{% extends "_layout.html" %}
{% block title %}I2P Status Notes for 2005-10-18{% endblock %}
{% block content %}I2P Status Notes for 2005-10-18
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi y'all, its Tuesday again
* Index
1) 0.6.1.3
2) Freenet, I2P, and darknets (oh my)
3) Tunnel bootstrap attacks
4) I2Phex
5) Syndie/Sucker
6) ??? [500+ symmetric nat bounty]
* 1) 0.6.1.3
Last Friday we pushed out a new 0.6.1.3 release, and with 70% of the
network upgraded, reports have been very positive. The new SSU
improvements seems to have cut down on unnecessary retransmissions,
allowing more efficient throughput at higher rates, and there
haven't been any major problems with the IRC proxy or Syndie
improvements to my knowledge.
One thing worth noting is that Eol has posted up a bounty for
symmetric NAT support on rentacoder[1], so hopefully we'll get some
progress on that front!
[1] http://rentacoder.com/RentACoder/misc/BidRequests
ShowBidRequest.asp?lngBidRequestId=349320
* 2) Freenet, I2P, and darknets (oh my)
We've finally wrapped up that 100+ message thread with a clearer
view of the two networks, where they fit, and what room we have for
further collaboration. I won't go into what topologies or threat
models they're best suited for here, but you can dig into the lists
if you want to know more. On the collaboration front, I bounced
toad some sample code for reusing our SSU transport which may be
helpful for the Freenet folks in the short term, and down the line
we may be working together to offer premix routing for Freenet users
in environments where I2P is viable. As Freenet progresses, we may
be able to get Freenet working on top of I2P as a client application
as well, allowing automated content distribution among users running
it (e.g. pushing around Syndie archives and posts), but we'll see
how Freenet's planned load and content distribution systems work
first.
* 3) Tunnel bootstrap attacks
Michael Rogers got in touch regarding some interesting new attacks
on I2P's tunnel creation [2][3][4]. The primary attack
(successfully mounting a predecessor attack during the entire
bootstrap process) is interesting, but not really practical - the
probability of success is (c/n)^t, with c attackers, n peers in
the network, and t tunnels built by the target (lifetime) - less
than the probability of an adversary taking over all h hops in a
tunnel (P(success) = (c/n)^h) after the router has built h tunnels.
Michael has posted another attack on the list which we're working
through at the moment, so you'll be able to follow that one up there
as well.
[2] http://dev.i2p.net/pipermail/i2p/2005-October/001005.html
[3] http://dev.i2p.net/pipermail/i2p/2005-October/001008.html
[4] http://dev.i2p.net/pipermail/i2p/2005-October/001006.html
* 4) I2Phex
Striker is making more progress on the upload bug, and reports are
that he has it pinned down. It will hopefully get into CVS tonight,
and will be released as 0.1.1.33 soon afterwards. Keep an eye on
the forum [5] for more info.
[5] http://forum.i2p.net/viewforum.i2p?f=25
The word on the street is that redzara is making pretty good
progress merging back in with the Phex mainline too, so hopefully
with Gregor's help we'll get things up to date soon!
* 5) Syndie/Sucker
dust has been crunching away with Sucker as well, with code getting
more RSS/Atom data into Syndie. Perhaps we can get Sucker and the
post CLI further integrated into Syndie, maybe even a web based
control to schedule imports of different RSS/Atom feeds into various
blogs. We shall see...
* 6) ???
There's lots going on beyond the above, but thats the main gist of
what I'm aware of. If anyone has any questions/concerns, or wants
to bring some other things up, swing on by the meeting tonight at
8PM UTC in #i2p!
=jr
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
iD4DBQFDVU4UWYfZ3rPnHH0RApF2AJ9Eqg55ydhqEjqSpkN5BSGb8MerYgCY16bD
rYTKbx3oDZr3XWp3E9s0ag==
=ga+r
-----END PGP SIGNATURE-----
{% endblock %}