[Talk-us] [Talk-us-newyork] Highway classification guidelines for New York State
Eric Patrick
txemt1 at gmail.com
Thu Sep 16 22:05:39 UTC 2021
Code for OSM routing engine which can be found here:
https://github.com/osmandapp/OsmAnd-resources/blob/0b8669d73a3befb7420a5d767a814871ae083de5/routing/routing.xml#L562-L574
It places trunk roads underneath motorway, which makes the routing engine
look for the quickest route to the trunk road. I believe some of us on this
list aren't placing that high of a value on trunk roads, because of the
British and what they call a "trunk road." if we assigned trunk roads to
Primary Arterials (which is what I've been proposing), the router will get
to that road as quickly as possible, if the end user needs to be routed
onto the road. Primary Arterials exist in rural areas, just as well as
urban areas, so why not keep them all the same. Is there really a
difference between the two? I've seen the argument about it would clutter
cities with trunk roads. Is that necessarily a bad thing? Cincinnati has a
lot of its downtown areas classified as Minor Arterials, while Augusta, Ga
has many of its downtown roads classified as Primary Arterial. If Broward
County, Florida had PA=trunk, there would be roughly 20 different roads,
with only 3 of them running through downtown Ft Lauderdale. Each state and
downtown area will differ, I'm 100% certain of that, but arbitrarily
assigning a road to be trunk because it's a US Highway doesn't do the end
user any justice.
In the Slack channel, Brian suggested making US-1 a trunk road from
Brunswick to the Canadian border. Maine has it classed as a Minor Arterial.
Maine has ME-9 classed as a Primary Arterial from 95 to the Canadian
border. If US-1 were turned into a trunk road and ME-9 were turned to a
Primary, the router would start routing people down that highway, adding 90
minutes to their commute.
Since Primary Arterials are at the top of the food chain (below
interstates) for functional classification, I believe trunk roads should
too for routing purposes.
Eric
On Thu, Sep 16, 2021 at 5:31 PM Paul Johnson <baloo at ursamundi.org> wrote:
> Aaand there it is, proving my point right on cue.
>
> On Thu, Sep 16, 2021 at 4:07 PM MoiraPrime <MoiraPrime at pm.me> wrote:
>
>> They're already numbers‽
>>
>> From most important to least important.
>>
>> 0 - Trunk
>> 1 - Primary
>> 2 - Secondary (the 2 is literally in the name)
>> 3 - Tertiary (the 3 is also in the name)
>>
>> This isn't rocket science.
>>
>>
>> Sent from ProtonMail mobile
>>
>>
>>
>> -------- Original Message --------
>> On Sep 16, 2021, 3:56 PM, Paul Johnson < baloo at ursamundi.org> wrote:
>>
>>
>> On Thu, Sep 16, 2021 at 2:26 PM Jmapb <jmapb at gmx.com> wrote:
>>
>>> I can't say exactly whose word I took on these definitions, or when or
>>> where -- probably some combination of the wiki, the mailing list,
>>> conversations, changeset comments, and other channels -- but I've been
>>> under the naive impression that this was a settled question. It seems it's
>>> not! If the nature of the low end of the highway classification hierarchy
>>> is indeed up for debate, I'd say the NY proposal's recommendations for
>>> anything below tertiary might be better replaced with a simple text that
>>> recommends following standard tagging practices and local conventions.
>>>
>> Probably could be fixed by replacing highway=whatever with a numerical
>> hierarchy but that sounds too much like actually fixing a broken tagging
>> scheme for OSM to actually handle based on previous discussions on fixing
>> broken parts of the tagging scheme.
>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-us/attachments/20210916/1f9fdb4e/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: roads.png
Type: image/png
Size: 319108 bytes
Desc: not available
URL: <http://lists.openstreetmap.org/pipermail/talk-us/attachments/20210916/1f9fdb4e/attachment-0001.png>
More information about the Talk-us
mailing list