Potentially dangerous nodes and ideas for a fix

[mdomsy (Matthew Domsy)]

This concerns Sir John A. Macdonald Parkway in Ottawa, Ontario, Canada, also (Old) Ottawa and similar style “streets” in general. If any of the below comments are an OSM issue please let me know.

1. Do all parkways and similar style roads in City Strides have two sets of nodes, a set for each traffic direction? For the parkway noted above, there is a set for both eastbound and westbound despite being a single “street”. I think this might be an OSM thing based on the pinned topics for this section. Making a single node from the two would likely have to happen in OSM I would think. Not sure that is the best option.

2. The parkway noted does not have pedestrian walkways/pathways immediately adjacent however multi-use pathways do exist, in some locations, 10 or more metres away. In many sections it is unsafe due to vehicle traffic. Would it be a major undertaking to increase the allowable range of GPS coordinates for those nodes?

3. The city closes the westbound traffic lanes (north side) during daylight hours at least one weekend day per week throughout the summer (summer daylight hours all weekend during COVID-19 pandemic). Accessing the parkway at that time is safe. Would it be possible to couple the nodes along the eastbound and westbound such that when a node on one side of the parkway is completed the nearest node on the other side of the parkway is also completed?

4. I suppose another option for future users would be to run the adjacent pathway, however far off, and manually mark the nodes as complete.

This will not affect me in anyway as I have completed all the nodes on that parkway. If other users in Ottawa, Canada or other users with similar circumstances in their city have an opinion on this I would like to hear them.

@JamesChevalier , what do you think?

[m_l_heywood (Mike Heywood)]

Sounds to me that if the parkway isn’t safe to run on foot then the foot=yes tag in OSM is incorrect. If this was changed then it would no longer pull through to CS. I’ve had a similar thing on some roads near me where in OSM they were tagged as Not Specified which meant that CS ignored them but other mapping software kept trying to send me along them

The Trouble with Trunk Roads

Its a difficult one this. I had the opposite issue recently where a mapping site I was using was insisting on using a dual carriageway as is was foot=Not Specified so I had to switch this road to foot=no to get sensible route suggestions. Therefore, simply switching on trunk roads may give too many false positives. Would the solution therfore be to allow trunk roads but exclude foot=not specified? Would need to do the work suggested earlier in the thread to update OSM to be foot=yes but that fe…

[hans1 (Hans Westerback)]

Seems it should be foot=no in OSM, but maybe when one direction is at timrs accessible on foot, it could be foot=permissive in that direction, and the other foot=no

[mdomsy (Matthew Domsy)]

Thanks for the responses. Leads to me to think that this is more an issue with OSM designations. My experience with mapping foot/bicycle routes through this area has always been on the multi-use pathways and never suggests the roadway. Based on the other examples provided and the consistency of opinion between the two of you I think it is reasonable to set foot=no in OSM.

I am new to OSM edits and found that the parkway is a “Road Route” made up of multiple sections. Looking at the wiki, I have added a tag to the Road Route, highway=motorway based on the Default routing tags for Access Permissions and another tag bicycle=permissive for the more adventurous souls.

Thanks again for the input and happy striding!

[Marty]

@mdomsy

Do all parkways and similar style roads in City Strides have two sets of nodes, a set for each traffic direction? For the parkway noted above, there is a set for both eastbound and westbound despite being a single “street”. I think this might be an OSM thing based on the pinned topics for this section. Making a single node from the two would likely have to happen in OSM I would think. Not sure that is the best option.

This is common for divided roads/streets. The purpose of nodes within OSM is to mark where the path of the road is. So if each direction is split with a median in between, it makes sense to have separate nodes for each side of road; even though it has a single street name. Don’t undo that in OSM just for CS!

For CS purposes it could potentially mean you have to run each side to pick up enough nodes. In my experience, if the median is safe, I’ll run in the median to try and pick up nodes on each side at same time.

Looking specifically at Sir John A. Macdonald Parkway on Google Maps, it does look like there is a nice, wide grass median. However as the others have mentioned, this road does look like it should have foot=no for most if not all of its sections. It’s primarily designed for car usage; I wouldn’t even ride my bike on it. Especially since there is a foot/cycle path next to it.

[mdomsy (Matthew Domsy)]

I agree Marty and have not modified the nodes in such a way. Unfortunately this parkway does not have a consistent and safe median and found myself in some miserable conditions that I don’t think others should have to submit to.

I have submitted an OSM edit for altering the Access Permissions to set highway=motorway and bicycle=permissive which marks the parkway as foot=no and some other mobility types to no as well. I have asked that the edit be reviewed as I think it should be a community decision since OSM is community based.

I appreciate your input!