[Talk-us] US Trunk road tagging

[Minh Nguyen]

Vào lúc 11:11 2021-05-05, Evin Fairchild đã viết:
> To be clear, my proposal is to tag only NHS routes that are on state and 
> US routes. Also, I don't think we should include intermodal connectors 
> either. That will hopefully cut down on the number of NHS routes in 
> urban areas.

By the way, while this more nuanced heuristic is far more reasonable, I 
don't think it completely solves the problem of road classification islands.

For example, State Route 73 is a bypass around Wilmington, Ohio. The 
bypass itself is 7 miles miles long and built to Interstate standards, 
so it would be uncontroversially highway=motorway. It's connected to 
Interstate 71 by a 4-mile-long, two-lane surface road, which is 
currently tagged highway=primary because it doesn't have any of the 
physical characteristics of an expressway. Thus, at low zoom levels, the 
bypass is a motorway island. According to NHS, both the surface road and 
the bypass are mere intermodal connectors (by virtue of connecting to a 
cargo airport). The proposed NHS-based heuristic would maintain the 
status quo, including the motorway island.

As another example, a controlled-access highway from Celina to 
Wapakoneta is being upgraded to Interstate standards in segments. The 
eastern segment from St. Marys to Wapakoneta, the highway is on the NHS 
as part of U.S. 33, but the western segment from St. Marys to Celina is 
a state route excluded from the NHS. I have mixed feelings about how 
disjointed we've made the overall highway, but an NHS-based heuristic 
would only make the contrast starker: instead of alternating between 
highway=motorway and highway=trunk, it would alternate between 
highway=motorway and highway=primary. Or maybe it would just be 
highway=primary, ignoring half the highway project and potentially 
influencing the route a motorist might take from Decatur, Indiana, to 
Wapakoneta.

I don't think these counterexamples detract all that much from your 
general point about NHS (or at least Other Principal Arterials) being a 
decent potential starting point for trunk classification, assuming we're 
OK with opening trunk up to non-expressways. But it seems like there 
would still be some need for case-by-case exceptions. Each exception 
would undermine the idea that NHS is the final arbiter of trunk 
classification and thus the solution to road classification edit wars.

-- 
minh at nguyen.cincinnati.oh.us