An additional note that I just remembered: in the New York-specific guidance, the Niagra Scenic Parkway (957A) is specifically called out as a candidate for highway=primary+expressway=yes. osm.wiki/Proposal:_New_York/Highway_Classification#3._highway.3Dprimary

Hi andrepoly, thanks for asking! The short answer is that the New York DOT gives the northern part of the Niagra Scenic Parkway an Arterial Classification Code (ACC) of 3, which has been determined by other mappers in New York to correspond to `highway=primary`. Even though this road is of a grade-separated construction it has a low speed limit and doesn't serve as a connection to large regional cities like most of the motorway network.

Here is info on the update US Highway classification guidance:
osm.wiki/United_States/2021_Highway_Classification_Guidance
and here are the New York-specific guidelines developed and discussed by other New York-based mappers which draw on the DOT's ACC classification for top-level highway classification.: osm.wiki/Proposal:_New_York/Highway_Classification

I've gotten a lot of feedback on this project by mappers in the OSMUS Slack's #local-newyorkstate channel: https://osmus.slack.com/archives/CJ4QKU40H/p1636421497086700

As I worked through reclassify the top-levels of highways in upstate New York I've been compiling a list exceptions, and places where the ACC values given by the DOT don't necessarily make sense to me:
osm.wiki/New_York/Highway_Classification/2021_Classification_Project
A portion of the Niagra Scenic Parkway south of I-190 and the Rainbow bridge has an ACC=2 which would likely make it `highway=trunk`+`expressway=yes`, but it then seems to peter out into a network of ACC=4/5 city streets at its southern end rather than obviously serving a top-level connection in the road network. It you have local knowledge of this roadway, feedback would be very welcome!

Ha ha! You beat me too it -- I was going to reach out in the OSMUS #local-newyorkstate for local guidance after finishing more of the secondary->primary upgrades in this area.

Here's a screen-shot of the ACC classifications from NewYorkState Streets that I'm using in the Kingston area:
osm.wiki/File:NYS_Streets_-_Kingston_NY_ACC_classification.png

Given that Delaware and Hasbrouck have an ACC of 4 I was hesitant to upgrade them for the duration of a project as I'm not that familiar with this area. I'm personally OK with a small trunk-gap with a construction detour on more minor roads, but if the alternate connection makes more sense to you I'm OK with that as well. :-)

Oops, I meant from I-86 to NY-31.

I've fixed a missing section of NY-37 west of Massena and some overlapping roads caused by this change-set in osm.org/changeset/112510817

This change is a rounding up of the Shelburne Falls-Hinesburg road from a FC of Town-maintained "Major collector" to secondary, but it allow the lower part of Spear, Irish Hill and Dorset which have an FC of Town-maintained "Minor arterial" to avoid being stubs. As well, connecting the larger town centers of Hinesburg and Shelburne via secondary makes sense from a connectivity standpoint.

Note that Irish Hill/Falls/Marsett Roads have a functional classification of "Major Collector", but I'm rounding up to prevent Spear from being a spur in the secondary network.

Thanks, Ewen! :-)

Years of software development has ingrained in me this mantra for version-control comments: What changed? Why did you change it?

See comments on osm.org/changeset/111570311

See comments on osm.org/changeset/111570311

Hi tcarlisle, welcome to mapping in Open Street Map! I've reviewed a number of your change-sets and see that you're adding a lot of great detail to the map. Thanks for all of the good contributions!

I want to flag an issue with this and several other similar change-sets that you might not be aware of. In this change-set you added landuse=retail and highway=pedestrian areas to "paint" end-zones, lettering, and symbols on a sports field. This is a practice known as "Tagging for the renderer" and something we want to avoid in OSM. See: osm.wiki/Tagging_for_the_renderer

When mapping in OSM we want to be both precise and accurate. In the case of these sports fields, there aren't actually commercial areas or pedestrian areas on this field, so we shouldn't add these fake areas in an attempt to represent paint colors. There are many many different systems that use OSM data in addition to the map-renderings shown on openstreetmap.org, and by adding fake data to get a certain effect in one map we potentially confuse other users of the data, such as a data-user comparing the density of residential to commercial areas in a city.

To fix these fields I recommend removing these fake areas that represent paint colors.

If you have more questions about this or other mapping topics you're welcome to reply here or join other mappers discussing all sorts of topics on the OSM-US Slack channel at https://slack.openstreetmap.us/ .

Best,
Adam

Hi Alex, I just noticed that this change-set demoted Middlebury from a Town to a Village. Even though the place is referred to in Town documents as "the village" I don't think this demotion is correct due to the level of services available in the village compared to other communities. I've started a thread in the #local-us-northeast channel of the OSMUS slack to get more regional input on the town/village distinction as there is a wide difference in application across NY, VT, NH, and ME. If you'd like to join in that discussion, head here:
https://osmus.slack.com/archives/CC0LMFWBH/p1631729815087000

Thanks, Joseph! Happy mapping!

Hi Joseph, please do not upgrade these sections to motorway. As described at osm.wiki/Vermont these sections are not up to full interstate standards. While they have grade separation they are mostly single-lane each direction aside from an exceedingly short approach to the large interchange, are otherwise not separated by direction, and low speed limit (50-55mph). Beyond this large interchange they are single-carriageway.

Please join us to discuss further in the #local-vermont channel of the OSM-US Slack (I see you've already joined the Slack). About a dozen mappers from the region have been discussing the top-level (motorway & trunk) road tagging in the region for months and have come to the consensus that these roads should be tagged as highway=trunk for their connectivity importance and expressway=yes for their enhanced construction level, but not highway=motorway as their construction level isn't high enough for a significant-enough distance to warrant this upgrade.

Hi Eric, thanks for all of the great land-cover mapping you've been doing! I hope you'll continue if you remain inspired as Vermont has lots of land without good land-cover coverage. :-) I spent a day in Orange County a couple weeks ago and was inspired to add a lot of land cover (mostly natural=wood) in Vershire and Chelsea that connects down to your beautiful work.

One tip: as you are mapping land-cover, it can be a good practice not to connect land-cover ways to town boundaries, roads, and other non-land-cover types of objects as doing so can make it harder to change those other objects in the future or introduce unexpected changes to them when further refining land-cover. In particular, tweaks to land-cover may accidentally drag town boundaries a bit in a way that can be hard to correct without downloading the boundary data from the state again and comparing its alignment.

Roads are less of a correction issue than boundaries, but years ago I made the mistake of splitting road-ways at each land-cover change so that I could use small parts of the road as the outer-ways of land-cover multipolygons. Unfortunately this made it very difficult to update road properties as each road was now dozens of little ways -- I'm still finding and fixing these a decade later. ;-)

I see you are on the OSM-US Slack -- feel free to reach out if you'd like to chat more on Vermont land-cover mapping or other topics. :-)

Cheers!
Adam

Please don't remove valid surface=* tags from road ways. Without an explicit tag it is impossible to know if a road is paved, unpaved, or just hasn't been examined yet. Especially in northern New England and other rural parts of the US (let alone the rest of the world) there are many significant roads that are unpaved, but for which surface data still needs to be added. Removing surface=* tags complicates that effort of getting complete data coverage.
---

Published using OSMCha: https://osmcha.org/changesets/109429933

Thanks, DUGA.

My changeset comment got automatically truncated when saving. The full comment:

Downgrading US7 near Bennington to Trunk. As discussed in previous changeset comments and replies, this highway is a "Super 7" structure that is not up to full motorway standards. While it does have many grade-separate intersections, traffic is either not directionally separated or only a single lane in either direction for all but a tiny portion of these routes. Speeds are lower than those of I-89 and I-91 as well. Depending on the consensus of highway=trunk use in the US, downgrading this to highway=primary may be warranted, but it is certainly not highway=motorway. I'm leaving it as highway=trunk now to indicate that it is "something more than primary" while it is not fully a motorway.

Welcome to editing OpenStreetMap, SGC-I. The edit looks good. I was able to pull up the same USGS topo layer you were using and verified the name there. In my area we have a lot of families named "Hurlburt" as well, so I would trust this spelling. Go ahead and add a bit more detail to your change-set comments if you are able as to *why* you made the change rather than just *what* the change is. Feel free to reach out if you have any questions about OSM or editing in Vermont. More resources for Vermont can be found at osm.wiki/Vermont#Resources . Cheers, Adam
---

Published using OSMCha: https://osmcha.org/changesets/96911789

Hi Ronnie, I'm interested in what data source you used to pick the name and type for this feature?

The Pulp Mill Covered Bridge is already mapped to the south at osm.org/way/19683796

Zooming in on the satellite imagery it looks like this is some sort of dam control structure. It spans the river but isn't a bridge.