The Lexington/Heppner area is probably big enough that it should be served by a primary road. OregonDOT traffic counts data indicate that OR 207 from the north is by far the most common route into the area. The OR 19 route also makes a lot of sense coming from the west (more-so than the direct OR 74 route since the latter is very slow and winding).

I had trouble avoiding a primary spur into Heppner though. The only real possible candidates are OR 74 to the east (way too circuitous and curvy to the point that going around via OR 207 is faster), Forest Road 53 to the east (a seasonal road that's closed most of the year), and OR 206/207 to the south (also both very circuitous/windy with almost no traffic). None of those make any sense to upgrade to primary, so a primary spur it is.

Whoops! Appreciate you catching that and fixing it; thanks!

Whoops, meant to say didn't exist, not did exist in my first sentence

If I-84 did exist, I would agree that US 30 would be trunk. However, I-84 does exist, and fully bypasses it for any through traffic. And I agree that the road is the only one going through the town...of 500 people. A town of 500 people is not a destination that needs to be served by a primary road at all. There's not a precise cutoff, but generally I use around 1000-2000 people as the cutoff for whether a town should be served by a primary road.

Take the town of Sumpter, Oregon as an example. It's a town in the middle of nowhere with a population of 200 people. I don't think anyone would argue that the one highway serving it should be upgraded to primary. And this is a similar example.

Could you explain why you upgraded a route that only serves a tiny town of 500 people to primary? US 30 here is absolutely not a major cross-country highway; it's been fully supplanted by I-84 for cross country travel.

Once I work through to this area in the Oregon highway classification work I'm doing right now, I will be downgrading this to secondary again.

This was a really hard one to get "right", as western Bend has a bunch of roads with roughly equal importance, quality, and traffic. I tried my best to make the primary/secondary/tertiary distinction in the area, but any and all refinements from others would be super helpful and welcome.

Wikipedia is not a reliable source for information on exact highway routings. Ground signage and state databases (https://maps.udot.utah.gov/uplan_data/documents/HRO/ for Utah) are much more reliable, although still not infallible.

Exactly. The only reason this stretch of road is primary is that it serves as the northbound off-ramp from I-15 to 3300 S. In fact, the fact that we have incomplete interchanges here is the only reason why it's primary. If there were an on-ramp to access 3300 S directly from I-15 northbound then that would be the primary route.

The reason this road is primary is that there is no corresponding interchange to 3300 S from northbound I-215 to the one from southbound I-215. 3300 S is the primary road in the area, and this part of Wasatch Boulevard serves as the access to 3300 S from I-215. Reverting this change as it breaks this primary connectivity (and the related change that upgrades 3900 S to primary since it doesn't serve a primary function here)

*SR-201, not SR-210 (which is the road up little cottonwood). The continuation of the freeway section of SR-201 is functionally I-80, not this surface street.

Really, I find that a highway designation is a pretty meaningless indicator of what functional class the road should be. There's a correlation for sure (highways are more likely to be primary), but the cause and effect is reversed. Just because a road is maintained by the state doesn't make it inherently more or less important than a road maintained by the city or county; the current classification guidelines are designed to take into account the road's characteristics and importance more than the designation.

Why did you upgrade this section 2100 S to primary? I see no world where the road is primary: it's been bypassed as a through route by SR-201 and I-80, it does not have exceptionally high traffic volumes compared to nearby secondary roads like 1300 S or 1300 E, it does not "feel" like a primary road in terms of things like speed limits and road widths, and it doesn't serve an important through route. I've reverted the classification to secondary.

This one I definitely disagree with (I'm the one who upgraded the northbound lanes to tertiary). The reason for this has to do with the one-way configuration of 600 S and 500 S in this area.

All westbound traffic on 600 S is forced to make a northbound turn onto 600 E. Due to the configuration of roads in the area this is not an uncommon movement to make. Thus, the northbound section of 600 E here serves more than just local traffic: it serves a decent portion of the traffic that wants to access either westbound 400 S or 500 S.

In contrast, the southbound section of 600 E in this area serves no such purpose: it's the definition of an unclassified road that serves no purpose beyond local access to the businesses along it (or a bike route, but that's irrelevant to road class). This is borne out in my experience driving and biking in this area: the northbound lane has easily at least twice the traffic of the southbound lane.

The most up to date guidance for classifying US roads is located here: osm.wiki/United_States/2021_Highway_Classification_Guidance, and this page states that "Lower classification [secondary or lower] levels should be applied to roads based on their relative importance to local transportation."

The classification that you mentioned works well in the eastern US or in Europe, but is a terrible fit for the western US due to how much lower density the region is. That's why the western US has been moving away from that, in my experience and from what I know.

So the purpose of these classifications is as follows: trunk roads are the most important routes intothe major towns and cities: my definition for this is approximately >20,000 people. Primary roads serve as links between the mid-sized towns (~2000-20000 people), while secondary roads link the smaller towns and serve the less-important links between larger towns and cities.

This leaves tertiary roads in rural areas to link unincorporated communities or somewhat important destinations (e.g. major parks) to each other, and also to represent roads that could reasonably be used as through roads. Tertiary roads can also be distinguished from lower classification roads by their characteristics: in general they're designed for higher-capacity travel than unclassified or residential roads, often including features like striping or higher speed limits.

Then unclassified for roads that really have no reason to be used for any through traffic that don't serve primarily residential usage (e.g. farming, recreation, logging, etc.), and residential for roads that just serve residential uses.

Does that help answer your question?

I should add that the reason it takes me so long to do this that I have to pretty much do it on a case by case basis if I want to have any accuracy :)

Land ownership boundaries are a portion of it. Combining that with aerial imagery, street-level imagery, and (in places I have it) local knowledge.

Technically there's a few roads on USFS, BLM, or state land that might be inaccessible, but these are few and far between so that vast majority of time any road or path on public land is going to be public.

With private land, some landowners appear to be fine with letting people cross their land, and some don't. Generally for private land I look for the presence of gates or no trespassing signs to confirm that the land is inaccessible. All the roads/tracks/paths I tag as private I've found something to indicate that the land is likely posted no trespassing or gated. I'm only one person so I'm bound to make the occasional mistake, but that's my methodology that I believe should be pretty thorough. Does that answer your question?

Note that some of the tracks I marked as private do cross public BLM land, but as far as I can tell there's no way to actually access these tracks without crossing private or closed tracks

For sure! I don't really follow the forums because 9/10 times the topics won't be related to the things that I map.

But I'm mostly aligned with you on the tagging method for these types of paths. If it's an informal path that I'm not super familiar with, I'll usually use highway=path, informal=yes, surface=ground. In my view the surface tag is one of the most important tags, as it seems to be one of the most common that external applications (e.g. Gaia GPS, Strava) make heavy use of from OSM data.

Side note, I see people using the incorrect surface=dirt tag in the Wasatch and it bugs me more than it should haha. Pretty much none of our paths save for a few flowy mountain bike trails are actually manicured dirt like the tag is designed for; it's almost always just whatever the natural ground happens to be in the area. I've been over time correcting these surface=dirt tags to surface=ground, but there's a ton of them and I haven't gotten to all of them yet.

In areas close to the Bonneville Shoreline Trail https://stateofthebst.org/explorer/ is a very helpful website for determining public access

Ok, I think I figured out what happened:

The name "Brushy Springs Trail" comes from the official USFS map, but the USFS map itself has the trail descending into the basin east of Box Elder Peak and south of Bob Stewart Peak. However, I was unable to find any trace of the trail on the Strava heatmap or on any aerial imagery, so it doesn't seem to actually exist in most places.

The part that remains on the map is the one part of the official Brushy Springs Trail that appears to actually exist nowadays, between the main ridge and Bob Stewart Peak. The extra bit that I accidentally deleted is not part of the Brushy Spring trail; seems like I probably meant to mark it as informal but instead deleted it entirely accidentally.

Oh weird not sure what happened there; looks like I somehow accidentally deleted some of the path unintentionally!