[Tagging] Should 'highway=incline[_steep]' be discouraged?

[Richard Bullock]

> On Mon, 28 Dec 2009, Roy Wallace wrote:
>> Also, incline=* is still mathematically valid for nodes to indicate
>> the instantaneous incline at that point, so I don't see a problem with
>> that.
>
> I might be old, I might have gone to school in the Dark Ages, but a point
> cannot have an incline.

But the point is part of a way, and the way *does* have a gradient at that 
point, which is what Roy meant by "instantaneous incline at that point" 
(assuming that a road's altitude is differentiable, a fairly safe 
assumption).

Most places I've ever seen where the road is very steep usually have a sign 
telling you how steep
In the UK they look like this one I came across in Leek in Staffordshire 
(which I stupidly decided to cycle in both directions) 
http://www.flickr.com/photos/19863303@N00/4224573437/
In the UK, older signs are often found with the gradient expressed as a 
ratio e.g. 1:5

Might be worth using incline = 20%, (or if the way is reversed incline 
= -20%), instead of just steep if the information is available (i.e. as a 
sign).

Presumably the gradient on the sign is the maximum gradient along the way in 
question, so might be some sense in marking the point where the gradient is 
a maximum as a node.

Having said that, if you are driving a laden truck, or indeed cycling etc., 
you might be interested about knowing that a length of way was steep, rather 
than just that the gradient at a point further up is x%. You may be able to 
turn off before reaching the steep section of road, but you won't know where 
it starts unless someone uses a way to tag the incline.