount and cross as a pedestrian)

>
> there are ~12,000 uses of cycleway=crossing  on ways according to taginfo, 
> but no documentation for the tag. 
>

https://wiki.openstreetmap.org/w/index.php?title=Tag:cycleway%3Dcrossing=no
 

 
redirects to https://wiki.openstreetmap.org/wiki/Tag:footway%3Dcrossing 
 that has some
documentation.

See also "footway=crossing and cycleway=crossing are sometimes used on ways 
which lead 
from a sidewalk to the crossing node (The node which has this highway=crossing 
tag) "
at https://wiki.openstreetmap.org/wiki/Tag:highway%3Dcrossing 
 - see
https://wiki.openstreetmap.org/w/index.php?search=cycleway%3Dcrossing
I found it by searching cycleway=crossing at 
https://wiki.openstreetmap.org/w/index.php?search 




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

Re: [Tagging] Subtag for place=locality?

[Christoph Hormann]

On Tuesday 16 April 2019, Mark Wagner wrote:
>
> There's a "place=locality" near me called "Seven Mile Airstrip". 
> Now, that's an interesting choice of names for the place, because
> there's no evidence that it was ever used for aviation.  The best
> guess I've seen for where the name came from is that it was intended
> as an auxiliary runway for Spokane Army Air Depot during World War
> II, and after construction was canceled, the name stuck around.
>
> What tag would you recommend for "thing people believe is the
> abandoned construction site for a runway that was never built"?

The crux about about abandoned:* is that it is usually only verifiable 
as long as physical remains are present.  I don't know this particular 
situation but it looks like that is not the case here.  The question 
you need to ask yourself is what the name currently refers to and tag 
accordingly.  Is it the name of a section of a road ("drive east along 
Seven Mile Airstrip"), the name of a neighborhood or parts of it ("i 
live in Seven Mile Airstrip"), the name of some kind of common area 
("lets have a barbecue tonight at Seven Mile Airstrip"), some patch of 
wilderness ("i went hunting yesterday and shot a rabbit at Seven Mile 
Airstrip").  If you can clearly give the named feature some kind of 
classification of what it is that could also apply to other similar 
places with different name elsewhere you should use or create a tag 
indicating that.  Only if that is not the case you might use the 
generic place=locality - but only if it is actually a verifiably 
locatable place and not just a name you have heard from your 
grandfather to apply to a place around here somewhere but you can't 
really specifiy what it refers to now.

If you look into the database you can find place=locality being used for 
a lot of very different things most of which you could clearly classify 
more precisely.  A tag like place=locality will likely always exist in 
OSM - even if this one is deprecated an alternative would be invented.  
But it should be used as sparsely as possible to make the data as 
meaningful as possible.

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

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

Re: [Tagging] Avoid using place=locality - find more specific tags instead

[Markus]

Hi

On Tue, 16 Apr 2019 at 09:40, Joseph Eisenberg
 wrote:
>
> Two of the examples need new tags created:
> 3 lakes with a name: needs a new tag, perhaps natural=lake_group as a
> multipolygon relation?

There is already a proposed and used type=group relation for all kind
of named groups:

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

Regards

Markus

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

Re: [Tagging] Avoid using place=locality - find more specific tags instead

[Warin]

On 16/04/19 17:38, Joseph Eisenberg wrote:

We recently discussed place=locality, and I now believe this tag
should be avoided, and perhaps deprecated.

To summarize, most of these features were added in North America from
GNIS imports; almost 20% are in Alaska alone (>200,000!), and they
were used for all sorts of features that are not populated places:
abandoned hamlets, former mining camps, construction sites, railroad
and highway junctions, former locations of Native Alaskan villages,
etc.


The key place is not only for populated places e.g.

place=island - some of these are not populated...

place=islet - many of these are not populated...

place=sea

place=ocean




Martin and Warin suggested to use abandoned:place=* for those which
were former place=hamlet, =village, isolated_dwelling, etc.

Several people mentioned ways they have used this tag for a "place
without population that has a name:" for example, to tag crossroads,
hills, a wood, a field, a pair islands, a group of a few lakes, an
informal landmark / route mark, an abandoned airstrip, a proposed
airstrip, etc.

However, most of these suggested uses have other tags that could be
more specific
crossroads: highway=junction
railway junction: railway=junction
hill: natural=peak or natural=ridge or natural=hill
wood: natural=wood
field: landuse=farmland or =meadow
islands: place=archipelago
airstrip: proposed:aerodrome=airstrip + abandoned=yes;


I don't think that fits the OSM use of the key proposed, I believe that is for 
a feature that is likely to be built..
in this particular case it was proposed but is now not proposed yet people 
still use the name.
OHM is the place to put history - not OSM, so proposed:aerodrome=airstrip with 
the start and end dates can go there...
but the name should remain in OSM as that is still 'in use'.


abandoned:aerodome=airstrip

Two of the examples need new tags created:
3 lakes with a name: needs a new tag, perhaps natural=lake_group as a
multipolygon relation?
An informal landmark (eg an old car wheel up on a tree) - perhaps
there is something for this already.


There is all ready ... place=locality : A named place that has no population.!!!



I believe that place=locality was a reasonable idea when it was
proposed in 2007, and few tags had been developed. But now, over 11
years later, we have more specific tags for almost everything that is
currently tagged this way.


But not for 'A named place that has no population'. Which is what  
place=locality is mean for.



My suggestion: check out all the features tagged with place=locality
in your area. If they have a more specific tag or a more precise tag
can be added, please remove the place=locality tag.


I have changed some tags from  place=hamlet/village to place=locality as I know 
that there has never ever been a population there ...

How many of these exist in 'your' area? Have you checked them and changed those 
appropriately?

The example I gave is one that cannot be migrated to a better tag .. as there 
is not other tag for it.
I have others too .. but I don't know the story behind them. Never the less 
they are in use as local names used for navigation.
I would object strongly to there removal.


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

Re: [Tagging] Avoid using place=locality - find more specific tags instead

[Anton Klim]

Deprecating a tag because it was misused would remove quite a lot of current 
osm tagging. 
I also think some of the examples you mention cannot be re-tagged without some 
proper research that’s not going to happen (do the locals call that place 
%name% because it was a crosssroads at some point or because there was a 
village nearby? Who knows). 
Using a lifecycle prefix is definitely a good idea, but there is simply no data 
to determine what to prefix in some cases. By all means, if one could add a 
more specific tag they should, but localities are a thing and place=locality 
can be used correctly. 

But it’s osm, so any tag you like, huh?

Ant

> 16/04/2019, 8:38, Joseph Eisenberg  написал(а):
> 
> We recently discussed place=locality, and I now believe this tag
> should be avoided, and perhaps deprecated.
> 
> To summarize, most of these features were added in North America from
> GNIS imports; almost 20% are in Alaska alone (>200,000!), and they
> were used for all sorts of features that are not populated places:
> abandoned hamlets, former mining camps, construction sites, railroad
> and highway junctions, former locations of Native Alaskan villages,
> etc.
> 
> Martin and Warin suggested to use abandoned:place=* for those which
> were former place=hamlet, =village, isolated_dwelling, etc.
> 
> Several people mentioned ways they have used this tag for a "place
> without population that has a name:" for example, to tag crossroads,
> hills, a wood, a field, a pair islands, a group of a few lakes, an
> informal landmark / route mark, an abandoned airstrip, a proposed
> airstrip, etc.
> 
> However, most of these suggested uses have other tags that could be
> more specific
> crossroads: highway=junction
> railway junction: railway=junction
> hill: natural=peak or natural=ridge or natural=hill
> wood: natural=wood
> field: landuse=farmland or =meadow
> islands: place=archipelago
> airstrip: proposed:aerodrome=airstrip + abandoned=yes;
> abandoned:aerodome=airstrip
> 
> Two of the examples need new tags created:
> 3 lakes with a name: needs a new tag, perhaps natural=lake_group as a
> multipolygon relation?
> An informal landmark (eg an old car wheel up on a tree) - perhaps
> there is something for this already.
> 
> I believe that place=locality was a reasonable idea when it was
> proposed in 2007, and few tags had been developed. But now, over 11
> years later, we have more specific tags for almost everything that is
> currently tagged this way.
> 
> My suggestion: check out all the features tagged with place=locality
> in your area. If they have a more specific tag or a more precise tag
> can be added, please remove the place=locality tag.
> 
> (If this results in the name no longer rendering in the
> Openstreetmap-carto, please check the list of issues and add a comment
> if you think that the feature should have a name label rendered on a
> general map: http://github.com/gravitystorm/openstreetmap-carto/issues
> )
> 
> Joseph
> 
> ___
> 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] Subtag for place=locality?

[Joseph Eisenberg]

It sounds like "Sled Harbor" is a trailhead, a place where you leave a
vehicle to start hiking on a footpath, or where you are picked up in a
vehicle after your hike, correct?

A few months ago we discussed the tag highway=trailhead - used over
1500 times: https://wiki.openstreetmap.org/wiki/Tag%3Ahighway%3Dtrailhead

Trailheads in remote parts of the USA are often no more than a named
place where a path meets a highway.

If the name also describes the clearing in the woods as well, this
might also be landuse=meadow (or natural=scrub, natural=wetland etc).

On 4/16/19, Kevin Kenny  wrote:
> There are named localities that have only the most tenuous of
> identifiable features.
>
> One example that I've visited is 'Sled Harbor'. It never had a
> population. It was just a place where the woods were open enough that
> loggers could store their sleds there in the summer. It's now right at
> the boundary between protected wilderness and International Paper
> land. Since there's an easement for the public to travel International
> Paper's road (well, logging track), it's the farthest that one can
> lawfully drive (well, force passage with a 4WD, when there isn't deep
> snow or mud) to pick up or drop off a party. Because of this, hikers
> still use the name. But it's really just a point where the
> highway=track crosses into the boundary=protected_area. There's no
> formal parking. It isn't the endpoint of the track, since it continues
> in farther to abandoned logging camps dating from before the state
> owned the Jessup River parcel. All that there is there is a sign
> saying something like, "no motor vehicles beyond this point."  It is
> still a place with a name.
>
> It did come in from GNIS as 'populated place,' which it is not and
> never was. Still, I don't see a good alternative to place=location for
> it, so I'm definitely against the idea of removing locations
> wholesale.
>
> ___
> 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] Avoid using place=locality - find more specific tags instead

[Joseph Eisenberg]

We recently discussed place=locality, and I now believe this tag
should be avoided, and perhaps deprecated.

To summarize, most of these features were added in North America from
GNIS imports; almost 20% are in Alaska alone (>200,000!), and they
were used for all sorts of features that are not populated places:
abandoned hamlets, former mining camps, construction sites, railroad
and highway junctions, former locations of Native Alaskan villages,
etc.

Martin and Warin suggested to use abandoned:place=* for those which
were former place=hamlet, =village, isolated_dwelling, etc.

Several people mentioned ways they have used this tag for a "place
without population that has a name:" for example, to tag crossroads,
hills, a wood, a field, a pair islands, a group of a few lakes, an
informal landmark / route mark, an abandoned airstrip, a proposed
airstrip, etc.

However, most of these suggested uses have other tags that could be
more specific
crossroads: highway=junction
railway junction: railway=junction
hill: natural=peak or natural=ridge or natural=hill
wood: natural=wood
field: landuse=farmland or =meadow
islands: place=archipelago
airstrip: proposed:aerodrome=airstrip + abandoned=yes;
abandoned:aerodome=airstrip

Two of the examples need new tags created:
3 lakes with a name: needs a new tag, perhaps natural=lake_group as a
multipolygon relation?
An informal landmark (eg an old car wheel up on a tree) - perhaps
there is something for this already.

I believe that place=locality was a reasonable idea when it was
proposed in 2007, and few tags had been developed. But now, over 11
years later, we have more specific tags for almost everything that is
currently tagged this way.

My suggestion: check out all the features tagged with place=locality
in your area. If they have a more specific tag or a more precise tag
can be added, please remove the place=locality tag.

(If this results in the name no longer rendering in the
Openstreetmap-carto, please check the list of issues and add a comment
if you think that the feature should have a name label rendered on a
general map: http://github.com/gravitystorm/openstreetmap-carto/issues
)

Joseph

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

Re: [Tagging] Subtag for place=locality?

[Joseph Eisenberg]

The definition of archipelago (borrowed from wikipedia): "also known
as an island group or island chain: a named chain, cluster or group of
closely related islands."

A "group" usually has more than 2 members, but I can't think of an
objective cut-off point above 2 or 3. If "three's a crowd" it's also a
group, no?

So I think it's reasonable for mappers using place=archipelago to
describe a group of as few as 2 or 3 islands.

Similarly, if you use a tag like "natural=lake_group" to describe
named groups of interconnected lakes, this could be used for as few as
2 or 3 lakes, or as many as hundreds.

Using place=locality wouldn't work in this situation, since it's used
on a single node, and this is not helpful for describing 2 islands.

On 4/16/19, Dave Swarthout  wrote:
>> Can you give an example of one of these groups of named islands? If they
> are close together and divided from other islands in the area, I would use
> “archipelago”.
>
> Here's a small group of only two islands that is definitely not an
> archipelago, (as I understand that term, i.e., a "chain" of islands), and
> have one name to describe both islands, the Leland Islands:
>
> https://www.openstreetmap.org/way/20287799#map=14/58.6562/-135.9916
>
> In this case, the original mapper didn't tag them as a multipolygon but
> applied the place=island tag to the group as a node. I fact, he didn't even
> bother to redraw the horrible PGS coastline to separate them into
> individual islands.
>
> Alaska has hundreds of these island groups.
>
> On Mon, Apr 15, 2019 at 7:12 PM Warin <61sundow...@gmail.com> wrote:
>
>> On 15/04/19 22:04, Joseph Eisenberg wrote:
>> > That's an interesting example. Was the wheel put there as a landmark
>> > or route marker, or just for fun?
>>
>> I don't know. I would assume as a landmark, to form a meeting place or a
>> simple navigational aid. I don't even know if the present wheel is the
>> original one.
>>
>> >
>> > If the tag "place=locality" didn't exist, how would you tag this?
>>
>> I'd ask here, that is one of the things this group is good for.
>>
>> >
>> > On 4/15/19, Warin <61sundow...@gmail.com> wrote:
>> >> As an example of a locality that has never had a population
>> >>
>> >> https://www.openstreetmap.org/node/117041320
>> >>
>> >> /The Wheel/ (a car wheel - no tyre) was originally mounted on a tree
>> >> by
>> >> bushwalkers to mark the hub of the Blue Labyrinth's ridges.
>> >>
>> >> No one has ever lived there. Plenty of people go past, and it still a
>> >> navigational feature.
>> >>
>> >> Fairly certain other localities have their stories to tell too.
>> >>
>> >>
>> >> n 15/04/19 17:23, Warin wrote:
>> >>>  From the original start of place=locality
>> >>>
>> >>> /All current place tags are for either populated areas, or for larger
>> >>> areas of County sized or bigger. The place=locality tag is useful for
>> >>> places that have a specific name, but do not necessarily have any
>> >>> geographic feature or population centre that could be used to attach
>> >>> a
>> >>> name tag to. /
>> >>>
>> >>> That to me suggest that places that locality can be a place that had
>> >>> population, or places that did not have a population.
>> >>>
>> >>> But, I agree, that places that had a population would be better
>> >>> tagged
>> >>> disused:/abandonded: place=hamlet/town/village/city
>> >>>
>> >>> I think that can go on the wiki for locality... under 'when not to
>> >>> use' with the others there.
>> >>> /
>> >>> /
>> >>> On 15/04/19 17:03, Martin Koppenhoefer wrote:
>> 
>>  sent from a phone
>> 
>>  On 15. Apr 2019, at 03:55, Joseph Eisenberg
>>  mailto:joseph.eisenb...@gmail.com>>
>> wrote:
>> 
>> > The most important value would be one for a locality that is a
>> > former
>> > populated place but no longer has a population.
>> 
>>  I’ve always understood the population part of the locality tag
>>  definition as a way of saying the place name does not relate to a
>>  settlement or dwelling (but it doesn’t necessarily mean nobody is
>>  living around there, it means this name is not for an inhabited
>>  place). A generic tag for a place name/ toponym, to be used where no
>>  specific tag has yet been developed.
>>  (e.g. we have specific tags for toponyms that refer to mountain
>>  peaks, wetlands, lakes, islands, deserts, caves, settlements, etc.
>>  so
>>  we don’t use locality for them)
>> 
>>  I’m not sure I’d support locality subtags, for lots of things a main
>>  tag might be more fitting with the established tagging system, but
>>  it
>>  depends on the actually proposed values.
>> 
>>  For ghost towns (settlements) I’ve found a lot tagged as
>>  abandoned:place=hamlet/village/town
>> 
>>  https://taginfo.openstreetmap.org/keys/abandoned:place#values
>> 
>>  which seems inline with the rest of our tagging and is by far more
>>  frequent than any “ghost” variations.
>> 

Re: [Tagging] Subtag for place=locality?

[Joseph Eisenberg]

That's a challenging one, but it's possible to use a lifecycle prefix
like proposed:aeroway=aerodrome or abandoned=yes?

If 2 prefixes can be added, you could use abandoned:proposed:aeroway=aerodrome

But I wonder if people are talking about the former proposed airstrip
when they refer to this location.

Is there a physical geography feature at this location as well,
perhaps a valley or plateau or natural meadow?

On 4/16/19, Mark Wagner  wrote:
> On Mon, 15 Apr 2019 13:25:14 +0200
> Christoph Hormann  wrote:
>
>> place=locality is currently used as a generic tag for anything with a
>> name for which no established more precise tag exists.
>>
>> This kind of contradicts the idea of OSM which would normally suggest
>> to invent a new tag then for the type of feature you have.
>> Subtagging the generic tag to make it less generic would kind of take
>> this to a whole new level.  You could take this even further and
>> suggest tagging everything in OSM something like 'feature=thing' and
>> then differentiating only through 'thing=*'.
>>
>> Long story short - to better differentiate what is currently tagged
>> place=locality the way to go is IMO to create more specific top level
>> tags (or use existing ones like the mentioned "disused:/abandonded:").
>
> There's a "place=locality" near me called "Seven Mile Airstrip".  Now,
> that's an interesting choice of names for the place, because there's no
> evidence that it was ever used for aviation.  The best guess I've seen
> for where the name came from is that it was intended as an auxiliary
> runway for Spokane Army Air Depot during World War II, and after
> construction was canceled, the name stuck around.
>
> What tag would you recommend for "thing people believe is the abandoned
> construction site for a runway that was never built"?
>
> --
> Mark
>
> ___
> 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] documenting cycleway=crossing

[John Willis via Tagging]

> highway=cycleway cycleway=crossing


I have been mapping more and more cycleways here in Japan, and there is a true 
need for cycleway=crossing. 

Unlike some other countries where the cycleway is more akin to a road, 
cycleways here in Japan are often treated like footpaths, and dead-end into 
sidewalks and use pedestrian crosswalks for crossing roads. I have no problem 
using relation links to incluse non-cycleway ways into a larger cyclepath 
replation, but often times a cyclepath crosses a road and there is heavily 
painted markings, such as a zebra crossing. 

there are ~12,000 uses of cycleway=crossing  on ways according to taginfo, but 
no documentation for the tag. 

https://taginfo.openstreetmap.org/tags/cycleway=crossing 


I understand that in some places the mental model of using a crossing for 
something that feels like a “road” is absurd, but for paths that are forced to 
use crosswalks and various pedestrian traffic control features, we need to 
document and codify the cycleway=crossing tag. when you use a zebra crossing to 
cross a road, not marking that as a “crossing” of some type feels wrong. 

Javbw


> On Jan 27, 2019, at 2:15 AM, Martin Koppenhoefer  
> wrote:
> 
> 
> 
> sent from a phone
> 
>> On 26. Jan 2019, at 15:17, Volker Schmidt  wrote:
>> 
>> The fact that the road is crossed by is crossed by a cycleway is already 
>> defined by the "highway" tags' values of the two crossing highways.
> 
> 
> +1, I would go with highway=crossing and crossing=type of crossing, on the 
> crossing node and highway=cycleway cycleway=crossing on the crossing part of 
> the cycleway (way)
> 
> Cheers, Martin 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

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