[OSM-newbies] tertiary dirt roads?

[Thomas Meller]

Wee-ll, no.
I don't think so.

I do not know the roads we were talking of, but I think I can imagine how they look like: wide, dirty, dusty, very straight, good for high speeds.
According to James' description, that is.

When I compare this to the tracks I know at home, there's no similarity.
And there is no real similarity to service roads, as these are not roads for passing-traffic by definition. So the renderers do it right not to show the latter in lowres renders.

There is no way to tag a road minor than these if they are rendered in lowres maps. So the map gets cluttered if they must be visible.
'Unclassified' does no good job, too.

I am aware, that we do not talk about correct rendering, but think of it this way: Once a renderer has a well-balanced ruleset it is difficult to change it, only to supply people in another country with another interpretation due to their local road-building culture.
There will never be a perfect match, but a road's local importance to find a route is the correct thing to decide on whether it shows on the map or not.

The current scheme looks not broad enough to match this case. To distinguish between those 'forest highways' and 'tertiary' roads is obviously necessary.
You would not drive such a 'road' with a Ferrari.
Would you?
But you could drive very fast with an ATV, as you can see far enough to avoid major potholes. Even faster than over a narrow tertiary, highly curved road. A sportscar would feel better there, though.

I don't think that the discussed speed-limit is a good solution.
I know the difference. In off-road sections where a car gets stuck, an endurist can ride fast whereas in other sections where the endurist struggles a 4x4 may not even care.
Once more I invite you to have a look on my wiki account's discussion page.
I'm still trying to figure it out.

Tagging roads only based on their physical characteristics is as bad as only to tag them based on their relative importance. The interpretation a routing or rendering software will show cannot easily depend on one or the other criteria, but on a combination of both (a simple description, I know). Else, you will end up with vast amounts of theme-specific data interpretation rules.
Babel's Tower ist not far away.

The goal must be to find a set of values which can be combined with others to allow for decisions on painting a way in a certain zoom level or to ignore it when routing. The interpretation is never absolute. It depends on the traffic means aimed at. It's usecase-based. Yet it must be standardized.
That leads to a standardized matrix people agree on.

Can I convince you?

OK, I think this is enough philosophy for the newbies list.
The current understanding and used techniques don't allow this all to develop in the near future. Step one first, then step two.

-------- Original-Nachricht --------
> Datum: Fri, 27 Nov 2009 12:34:49 -0800
> Von: Paul Johnson <baloo at ursamundi.org>
> An: newbies at openstreetmap.org
> Betreff: Re: [OSM-newbies] tertiary dirt roads?

> Thomas Meller wrote:
> 
> > OK, I see the need for a new way to tag roads.
> 
> I'm not sure that's what we're running into.  I think the OP is trying
> to solve a renderer problem with tagging.  It really is possible to go
> hundreds of miles in the western US and northern Canada without ever
> touching a route more major than highway=track.  Get yourself a used
> pickup truck and a flannel shirt and you could film every Nickelback
> video ever made out there!
> 
> 
> 
> _______________________________________________
> newbies mailing list
> newbies at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/newbies

-- 
http://www.openstreetmap.org/?lat=47.172&lon=7.4395&zoom=14&layers=0B00FTFTT&mlat=47.16677&mlon=7.43513

Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/chbrowser