"Manfred Brenneisen" <[email protected]> writes:

> Thinking about maxspeed:practical, which can be used to initialize
> maxspeed in some cases, I thought about rural areas, which do not have
> any maxspeed, or maxspeed:practical information included in OSM data:

I think we do need tags for
  maximum speed that tends to be reasoable
  typical speed one can go


> The new curviness() function (useful for ways).
>
> Such a function could e.g. return the sum of direction change at every
> point, divided by the way&#39;s length.
>
> This could give an estimation for the maximum practical speed.

This could be useful to drop the speed absent tags, but I would worry
that it would be very sensitive to how well the road is represented.

As a completely alternative approach, one could process tracks to obtain
maxspeed:typical values for ways.

A related concern is if some ways (the more commmonly-used ones) are
tagged, then the untagged, even scarier roads will be preferred by
routing.

Attachment: pgp1Lql1Wl3wK.pgp
Description: PGP signature

_______________________________________________
mkgmap-dev mailing list
[email protected]
http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Reply via email to