[OLPC Networking] Path discovery sanity

[Aaron Kaplan]

On May 14, 2008, at 12:43 AM, Ricardo Carrano wrote:

> I totally agree with that:
>
> actually the far bigger effect and difference that I saw between
> different protocols is
> a) the quality of the implementation and algorithm choice
> we are just now (apparently) free of serious bug
>
congrats :)

> and b) how
> you chose your metric and
> we are still in the beginning of this tuning
>
has anybody here experimented with MIC already?
Basically in OLSR now the metric calculation moved to a plugin  
architecture - cudos and thanks and full recognition go to Henning  
Rogge - with ETT being the first potential metric which we could  
support. But MIC seems to have some nice properties.
Any other metrics that you would highly recommend?

> c) how careful you are with your
> "airtime" (i.e. how much of the precious spectrum you conserve
> without flooding it (*)).
> we are still removing some airtime eaters.
>
good!
Please do report, it is interesting how you do this.
:)

>

---
there's no place like 127.0.0.1