Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[liberationtech] Tahrir

Ian Clarke ian.clarke at gmail.com
Sat Dec 24 21:22:08 PST 2011


Hi all, just wanted to introduce myself, and share a bit more about Tahrir,
which I believe has been discussed here recently.

I'm the founder and coordinator of the Freenet project, perhaps the
earliest P2P anti-censorship technology, the first version being released
in March 2000.  Freenet has been under active development ever since,
employing one full-time developer for much of the decade since we began.
 In a recent survey<http://freedomhouse.org/images/File/special_reports/LOtF_China.pdf>of
Chinese anti-censorship tool users, Freenet came out on top, so it
seems
the project is still very relevant today.

One of my more recent projects is called Tahrir.  Tahrir was inspired by my
conversations over the years about the real needs of democracy activists in
a variety of countries that lack freedom of speech, and over a decade of
experience building anti-censorship tools (and many many mistakes doing
so!).

I've realized that many anti-censorship tools lean heavily towards
attacking the problems their developers find intellectually stimulating,
rather than those actually experienced by people in the trenches.  We've
certainly been guilty of that from time to time with Freenet.

One example of this is that bandwidth can be very limited in many of the
countries we are interested in helping, particularly China, and yet most
P2P systems treat bandwidth like a near-infinite resource.

Tahrir aims to tackle this and other problems head-on.  It is an
ultra-low-bandwidth secure, decentralized microblogging platform.  The
initial release is intended to have functionality similar to Twitter, but
it will be extensible well beyond that.

Initially it will be a desktop app that requires a (slow) internet
connection, but down the line it can run on mobile phones, and operate in
an "ad-hoc" mode relaying over wifi and other short-range connections.

I've been working on Tahrir since March, although other pressures on my
time have prevented me from making much progress in recent months.

My approach has been ground-up and I've attempted to stick closely to
modern best-practices for Java coding.  So far I've implemented:

   - A low-level encrypted reliable UDP transport capable of UDP
   hole-punching
   - A very space-efficient serialization mechanism to convert POJOs into
   bytes and back again
   - A very convenient RPC mechanism built on top of the above two
   mechanisms (similar to the approach used by GWT)
   - A convenient set of cryptographic primitives that integrates nicely
   with the above

You can read more about this architecture here:
https://github.com/sanity/tahrir/wiki/Architecture

I have been keeping relatively quiet about my work on Tahrir so as not to
distract too much attention from Freenet, but I really want to move things
forward after several months of dormancy.  To do this I primarily need
smart Java coders, who I will be happy to bring up to speed on the
codebase.  The top priority is to get to a working prototype ASAP, and then
begin iteratively improving from there (in line with the "release early,
release often" open source mantra).

You can learn a lot more about what I've done to-date here, in addition to
checking out the source code: https://github.com/sanity/tahrir/wiki

I'm also very happy to answer any questions anyone might have about Tahrir,
and I hope that some of you are interested in helping.

Ian.

-- 
Ian Clarke
Blog: http://blog.locut.us/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/liberationtech/attachments/20111224/72954692/attachment.html>


More information about the liberationtech mailing list