"'Xavier' via Osmand" <[email protected]> writes: > On Fri, Dec 07, 2018 at 10:01:15AM -0500, Greg Troxel wrote:
>>Then there is maxspeed:typical. > > The OSM wiki has no reference for maxspeed:typical. It does have > maxspeed:practical > (https://wiki.openstreetmap.org/wiki/Key:maxspeed:practical). Perhaps > this is the tag you are referencing? I was sort of referring to a tag that might be, not being careful about the tag that is. >>The real question for routing is "at what speed can we expect that a >>reasonable driver will traverse whis segment". > > The problem there, of course, is determining what is "reasonable", > which is also a concern put forth in the OSM Wiki. In this context, > "reasonable" is highly subjective. For a competent driver, in a > sports car with good handling, their 'reasonable' traversal speed of a > curvy road will be much higher than a new driver in a vehicle with > poor handling. Even the same competent driver but in the car with > poor handling, will traverse more slowly than in the car with good > handling. So no matter what one does, or which starting point one > starts from, the computation will be wrong for some group or another > of drivers. That's very true, but in my view a second-order effect. But that's why I am focusing on a range of speeds that 90% of the drivers are within, and the speed at which traffic typically flows. Perhaps it's the median speed when there is no congestion and no bad weather. I'm talking about a case where it would be bizarre for a car to be going at the advisory speed of 20 mph. Super cautious (but not incompetent) would be 35 mph, normal 40 mph, higher performance car 50mph, sports car would be fine at 60 mph. Hence I would like 40 mph to be used for routing, as the speed that if you have to pick one, is a pretty good indicator of normal drivers in normal cars. This is only an issue because the advisory signs are out of touch with reality. It's certainly true that for any single speed, trucks will go slower and sports cars faster (if the speed is about handling, as opposed to enforcement). But that's a general trend, and it doesn't usually lead to bad routes. -- You received this message because you are subscribed to the Google Groups "Osmand" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
