Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Di., 27. Okt. 2020 um 00:38 Uhr schrieb Dave F via Tagging <
tagging@openstreetmap.org>:

>
>
> On 26/10/2020 23:26, Martin Koppenhoefer wrote:
> >
> > crossing_ref as far as I have understood the tag, is not about the
> > type of crossing,
>
> I think you've misunderstood.



then I am happy I never used this tag. If this is about the type of
crossing, and crossing=* is also about the type of crossing, why should I
prefer one over the other? Or should I put both tags?
People have told me the crossing_ref was for the zebra markings, and I have
seen a lot of people putting them on traffic light crossings with zebra
markings.

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Dave F via Tagging]

On 26/10/2020 23:26, Martin Koppenhoefer wrote:


crossing_ref as far as I have understood the tag, is not about the 
type of crossing,


I think you've misunderstood.

DaveF

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Mo., 26. Okt. 2020 um 17:12 Uhr schrieb Dave F <
davefoxfa...@btinternet.com>:

> 'Zebra' shouldn't be use on the primary tag 'crossing'
>
> crossing_ref was created for use within the UK because many parts of the
> rest of the world didn't understand what was meant by 'zebra'. It is, after
> all, a nickname:
>
> https://wiki.openstreetmap.org/wiki/Key:crossing_ref
>


crossing_ref as far as I have understood the tag, is not about the type of
crossing, but about the presence of zebra road markings. At least around
here, they are common also on traffic light controlled pedestrian crossings.

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Dave F via Tagging]

'Zebra' shouldn't be use on the primary tag 'crossing'

crossing_ref was created for use within the UK because many parts of the 
rest of the world didn't understand what was meant by 'zebra'. It is, 
after all, a nickname:


https://wiki.openstreetmap.org/wiki/Key:crossing_ref

DaveF



On 16/09/2020 15:33, Martin Koppenhoefer wrote:



Am Mi., 16. Sept. 2020 um 16:27 Uhr schrieb Dave F via Tagging 
mailto:tagging@openstreetmap.org>>:


I thought the correct tag for this was crossing_ref. Have you
cross checked to see if they've been swapped instead of removed?



crossing_ref is a different kind of beast, as some people use it to 
tell whether there are zebra markings (can also apply to traffic light 
controlled crossings).


Frankly, I do not like the tag for zebra crossings, because this 
approach requires me to set 3 tags (one of crossing=zebra / marked / 
uncontrolled(?)  +, crossing_ref=zebra + highway=crossing, on every 
zebra crossing while I could use 2 and be done (highway=crossing with 
crossing=zebra).



Cheers
Martin


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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Fr., 18. Sept. 2020 um 09:41 Uhr schrieb Peter Elderson <
pelder...@gmail.com>:

> Changing to crossing=marked then specifying that it's a zebra just makes
> it more work, and harder to interpret.
>


+1, if you don't know the implications of crossing=zebra, then you don't
know them either for crossing=marked, marked=zebra. The latter is just a
more complicated way of telling the same, and it mostly leads to less
information because the marked=zebra (or whatever similar tag to tell the
markings are zebra marking) is often missing.

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Peter Elderson]

>
> Maybe crossing=marked + marked=???
> where ??? is the "type" of crossing - UK_zebra (as well as all their other
> birds & animals!), US_zebra, EU_zebra & so on, so if you know exactly what
> it is you can specify, but if you can only see a crossing marked there, you
> can just call it a marked crossing?
>

Isn't it obvious from the location what the country is?

Crossings can have all sorts of markings. Most markings just indicate that
it's a crossing, i.e. a preferred location for pedestrians to cross the
road. Those markings result in the *=crossing tag.

UK has a rich wildlife of crossings, but afaik all have controls except the
zebra, so you can tag the control if you think it's worthwhile.

Priority is regarded as a mappable attribute. It seems that, in many
countries, zebra striping is used to grant priority to pedestrians. If
mappers do not think crossing=zebra combined with the location is clear
enough about the priority, then I suggest to add a tag for priority.

In Nederland, zebra also implies drivers are not allowed to park on or
within a certain distance from the crossing. If that's deemed important and
should not just be implied by the zebra tag, it should be tagged on the
section of the road itself.

In Nederland, pedestrians are not allowed to cross the road near the zebra
within a certain distance (50 m or so).
This is important for pedestrian routing. I think this is handled already
by the presence of the way for pedestrians.  I may be wrong, but I believe
routing over a way is preferred to crossing wildly over a road.

In short, I think crossing=zebra says what you see, the implications can be
read from the country the location is in,  and if that's not sure enough
tag the implications separately and precisely.

Changing to crossing=marked then specifying that it's a zebra just makes it
more work, and harder to interpret.


Vr gr Peter Elderson___
> 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] automated edits seem to remove crossing=zebra drastically

[Tod Fitch]

> On Sep 17, 2020, at 9:30 AM, Matthew Woehlke  wrote:
> 
> On 17/09/2020 10.07, Shawn K. Quinn wrote:
>> On 9/17/20 08:15, Matthew Woehlke wrote:
>>> It's also atrocious because it can *only* be verified by survey. As
>>> much as we prefer surveys, the reality is that a lot of mapping
>>> happens just from aerials, where crossings (both marked and, in some
>>> cases, unmarked) can be seen, but signals cannot.
>> I have mapped many traffic signals (and, for that matter, stop and yield
>> signs) based on shadows visible on the satellite photos. If you look
>> carefully enough (Bing and Mapbox Satellite at least), they are there.
>> (Local knowledge helps too in some cases.)
> 
> *Traffic* lights I can buy. I am more suspicious of the claim that you can 
> tell whether they have pedestrian crossing signals or not, or that you can 
> reliably identify other signage based solely on outline. *Maybe* if you get 
> lucky and have a very clear shadow at the right angle, but if you try to tell 
> me you can identify https://www.openstreetmap.org/node/7695704414 (n.b. a 
> yield sign) from a shadow in aerial imagery, I am going to be deeply 
> suspicious ;-).
> 

Not from the signs or shadows of the signs, but in my area the pavement 
markings can often tell you if it is a stop or yield. Some times it is easy 
(“STOP” or “YIELD” painted on the pavement). But it seems that newer road work 
uses a different style limit line for a stop versus yield.

Back to the original topic: I am not really sure what, if any, the US version 
of a “zebra" crossing is versus a “marked” crossing. So I usually just tag as 
“marked” as that seems to be the more generic item.

The crossing you linked to *might* be an example of a US “zebra” crossing. Can 
anyone verify that for me. Also, there are no tags on the intersection node 
itself. Should there be? I have assumed that there should so that vehicle based 
navigation would have the information needed to advise the driver of particular 
type of crossing ahead.

Cheers!



signature.asc
Description: Message signed with OpenPGP
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Graeme Fitzpatrick]

On Fri, 18 Sep 2020 at 06:58, Paul Allen  wrote:

>
> Maybe "zebra crossing" should be a region-specific editor
> preset and generates a tag(s) indicating priority of the pedestrian
> and under what circumstances: "crossing, indicated by surface
> markings, pedestrian has priority after stepping onto the
> crossing" with "surface markings are straight zebra."  Something
> like that.
>

Maybe crossing=marked + marked=???
where ??? is the "type" of crossing - UK_zebra (as well as all their other
birds & animals!), US_zebra, EU_zebra & so on, so if you know exactly what
it is you can specify, but if you can only see a crossing marked there, you
can just call it a marked crossing?

Thanks

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Paul Allen]

On Thu, 17 Sep 2020 at 19:29, Peter Elderson  wrote:

> In Nederland, the zebra is a very clear and specific type of crossing with
> legal rules including yield to pedestrians walking on or even toward the
> zebra.
>

I think we may end up having to make a distinction between the pattern
of the markings and what those markings signify.

In the UK the pedestrian has priority when they step on the crossing.
In most of Europe pedestrians have priority when they one the curb
about to step onto the crossing.  There may be other countries where
a crossing has zebra markings but pedestrians do not have
priority.

In the US zebra markings are the most common way of marking
crosswalks but other markings are also used.  If I read things
right, all US states give pedestrians priority when they
enter a crosswalk, however marked.  The US also has several
variants of the zebra.  Some states give pedestrians priority
if they enter unmarked crosswalks.

Maybe "zebra crossing" should be a region-specific editor
preset and generates a tag(s) indicating priority of the pedestrian
and under what circumstances: "crossing, indicated by surface
markings, pedestrian has priority after stepping onto the
crossing" with "surface markings are straight zebra."  Something
like that.

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Do., 17. Sept. 2020 um 18:32 Uhr schrieb Matthew Woehlke <
mwoehlke.fl...@gmail.com>:

>
> *Traffic* lights I can buy. I am more suspicious of the claim that you
> can tell whether they have pedestrian crossing signals or not,



usually pedestrian crossings are marked, and depending on the imagery
resolution and width of the markings you might be able to see it. In my
area these are marked with zebra markings, and you can try to see whether
there are also stop lines for cars (i.e. traffic signals).



> or that
> you can reliably identify other signage based solely on outline.



I have seen some stop signs with additional "stop" road markings that were
clearly visible from above.
Stop-signs and give way signs use different road markings around here. The
real question in my context is: does it still apply, or has the situation
changed since these pictures were taken.

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Do., 17. Sept. 2020 um 02:45 Uhr schrieb Taskar Center :

> 1) How is this shared space controlled? ...
>
> 2) How is the space demarcated? A crossing may be demarcated by a number
> of different ground markers, 
>
> 3) How can a pedestrian call up the signal ...
>
> 4) who is sharing the way (also a bicycle crossing, animal crossing, etc)?
>
> 5) How is the space connected to the rest of the transportation layer?
>
...

>
> I think crossing=marked/unmarked was a really good step in the direction
> of getting resolution and refinement on at least one of these questions
> above.
>



really? Which of these questions do you see answered?

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Matthew Woehlke]

On 17/09/2020 15.50, Shawn K. Quinn wrote:

On 9/17/20 11:30, Matthew Woehlke wrote:

*Maybe* if you get lucky and have a very clear shadow at the right
angle, but if you try to tell me you can identify
https://www.openstreetmap.org/node/7695704414 (n.b. a yield sign)
from a shadow in aerial imagery, I am going to be deeply suspicious
;-).


Are you sure you didn't mean node 42164543 or something west of it? That
one, I'd need to survey or see street-level imagery to be confident
enough to map it. The shadow, if present, is overlaid by another in the
area. Nodes 6393986190 and 6393985684 do have the "shark's teeth" line
used with yield signs (which I did add just now).


Ah, I think I see what happened... iD really wants me to tag the node as 
something, and I was just seeing "yield" at the top of its suggestion 
list. I think I didn't notice because every other node I as looking at 
just showed up as a "point", apparently because they belonged to more 
than one way.


Also, I was trying to figure out why an apparently unnecessary node was 
there, which probably helped mislead me into thinking it was tagged. I'm 
pretty sure I'd intended to split between crossing and just sidewalk 
there, as discussed earlier in this thread (also why I was looking at 
that particular spot in the first place) and botched it. (Fixed, now.)


--
Matthew

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Shawn K. Quinn]

On 9/17/20 11:30, Matthew Woehlke wrote:
> On 17/09/2020 10.07, Shawn K. Quinn wrote:
>> On 9/17/20 08:15, Matthew Woehlke wrote:
>>> It's also atrocious because it can *only* be verified by survey. 
>>> As much as we prefer surveys, the reality is that a lot of 
>>> mapping happens just from aerials, where crossings (both marked 
>>> and, in some cases, unmarked) can be seen, but signals cannot.
>> 
>> I have mapped many traffic signals (and, for that matter, stop and 
>> yield signs) based on shadows visible on the satellite photos. If 
>> you look carefully enough (Bing and Mapbox Satellite at least), 
>> they are there. (Local knowledge helps too in some cases.)
> 
> *Traffic* lights I can buy. I am more suspicious of the claim that 
> you can tell whether they have pedestrian crossing signals or not,
> or that you can reliably identify other signage based solely on 
> outline.

In Texas (possibly elsewhere in the US) a crossing is legally considered
signal controlled even if there is only a three-colored traffic light
and  there is not a specific orange hand/white man  pedestrian signal.
Yes, it may differ elsewhere.

> *Maybe* if you get lucky and have a very clear shadow at the right 
> angle, but if you try to tell me you can identify 
> https://www.openstreetmap.org/node/7695704414 (n.b. a yield sign) 
> from a shadow in aerial imagery, I am going to be deeply suspicious 
> ;-).

Are you sure you didn't mean node 42164543 or something west of it? That
one, I'd need to survey or see street-level imagery to be confident
enough to map it. The shadow, if present, is overlaid by another in the
area. Nodes 6393986190 and 6393985684 do have the "shark's teeth" line
used with yield signs (which I did add just now).

-- 
Shawn K. Quinn 
http://www.rantroulette.com
http://www.skqrecordquest.com

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Peter Elderson]

In Nederland, the zebra is a very clear and specific type of crossing with
legal rules including yield to pedestrians walking on or even toward the
zebra.

I think this will continue to be the case even after Europe leaves the
British Union.

Vr gr Peter Elderson


Op do 17 sep. 2020 om 20:12 schreef Matthew Woehlke <
mwoehlke.fl...@gmail.com>:

> On 17/09/2020 13.44, Tod Fitch wrote:
> >> On Sep 17, 2020, at 9:30 AM, Matthew Woehlke wrote:
> >> On 17/09/2020 10.07, Shawn K. Quinn wrote:
> >>> On 9/17/20 08:15, Matthew Woehlke wrote:
>  It's also atrocious because it can *only* be verified by survey. As
>  much as we prefer surveys, the reality is that a lot of mapping
>  happens just from aerials, where crossings (both marked and, in some
>  cases, unmarked) can be seen, but signals cannot.
> >>> I have mapped many traffic signals (and, for that matter, stop and
> yield
> >>> signs) based on shadows visible on the satellite photos. If you look
> >>> carefully enough (Bing and Mapbox Satellite at least), they are there.
> >>> (Local knowledge helps too in some cases.)
> >>
> >> *Traffic* lights I can buy. I am more suspicious of the claim that
> >> you can tell whether they have pedestrian crossing signals or not,
> >> or that you can reliably identify other signage based solely on
> >> outline. *Maybe* if you get lucky and have a very clear shadow at
> >> the right angle, but if you try to tell me you can identify
> >> https://www.openstreetmap.org/node/7695704414 (n.b. a yield sign)
> >> from a shadow in aerial imagery, I am going to be deeply suspicious
> >> ;-).
> >
> > Not from the signs or shadows of the signs, but in my area the
> > pavement markings can often tell you if it is a stop or yield. Some
> > times it is easy (“STOP” or “YIELD” painted on the pavement). But it
> > seems that newer road work uses a different style limit line for a
> > stop versus yield.
>
> Ah, that's fair; I was under the impression we were talking about
> *signs*. Possibly because most of the yields I see are to yield to other
> *vehicles*, not pedestrians. (I *have* seen "yield to pedestrians", now
> that I think about it, but not sure I've ever seen *lane markings* where
> it's clear that what you are supposed to yield for is pedestrians. Other
> than crosswalks, anyway. Which... makes me wonder if
> "crossing=uncontrolled" is even correct; even more reason to not use
> that! My understanding was "uncontrolled" meant by traffic signals, but
> now I'm not so sure.)
>
> I've tagged some yields based on lane markings myself, e.g.
> https://www.openstreetmap.org/node/7714853074.
>
> > Back to the original topic: I am not really sure what, if any, the
> > US version of a “zebra" crossing is versus a “marked” crossing. So I
> > usually just tag as “marked” as that seems to be the more generic
> > item.
>
> Likewise. Even the wiki notes that this is unclear "outside the UK" (as
> I previously observed).
>
> > The crossing you linked to *might* be an example of a US “zebra”
> > crossing. Can anyone verify that for me. Also, there are no tags on
> > the intersection node itself. Should there be? I have assumed that
> > there should so that vehicle based navigation would have the
> > information needed to advise the driver of particular type of
> > crossing ahead.
>
> As I understand it, yes, and I've tagged that in other places (e.g. the
> above example). I actually have no idea why that node is marked as a
> yield; I don't think there's actually a yield there, but I'm hesitant to
> just delete it (even though apparently I'm the one that added it).
> Unfortunately I can't go survey it right now. (Have to try to remember
> to do that when/if I ever make it back to that Cracker Barrel :-).)
>
> --
> Matthew
>
> ___
> 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] automated edits seem to remove crossing=zebra drastically

[Matthew Woehlke]

On 17/09/2020 13.44, Tod Fitch wrote:

On Sep 17, 2020, at 9:30 AM, Matthew Woehlke wrote:
On 17/09/2020 10.07, Shawn K. Quinn wrote:

On 9/17/20 08:15, Matthew Woehlke wrote:

It's also atrocious because it can *only* be verified by survey. As
much as we prefer surveys, the reality is that a lot of mapping
happens just from aerials, where crossings (both marked and, in some
cases, unmarked) can be seen, but signals cannot.

I have mapped many traffic signals (and, for that matter, stop and yield
signs) based on shadows visible on the satellite photos. If you look
carefully enough (Bing and Mapbox Satellite at least), they are there.
(Local knowledge helps too in some cases.)


*Traffic* lights I can buy. I am more suspicious of the claim that 
you can tell whether they have pedestrian crossing signals or not, 
or that you can reliably identify other signage based solely on 
outline. *Maybe* if you get lucky and have a very clear shadow at 
the right angle, but if you try to tell me you can identify 
https://www.openstreetmap.org/node/7695704414 (n.b. a yield sign) 
from a shadow in aerial imagery, I am going to be deeply suspicious

;-).


Not from the signs or shadows of the signs, but in my area the 
pavement markings can often tell you if it is a stop or yield. Some 
times it is easy (“STOP” or “YIELD” painted on the pavement). But it

seems that newer road work uses a different style limit line for a
stop versus yield.


Ah, that's fair; I was under the impression we were talking about 
*signs*. Possibly because most of the yields I see are to yield to other 
*vehicles*, not pedestrians. (I *have* seen "yield to pedestrians", now 
that I think about it, but not sure I've ever seen *lane markings* where 
it's clear that what you are supposed to yield for is pedestrians. Other 
than crosswalks, anyway. Which... makes me wonder if 
"crossing=uncontrolled" is even correct; even more reason to not use 
that! My understanding was "uncontrolled" meant by traffic signals, but 
now I'm not so sure.)


I've tagged some yields based on lane markings myself, e.g. 
https://www.openstreetmap.org/node/7714853074.



Back to the original topic: I am not really sure what, if any, the
US version of a “zebra" crossing is versus a “marked” crossing. So I
usually just tag as “marked” as that seems to be the more generic
item.


Likewise. Even the wiki notes that this is unclear "outside the UK" (as 
I previously observed).



The crossing you linked to *might* be an example of a US “zebra”
crossing. Can anyone verify that for me. Also, there are no tags on
the intersection node itself. Should there be? I have assumed that
there should so that vehicle based navigation would have the
information needed to advise the driver of particular type of
crossing ahead.


As I understand it, yes, and I've tagged that in other places (e.g. the 
above example). I actually have no idea why that node is marked as a 
yield; I don't think there's actually a yield there, but I'm hesitant to 
just delete it (even though apparently I'm the one that added it). 
Unfortunately I can't go survey it right now. (Have to try to remember 
to do that when/if I ever make it back to that Cracker Barrel :-).)


--
Matthew

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Matthew Woehlke]

On 17/09/2020 10.07, Shawn K. Quinn wrote:

On 9/17/20 08:15, Matthew Woehlke wrote:

It's also atrocious because it can *only* be verified by survey. As
much as we prefer surveys, the reality is that a lot of mapping
happens just from aerials, where crossings (both marked and, in some
cases, unmarked) can be seen, but signals cannot.


I have mapped many traffic signals (and, for that matter, stop and yield
signs) based on shadows visible on the satellite photos. If you look
carefully enough (Bing and Mapbox Satellite at least), they are there.
(Local knowledge helps too in some cases.)


*Traffic* lights I can buy. I am more suspicious of the claim that you 
can tell whether they have pedestrian crossing signals or not, or that 
you can reliably identify other signage based solely on outline. *Maybe* 
if you get lucky and have a very clear shadow at the right angle, but if 
you try to tell me you can identify 
https://www.openstreetmap.org/node/7695704414 (n.b. a yield sign) from a 
shadow in aerial imagery, I am going to be deeply suspicious ;-).


--
Matthew

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Paul Allen]

On Thu, 17 Sep 2020 at 15:09, Shawn K. Quinn  wrote:

>
> I agree that the current presets available in JOSM are a bit of a botch,
> particularly "uncontrolled" for crossings technically controlled by a
> sign. "Marked" may be better but we still have the issue of changing a
> lot of previously tagged crossings. I think "island" is already covered
> by traffic_calming=island, no?
>

No.   https://wiki.openstreetmap.org/wiki/Key:crossing:island

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Clifford Snow]

On Thu, Sep 17, 2020 at 7:08 AM Shawn K. Quinn  wrote:

>
> I agree that the current presets available in JOSM are a bit of a botch,
> particularly "uncontrolled" for crossings technically controlled by a
> sign. "Marked" may be better but we still have the issue of changing a
> lot of previously tagged crossings. I think "island" is already covered
> by traffic_calming=island, no?
>

Where I lived in Seattle, we have these traffic calming islands. They are
basically just a physical small island in the center of an intersection
usually planted with small scrubs. From aerial imagery they might look like
a roundabout, but technically they are not. The purpose is to slow traffic.
Often there isn't even a place for a pedestrian. They play hell with trucks
trying to make deliveries.

I agree that JOSM's presets are insufficient. I've made my own for mapping
pedestrian crossings.

Best,
Clifford
-- 
@osm_washington
www.snowandsnow.us
OpenStreetMap: Maps with a human touch
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Shawn K. Quinn]

On 9/17/20 08:15, Matthew Woehlke wrote:
> It's also atrocious because it can *only* be verified by survey. As
> much as we prefer surveys, the reality is that a lot of mapping
> happens just from aerials, where crossings (both marked and, in some
> cases, unmarked) can be seen, but signals cannot.

I have mapped many traffic signals (and, for that matter, stop and yield
signs) based on shadows visible on the satellite photos. If you look
carefully enough (Bing and Mapbox Satellite at least), they are there.
(Local knowledge helps too in some cases.)

> As someone who's generated a fair number of "uncontrolled" crossings
> because that was the only "blessed" tag, I would much prefer
> separating the presence or absence of features that can be verified
> in an aerial (marked, unmarked, striped, island, ...) from whether or
> not signals are present.

I agree that the current presets available in JOSM are a bit of a botch,
particularly "uncontrolled" for crossings technically controlled by a
sign. "Marked" may be better but we still have the issue of changing a
lot of previously tagged crossings. I think "island" is already covered
by traffic_calming=island, no?

-- 
Shawn K. Quinn 
http://www.rantroulette.com
http://www.skqrecordquest.com

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Matthew Woehlke]

On 16/09/2020 20.40, Taskar Center wrote:

crossing has been a very poor tag because it seems to be the kitchen
sink for all the questions pertaining to crossings... Many of the
attributes that get values in "crossing" are potentially overlapping
and not mutually exclusive, causing a lot of confusion and poorly
tagged crossings. Nevertheless, specifying crossings is very
important because it's a highly contested street region.

The crossing tag has held many values that may overlap, and we should
once and for all split out all these different tags so we can be
mapping what we mean and mean what we map. Questions we should be
answering when mapping a crossing: 1) How is this shared space
controlled? A crossing is a high risk environment where traversal is
shared between cars and pedestrians (they are of unequal footing). So
the type of 'control' and 'right of way' in that space is important
to specify. 'uncontrolled' is a very bad tag in this direction
because it has an actual legal, non-intuitive meaning and many
mappers mistakenly think a crossing that has no traffic signal is
uncontrolled- so that's a really bad tag. crossing_control= ?


It's also atrocious because it can *only* be verified by survey. As much 
as we prefer surveys, the reality is that a lot of mapping happens just 
from aerials, where crossings (both marked and, in some cases, unmarked) 
can be seen, but signals cannot. As someone who's generated a fair 
number of "uncontrolled" crossings because that was the only "blessed" 
tag, I would much prefer separating the presence or absence of features 
that can be verified in an aerial (marked, unmarked, striped, island, 
...) from whether or not signals are present.



3) How can a pedestrian call up the signal and how can they sense
whose right of way is currently allowed? Is there a call button? Does
it chirp, speak out,  vibrate?


Id' be careful with this one; I've read that those buttons are often 
placebos. I suppose if we're just mapping whether a button is present or 
not, that's okay, but just because there *is* a button doesn't 
necessarily mean it has to be pressed.


--
Matthew

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[António Madeira]

I do believe that uncontrolled should be deprecated in favour of marked,
which iD already did. I also agree that marked/unmarked was a good
improved in the crossing scheme, but it should be cleared on the wiki
page, which seems to favour the uncontrolled tag.
About your considerations:

1 - That depends on the country. For example, in Portugal, all crossings
have right of way over vehicles. So, marking a crossing is the same
whatever the type you map (besides unmarked, of course)

2 - I think there area already tags for all that. You can check them
here under "Additional tags":
https://wiki.openstreetmap.org/wiki/Key:crossing

3 - Same as previous point.

4 - In the same page, under "Mode of transport"

5 - There's also reference to that in the page, but I agree this is not
very clear and is scattered in several wikis, like this:
https://wiki.openstreetmap.org/wiki/Tag:footway%3Dcrossing

Regards,
António


Às 21:40 de 16/09/2020, Taskar Center escreveu:

Hi,

crossing has been a very poor tag because it seems to be the kitchen
sink for all the questions pertaining to crossings...
Many of the attributes that get values in "crossing" are potentially
overlapping and not mutually exclusive, causing a lot of confusion and
poorly tagged crossings. Nevertheless, specifying crossings is very
important because it's a highly contested street region.

The crossing tag has held many values that may overlap, and we should
once and for all split out all these different tags so we can be
mapping what we mean and mean what we map.
Questions we should be answering when mapping a crossing:
1) How is this shared space controlled? A crossing is a high risk
environment where traversal is shared between cars and pedestrians
(they are of unequal footing). So the type of 'control' and 'right of
way' in that space is important to specify. 'uncontrolled' is a very
bad tag in this direction because it has an actual legal,
non-intuitive meaning and many mappers mistakenly think a crossing
that has no traffic signal is uncontrolled- so that's a really bad tag.
crossing_control= ?

2) How is the space demarcated? A crossing may be demarcated by a
number of different ground markers, it may also be physically
demarcated from other street environments by raised footway, tactile
paving or reflectors.
crossing_ref=? (for visual demarcation)
additional tag for physical demarcation?
(I'm in disagreement with those saying it's superfluous or hard to tag
this way)

3) How can a pedestrian call up the signal and how can they sense
whose right of way is currently allowed?
Is there a call button? Does it chirp, speak out, vibrate?

4) who is sharing the way (also a bicycle crossing, animal crossing, etc)?

5) How is the space connected to the rest of the transportation layer?
to the pedestrian layer? Crossings should really only extend from curb
to curb, so that the kerb could be properly tagged for its physical
characteristics. The habit of extending crossings all the way into and
overlapping with sidewalk spaces is a pretty bad idea considering
those are protected pedestrian spaces and have very semantic meaning
to pedestrians than the high risk crossing environment.

I think crossing=marked/unmarked was a really good step in the
direction of getting resolution and refinement on at least one of
these questions above. We should now move together to refine the
definitions and values for these other questions...

Best,
Anat



Sent from my mobile. Please excuse brevity and typos.
On Wed, Sep 16, 2020 at 4:41 PM Clifford Snow mailto:cliff...@snowandsnow.us>> wrote:



On Wed, Sep 16, 2020 at 2:46 PM Graeme Fitzpatrick
mailto:graemefi...@gmail.com>> wrote:




I must admit that I only do crossings as =traffic_signals;
=marked (by itself) for zebra crossings; & =unmarked where
there is provision to cross the road but no signage or roadway
markings on any sort.


I do crossings as crossing=marked/unmarked. I believe software
should be able to identify if the crossing has a stop sign or
traffic signal. Pedestrian walk/don't walk are low on my radar
right now.

I stopped using zebra since they seemed more appropriate for a
crossing in England than where I live in the US.
Crossing=marked/unmarked the only thing I see where I map them.

BTW - I believe in the US hitting a pedestrian in a marked
crossing is illegal most everywhere. In some cities, drivers seem
to believe they have the right of way over pedestrians, even if
they are in a marked crossing.

In another country I've spent some time in, cars definitely have
the right of way over pedestrians.
--
@osm_washington
www.snowandsnow.us 
OpenStreetMap: Maps with a human touch
___
Tagging mailing list
Tagging@openstreetmap.org 
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Taskar Center]

Hi,

crossing has been a very poor tag because it seems to be the kitchen sink for 
all the questions pertaining to crossings...
Many of the attributes that get values in "crossing" are potentially 
overlapping and not mutually exclusive, causing a lot of confusion and poorly 
tagged crossings. Nevertheless, specifying crossings is very important because 
it's a highly contested street region.

The crossing tag has held many values that may overlap, and we should once and 
for all split out all these different tags so we can be mapping what we mean 
and mean what we map.
Questions we should be answering when mapping a crossing:
1) How is this shared space controlled? A crossing is a high risk environment 
where traversal is shared between cars and pedestrians (they are of unequal 
footing). So the type of 'control' and 'right of way' in that space is 
important to specify. 'uncontrolled' is a very bad tag in this direction 
because it has an actual legal, non-intuitive meaning and many mappers 
mistakenly think a crossing that has no traffic signal is uncontrolled- so 
that's a really bad tag.
crossing_control= ?

2) How is the space demarcated? A crossing may be demarcated by a number of 
different ground markers, it may also be physically demarcated from other 
street environments by raised footway, tactile paving or reflectors.
crossing_ref=? (for visual demarcation)
additional tag for physical demarcation?
(I'm in disagreement with those saying it's superfluous or hard to tag this way)

3) How can a pedestrian call up the signal and how can they sense whose right 
of way is currently allowed?
Is there a call button? Does it chirp, speak out,  vibrate?

4) who is sharing the way (also a bicycle crossing, animal crossing, etc)?

5) How is the space connected to the rest of the transportation layer? to the 
pedestrian layer? Crossings should really only extend from curb to curb, so 
that the kerb could be properly tagged for its physical characteristics. The 
habit of extending crossings all the way into and overlapping with sidewalk 
spaces is a pretty bad idea considering those are protected pedestrian spaces 
and have very semantic meaning to pedestrians than the high risk crossing 
environment.

I think crossing=marked/unmarked was a really good step in the direction of 
getting resolution and refinement on at least one of these questions above. We 
should now move together to refine the definitions and values for these other 
questions...

Best,
Anat



Sent from my mobile. Please excuse brevity and typos.
> On Wed, Sep 16, 2020 at 4:41 PM Clifford Snow  wrote:
> 
> 
>> On Wed, Sep 16, 2020 at 2:46 PM Graeme Fitzpatrick  
>> wrote:
>> 
>> 
>> 
>> I must admit that I only do crossings as =traffic_signals; =marked (by 
>> itself) for zebra crossings; & =unmarked where there is provision to cross 
>> the road but no signage or roadway markings on any sort.
> I do crossings as crossing=marked/unmarked. I believe software should be able 
> to identify if the crossing has a stop sign or traffic signal. Pedestrian 
> walk/don't walk are low on my radar right now.
> 
> I stopped using zebra since they seemed more appropriate for a crossing in 
> England than where I live in the US. Crossing=marked/unmarked the only thing 
> I see where I map them. 
> 
> BTW - I believe in the US hitting a pedestrian in a marked crossing is 
> illegal most everywhere. In some cities, drivers seem to believe they have 
> the right of way over pedestrians, even if they are in a marked crossing. 
> 
> In another country I've spent some time in, cars definitely have the right of 
> way over pedestrians. 
> -- 
> @osm_washington
> www.snowandsnow.us
> OpenStreetMap: Maps with a human touch
> ___
> 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] automated edits seem to remove crossing=zebra drastically

[Clifford Snow]

On Wed, Sep 16, 2020 at 2:46 PM Graeme Fitzpatrick 
wrote:

>
>
>
> I must admit that I only do crossings as =traffic_signals; =marked (by
> itself) for zebra crossings; & =unmarked where there is provision to cross
> the road but no signage or roadway markings on any sort.
>
>
> I do crossings as crossing=marked/unmarked. I believe software should be
able to identify if the crossing has a stop sign or traffic signal.
Pedestrian walk/don't walk are low on my radar right now.

I stopped using zebra since they seemed more appropriate for a crossing in
England than where I live in the US. Crossing=marked/unmarked the only
thing I see where I map them.

BTW - I believe in the US hitting a pedestrian in a marked crossing is
illegal most everywhere. In some cities, drivers seem to believe they have
the right of way over pedestrians, even if they are in a marked crossing.

In another country I've spent some time in, cars definitely have the right
of way over pedestrians.
-- 
@osm_washington
www.snowandsnow.us
OpenStreetMap: Maps with a human touch
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Peter Elderson]

In Nederland zebra crossings are very common, and go by the name zebra.This is 
also the name used in legislation. Zebra crossings give priority to pedestrians 
crossing the street on the zebra. Hm how should this be tagged... maybe 
crossing=pathtocrosstheroadmarkedwithstripeslikeazebratograntprioritytopedestrians?

Peter Elderson

>> Op 16 sep. 2020 om 23:47 heeft Graeme Fitzpatrick  
>> het volgende geschreven:
> 
> 
> 
>> On Wed, 16 Sep 2020 at 20:01, Martin Koppenhoefer  
>> wrote:
>> while the very generic crossing=marked, which was quite unpopular before 
>> (2013-2018 below 6000 uses) now went through the roof and is leading the 
>> tagstats with more than 1 million uses.
> 
> You may find that it is partly, at least, iD's "fault"? Crossings now "error" 
> in iD to say that "this street crosses an unmarked crossing", despite the 
> crossing being mapped & tagged? eg 
> https://www.openstreetmap.org/edit#map=19/-28.06439/153.43854
> 
> The "fix" inserts an "Unmarked Crossing" node on the junction of the street & 
> crossing.
> eg https://www.openstreetmap.org/node/7914347813
> 
>> What do you think about it, shouldn't we be encouraging people to use more 
>> specific tags like crossing=zebra or crossing=traffic_signals instead?
> 
> I must admit that I only do crossings as =traffic_signals; =marked (by 
> itself) for zebra crossings; & =unmarked where there is provision to cross 
> the road but no signage or roadway markings on any sort.
> 
> Thanks
> 
> Graeme
> 
> ___
> 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] automated edits seem to remove crossing=zebra drastically

[Graeme Fitzpatrick]

On Wed, 16 Sep 2020 at 20:01, Martin Koppenhoefer 
wrote:

> while the very generic crossing=marked, which was quite unpopular before
> (2013-2018 below 6000 uses) now went through the roof and is leading the
> tagstats with more than 1 million uses.
>

You may find that it is partly, at least, iD's "fault"? Crossings now
"error" in iD to say that "this street crosses an unmarked crossing",
despite the crossing being mapped & tagged? eg
https://www.openstreetmap.org/edit#map=19/-28.06439/153.43854

The "fix" inserts an "Unmarked Crossing" node on the junction of the street
& crossing.
eg https://www.openstreetmap.org/node/7914347813

What do you think about it, shouldn't we be encouraging people to use more
> specific tags like crossing=zebra or crossing=traffic_signals instead?
>

I must admit that I only do crossings as =traffic_signals; =marked (by
itself) for zebra crossings; & =unmarked where there is provision to cross
the road but no signage or roadway markings on any sort.

Thanks

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Grzegorz Szymaszek]

https://wiki.openstreetmap.org/wiki/ID/Controversial_Decisions#Changing_crossing.3Dzebra_to_crossing.3Dmarked


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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[António Madeira via Tagging]

The problem, I believe is with iD's presets.
When I started mapping some years ago I always marked crossings as
zebras, then iD changed the preset to crossing =marked and I believe
that's what you're seeing with the increasing number of this tag.
Although iD presents the type selector within that element, with
"uncontrolled", "traffic_signs", "unmarked", "zebra", "no", "toucan",
"pelican" and others, most mappers just leave the first value, which is
"marked".


Às 11:33 de 16/09/2020, Martin Koppenhoefer escreveu:



Am Mi., 16. Sept. 2020 um 16:27 Uhr schrieb Dave F via Tagging
mailto:tagging@openstreetmap.org>>:

I thought the correct tag for this was crossing_ref. Have you
cross checked to see if they've been swapped instead of removed?



crossing_ref is a different kind of beast, as some people use it to
tell whether there are zebra markings (can also apply to traffic light
controlled crossings).

Frankly, I do not like the tag for zebra crossings, because this
approach requires me to set 3 tags (one of crossing=zebra / marked /
uncontrolled(?)  +, crossing_ref=zebra + highway=crossing, on every
zebra crossing while I could use 2 and be done (highway=crossing with
crossing=zebra).


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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

sent from a phone

> On 16. Sep 2020, at 20:22, António Madeira  wrote:
> 
> The problem, I believe is with iD's presets.



thank you for the hint, I think you’re right.

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Mi., 16. Sept. 2020 um 16:27 Uhr schrieb Dave F via Tagging <
tagging@openstreetmap.org>:

> I thought the correct tag for this was crossing_ref. Have you cross
> checked to see if they've been swapped instead of removed?
>


crossing_ref is a different kind of beast, as some people use it to tell
whether there are zebra markings (can also apply to traffic light
controlled crossings).

Frankly, I do not like the tag for zebra crossings, because this approach
requires me to set 3 tags (one of crossing=zebra / marked /
uncontrolled(?)  +, crossing_ref=zebra + highway=crossing, on every zebra
crossing while I could use 2 and be done (highway=crossing with
crossing=zebra).


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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Dave F via Tagging]

On 16/09/2020 14:59, Jeremy Harris wrote:

On 16/09/2020 14:26, Matthew Woehlke wrote:

On 16/09/2020 05.57, Martin Koppenhoefer wrote:

I noticed that crossing=zebra tag usage is drastically shrinking while
the
very generic crossing=marked, which was quite unpopular before (2013-2018
below 6000 uses) now went through the roof and is leading the tagstats
with
more than 1 million uses. What do you think about it, shouldn't we be
encouraging people to use more specific tags like crossing=zebra or
crossing=traffic_signals instead?

My understanding is that crossing=zebra is deprecated in favor of
crossing=uncontrolled / crossing=traffic_signals. In particular, my
understanding is that they are synonymous for (almost¹) all practical
purposes. (Also, that crossing=marked is not desired either...)

Please explain how crossing=marked is "very generic" and what value
crossing=zebra adds.

In the UK, at least, there is a legal distinction: motor traffic
is required to give way to pedestrians *waiting to cross* at a
zebra crossing; this does not apply for crossings that are marked
(and have feature useful to pedestrians such as refuge islands
and dropped kerbs).


Indeed; which is why crossing_ref=zebra (which is the correct tag for 
this) should not be tagged as 'uncontrolled'. The presence of a 
pedestrian controls the motor traffic.


DaveF




I could imagine, for example, sight-challenged pedestrians wanting
to know about zebra-crossings specifically.





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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Mi., 16. Sept. 2020 um 15:26 Uhr schrieb Matthew Woehlke <
mwoehlke.fl...@gmail.com>:

> My understanding is that crossing=zebra is deprecated in favor of
> crossing=uncontrolled / crossing=traffic_signals.



there are many issues with "uncontrolled", especially if you use it to
intend a zebra crossing (as road markings are a kind of control).
crossing=zebra and traffic_signals are not synonymous.

What strikes me is the raise of "marked", which used to be a niche tag for
crossings that were somehow marked but were not describable with the other
tags, and now it is the leading value for "crossing".



>
> Please explain how crossing=marked is "very generic" and what value
> crossing=zebra adds.
>


crossing=zebra is about a zebra crossing, which is a typical kind of
pedestrian crossing in many countries (i.e. no traffic lights, zebra
markings, possibly zebra crossing signs, according to jurisdiction).
crossing=marked is about any marked crossing.



>
> Additionally, crossing=zebra is not an approved tag (according to the
> wiki),

and "It is not always clear what the intended meaning is when
> used outside of the UK". This doesn't seem like a tag we should be
> encouraging.
>


the wiki should be updated. It means a zebra crossing, and I do not believe
there is ambiguity in this, in the UK or in Europe.


> (¹ Pedantically, I suppose you could argue that crossing=zebra refers to
> a specific *form* of marking, i.e. repeated white stripes, while the
> approved crossing=uncontrolled could include crossings marked only by
> two parallel white lines. However, I would question the value added by
> mapping that distinction.)



the "approved" crossing=uncontrolled has bugged many mappers for years, and
I believe the current idea about the tag is that it should be avoided. The
word implies without markings (although a different meaning is defined).

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Dave F via Tagging]

I thought the correct tag for this was crossing_ref. Have you cross 
checked to see if they've been swapped instead of removed?


DaveF

On 16/09/2020 10:57, Martin Koppenhoefer wrote:
I noticed that crossing=zebra tag usage is drastically shrinking while 
the very generic crossing=marked, which was quite unpopular before 
(2013-2018 below 6000 uses) now went through the roof and is leading 
the tagstats with more than 1 million uses. What do you think about 
it, shouldn't we be encouraging people to use more specific tags like 
crossing=zebra or crossing=traffic_signals instead?


___
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] automated edits seem to remove crossing=zebra drastically

[Jeremy Harris]

On 16/09/2020 14:26, Matthew Woehlke wrote:
> On 16/09/2020 05.57, Martin Koppenhoefer wrote:
>> I noticed that crossing=zebra tag usage is drastically shrinking while
>> the
>> very generic crossing=marked, which was quite unpopular before (2013-2018
>> below 6000 uses) now went through the roof and is leading the tagstats
>> with
>> more than 1 million uses. What do you think about it, shouldn't we be
>> encouraging people to use more specific tags like crossing=zebra or
>> crossing=traffic_signals instead?
> 
> My understanding is that crossing=zebra is deprecated in favor of
> crossing=uncontrolled / crossing=traffic_signals. In particular, my
> understanding is that they are synonymous for (almost¹) all practical
> purposes. (Also, that crossing=marked is not desired either...)
> 
> Please explain how crossing=marked is "very generic" and what value
> crossing=zebra adds.

In the UK, at least, there is a legal distinction: motor traffic
is required to give way to pedestrians *waiting to cross* at a
zebra crossing; this does not apply for crossings that are marked
(and have feature useful to pedestrians such as refuge islands
and dropped kerbs).

I could imagine, for example, sight-challenged pedestrians wanting
to know about zebra-crossings specifically.


-- 
Cheers,
  Jeremy

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Matthew Woehlke]

On 16/09/2020 05.57, Martin Koppenhoefer wrote:

I noticed that crossing=zebra tag usage is drastically shrinking while the
very generic crossing=marked, which was quite unpopular before (2013-2018
below 6000 uses) now went through the roof and is leading the tagstats with
more than 1 million uses. What do you think about it, shouldn't we be
encouraging people to use more specific tags like crossing=zebra or
crossing=traffic_signals instead?


My understanding is that crossing=zebra is deprecated in favor of 
crossing=uncontrolled / crossing=traffic_signals. In particular, my 
understanding is that they are synonymous for (almost¹) all practical 
purposes. (Also, that crossing=marked is not desired either...)


Please explain how crossing=marked is "very generic" and what value 
crossing=zebra adds.


Additionally, crossing=zebra is not an approved tag (according to the 
wiki), and "It is not always clear what the intended meaning is when 
used outside of the UK". This doesn't seem like a tag we should be 
encouraging.


(Feel free to disagree with the above, but in that case, the correct 
solution is to a) seek approval for the tag and b) clarify the 
documentation.)


(¹ Pedantically, I suppose you could argue that crossing=zebra refers to 
a specific *form* of marking, i.e. repeated white stripes, while the 
approved crossing=uncontrolled could include crossings marked only by 
two parallel white lines. However, I would question the value added by 
mapping that distinction.)


--
Matthew

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

sent from a phone

> On 16. Sep 2020, at 14:25, Supaplex  wrote:
> 
> Do you have examples where "zebra" is changed automatically? Where and who 
> and why?


I have seen it only sporadically and have contacted the mappers in some cases, 
the same for marked when there were traffic lights. I have been writing here 
because of the bend in the tagging curve:
https://taghistory.raifer.tech/#***/crossing%20/&***/crossing/zebra

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Supaplex]

I would appreciate using crossing=zebra! (instead of crossing=marked +
crossing_ref=zebra, so I have tagged it so far.) But I can't imagine
that people use or change "marked" instead of "traffic_signals". Or have
you observed this somewhere? For me "marked" would be something like
"paved" for Key:surface: a value which could be specified more exactly
in most cases (especially using "zebra" instead).

Do you have examples where "zebra" is changed automatically? Where and
who and why?

In rare cases, however, there are still crossings with special
structural forms, where "marked" could be a useful value for greater
generalisation, e.g. in this case:
https://wiki.openstreetmap.org/wiki/File:%C3%9Cberwegmarkierung_B%C3%BCrgerstra%C3%9Fe.jpg
(because of the surface, not the markings - and it's not a traffic calming)

In Berlin we are experimenting with a few extensions, by the way, see
[de]:
https://wiki.openstreetmap.org/wiki/Berlin/Verkehrswende/Fu%C3%9Fwege#Gehweg.C3.BCberg.C3.A4nge

Alex


Am 16.09.20 um 11:57 schrieb Martin Koppenhoefer:
> I noticed that crossing=zebra tag usage is drastically shrinking while the
> very generic crossing=marked, which was quite unpopular before (2013-2018
> below 6000 uses) now went through the roof and is leading the tagstats with
> more than 1 million uses. What do you think about it, shouldn't we be
> encouraging people to use more specific tags like crossing=zebra or
> crossing=traffic_signals instead?
>
>
> ___
> 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] automated edits seem to remove crossing=zebra drastically

[ael]

On Wed, Sep 16, 2020 at 12:40:06PM +0200, Martin Koppenhoefer wrote:
> Am Mi., 16. Sept. 2020 um 12:36 Uhr schrieb ael :
> 
> > Yes. ISTR that the last time I tried to mark a crossing, zebra wasn't a
> > option in the presets. But my memeory may be at fault.
> >
> 
> which editor are you using?

I should have said josm. But I just checked and zebra was available as a
subtag. I was mapping a crossing with traffic lights so zebra was not
appropriate there. So my memory was at fault. Apologies for the noise.

ael


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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

Am Mi., 16. Sept. 2020 um 12:36 Uhr schrieb ael :

> Yes. ISTR that the last time I tried to mark a crossing, zebra wasn't a
> option in the presets. But my memeory may be at fault.
>


which editor are you using?

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

Re: [Tagging] automated edits seem to remove crossing=zebra drastically

[ael]

On Wed, Sep 16, 2020 at 11:57:58AM +0200, Martin Koppenhoefer wrote:
> I noticed that crossing=zebra tag usage is drastically shrinking while the
> very generic crossing=marked, which was quite unpopular before (2013-2018
> below 6000 uses) now went through the roof and is leading the tagstats with
> more than 1 million uses. What do you think about it, shouldn't we be
> encouraging people to use more specific tags like crossing=zebra or
> crossing=traffic_signals instead?

Yes. ISTR that the last time I tried to mark a crossing, zebra wasn't a
option in the presets. But my memeory may be at fault.

ael


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

[Tagging] automated edits seem to remove crossing=zebra drastically

[Martin Koppenhoefer]

I noticed that crossing=zebra tag usage is drastically shrinking while the
very generic crossing=marked, which was quite unpopular before (2013-2018
below 6000 uses) now went through the roof and is leading the tagstats with
more than 1 million uses. What do you think about it, shouldn't we be
encouraging people to use more specific tags like crossing=zebra or
crossing=traffic_signals instead?
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging