Re: Servers Should Use a Secure Mix Algorithm

[Laurent Fousse <laurent@xxxxxxxxxx>]

Hello,

* Colin Tuckley [Thu, Mar 02, 2006 at 08:46:12AM +0000]:
> 3) The system is *not* reliable at the moment, for those of us trying to
> track down bugs this makes it worse. The NymBaron team for instance would
> really prefer a timed mix as it makes their testing faster.

That's true. We can still do testing with a slow/unreliable network if
we short-circuit most (or all) nodes, but that's only feasible by
users who are willing to test the client *and* install a server. So it
is harder for users to test and we lose testing that way.

Admittedly, we are in a phase where the pace of bugreports is faster
than the fixing rate, so it's not that much of an issue currently.

* Marco A. Calamari [Thu, Mar 02, 2006 at 10:06:51AM +0100]:
> This list, as the Nymbaron one, is almost dead, no talk between
> developers,
>  no news, questions or proposals. I suppose most of the developer
> communications
>  are done in different ways (chat, private mail) but this is in
>  my humble opinion a REAL BAD THING.
> Investing a little time in communication between developer & power user
>  is a good investment.

As far as nymbaron is concerned, there is no hidden discussion going
on (and I believe it's the same for mixminion). You are welcome to
join the #nymbaron channel on irc.oftc.net where most of the current
development happen (mostly between Steve and I). More in-depth
discussion and proposals *are* sent to the list (ok, we didn't forward
the "let's move the packet format to S-EXP" discussion at the time,
but it's not really mature enough).

Back to the topic mixed pool vs timed : since we can mostly cut the
actual network for nymbaron testing (or we could have a small group of
server with timed algorithm and a very short interval) I don't mind
the change -- I just won't follow it for my nodes yet.

Laurent.