Actually, it looks like someone already changed the access tags on many of the roads in Changeset #151651659.

This change was probably made around the time the base started issuing visitor passes to prevent routing applications from sending people through the controlled areas of the base, which would be inconvenient for anyone who does not already have a pass. Previously, anyone could enter the base by presenting some form of identification.

I would suggest tagging most of these roads to access=permit. That tag was not commonly used when I made this edit. However, I am leary of reasons (1) and (2), since the common guidance in OSM that I've always known is to avoid tagging for the renderer.

The New Mexico classification page defined Window Rock as a regional center, because of the large volume of traffic between it and Gallup, and its status as the capital of the Navajo Nation. In hindsight, that might not have been the best decision. Since the two cities are so close, Window Rock might as well be a suburb of Gallup.

Regarding the rest of that route, I would first question that Page is an important enough regional center to have `place=city` status and a node on the trunk network. It is not defined as a micropolitan statistical area by the US Census, and I'd guess that people from places like Kayenta, Tuba City, and Kanab aren't driving to Page for much because they aren't much farther from a bigger city. It is hard to tell without knowing the area better. Do you have ties to Northern Arizona or Southern Utah? Otherwise, I think the only way you could justify this is digging into American Community Survey data from the Census to make a case that this is a regional center, if the data are available in that level of granularity. Without that kind of justification, making Page a city seems like tagging for the renderer.

Second, there are many small cities throughout the West that are considered regional centers on the Trunk network but don't have direct connections tagged as `highway=trunk`. While in nearly all cases once could identify some road connecting any two cities, at some point it just isn't reasonable. Long distances or mountainous terrain can reduce the importance of the link between two cities. If we don't make a judgement call at some point, we'll have lots of minor highways that aren't in great shape nor very well traveled tagged as trunk, and the tag will lose meaning. In my opinion, small cities, such as centers of Micropolitan Statistical Areas, need connectivity to the the Interstate Highway System, a major metro area, and possibly a direct path to a major port. Gallup, Page, and St. George already have all three, so I have to question the how important the link between Gallup and St. George really is for those cities.

Finally, for the reasons stated above, there may be dead-ends or spurs in the trunk network for good reasons. So please, do not tag something as trunk just to "extend an existing trunk route."

I think it's safe to say that page is past Draft status by now. Removing the draft tag seems reasonable to me.

I must have been thinking when deciding to only make US 180 into Silver City that most commerce from that city would feed into Las Cruces and El Paso making NM 90 less important. It would be interesting to see if there are any data on the amount of trade between cities and counties that we could look at, but for now, I think this edit is reasonable. It would be nice to keep the documentation up to date where we have it, though.

I have to question the validity of this change. A few reasons:
- AADTs between Window Rock and the NM State Line are significantly higher than anywhere west of Window Rock.
- ADOT considers this highway a minor arterial, not a principal arterial.
- What regional centers does this route connect?
- Having done a cursory search through mailing lists, OSMUS Slack, and the wiki, it appears that no consultation was done with the community regarding this chahge.

Hi Joseph,

Can you point to any correspondence with the community or updating tagging guidance regarding this classification change to trunk? Last I checked, New Mexico agreed on this trunk network: osm.wiki/New_Mexico/Highway_Classification#Route_Roster, which does not include this stretch of NM 90. If you want to make a change to the trunk network, could you please leave a note in that wiki page's discussion board as well as the pertinent mailing lists and Slack channels before making the change, and then after getting a consensus or no feedback after a few days, document the change in that route roster? A lot of thought was put into that list to prevent all the major highways from getting retagged every couple of years.

Thank you.

Taos is the center of a micropolitan statistical area of about 32,000 people. 6,000 is only the population of the city limits. Since the drive to Santa Fe is 2 hours, it is definitely not a suburb. Instead, it is a hub for a large portion of Northern New Mexico. Its cultural significance and proximity to skiing also attract visitors from all over the world.

For these reasons, it was identified as a regional center in the New Mexico Highway Classification Guidance (osm.wiki/New_Mexico/Highway_Classification). The guidance calls Taos a city (which I think is generally synonymous with regional center). Since the guidance has been up for some time, and no one from New Mexico seemed opposed to this, I went ahead and changed Taos to city on the map.

This was discussed in changeset osm.org/changeset/115451427#map=11/34.5126/-112.3475

I must have been thinking that it didn't continue as a primary road past the junctions of AZ 69 and AZ 89A, so it really only served neighborhoods in Prescott Valley. Since you seem to have more local knowledge of the Prescott area, and it sounds like the road provides connectivity to more than the neighborhoods along it, I will go ahead and change it back to primary.

Can you provide justification for the highway classification changes you made? North Roadrunner Parkway and Settler Pass, in particular, look suspicious. Living here, I know that those are definitely not secondary. They are feeder roads providing access to neighborhoods. They should be tertiary. Likewise, Engler Road (osm.org/way/170524639) does not currently connect the Park Hill subdivision to any other thoroughfare or neighborhood. This makes any claim that it is tertiary seem dubious.

DannyES, thank you for sharing that link! I assume you have seen the discussion about this on Slack. Your proposal mostly aligns with the current one there.

OSM does not use the federal functional highway classification standards (see osm.wiki/United_States/Tags#Road_classifications). "Urban Arterials" were traditionally mapped as secondary, unless they extend outside the city to connect to nearby towns, or are obviously a better option than the rest of the arterial road network to get across the metropolitan area. This edit restores Tucson to that standard, until a road classification standard is developed for Arizona. Please see the #local-arizona channel in the OSMUS slack for more discussion about this.

As I had clearly noted in my changeset comment (osm.org/changeset/84439281) and in the building outline with razed:building=yes tagging, this building was demolished. Please be mindful of this before re-adding demolished buildings. Satellite imagery is not always current in many areas.

It looks like you changed the tagging on the M and the L from man_made=geoglyph to the less accurate highway=pedestrian. Changing the tagging of features to something less accurate is never allowed, even if doing so makes it show up on the renderer [1]. Please do a wiki search on any tags that you are unfamiliar with before changing them.

[1] osm.wiki/Tagging_for_the_renderer

Hi mueschel,
Thanks for pointing this out. I removed the tags leftover from the shapefile and added the ways to the Zuni Reservation relation.
osm.org/changeset/92443878

These changes are imported from the USGS woodland tint database. See https://www.sciencebase.gov/catalog/item/5b6fa9fbe4b0f5d57878e755 . As a work of the United Stated Federal Government, these data are in the public domain. I did not make an import page for this changeset; I didn't know I needed to. Aside from documentation (in the wiki, presumably), is there anyone else I need approval from to keep this change, and do future imports?

Thanks,

Andrew