Re: [Tagging] railway=platform nodes at tram stops

2018-02-20 Thread Jo
railway=platform unambiguously refers to an actual platform, mapped as a
way or as an area.

public_transport=platform + tram=yes (+railway=tram_stop) can be mapped on
an isolated node (so not part of the railway=* way).

It would be my preference to ONLY map it on (isolated) nodes. That's what I
do for bus stops public_transport=platform + bus=yes (+highway=bus_stop).
Nodes should be preferred, as they have coordinates. And all the extra
details like name, ref, route_ref, operator, network, zone should be mapped
ONCE on these nodes.

If we can move to such a scheme it would simplify public transport mapping
in OSM a lot.

At some point it was decided to use public_transport=stop_position for a
node which is part of the highway/railway and public_transport=platform for
the isolated node next to the highway where the passengers wait. Maybe it
would have been better to use public_transport=pole for this purpose, or
public_transport=passengers or
public_transport=abstract_approximate_position_on_node_next_to_way_containing_all_the_details_for_this_stop_and_these_are_the_nodes_that_represent_the_stops_in_the_route_relations.
But we use public_transport=platform, regardless of whether an actual
platform is present.

If there is an actual platform, it can be mapped as a way or area, with
possibly tactile_paving=yes, wheelchair=yes, but no need to repeat the name
on them and no need to add them to the route relations. They can be added
to stop_area relations.

Polyglot

2018-02-20 22:28 GMT+01:00 Selfish Seahorse :

> > in osm, platform is where people wait to take the train.
> > people always are "somewhere" before taking the train.
> > osm platform <> irl plateform
>
> Do you mean `public_transport=platform` [^1], `railway=platform` [^2] or
> both?
>
> If `railway=platform` means the same as `public_transport=platform`,
> then the wiki page about `railway=platform` needs an update, because
> it currently reads 'a railway platform'.
>
> [^1]:  >
> [^2]: 
>
> >  > how can we know whether there is a platform?
> >
> > by geometry. if it's a way, it is a "real" plaform.
> > if not, it's unknown.
>
> 'Unknown' means we don't know anything. If `railway=platform` really
> is identical to `public_transport=platform` this would mean that there
> is no way to define that there is no platform.
>
> > maybe a better value should be "embarkation point" but it's "too late"
> > to change a so common value.
>
> Why not just 'stop'?
>
> It's never too late to correct a mistake ...
>
>
> On 20 February 2018 at 22:11, marc marc  wrote:
> > Le 20. 02. 18 à 21:50, Selfish Seahorse a écrit :
> >> On 20 February 2018 at 21:24, marc marc 
> wrote:
> >>> Le 20. 02. 18 à 21:17, Selfish Seahorse a écrit :
>  If no one objects, I'd like to change the wiki page [^1] so that
>  `railway=platform` should not be used on nodes.
> >>>
> >>> I objets and request to do the opposite :
> >>> remove the "forbiden" node for railway=platform
> >>>
> >>> in almost all osm objects, it makes sense to use a node if
> >>> you don't know the precise geometry or if the object has
> >>> a small geometry
> >>> or
> >>> allow to use a way when a more precise geometry is desired.
> >>
> >> I'm sorry, I wasn't precise: I'd like to change the wiki page so that
> >> `railway=platform` should not be mandatory for tram stops without
> >> platform.
> >
> > in osm, platform is where people wait to take the train.
> > people always are "somewhere" before taking the train.
> > osm platform <> irl plateform
> > maybe a better value should be "embarkation point" but it's "too late"
> > to change a so common value.
> >
> >  > how can we know whether there is a platform?
> >
> > by geometry. if it's a way, it is a "real" plaform.
> > if not, it's unknown.
> >
> > Regards,
> > Marc
> > ___
> > Tagging mailing list
> > Tagging@openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/tagging
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging


Re: [Tagging] railway=platform nodes at tram stops

2018-02-20 Thread Selfish Seahorse
> in osm, platform is where people wait to take the train.
> people always are "somewhere" before taking the train.
> osm platform <> irl plateform

Do you mean `public_transport=platform` [^1], `railway=platform` [^2] or both?

If `railway=platform` means the same as `public_transport=platform`,
then the wiki page about `railway=platform` needs an update, because
it currently reads 'a railway platform'.

[^1]: 
[^2]: 

>  > how can we know whether there is a platform?
>
> by geometry. if it's a way, it is a "real" plaform.
> if not, it's unknown.

'Unknown' means we don't know anything. If `railway=platform` really
is identical to `public_transport=platform` this would mean that there
is no way to define that there is no platform.

> maybe a better value should be "embarkation point" but it's "too late"
> to change a so common value.

Why not just 'stop'?

It's never too late to correct a mistake ...


On 20 February 2018 at 22:11, marc marc  wrote:
> Le 20. 02. 18 à 21:50, Selfish Seahorse a écrit :
>> On 20 February 2018 at 21:24, marc marc  wrote:
>>> Le 20. 02. 18 à 21:17, Selfish Seahorse a écrit :
 If no one objects, I'd like to change the wiki page [^1] so that
 `railway=platform` should not be used on nodes.
>>>
>>> I objets and request to do the opposite :
>>> remove the "forbiden" node for railway=platform
>>>
>>> in almost all osm objects, it makes sense to use a node if
>>> you don't know the precise geometry or if the object has
>>> a small geometry
>>> or
>>> allow to use a way when a more precise geometry is desired.
>>
>> I'm sorry, I wasn't precise: I'd like to change the wiki page so that
>> `railway=platform` should not be mandatory for tram stops without
>> platform.
>
> in osm, platform is where people wait to take the train.
> people always are "somewhere" before taking the train.
> osm platform <> irl plateform
> maybe a better value should be "embarkation point" but it's "too late"
> to change a so common value.
>
>  > how can we know whether there is a platform?
>
> by geometry. if it's a way, it is a "real" plaform.
> if not, it's unknown.
>
> Regards,
> Marc
> ___
> 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] railway=platform nodes at tram stops

2018-02-20 Thread marc marc
Le 20. 02. 18 à 21:50, Selfish Seahorse a écrit :
> On 20 February 2018 at 21:24, marc marc  wrote:
>> Le 20. 02. 18 à 21:17, Selfish Seahorse a écrit :
>>> If no one objects, I'd like to change the wiki page [^1] so that
>>> `railway=platform` should not be used on nodes.
>>
>> I objets and request to do the opposite :
>> remove the "forbiden" node for railway=platform
>>
>> in almost all osm objects, it makes sense to use a node if
>> you don't know the precise geometry or if the object has
>> a small geometry
>> or
>> allow to use a way when a more precise geometry is desired.
> 
> I'm sorry, I wasn't precise: I'd like to change the wiki page so that
> `railway=platform` should not be mandatory for tram stops without
> platform.

in osm, platform is where people wait to take the train.
people always are "somewhere" before taking the train.
osm platform <> irl plateform
maybe a better value should be "embarkation point" but it's "too late" 
to change a so common value.

 > how can we know whether there is a platform?

by geometry. if it's a way, it is a "real" plaform.
if not, it's unknown.

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


Re: [Tagging] railway=platform nodes at tram stops

2018-02-20 Thread Selfish Seahorse
> Since the Wiki defines the 'platform' as 'the place where passengers
> wait for the vehicle', it doesn't have to be anything in particular. A
> patch of sidewalk could indeed be a light rail 'platform'.

This is only true for `public_transport=platform` [^1] but not for
`railway=platform`: the latter tag is defined as 'a railway platform',
that is, a *real* platform.

If we tag every tram stop with `railway=platform`, how can we know
whether there is a platform? Do we need yet another tag, e.g.
`public_transport=platform` + `tram=yes` + `railway=platform` +
`platform=no`?

[^1]: 
[^2]: 


On 20 February 2018 at 21:30, Kevin Kenny  wrote:
> On Tue, Feb 20, 2018 at 3:17 PM, Selfish Seahorse
>  wrote:
>> iD's new 'Tram Stop / Platform' preset adds a `railway=platform` tag
>> even on nodes. I thought this were a bug [^1], but the wiki [^2] says:
>>
>>> platforms: `public_transport=platform` + `railway=platform` If the platform 
>>> is just a pole with a sign and the tram stops on the road without a 
>>> physical platform, use a single node to map the "platform".
>>
>> It seems wrong to me to tag a tram stop (not the stop position) with
>> `railway=platform` if there is no platform, but only a sidewalk or not
>> even that. (It's strange enough already that
>> `public_transport=platform` doesn't mean 'platform' when used on a
>> node.) Besides, the wiki is contradictory, because according to the
>> wiki page about `railway=platform` [^3], this tag should not be used
>> on nodes.
>>
>> If no one objects, I'd like to change the wiki page [^1] so that
>> `railway=platform` should not be used on nodes.
>
> Since the Wiki defines the 'platform' as 'the place where passengers
> wait for the vehicle', it doesn't have to be anything in particular. A
> patch of sidewalk could indeed be a light rail 'platform'.
>
> Also, it's pretty common to permit nodes as surrogates for area
> features when the area is not known, not surveyed, not determinate,
> ... It's been a common practice in some quarters to drop in a node
> that means, "I know there is a building/park/whatever here" and then
> replace it with the area feature when the footprint is known. In some
> cases, the boundary of the area is indefinite - how close to the post
> you stand at a bus stop is up to you and your ability to catch the
> driver's eye.
>
> ___
> 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] railway=platform nodes at tram stops

2018-02-20 Thread Kevin Kenny
On Tue, Feb 20, 2018 at 3:17 PM, Selfish Seahorse
 wrote:
> iD's new 'Tram Stop / Platform' preset adds a `railway=platform` tag
> even on nodes. I thought this were a bug [^1], but the wiki [^2] says:
>
>> platforms: `public_transport=platform` + `railway=platform` If the platform 
>> is just a pole with a sign and the tram stops on the road without a physical 
>> platform, use a single node to map the "platform".
>
> It seems wrong to me to tag a tram stop (not the stop position) with
> `railway=platform` if there is no platform, but only a sidewalk or not
> even that. (It's strange enough already that
> `public_transport=platform` doesn't mean 'platform' when used on a
> node.) Besides, the wiki is contradictory, because according to the
> wiki page about `railway=platform` [^3], this tag should not be used
> on nodes.
>
> If no one objects, I'd like to change the wiki page [^1] so that
> `railway=platform` should not be used on nodes.

Since the Wiki defines the 'platform' as 'the place where passengers
wait for the vehicle', it doesn't have to be anything in particular. A
patch of sidewalk could indeed be a light rail 'platform'.

Also, it's pretty common to permit nodes as surrogates for area
features when the area is not known, not surveyed, not determinate,
... It's been a common practice in some quarters to drop in a node
that means, "I know there is a building/park/whatever here" and then
replace it with the area feature when the footprint is known. In some
cases, the boundary of the area is indefinite - how close to the post
you stand at a bus stop is up to you and your ability to catch the
driver's eye.

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


Re: [Tagging] railway=platform nodes at tram stops

2018-02-20 Thread marc marc
Hello,

Le 20. 02. 18 à 21:17, Selfish Seahorse a écrit :
> If no one objects, I'd like to change the wiki page [^1] so that
> `railway=platform` should not be used on nodes.

I objets and request to do the opposite :
remove the "forbiden" node for railway=platform

in almost all osm objects, it makes sense to use a node if
you don't know the precise geometry or if the object has
a small geometry
or
allow to use a way when a more precise geometry is desired.

I find it wrong to consider either wrong

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


[Tagging] railway=platform nodes at tram stops

2018-02-20 Thread Selfish Seahorse
Hello

iD's new 'Tram Stop / Platform' preset adds a `railway=platform` tag
even on nodes. I thought this were a bug [^1], but the wiki [^2] says:

> platforms: `public_transport=platform` + `railway=platform` If the platform 
> is just a pole with a sign and the tram stops on the road without a physical 
> platform, use a single node to map the "platform".

It seems wrong to me to tag a tram stop (not the stop position) with
`railway=platform` if there is no platform, but only a sidewalk or not
even that. (It's strange enough already that
`public_transport=platform` doesn't mean 'platform' when used on a
node.) Besides, the wiki is contradictory, because according to the
wiki page about `railway=platform` [^3], this tag should not be used
on nodes.

If no one objects, I'd like to change the wiki page [^1] so that
`railway=platform` should not be used on nodes.

[^1]: 
[^2]: 
[^3]: 

Regards

SelfishSeahorse

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


Re: [Tagging] reviving hollow way

2018-02-20 Thread Andy Mabbett
On 19 February 2018 at 09:00, Philip Barnes  wrote:

> As a native English speaker I have never heard the term Hollow Way

Well, as a native English speaker, I have. It's what gave the name to
"Holloway Head", in Birmingham:

   https://www.openstreetmap.org/way/430568379

   
https://billdargue.jimdo.com/placenames-gazetteer-a-to-y/places-h/holloway-head/

which is an historic street name, not a neologism.

That said, I have no view regarding the use of the term in tagging.

-- 
Andy Mabbett
@pigsonthewing
http://pigsonthewing.org.uk

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


Re: [Tagging] reviving hollow way

2018-02-20 Thread Kevin Kenny
On Tue, Feb 20, 2018 at 9:14 AM, Philip Barnes  wrote:
> As Andy says Hollow Way is an archaic term, of which the Oxford English
> Dictionary contains many.
>
> More contemporary organisations use the term sunken lane for example
> The National Trust
> https://www.nationaltrust.org.uk/features/what-are-sunken-lanes
>
> http://shropshirehistory.com/comms/romanroads.htm

None of the terms was familiar to me (and I'm a native speaker of
American English who recognizes that the USAians and Britons are
sundered by a common language). But 'sunken lane' at least made
sense as having a natural meaning as a pair of common words;
'hollow way', I'd have had to look up to make sure that its technical
meaning matched my understanding.

Contrary to popular belief, we do have them around here on the
old carriageways and turnpikes - it doesn't take a history back to
Roman or medieval times to erode them. Even 120 years of weather can
exact a toll. I'd be tempted to change the tagging on
https://flic.kr/p/oNCZVD https://www.openstreetmap.org/way/516500341
since that's exactly what happened there.

Off topic:

That is still a public right-of-way used by hikers; the rangers will
occasionally take a Jeep on it as far as the West Stony Creek
campsite gate. The guy who owns the land on both
sides thinks that because the state bought the failed homesteads
farther in, the road is abandoned. He occasionally gates it, and
the state tears down the gate. He sicked a dog on me once
when I was signing in at the trailhead. A whack across the snout
with the grip of a trekking pole made the dog lose interest in
the project - particularly since the dog could see that the
pole had a sharp end, too.

The last time I was there, the dog wasn't, and the sign at the
(open) gate was changed to read "Public easement on
private land - No motor vehicles -STAY ON THE TRAIL!" so maybe
The Powers That Be had a word with the guy.

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


Re: [Tagging] Proposed features - RFC 2 - Pressurized waterways

2018-02-20 Thread Janko Mihelić
pon, 19. velj 2018. u 19:18 François Lacombe 
napisao je:

>
> Are we talking about a new value like waterway=aqueduct ?
>

I would really like a new waterway value because the ones we have are too
restricting. "River" and "stream" cover natural waterways, and man made
values are:

1. canal - big enough to be used by boats
2. drain - for carrying superfluous water away, concrete walls
3. ditch - the same as drain but without concrete

We need a value for man made waterways that aren't used for carrying water
away, but for bringing water somewhere.

Waterway=aqueduct looks ok to me, although, in the Wikipedia page for
aqueduct, bringing water to hydro power stations is never mentioned. But I
think it's ok to use that value for that use.

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


Re: [Tagging] reviving hollow way

2018-02-20 Thread Kevin Kenny
 On 02/20/2018 08:57 AM, Martin Koppenhoefer wrote:

On 20. Feb 2018, at 00:57, Dave F 
 wrote:

But not for its original purpose, as it is in this cae

in the original “purpose”, yes. A historic=memorial is and likely ever
was a memorial. An archaeological site was something else in the past,
but it wasn’t called archaeological site at that time. The history tag
in OSM is what it currently is, not what it was.

That bald statement doesn't capture the subtleties of repurposed things. It
does not handle the case, for instance, of a building that is historically
signficant (and listed) for being the first district school in the area,
later was a meeting hall reenported to have be a meeting place of the
rebels in the Helderberg War, and is now the office of a security company.
Schools in the time it was built looked very much like private houses,
although the brickwork has a cornerstone bearing the date it was laid
(uncommon in private houses here), and the door lintel says 'District
School No. 4' [1] (but is obscured by a portico that was added later).

What it is now is building=office, and it doesn't look much different from
neighbouring small offices (which aped its style - and the building in
question has been remodeled so many times that it's now inauthentic for any
single period). Still, there's an explanatory marker out front and a little
display of old artifacts in the entryway, and it's on the National Register
of Historic Places. Simply saying, "what it is now is a building=office',
while true, loses information.

[1] yes, the number is '4' - '1', '2', and '3', while planned earlier, were
actually constructed later.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging


Re: [Tagging] reviving hollow way

2018-02-20 Thread Dave F



On 20/02/2018 07:40, joost schouppe wrote:
Some of the most used historical tags are for things that are just 
old, not necessarily disused or with another use than the original one.


As I said, everything has a history.

Wayside cross and shrine, monuments, memorials, castles etc. It just 
seems to signify a special relationship with history, not much more 
than that.


Shrines/Memorials are a bit of a strange one. Their current use is the 
same as originally planned. What's 'historic' is the event they're 
memorializing.



DaveF

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


Re: [Tagging] reviving hollow way

2018-02-20 Thread Dave F

...and water.

As the

On 20/02/2018 14:30, Philip Barnes wrote:

In this case, they are old and have a history. They started as tracks and 
usually still are.


Which is why they should be tagged as track/footway etc.


  They are sunken purely by the passage of time, wear from feet, hooves and 
cartwheels.


... and water.

DaveF


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


Re: [Tagging] reviving hollow way

2018-02-20 Thread Philip Barnes
In this case, they are old and have a history. They started as tracks and 
usually still are. They are sunken purely by the passage of time, wear from 
feet, hooves and cartwheels.

Phil (trigpoint) 

On 20 February 2018 13:57:33 GMT+00:00, Martin Koppenhoefer 
 wrote:
>
>
>sent from a phone
>
>> On 20. Feb 2018, at 00:57, Dave F 
>wrote:
>> 
>> But not for its original purpose, as it is in this cae
>
>
>in the original “purpose”, yes. A historic=memorial is and likely ever
>was a memorial. An archaeological site was something else in the past,
>but it wasn’t called archaeological site at that time. The history tag
>in OSM is what it currently is, not what it was.
>
>cheers,
>Martin 
>___
>Tagging mailing list
>Tagging@openstreetmap.org
>https://lists.openstreetmap.org/listinfo/tagging

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

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


Re: [Tagging] reviving hollow way

2018-02-20 Thread Philip Barnes
On Mon, 2018-02-19 at 20:20 +, Steve Doerr wrote:
> On 19/02/2018 09:00, Philip Barnes wrote:
> 
> > As a native English speaker I have never heard the term Hollow
> > Way, 
> > however reading the description it seems that this proposal is 
> > describing what is called a Sunken Lane.
> 
> Might need a bit more research as both are fairly obscure terms, I
> would 
> imagine. For what it's worth, 'hollow-way' [sic] is mentioned in the 
> Oxford English Dictionary, while 'sunken lane' isn't, as far as I can
> see.
> 
It not a particularly obscure landscape feature, the rural landscape is full
of them many of them and they form parts of our rights of way network and my
experience is they are called sunken lanes. 

As Andy says Hollow Way is an archaic term, of which the Oxford English
Dictionary contains many. We cannot 

More contemporary organisations use the term sunken lane for example
The National Trust  
https://www.nationaltrust.org.uk/features/what-are-sunken-lanes

http://shropshirehistory.com/comms/romanroads.htm


Phil (trigpoint) 



-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging


Re: [Tagging] reviving hollow way

2018-02-20 Thread Martin Koppenhoefer


sent from a phone

> On 20. Feb 2018, at 00:57, Dave F  wrote:
> 
> But not for its original purpose, as it is in this cae


in the original “purpose”, yes. A historic=memorial is and likely ever was a 
memorial. An archaeological site was something else in the past, but it wasn’t 
called archaeological site at that time. The history tag in OSM is what it 
currently is, not what it was.

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


Re: [Tagging] reviving hollow way

2018-02-20 Thread Andy Townsend

On 20/02/2018 07:40, joost schouppe wrote:
Some of the most used historical tags are for things that are just 
old, not necessarily disused or with another use than the original 
one. Wayside cross and shrine, monuments, memorials, castles etc. It 
just seems to signify a special relationship with history, not much 
more than that.

...

As an aside, I did go through the usage of "historic" in the UK and 
Ireland, and it's a bit of a hodge-podge between "things that were once 
something but there's nothing there now" (effectively used as a a 
lifecycle tag), "things that are still something" (e.g. memorial), and 
"things that are still something, but are important because of what they 
were" (castle etc.).  I ended up with a list of things to treat as 
buildings, things to treat as "almost buildings", things that probably 
aren't buildings and some specific other checks (e.g. "chimney"):


https://github.com/SomeoneElseOSM/SomeoneElse-style/blob/master/style.lua#L1509

Essentially, it's a complete mess.  I was glad that I was able to do 
something useful with e.g. 
https://map.atownsend.org.uk/maps/map/map.html#zoom=17=52.657347=-7.12672 
- you can find a few grains of wheat in among the chaff but most of 
what's in there is (like "hollow_way") a bit too idosyncratic to be even 
thinking about creating a regular wiki page for.





I still haven't heard of a procedure to vote on tag A over tag B, so 
I've invented my own. You can vote here:


https://framadate.org/ApOlIj5ePZvrTjz8


Misses a "none of the above" option.

Best Regards,
Andy



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