Okay, I looked for it as I drove by, but couldn't see anything but bush there. It's odd that nothing seems to be visible there on aerial imagery either, but maybe that's due to shadows. There was definitely a driveway a few hundred feet farther north. If you have local knowledge that it exists here, then I'm fine with it staying.

I removed most of the cases where you added "Vancouver Island Trail" or another name to a way where that certainly or almost-certainly is not the name. This included some remote logging roads where they likely have some other name, and are not dedicated solely to the trail. Route relations are a better way to handle multi-way trails like this, and the VI Trail relation already covers it. If you'd like the trail to be more visible in a specific application, it should be up to the maker of that application to provide a way to view the route, not for us to add artificial names where they aren't accurate.

I can't find any evidence that the trail's name has changed. The "Humpback Connector" name was applied to the entire Victoria to Langford trail in 2011 (https://www.vicnews.com/news/regional-trail-gets-news-name-12860), and I checked and confirmed that the signage at the newest section (Esquimalt to Catherine) still uses the "E&N Rail Trail - Humpback Connector" name as of yesterday (https://i.imgur.com/4fhtLBE.jpeg at the Mary Street crossing). I've restored the name of the trail.

Please don't add marketing text to the object. I've removed that again, and updated a couple of the other tags to better match the typical OSM usage.

This needs some re-working. Some of these are marking ridges, so they probably shouldn't be tagged as peaks at all, let alone a single fixed elevation value. One is a blatant mis-tagging of a tree near Port Alberni.

It's tagged as a park, so I would suggest changing the tagging to natural=tree instead. Also, unless this tree's name is actually "Wildlife Tree", the name tag should be removed.

I'm sorry for having to do this. Unfortunately, it would take a bunch of work to map all these loops again, so reverting is the easiest route. I see you've been doing a lot of great edits around Victoria, and the vast majority have been high-quality with nothing that I would have done differently. If I get a chance, I'll go through these changesets I had to revert and re-map the other changes you did, like aligning roads.

This changeset and a few after it removed a lot of data by deleting mapped loop ways and replacing them with less-detailed nodes tagged with highway=turning_loop. The node method is considered a first-step way of indicating that there's a turning loop, whereas a fully-mapped loop way is considered more complete. Since these changesets removed so much data, I was forced to revert them in their entirety. I re-mapped many of the other changes that were made in this changeset, but I haven't re-mapped the changes from the others. Many of the changes north of roughly Kenmore/San Juan will need to be made again.

The circumference and height tags are assumed to be in meters unless otherwise specified. It's been a bit since I've looked at these trees, so 17 and 20 metres tall could be correct (they ARE sequoias, after all!). However, 9 or 10 metres around the trunk seems very high. This measurement might need to be double-checked.

I've reverted both this and the subsequent changeset that removed relations for no apparent reason. I then spent 30 seconds to resolve the two trivial issues with them.

It isn't your job to resolve issues like this around the world. If you don't know how to properly troubleshoot and fix a relation, or don't have sufficient local knowledge to do so, it's fine to leave it as-is and allow a local or a more experienced contributor to fix the issue.

This path isn't primarily a cycleway, but is rather a shared-use pathway with equal access for both bikes and pedestrians. highway=path combined with the designated users better describes it, so I've restored these to the previous classification.

The newest pathway that parallels Dallas Road is not primarily a cycleway, but rather a shared-use pathway with designated use by both bikes and foot traffic. I've changed this back to highway=path, as well as the ones around the Johnson Street Bridge which are the same situation.

I also updated a couple of historic=monument to historic=memorial. Monuments are defined in OSM as being large structures that you can go inside, like the Lincoln Memorial or the Taj Mahal. These cairns and plaques are obviously much smaller than that.

I'm curious where you got the name "Dallas Road Waterfront Trail". You're the second contributor recently who has added this name (and there are now two of them in OSM). I haven't seen any signs or city documents referring to either of these trails by this name. The only places I have seen it are on Google Maps, or secondary sources (e.g. Alltrails, community-generated maps, etc.), that likely pulled the name from Google Maps. We can leave the name in place if people generally know both these parallel trails by that name, but we should remove it if the only reason it's there is because it was on Google Maps.

Only the northeastern part of this area is Promenade Park. The southwestern portion is a separate, undeveloped lot with the address 499 Royal Bay Drive. I've restored the park to its correct size.

There isn't a book shop here. I had already updated the previous one you added to the correct tag of amenity=public_bookcase. I've removed this new "shop".

The StatsCan dataset does not contain a road by this name here, but Google Maps does, so I suspect that's where this data came from.

The StatsCan dataset does not contain a road by this name here, but Google Maps does, so I suspect that's where this data came from.

The StatsCan dataset does not contain roads by these names here, but Google Maps does, so I suspect that's where this data came from.

The StatsCan dataset does not contain a road by this name here, but Google Maps does, so I suspect that's where this data came from.

The StatsCan dataset does not contain a road by this name here, but Google Maps does, so I suspect that's where this data came from.

The StatsCan dataset does not contain roads by these names in these locations, but Google Maps does, so I suspect that's where this data came from.