I am as well, but your tagging goes against the established definition. Please check the wiki page I linked

In this case, it is definitely not leisure=park. It looks like the entirety of Cranberry Lake Preserve is wooded or otherwise natural.

Also, who is "we"? Are you contributing for a specific app or data consumer?

Essentially, if the place is left mostly natural and has a focus on hiking, it should probably be a leisure=nature_reserve. If the space is fully managed, with little or no natural forest, and has amenities such as sports fields, it should probably be a leisure=park.

Hi, and thanks for contributing.

It looks like you have been changing a lot of features from leisure=nature_reserve into leisure=park. I believe that in these cases, leisure=nature_reserve was the correct tag; the OSM definition for leisure=park is more strict than how the term is typically used in the US. Please take a look at the wiki page (especially the "Mapping history in the USA" section) to see what I mean: leisure=park

Also, please take a look at https://americanamap.org, since it differentiates trunk surface roads from trunk expressways.

Since I haven't heard back, I am going to reapply these changes. As with the Taconic, the roads that are borderline cases for motorway (e.g. Saw Mill, Bronx River Parkway) should not have their classification changed without community consensus.

If you take issue with any of the non-classification reverts, please let me know which ones and I can reapply your edits if they look good.

Yes, I have, but I read it again to brush up on it. Under what it says, the Taconic would be a borderline case even in the controlled access segments. However, the page says that highways with at-grade intersections should be downgraded to highway=trunk.

More importantly, the page also says the following:
These borderline cases have been the subject of edit wars. As such, roads that are "borderline" as described above (motorway islands and slightly degraded expressways) must not be changed from their current classification without consultation with the community to determine whether a consensus exists. Changes to the classification of these edge case roads without community discussion may be summarily reverted.

As such, I am reverting this to as it was before. If you want to change the classification, please consult the community in places like the OSM community forum, the OSMUS Slack, or the OSM Discord. I would be willing to make the classification changes myself if a new consensus is formed.

As mentioned, RIROs and full intersections are incompatible with motorway, so at the very least, the segments of the Taconic that have those can not be motorway. If you want to make an argument to make a fully controlled access segment motorway, that is one thing, but even then, the overall quality of the road is somewhat below a typical motorway. highway=trunk + expressway=yes is a fair combination in this case.

The Taconic is limited access, not fully controlled access. Therefore it is not fully a motorway. An argument can be made to make certain controlled access segments of it a motorway, like is currently done at the southern and northern ends, but you set the whole thing as motorway even through full intersections.

Please note that all of the trunk segments have expressway=yes, which makes trunk an adequate classification in this case.

The parkways do not all have RIROs, especially not ones as blatant as on Ocean Parkway and the Taconic. Your changesets have set multiple roads as motorway straight through full intersections (see osm.org/#map=19/41.180799/-73.758318 osm.org/#map=18/41.974509/-73.749359), which is completely unacceptable for a motorway. Additionally, it is well agreed upon that in the US, RIROs are not allowed on motorways (expressway=*#How_to_decide). expressway=yes is what distinguishes a surface road from an expressway in cases where a road doesn't qualify for highway=motorway.

Notably, I have the Saw Mill as a trunk expressway from the Cross County all the way to the northern end, since even in the controlled access segments, the design is deficient compared to the other nearby motorways (e.g. Thruway, Sprain, Hutch). Also, I set Mosholu Parkway to a primary expressway because based on the way it is used (using local knowledge), it is not important enough to be at the trunk/motorway level.

Otherwise, you made the BRP a motorway from the Sprain to Scarsdale Road, which I disagree with given local knowledge of how that road is to drive on. Just because it is controlled access for a short period of time doesn't mean that it is motorway quality.

I reviewed each of these changesets by hand before reverting, so please let me know which ones are proper in your eyes and I will be happy to discuss further.

See the note on the I-787 ways for the rationale there. As for NY-43, it is a route unto itself, and the signage on the ground seems more like the ramps don't start until the road diverges.

Hi,

Unfortunately, I've had to revert this changeset and several other similar ones that you've made in the area lately. Ocean Parkway is definitely not a motorway, as it has uncontrolled property access in many places. In addition, it is not important enough to be at the trunk/motorway level.

Please see changeset/169939717 for more details and to discuss any of these changes further.

Reverted; see changeset/169746353 for discussion

Reverted in 169940604. The Taconic is limited access, with partial controlled access. Motorway requires full controlled access, i.e. no intersections including RIROs. Currently I have the southern cutoff at US-6 because the road narrows significantly at this point and the RIROs begin shortly north of here.

Note that expressway=yes is used here on the trunk segments, which denotes the limited access nature of the road.

The primary issues with these was the overuse of motorway (e.g. setting all of Ocean Parkway to motorway), and removing `_link` from some highway ramps. Please discuss here if you disagree with any of these reverts.

The following changesets were reverted in this changeset:
169931511
169931494
169931358
169931597
169931278
169931257
169887702
169839373
169838893
169838539
169795549
169802698
169826615
169748025
169747422
169747403
169747362
169747329
169740359
169739861
169698980
169698856
169698787
169687763

I was changing NY-119 to primary, so I selected all members in that relation and changed them together. When checking my work, I noticed that this part had not been changed, so I looked at it and saw that it wasn't part of any relations.

Do you download data from Overpass? Otherwise if you use the normal way split function, the relations should carry over.

Same thing with way/1351485283

Hi,

Somehow the relations didn't carry over when you split this way: way/1351485284

I'll fix it; I just wanted to point it out so that you could figure out what might have been the cause.

The comment here has a typo; I meant "deletion"

Hello, and thanks for contributing.

I've reverted this change since it goes against the 2021 trunk/motorway classification guidelines, which impose additional requirements on what is allowed to be either of those classifications. In this case, the road definitely is controlled-access (i.e. a freeway), but it does not directly connect to any other trunk or motorway. As such, it is not part of the interconnected trunk/motorway network, and shouldn't be that classification. You can read more about the guidelines here: osm.wiki/United_States/2021_Highway_Classification_Guidance

The road already was and still is tagged with expressway=yes, which conveys the upgraded construction. If there are signs that explicitly say that foot and bicycle traffic is prohibited, you may tag the road with foot=no and/or bicycle=no to convey this.