Hello, your recent four changesets broke a number of relations, particularly 1996621, 9349818, 13111721, 13111489, 13111490, proper relations were deleted, for example, 6813178, and you opted to delete border segments that were named differently, 1227045532, 1227045533, 1227045534, to name a few. I'm sorry, but I had to undo these changesets.

"Most routing software will span a gap of a few metres for routing purposes", SK53 writes. Would not parking=street_side be the most suitable for these parking spots?

The newly mapped roads disconnected from the road system, are they really residential?

Both, multypoligon 14500792 and way/312363760, which was an outer member of the multypoligon were tagged as building=yes. This "building inside building" mapping did not look appealing to me, so I opted to delete the multypoligon and keep the way for the outline of the center. Now inside the outline, there are two ways tagged building:part. One of them, 1088985008 you have created and the smaller one, 1238723239 I added so that these two parts cover the entire outline. Such mapping is described at osm.wiki/Simple_3D_Buildings The structure aside, another messy part that is in the news is that Cole's center is probably out of business by now.

Please take a look at building:part key. Here way/1236167006 describes the outline of the building and way/473202641 and way/1220095055 reflect parts of the building that are different in height.

If a footway crossing a building, the editor that I use, JOSM, gives a warning, not an error. I have now split the footway into parts and the part under the roof is now tagged with covered=yes.

Still, one thing is whether a road is private or not, and the other is whether access to the road is private or not. The following document says "Since the coming into force of section 70 of the Municipal Powers Act, a local municipality may maintain a private road open to the public by permission of the owner or occupant, on a request by a majority of owners or occupants of the abutting property" https://www.chelsea.ca/application/files/7315/5492/7544/politiquedentretiendescheminsprivesrevisee08-08-2017resolution266-17.pdf

Why would access to this road be private? I counted 4 houses on this road, so it is definitely not a driveway. I don't see "private" signs or a gate on available images.

First of all, thank you for your contributions to OpenStreetMap, a joint-effort project. Second, please don’t assume that someone was correcting the errors pointed out by an OSM quality assurance tool (in this case, OSM Inspector) ”blindly”. I was not. Third, the closed ways that I corrected had repeated segments. I did notice that on the OSM map they were rendered as desired, but it does not mean that all other applications that use OSM data would be able to process them. OSM Inspector did not validate them as proper. Therefore, I chose to improve them by removing the repeated segments and adding inner rings. Fourth, I did create multipolygons (in this changeset, relation/10717394 that consists of one outer and three inner rings) fully in line with osm.wiki/Relation:multipolygon I’m not sure I understand the specific issues you are facing with them.

Hi RobNicholson. Let me share my experience with making small corrections to trails. I was also recording my trips as I was taking them and then I was adjusting the trails accordingly. Sometimes I was taking the same trail twice and then did the adjustments based on their average. This all ended when I discovered high-resolution heatmaps on Strava. For popular trails like the one around Pink Lake at Strava there are typically recordings from twenty or more people. Averaging across such a relatively large number of tracks makes a lot of sense as it reduces the individual GPS position errors. All I am suggesting is to give Strava a try for a better mapping. Low-resolution Strava heatmaps are available without registration. High-resolution Strava heatmaps require a Strava login.

Hello! Please note that Bing aerial imagery is outdated for this area. Maxar Premium Imagery is more recent. Based on it I put back building=yes for Strøm spa.

Well, there are two things here. One thing is whether this road is known as Bacon Road. I say, yes, because there are recent references to it under such name in online search and it was originally imported here under this name from US Census TIGER database. A new user extended it further north because he claims he actually lives there and it was taking a part of his driveway. Additionally, I have just checked at the official state site http://geodata.vermont.gov/ that Bacon Road is indeed shown on their map and its shape matches the way it is currently drawn on OSM. Another thing is whether this road is residential or not. I would say, yes, because it is used for local traffic mainly by residents to get to their houses. You have changed it to a service road but if not residential I’m in doubts about the main service this road provides. In your changeset you wrote that the source of your information was "knowledge". Would you be willing to reveal the knowledge you have about this road?

Hello! Are you sure that way/19701179 is not Bacon Road in Waterville, Vermont?

Hello! Could you please take another look at this changeset? It seems you tried to combine two wood relations (9134979 and 9135276) into one (10289025), but then you probably need to take all inner rings from both woods and put them into that new wood.

Welcome to OpenStreetMap! Thank you for your first contributions! Please be sure that new roads do not overlap with the existing roads as this would create problems for navigation software. I had to remove such duplicates here, but there remains to be many areas where auxiliary roads such as driveways or parking aisles are still missing.

Welcome to OpenStreetMap and thank you for your first contributions! Please be sure to connect new roads to the existing road network.

Hi Zzptichka! TCT relation in OSM is currently specified as a bike route, but some parts of TCT are not suitable for biking. Particularly, in Gatineau park a few trails that are part of TCT have "not for biking" signs. For this reason they were excluded from the TCT relation in OSM. Among possible solutions, TCT could be reclassified as route=foot. Would you be willing to do it?

Hello. In the park relation there are two very similar outer ways 507025219 and 689335542. This cannot be right. Only one of the two is expected there. Which one would you pick? Both of them are not so good as per park's website the park occupies 525 hectares but measuring it with either of the two ways gives a larger area.

I removed area=15.59 key from the park relation as per osm wiki the area key should be either yes or no. There is a rarely used key sqkm in case you wanted to indicate the size of the area.

Hello! Thank you for improving the map. Additional roads, including driveways and parking roads are expected to be connected to existing roads. i.e., they have to have a connecting node with them. Navigational software would then be able to provide meaningful driving direction. Thanks again!

Welcome to OpenStreetMap! Nobody managed to learn how to walk without falling. Perhaps it's a good idea to start with small steps and save chagesets with a relatively small number of changes so that correct changesets can be distinguished from erroneous. In this changeset, unfortunately, you have broken a wood relation/721456, so that this wood is no longer shown on the map. A wood typically consists of a large outer area with multiple inner areas that are all expected to be closed areas. Because fixing the broken wood does not seem to be trivial, I’m simply going to undo your changeset so that you can give it another try. Please include useful changeset comments that summarize your contribution. All the best!