[Routing] Fast Routing Engine - State of Play

[Dennis Luxen]

> in munich there was a small research project applying (Geisberger) 
> contraction hierarchies with osm data to do energy efficient routing.
>  We ran into some issues with negative edge weights (e.g. electronic 
> car downhill) What are your views on this?

That's easy to fix. Add a large enough constant to all of your edges
until all are non-negative (or non-zero). Electric cars are all about
potentials.

> @Brandon @Denis (How) does your implementations differ from Robert 
> Geisbergers implementation?

At the moment, the preprocessing is still the same. That's going to
change as well.

Brandon already mentioned in another email that he has is own
implementation.

--Dennis