Pretty sure bicycles and horses are not allowed on most of the "hiking trails" in the WMNF. E.g. Fishin' Jimmy Trail now has horse=yes and bicycle=yes with your changeset. Was this added in error? If not would appreciate a reference to support this change. Thanks.

I'm curious: what is the "TR" ref referring to? E.g. on Kinsman Ridge Trail it's tagged with ref=TR 31030. Do you have a link that supports this desgination? Thanks.

It would seem most/all of these newly added trails in the WMNF are illegal. Do you have any references to support a "bicycle=yes" tag? Thanks.

...and per wilderness area laws, people should not be creating new trails. There is controversy on what should and should not get mapped. But I believe we as an OSM community should not be mapping bushwhacks as it encourages/facilitates behavior that is against public policy.

Anyone can walk anywhere on these public lands. But mapping these bushwhacks is akin to "creating new trails" because it encourages dense usage/erosion of a strip of land.

Hello. I don't believe these types of bushwhacks should be added to OSM. They are not official trails. They are not marked, etc. Strava heatmap data alone is not sufficient to warrant mapping in OSM. Have you personally seen evidence of these trails? Even if there is a herd path somewhat visible, what do you see is the value of mapping this? Thanks.

I said nothing about liability. And I'm not hostile about armchair mapping. I just think there's a high risk of bad mapping when singularly using Strava data without a secondary reference (personal knowledge, another map, etc.) My people getting lost comment is a sentiment that guides my OSM mapping. "Would I be comfortable if a friend used this data to plan their hiking trip" etc.

From the wiki (osm.wiki/Armchair_mapping)

"Take particular care before attempting to map hiking trails from your armchair. Someone who follows a noexit=* mapped road in their car, will likely just curse their GPS and take another route. However, some serious obstacles in forest areas are easy to miss from aerial imagery. Remember the map user can't tell that one trail has been mapped by a hiker who has done it, and another trail may have been mapped by a virtual hiker who may have never been near the trail."

Allow me to end this on a friendly note: I see you have a ton of edits. Thanks for your contribution to the OSM community.

It should bug you. It's irresponsible to map trails just from Strava that you have not seen evidence of firsthand and/or you are able to verify against maps for an organization with some oversight responsibility. Not to be overly dramatic, but how would you feel if someone got lost or hurt trying to follow these "trails" that you mapped? We should strive to be curators of useful knowledge, not just blindly copying over lines from Strava.

I get a bit worked up about this sort of thing because this data gets sucked up by the likes of Gaia and AllTrails and becomes part of their ecosystem, often without the nuances of the more esoteric OSM tagging. Even if it gets deleted in OSM, it may take a long time before that change is reflected in the third-party mapping apps.

So....please remove ASAP. Thanks.

Hello. I believe a portion of the trail you deleted has a trail easement. It is listed on Lincon Conservation Land trail map: https://lincolnconservation.org/wp-content/uploads/2017/06/2013_TrailMap10-27.pdf

Do you have information that shows this map is no longer correct?

Hi Tom. I'm a fellow local Belmont OSM mapper. I've spruced up a lot of MA "public lands" this past year in OSM, including much of Belmont. FYI: your changeset messed up a landcover multipolygon I had created. No biggie, but just wanted to bring it to your attention. You may want to consider learning JOSM as it's far more powerful. It seems you're a new mapper? I'm happy to chat more if you're so inclined. Cheers.
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/103998791

Hello. I reached out to you as a courtesy before about Dana. I didn't hear from you after some time so went ahead with some changes that I think are more accurate. I see that you have added Dana back to the map. What is your reasoning? Dana is no longer a town. "Dana" or "East Quabbin" are not administrative divisions on DCR's map. https://www.mass.gov/doc/quabbin-reservoir-watershed-system-public-access-summary/download

I understand and appreciate the historical significance of Dana but think the existing historical OSM tagging is sufficient around Dana Commons.

Again, just trying to understand your reasons. OSM is a community effort and dialogue with other mappers is important. Thanks.

FYI: I had access to MassGIS L3 parcel shapefiles so made it more accurate.

You tagged private parcels as "leisure=nature_reserve" which will likely mean they will show up on third-party maps. You also removed the "landuse=conservation" tag which has been used a lot by multiple mappers in MA.

Why?
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/101946207

There are wetlands in this area that you are mass mapping as "natural=wood". What is the need to "paint" everything green? Are you just tagging for the renderer?

If you do have the desire to do detailed landcover mapping, then you should consult more sources to ensure everything you're mapping is indeed "natural=wood". Wetland data is available from MassGIS.
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/102189888

This private land is definitely not a "landuse=recreation_ground". Please consult the wiki to make sure you're getting the proper meaning. In many cases the colloqualial American Engilish intrepretation of tag names are quite different than the intended meaning.
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/102200759

"Dufresne Recreation Area" is not a landuse=recreation_ground according to the wiki.
osm.wiki/Tag:landuse%3Drecreation_ground

Why do you feel the need to remove landuse=conservation? It is used heavily in MA and is not depreciated.

Generally speaking, landcover mapping tags, like "natural=wood" should not go on the ways/multiploygons that define a parcel boundary...in this case the state forest. Furthermore there are some small wetland areas on the parcel so it's not true that the whole area is a "natural=wood".

I'd apreciate a response to a comment I made on one of your other changeset a while back. Did you not see it? Thanks.

Please consider keeping your changesets to a much smaller geographic area. Makes it much easier to comment on and/or revert if there are any issues. Thanks.
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/101937578

There's a couple of things going on.

A decade ago or more, someone did a big import of MassGIS "Open Space" data. This contained both public parcels like state parks but also privately owned parcels that have conservation restrictions (CRs) but NO public access. Someone made the (questionable) decision a while ago to tag these as "leisure=nature_reserve" or "recreation_ground". Bad idea IMHO. They get sucked up by third-party apps like Gaia and AllTrails and they get rendered in green so they LOOK like there is public access. I've been slowly removing these types of private CRs because I don't think they have much value to the average data consumer and furthermore they are ripe for misunderstandings. I believe this "Bakerfields" parcel has always been private, it just has a permanent CR on it too. It was good you removed it.

As far as land cover: land cover mapping in OSM is pretty messed up IMHO. The "nature=wood" rendering has two components: a solid green fill color and a different render layer for those little trees. For reasons I don't completely understand those two components get treated differently. Mapping a pond on top of a large "wood" area will put the blue of the pond on top of the green, but the "trees" icons get mapped over the pond. Makes no sense to me.

The "right" way to do land cover mapping involves complex "multipolygon" relations where you "cut out" every single little pond and marsh and field and whatever from the larger "wood".

In this example, you should have modified the existing relation 12422350 to do what you want. You'll likely find JOSM is a far superior tool for doing this type of thing.

Land cover mapping in the US is somewhat divisive with the OSM community. I personally largely don't do it because almost everything is covered with trees in MA anyways. What's the value? Especially considering land cover mapping is now more-and-more automated with AI.

Anyways...unfortunately OSM still has a long way to go to make all of this more straightforward. Mapping practice norms are somewhat varied...even within the state.

Happy mapping...

Hello. What is your goal with this changeset? For instance, your new relation 12487256 is basically a duplicate of preexisting relation 12422350. It's redundant data and messy. I'm happy to chat further regarading your mapping goals. Thanks.

Hello. I don't believe you're landcover mapping is being done in the best way. For instance on the little islands in the Quabbin: it looks like you're making a duplicate of the closed way with a slight offset, then adding your landcover information as well as redundant information about the island name, access, etc. that already exists. A better way is to use multipolygons. Basically, for every piece of data (way geometry, "name", "access") there should only be one instance. I notice you're using the iD editor. You'll probably find JOSM editor a much more capeable tool for the complex land cover mapping you're doing. My personal interest is in fixing up the boundaries of our public spaces so they map better in third-party apps. IMHO some of your mapping choices is making things pretty messy. Let me know your thoughts. I'm sure we can find some consensus so we both can attain our mapping goals. Thanks.