[j-nsp] Hash algorithms for LAG

[Pavel Lunin]

> It depends on a platform. For M/MX/T the hashing algorithm is considered
> > to be a kind of business secret (said to be patented, etc). For EX it's
>
> Just to make an important point, it's either a secret *or* it's
> patented, part of the point of a patent is that you publish your
> invention in a way that others can replicate once it expires (true that
> rarely happens these days)
>

Sounds alright, though I'm not a specialist in this. But I've heard the
word 'patented' from someone from Juniper, this is why I used it. It's been
a few years ago and I even went to the trouble of skimming through the US
patent DB (very quickly). Found a bunch of Juniper's patents but none
looking to be related to this. So if nobody can justify, let's consider
it's a secret :)