Re: [babel] [Babel-users] About Babel-RTT

I'll note that while a theoretical explanation/study sounds interesting,
it's not required for publication of an IETF document. The fact that
there is a real-world deployment of this technology is much more
useful to me than whether or not it's theoretically sound. So I'd be
in favor of advancing the document without the theoretical analysis,
if someone has the time to document the "algorithms needed to
make it work well" in the draft.

Note that I do not qualify as a "competent dynamic systems theorist",
I am merely an engineer that likes protocols that work, like Babel :-)

David

On Tue, Jul 28, 2020 at 9:26 AM Juliusz Chroboczek <jch@irif.fr> wrote:

> Dear Ronald,
>
> [Both Babel MLs in CC]
>
> At the WG meeting, you asked about the future of the Babel-RTT draft.
>
> The protocol extension isn't going away.  It's been used in production for
> many years at Nexedi, where it works beautifully.  No other
> production-quality
> routing protocol known to me has this kind of automation.
>
> As to the draft, the situation is somewhat less good.  Baptiste Jonglez,
> who did the initial implementation, does not appear to be interested in
> working on the subject.  The current draft merely describes the packet
> format, it doesn't fully describe the algorithms needed to make it work
> well.  As to the algorithms, we did determine experimentally that they
> work well, but we haven't been able to get a satisfactory theoretical
> explanation of why that is the case.
>
> So in order to make it into a peer-reviewed paper or an internet-draft,
> I would need help from a competent dynamic systems theorist.  Any help
> with that would be welcome.
>
> -- Juliusz
>
> _______________________________________________
> Babel-users mailing list
> Babel-users@alioth-lists.debian.net
> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/babel-users