Hallo zusammen
Ich beziehe mich auf eine frühere Diskussion um Fusswege auf Plätzen und anderen bebauten Flächen, http://lists.openstreetmap.ch/ pipermail/talk-ch/2016-October/003805.html.
Wie in den Mails von 2016 schon erwähnt, sind Fusswege wie < https://www.openstreetmap.org/way/68187643%3E keine wirklichen Wege, sondern dienen wohl in erster Linien einem besseren Routing. Falls aber - wie am Bahnhof Bern oder anderen Orten - viele dieser Wege erfasst sind, wird die Karte aus meiner Sicht unübersichtlich. Die Wege lassen sich attributiv nicht von anderen Wegen unterscheiden.
Daher meine Frage: Ist es sinnvoll, sich hier auf einen tag zu einigen, der diese Wege explizit bezeichnet? Kandidaten könnten sein:
* highway:virtual=pedestrian (siehe http://wiki.openstreetmap.org/ wiki/Proposed_features/virtual_highway und http://wiki.openstreetmap.org/ wiki/Talk:Proposed_features/virtual_highway#Easier_scheme) * ignore_for_mapping (ähnlich zu http://wiki.openstreetmap.org/ wiki/Key%3Aignore_for_routing)
Gibt es andere Ideen? Oder gibt es Gründe, solch einen tag nicht einzuführen?
Danke und Gruss Stephan
Bonjour,
Le 05. 10. 17 à 10:17, com@heuel.org a écrit :
Wie in den Mails von 2016 schon erwähnt, sind Fusswege wie https://www.openstreetmap.org/way/68187643 keine wirklichen Wege, sondern dienen wohl in erster Linien einem besseren Routing. Falls aber
- wie am Bahnhof Bern oder anderen Orten - viele dieser Wege erfasst
sind, wird die Karte aus meiner Sicht unübersichtlich. Die Wege lassen sich attributiv nicht von anderen Wegen unterscheiden.
Daher meine Frage: Ist es sinnvoll, sich hier auf einen tag zu einigen, der diese Wege explizit bezeichnet? Kandidaten könnten sein:
- highway:virtual=pedestrian (siehe
http://wiki.openstreetmap.org/wiki/Proposed_features/virtual_highway und http://wiki.openstreetmap.org/wiki/Talk:Proposed_features/virtual_highway#Ea...
- ignore_for_mapping (ähnlich zu
Il y a aussi la possibilité de rendre routable les area highway=pedestrian https://www.openstreetmap.org/relation/5117701 https://wiki.openstreetmap.org/wiki/Key:area#Highway_areas ca c'est cela le fond du problème. les chemins virtuels nécessiterait de faire un chemin entre tous les points de la place
Cordialement, Marc
Hi
2017-10-06 18:50 GMT+02:00 marc marc marc_marc_irc@hotmail.com:
Il y a aussi la possibilité de rendre routable les area highway=pedestrian https://www.openstreetmap.org/relation/5117701
Pedestrian relations like this are not optimal, because (1) they don't work for routing - only closed ways with area=yes do partially (some applications route around the edges, others don't at all) - and (2) even if the the routing would work for pedestrian relations, they imply that only pedestrians are allowed on that area and that they can move freely, which is not true, because there are also roads.
It might be better to replace this relation with a closed way mapped on the actual pedestrian-only area, like [this](https://www.openstreetmap.org/way/530632850). Thus, some virtual footway could be removed.
Besides, I think [this area](https://www.openstreetmap.org/#map=19/46.94739/7.44069) with all those virtual crossing footways could need a simplification.
On 06.10.2017 21:37, Selfish Seahorse wrote:
and (2) even if the the routing would work for pedestrian relations, they imply that only pedestrians are allowed on that area
Is that so, even if the roads are also mapped?
Currently, Mapzen and GraphHopper car routers on osm.org navigate along the street right through the area of highway=pedestrian multipolygon https://www.openstreetmap.org/relation/5117701 :
https://www.openstreetmap.org/directions?engine=graphhopper_car&route=46...
https://www.openstreetmap.org/directions?engine=mapzen_car&route=46.9479...
(OSRM currently seems broken on osm.org unrelated to the Bern location. It works on http://routing.osm.ch/?z=18%C2%A2er=46.949028%2C7.440560&loc=46.947750%2... , though and also routes right through that area.)
Cheers, Raphael
On 6 October 2017 at 22:23, Raphael Das Gupta (das-g) lists.openstreetmap.ch@raphael.dasgupta.ch wrote:
Is that so, even if the roads are also mapped?
Currently, Mapzen and GraphHopper car routers on osm.org navigate along the street right through the area of highway=pedestrian multipolygon
Sorry, my message wasn't clear. What I meant was that pedestrian relations don't work for *pedestrian routing* and that pedestrian areas imply that pedestrians can move freely on that area, which is not true, because there are also roads (this is rather a logical than a rendering/routing issue).
As an example, GraphHopper and Maps.me route pedestrians through Münsterhof in Zürich (only along the edges though) [1] but not through Lindenplatz [2]. Münsterhof is tagged as a closed way, Lindenplatz as a multipolygon. Mapzen isn't able to route pedestrians through both squares. [3][4]
1. http://www.openstreetmap.org/directions?engine=graphhopper_foot&route=47... 2. http://www.openstreetmap.org/directions?engine=graphhopper_foot&route=47... 3. http://www.openstreetmap.org/directions?engine=mapzen_foot&route=47.3703... 4. http://www.openstreetmap.org/directions?engine=mapzen_foot&route=47.3873...
IMHO there should be no tagging for the routing enginges, very much like there should be no tagging for the renderer. In fact, I did found almost no tags related to "virtual" paths in taginfo. I'm confident that routing engines with pedestrian profiles will have optimized routing over areas/plazas in urban areas rather sooner than later (see "area/plaza routing for pedestrian/bike"). One of challenges routing engines currently face, is that until now their preprocessing did not include areas (polygons, multipolygon relations), since they concentrated on linear geometries.
:Stefan
P.S. BTW It's also an issue when the pedestrian route starts or ends in areas.
2017-10-06 23:32 GMT+02:00 Selfish Seahorse selfishseahorse@gmail.com:
On 6 October 2017 at 22:23, Raphael Das Gupta (das-g) lists.openstreetmap.ch@raphael.dasgupta.ch wrote:
Is that so, even if the roads are also mapped?
Currently, Mapzen and GraphHopper car routers on osm.org navigate along the street right through the area of highway=pedestrian multipolygon
Sorry, my message wasn't clear. What I meant was that pedestrian relations don't work for *pedestrian routing* and that pedestrian areas imply that pedestrians can move freely on that area, which is not true, because there are also roads (this is rather a logical than a rendering/routing issue).
As an example, GraphHopper and Maps.me route pedestrians through Münsterhof in Zürich (only along the edges though) [1] but not through Lindenplatz [2]. Münsterhof is tagged as a closed way, Lindenplatz as a multipolygon. Mapzen isn't able to route pedestrians through both squares. [3][4]
- http://www.openstreetmap.org/directions?engine=graphhopper_foot&route=47...
- http://www.openstreetmap.org/directions?engine=graphhopper_foot&route=47...
- http://www.openstreetmap.org/directions?engine=mapzen_foot&route=47.3703...
- http://www.openstreetmap.org/directions?engine=mapzen_foot&route=47.3873...
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
Thank you for all the feedback so far.
I think that tagging "logical" footpaths as being logical is _not_ "tagging for the <your use case>". If I interpret [1] correctly, this phrase refers to the mis-use of tags. The mis-used tag has no semantic meaning for the original object. Example: Using `natural=wood` to obtain a green area, although the object has nothing to do with wood.
On wiki pages like [2], paths for pedestrians always refer to a physical linear structure. I haven't found a documentation on how to deal with paths which do not refer to a physical linear structure. The information whether a footway is "real" path or a "virtual" resp. "logical" path, is a semantic property of the mapped object. I don't want to delete those paths, as it seems that too many people rely on them. But I still think there needs to be a method to identify such routes.
As mentioned in an earlier post, one way to identify "logical" or "virtual" paths is with a relation to an area tagged `highway=pedestrian`. To me, this is a bit complicated and may cause unintended side effects when editing any object in the relation.
So, as far as I can tell, there are three options:
1. do nothing 2. assign a special tag to a "virtual" footpath 3. identify a "virtual path" by relation to an area tagged `highway=pedestrian`.
Any thoughts on these options?
Thanks Stephan
[1]: http://wiki.openstreetmap.org/wiki/Tagging_for_the_renderer [2]: http://wiki.openstreetmap.org/wiki/Tag:highway%3Dfootway
2017-10-08 15:14 GMT+02:00 Stefan Keller sfkeller@gmail.com:
IMHO there should be no tagging for the routing enginges, very much like there should be no tagging for the renderer. In fact, I did found almost no tags related to "virtual" paths in taginfo. I'm confident that routing engines with pedestrian profiles will have optimized routing over areas/plazas in urban areas rather sooner than later (see "area/plaza routing for pedestrian/bike"). One of challenges routing engines currently face, is that until now their preprocessing did not include areas (polygons, multipolygon relations), since they concentrated on linear geometries.
:Stefan
P.S. BTW It's also an issue when the pedestrian route starts or ends in
areas.
2017-10-06 23:32 GMT+02:00 Selfish Seahorse selfishseahorse@gmail.com:
On 6 October 2017 at 22:23, Raphael Das Gupta (das-g) lists.openstreetmap.ch@raphael.dasgupta.ch wrote:
Is that so, even if the roads are also mapped?
Currently, Mapzen and GraphHopper car routers on osm.org navigate along the street right through the area of highway=pedestrian multipolygon
Sorry, my message wasn't clear. What I meant was that pedestrian relations don't work for *pedestrian routing* and that pedestrian areas imply that pedestrians can move freely on that area, which is not true, because there are also roads (this is rather a logical than a rendering/routing issue).
As an example, GraphHopper and Maps.me route pedestrians through Münsterhof in Zürich (only along the edges though) [1] but not through Lindenplatz [2]. Münsterhof is tagged as a closed way, Lindenplatz as a multipolygon. Mapzen isn't able to route pedestrians through both squares. [3][4]
http://www.openstreetmap.org/directions?engine=graphhopper_foot&route=47...
http://www.openstreetmap.org/directions?engine=graphhopper_foot&route=47...
http://www.openstreetmap.org/directions?engine=mapzen_foot&route=47.3703...
http://www.openstreetmap.org/directions?engine=mapzen_foot&route=47.3873...
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
OSM relies verifiability. That means everything can be proven true or false on the ground. (http://wiki.openstreetmap.org/wiki/Verifiability). IMHO mapping logical or virtual ways contradict this basis concept. Such a way would lack of observable features and because of this it would be always somewhat arbitrary.
I assume, these ways have been mapped to overcome routing deficiencies. In such a case, I see it as <mapping for the router>. (The router did not produce a nice route over an area, therefore someone added ways to make it happen).
My preferred option is: Do not introduce any tagging without observable features. Let the dev team of the router enhance there engine to resolve its short comings (Maybe you want to push them a little).
Am 10.10.2017 um 09:32 schrieb com@heuel.org:
Thank you for all the feedback so far.
I think that tagging "logical" footpaths as being logical is _not_ "tagging for the <your use case>". If I interpret [1] correctly, this phrase refers to the mis-use of tags. The mis-used tag has no semantic meaning for the original object. Example: Using `natural=wood` to obtain a green area, although the object has nothing to do with wood.
On wiki pages like [2], paths for pedestrians always refer to a physical linear structure. I haven't found a documentation on how to deal with paths which do not refer to a physical linear structure. The information whether a footway is "real" path or a "virtual" resp. "logical" path, is a semantic property of the mapped object. I don't want to delete those paths, as it seems that too many people rely on them. But I still think there needs to be a method to identify such routes.
As mentioned in an earlier post, one way to identify "logical" or "virtual" paths is with a relation to an area tagged `highway=pedestrian`. To me, this is a bit complicated and may cause unintended side effects when editing any object in the relation.
So, as far as I can tell, there are three options:
- do nothing
- assign a special tag to a "virtual" footpath
- identify a "virtual path" by relation to an area tagged
`highway=pedestrian`.
Any thoughts on these options?
Thanks Stephan
2017-10-12 7:50 GMT+02:00 Nzara nzaraosm@gmail.com:
OSM relies verifiability. That means everything can be proven true or false on the ground. (http://wiki.openstreetmap.org/wiki/Verifiability). IMHO mapping logical or virtual ways contradict this basis concept. Such a way would lack of observable features and because of this it would be always somewhat arbitrary.
I assume, these ways have been mapped to overcome routing deficiencies. In such a case, I see it as <mapping for the router>. (The router did not produce a nice route over an area, therefore someone added ways to make it happen).
My preferred option is: Do not introduce any tagging without observable features. Let the dev team of the router enhance there engine to resolve its short comings (Maybe you want to push them a little).
Verifiability is a fair point and actually an argument for deleting these features now. But these ways already exist quite often. If you don't want to delete them, then you could argue that the new tag would actually be the only verifiable information for this object (because I can observe on the ground that a specific way does not correspond to an actual path).
Meanwhile, we have the problem that the virtual ways clutter the map renderer. I agree that therefor the new tag could be seen as <mapping for the renderer>. But I still would propose to introduce a new tag as long as these logical ways need to exist - the added information is verifiable. And regarding the timeframre: There are more than one router engine and probably some will rely on these ways in the foreseeable future.
Thanks again for all the input Stephan
Am 12.10.2017 um 07:50 schrieb Nzara:
OSM relies verifiability. That means everything can be proven true or false on the ground. (http://wiki.openstreetmap.org/wiki/Verifiability). IMHO mapping logical or virtual ways contradict this basis concept. Such a way would lack of observable features and because of this it would be always somewhat arbitrary.
I assume that physical guidelines printed on the ground for visual impaired people could be considered as verifiable as well
http://www.basel-unterwegs.ch/de/360/2017/7_2017/die-stadt-mit-vier-sinnen-e...
http://www.sbv-fsa.ch/de/leitlinien
cheeers, h.
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
Am 12.10.2017 um 14:49 schrieb Andreas Bürki:
I assume that physical guidelines printed on the ground for visual impaired people could be considered as verifiable as well
Of course, I remember http://wiki.openstreetmap.org/wiki/Key:tactile_paving . Other tags maybe available too.
Hi,
Tactile paving is of course a reason to map and it seems to be used also in OSM.
"Virtual" as an key or value seems not in use yet at all: * http://taginfo.openstreetmap.orgtags/highway=virtual * http://taginfo.openstreetmap.org/tags/highway:virtual=pedestrian
Unfortunately all more well known routers currently don't support any virtual attribute nor pedestrian/bike area routing. At least they are aware of that: * Valhalla: https://github.com/valhalla/valhalla/issues/42 (nice comment uf Züritütsch). * OSRM: https://github.com/Project-OSRM/osrm-backend/issues/64 * Graphhopper: https://github.com/graphhopper/graphhopper/issues/82
So even if we would add highway=virtual, the routers would not recognize that. In fact, some routers discourage (like me) to add e.g. highway=footway (nor highway=virtual) by saying "do not map for the router!" (see OSRM issue 64). I'm unhappy to see non-existing ways in OSM, and would be very reluctant to add them.
:Stefan
2017-10-12 19:36 GMT+02:00 Nzara nzaraosm@gmail.com:
Am 12.10.2017 um 14:49 schrieb Andreas Bürki:
I assume that physical guidelines printed on the ground for visual impaired people could be considered as verifiable as well
Of course, I remember http://wiki.openstreetmap.org/wiki/Key:tactile_paving . Other tags maybe available too.
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
2017-10-05 10:17 GMT+02:00 com@heuel.org:
Wie in den Mails von 2016 schon erwähnt, sind Fusswege wie https://www.openstreetmap.org/way/68187643 keine wirklichen Wege, sondern dienen wohl in erster Linien einem besseren Routing. Falls aber - wie am Bahnhof Bern oder anderen Orten - viele dieser Wege erfasst sind, wird die Karte aus meiner Sicht unübersichtlich. Die Wege lassen sich attributiv nicht von anderen Wegen unterscheiden.
By the way, this is the same reason I prefer not to map sidewalks as separate ways: in residential areas without marked pedestrian crossings, you have to map virtual footway=crossing to not break routing, which results in a confusing map.
Hoi Stephan
Gute Frage. Fusswege sollte man m.E. nicht für den Router taggen - so, wie die Regel gilt, dass man nicht für den Renderer taggen soll. Tatsächlich gibt es keine "virtual" Tags in Gebrauch, wenn ich die aktuelle Taginfo-Statistik anschaue. Ich bin zuversichtlich, dass Routing Engines mit Fussgänger-Profil bald auch über Flächen navigieren können (siehe "area/plaza routing for pedestrian/bike). Auch das Fussgänger-Routing in Flächen starten/enden zu können ist ein Problem. Ich betreue aktuell selber eine Semesterarbeit zu diesem Thema u.a. in Zusammenarbeit mit search.ch.
:Stefan
Am 6. Oktober 2017 um 21:52 schrieb Selfish Seahorse selfishseahorse@gmail.com:
2017-10-05 10:17 GMT+02:00 com@heuel.org:
Wie in den Mails von 2016 schon erwähnt, sind Fusswege wie https://www.openstreetmap.org/way/68187643 keine wirklichen Wege, sondern dienen wohl in erster Linien einem besseren Routing. Falls aber - wie am Bahnhof Bern oder anderen Orten - viele dieser Wege erfasst sind, wird die Karte aus meiner Sicht unübersichtlich. Die Wege lassen sich attributiv nicht von anderen Wegen unterscheiden.
By the way, this is the same reason I prefer not to map sidewalks as separate ways: in residential areas without marked pedestrian crossings, you have to map virtual footway=crossing to not break routing, which results in a confusing map. _______________________________________________ talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
-
Andreas Bürki -
com＠heuel.org -
marc marc -
Nzara -
Raphael Das Gupta (das-g) -
Selfish Seahorse -
Stefan Keller