[i2p] I2P Consumer Caveat

[David McNab]

jrandom wrote:
>A while back, most people were pushing 1-2KBps lifetime,
>and at another release, people were pushing 8-12KBps.
>Right now it seems 3-6KBps is the norm.

IMHO, the fate of I2P as a mass-deployable p2p framework, depends
critically on getting it back down to 1-2kB/s.

This, or ensuring that if a user sets a 1-2kB/s throttle, then s/he can
enjoy reliable connections between local and remote endpoints.

3-6kB/s translates to 15-30GB/month.

1kB/s translates to 5GB/month.

I might suggest that we set a goal that any given I2P router should be
(reluctantly) capable of delivering reliable connections between local
and remote endpoints with no more than 2-5GB/month overhead (plus of
course, any payload data).

Imagine the scenario of human rights activist Zahid Saleam, running an
eepsite exposing major high-level corruption within Saudi Arabia.

I2P is ideal for him, since he can run n servers all listening on the
same dest, and mirroring the same content. Additionally, each server
could be listening on its own unique additional dest.

Then, everyone browsing over eeproxy to the common dest will hit
whatever server is available. Also, Zahid can keep all servers up to
date via their unique dests (and thus know if any of the servers have
been compromised).

Unmonitored bandwidth (contrabandwidth) in Saudio Arabia is only
available from the black market, and in many cases may be throttled to
33k6b/s or even 19k2.

I2P's ability to work reliably in such a situation is in my mind a deal
maker/breaker.

If the guarantee of basic endpoint reliability in low-bandwidth
scenarios requires deep architectural surgery to the I2P router, then in
my mind it's well justified. Apologies that I'm just a client hacker,
and haven't yet built any skills with I2P router internals.

>Our anonymity goals have us aiming at state level adversaries, but
>to pull that off, we've got to get it working and kicking ass in
>less hostile regimes first.

Trivial in less hostile regimes to simulate the bandwidth scarcity
likely to be encountered in 'hotter' deployment areas.

Please, let's not allow I2P to suffer the fate of Freenet.

--
Cheers
David


_______________________________________________
i2p mailing list
i2p at i2p.net
http://i2p.dnsalias.net/mailman/listinfo/i2p

----- End forwarded message -----
--
Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net

[demime 1.01d removed an attachment of type application/pgp-signature]