Exclude railway=plaform + highway=footway/pedestrian match from "Crossing highway/railway"

[mkoniecz]

What steps will reproduce the problem?

1. Encounter mapped railway=platform with highway=footway line also mapped ​https://www.openstreetmap.org/way/234077303
2. Run validator

What is the expected result?

No complaints

What happens instead?

• Crossing highway/railway (1)

Please provide any additional information below. Attach a screenshot if possible.

See https://josm.openstreetmap.de/ticket/16919 that was equivalent for bus stops

URL:https://josm.openstreetmap.de/svn/trunk
Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b
Last:Changed Date: 2018-11-25 21:25:56 +0100 (Sun, 25 Nov 2018)
Build-Date:2018-11-25 21:15:39
Revision:14454
Relative:URL: ^/trunk

Identification: JOSM/1.5 (14454 en) Linux Ubuntu 16.04.5 LTS
Memory Usage: 511 MB / 869 MB (69 MB allocated, but free)
Java version: 1.8.0_191-b12, Oracle Corporation, Java HotSpot(TM) 64-Bit Server VM
Screen: :0.0 1920x1080
Maximum Screen Size: 1920x1080
Dataset consistency test: No problems found

Plugins:
+ OpeningHoursEditor (34535)
+ buildings_tools (34724)
+ continuosDownload (82)
+ imagery_offset_db (34641)
+ measurement (34529)
+ reverter (34552)
+ todo (30306)

Last errors/warnings:
- W: No configuration settings found.  Using hardcoded default values for all pools.

[skyper]

How should a routing software route without connection? Is it too hard to connect these objects with a common node ? I see different mapping which only connects the platform with short footways.

[mkoniecz]

"Is it too hard to connect these objects with a common node ?" - more of irritating. There are some people that love to add (quite suspect) areas for tram stops and bus stops, without adding this connections.

"How should a routing software route without connection?" - in the same way as it deals with other objects unattached to footways, like tram stop nodes, shops, bus stop nodes?

[reichg]

In my opinion, for this ​scenario presented in the description of this ticket wouldn't it make sense to completely remove the line going through the area? This is redundant because if we have a railway=platform area which is more geometrically AND geographically correct than a node, the routing software would create the shortest path through the area which is darn near the line that is going through it right now with highway=footway. That means I don't think that this scenario should be excluded from the validation test.

Not only that but think about the future of mapping software with 3D virtualizations of locations. Keeping this validation rule helps scale OSM for future use, if we even care about that. :)

Juist my two cents.

[mkoniecz]

wouldn't it make sense to completely remove the line going through the area?

No, in the same way as one should not remove highway=residential after mapping area:highway=residential

highway=footway is a linear representation of footway and is needed.

I am not aware about any actually working software converting area mapping to linear representation. And anyway OSM data should stay usable without massive preprocessing.

[anonymous]

Replying to mkoniecz:

wouldn't it make sense to completely remove the line going through the area?

No, in the same way as one should not remove highway=residential after mapping area:highway=residential

highway=footway is a linear representation of footway and is needed.

I am not aware about any actually working software converting area mapping to linear representation. And anyway OSM data should stay usable without massive preprocessing.

I don’t mind either way, just want to know if I need to make a change or not.

[reichg]

any update?

[reichg]

Can we confirm actions that need to be taken on this ticket please?

[reichg]

Still feeling like this warning should not be removed? Thoughts?

[Famlam]

I'd say that highway=*s and *=platforms *should* be connected by a node

[anonymous]

Replying to Famlam:

I'd say that highway=*s and *=platforms *should* be connected by a node

Yup, that would mean no actions need to be taken on this ticket I believe.

[reichg]

lets close this ticket then?

[skyper]

Replying to reichg:

lets close this ticket then?

Maybe, voting as possible (when logged in) on top of each wiki page and ticket can help to make a decision.

[reichg]

Maybe we can try to get this closed by the next milestone?