I'm not sure if you've tagged Hummingbird Place as well as it could be. I've not visited this facility, but my understanding is that this is a supported housing facility, built using the modular housing technique. I do not think it is right to call it a group home, or to say it is for homeless people. If it is housing, not a shelter, then it is for formally homeless people, and they now have housing. I'd like to remove the social facility tags, as I did with Sanford Apartments, a facility I'm more familiar with.

I just double checked, and I was correct with the southern end of the pedestrian section correct.

Are these traces based on the Esri Imagery, as the changeset tags state? It contains many details that are not visible in the imagery, so it seems that there is more going on here.

I wanted to let you know that I've retagged the obelisks you've added. The way you tagged them seems to be meant for very large monuments. I think the way I've tagged them in a way that makes more appropriate.

By the way, have you seen this article? https://scoutmagazine.ca/2017/01/04/you-should-know-the-explanation-behind-the-odd-granite-obelisks-of-mount-pleasant/

An interesting piece about these markers.

This edit adds the last of the sidewalk traces to Vancouver proper, except for a few odds and ends that I missed.

Hi Andrea, I wanted to let you know that I made some changes to how you mapped Sanford Apartments, and the Resource Centre. I work in the field, and have been inside both facilities, and have some knowledge about them. Here is a link to my changeset, let me know if you'd like more info about my edit. changeset/104907827#map=19/49.26562/-123.14131

I've deleted some of the maxspeed tags as I've not received a response.

It's not just common practice for buildings with one address, it's also neater, and more structurally logical.

Thanks for the work on sidewalks! Looks good. I was wondering if you have looked at the Esri imagery? It seems quite a bit better than the Bing Imagery you've been using. You can find it in the background settings in iD.

This changeset creates a new relation (relation/11075507/history#map=11/49.1175/-122.6994&layers=Y), ostensibly for a bike route, but in reality containing many different routes, and uses the `name` tag "Local Surrey Bike routes". Can you provide any evidence that this is the official name, and explain why so many routes are placed in the same relation?

That makes sense, thanks for the answer Ken.

Hi DunbarLoop, I see this changeset adds two info boards (node/7527083855 and node/7527083854) one with the `name` tag in English and one in Japanese. Would it make sense to use a single node with the `name:jp` tag for the Japanese name?

The source of this changeset is just what the changeset source tag states, satellite imagery and Strava heatmap. It was not based on a field survey.

The addresses in this changeset include many buildings tagged with the same addresses. I've fixed a few of these with this edit: changeset/96215843

Thank you Aglis, it looks great!

Good call changing Zero Bridge to `highway=pedestrian`, it better corresponds with reality than `highway=footway`.

I often see sidewalks tagged `highway=cycleway`, by editors that seem to value mapping features of interest to cyclists. I think it is unhelpful to map bicycle infrastructure as better than reality, tagging the sidewalks of the Knight Street Bridge as cycleways is an example. The sidewalks seem to be the same age as the bridge, and give little consideration to cyclists, as they are not wide enough to pass a pedestrian. The signage directing cyclists to use the sidewalks does not change the fact that they are sidewalks. The exception are the sections of actual cycleways that lead northbound cyclists down to SE Marine Dr.

It is not useful to map them all the same, as much as I'd like there to be good cycleways crossing the Knight St Bridge.

I checked the addresses on two buildings, and found the addresses added by this changeset were wrong:

way/324172370
way/324172363

The buildings had two and three addresses in reality, but a single one from this changeset. As with most (all?) changesets from this user, there is no source tag.

I should have tagged the source of this changeset as "survey", as I checked the addresses in person, and made notes.

I see what you mean about `highway=cycleway` and `highway=footway` prioritizing one mode of transport over another, but I don't think it's quite accurate. It seems to me that ways on OSM default to the largest vehicle that can use it. A road that can be accessed by foot or by motor-vehicle is tagged as a road of some sort, even if it is used more by foot (of course if a separate sidewalk exists then we can map it separately, but that's another senario).

From what you describe (as I'm afraid I've apparently forgotten that section of the bike path) I think we should tag it as `highway=cycleway`, `foot=yes`, and `segregated`=*. This has been the approach that I've seen taken for multi-use paths elsewhere in the Lower Mainland.

Does that make seem reasonable to you?