[Tagging] Making public transit tagging orthogonal across the map

Minh Nguyen minh at nguyen.cincinnati.oh.us
Sat Jan 2 02:17:47 UTC 2021


Vào lúc 16:53 2021-01-01, Paul Johnson đã viết:
> On Fri, Jan 1, 2021 at 3:03 PM Joseph Eisenberg 
> <joseph.eisenberg at gmail.com 
> <mailto:joseph.eisenberg at gmail.com>> wrote:
> 
>     While I can understand defining a route as separate from the road
>     itself, I think it should be remembered that most roads and streets
>     (which we map as ways with highway=primary/secondary/etc) have a
>     name, even if that name is similar to the reference number.
> 
> 
> They don't, always, though.  That's why noname=yes is a perfectly valid 
> tag. If you think "Highway 64" is a name for a highway, then, no, you're 
> absolutely wrong in a definite sense; noname=yes applies.

Are we still talking about how to tag a road route relation, or more 
generally a highway way?

"Highway 64" _can_ be an idiosyncratic name for a highway, in that a 
data consumer would otherwise have no way to know that it should be 
called that in one county but "U.S. Highway 64" in the next county over. 
And when U.S. 64 gets rerouted, then it becomes "Old Highway 64", 
including the short connector to the new road that was never ref=US 64 
to begin with and shouldn't be tagged old_ref=US 64.

I could stand behind tagging a route relation as ref=64 noname=yes 
official_name=John Smith Memorial Highway but half its member ways as 
name=Highway 64 and the other half as name=U.S. Highway 64. It's all the 
same route, after all.

There's an even stronger case for tagging numbered names on ways 
belonging to concurrent routes. [1] If you're saying these numbered 
names should be tagged but go in a tag other than name=*, like 
systematic_name=*, then we can debate about where to draw the line.

[1] 
https://lists.openstreetmap.org/pipermail/talk-us/2014-December/013943.html


-- 
minh at nguyen.cincinnati.oh.us




More information about the Tagging mailing list