Should pedestrian directions allow following expressway=yes ways? #7494
-
|
Hello, Recently I was trying to help with a concern about pedestrian access to a highway reported at: https://www.openstreetmap.org/user/Jennifer%20Pownall/diary/408540 I misunderstood the problem, but noticed something else while investigating. With the FOSSGIS OSRM (v5.27.1) routing used by the openstreetmap.org website, I found a case where pedestrian directions follows the BC7 (Lougheed Highway) along a section with expressway=yes tagged: In this example: there doesn't seem to be much pedestrian pathing available in the area as a preferable way to cross the highway. I've created a nearby note to check with local mappers about that. My question is: should By comparison, osrm-backend/profiles/foot.lua Lines 78 to 82 in a27573f Thanks, |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 2 replies
-
|
Thanks for flagging this. After consulting with the wiki, express ways shouldn't be accessible to pedestrians. Will check if we have a test for the pedestrian profile to ensure this. |
Beta Was this translation helpful? Give feedback.
Ok, updating my previous assessment. It seems
expressway=yesis descriptive on prescriptive. It describes a road's physical characteristics. It does not automatically mean "no pedestrians".The right fix is upstream in the data. The existing routing logic does the right thing: explicit
foot=nooraccess=noon the way blocks it.Adding
expressway=yesto avoid would be reaching beyond the tag's defined meaning.