[mythtv-users] What would it take to break encrypted QAM

[Mache Creeger]

We have DVD ripping for personal use today because someone 
successfully broke its encryption scheme and many of us having been 
bemoaning the fact that unencrypted QAM is going away to be replaced 
by light encryption (potentially DES-56 - 
http://anandtech.com/weblog/showpost.aspx?i=637). While I would never 
advocate that anyone break national or local laws like DMCA, I 
believe it would be an interesting academic exercise to 
hypothetically evaluate what it would take to crack future DES-56 
enabled CATV distribution architectures.

DTAs are highly cost reduced. They have to be pretty simple devices 
and cannot have a lot going on inside. Its been at least 10 years 
since the first DES-56 cracking challenges were put on by folks like 
EFF and RSA. With the advent of specialized FPGAs 
(http://en.wikipedia.org/wiki/Custom_hardware_attack), MapReduce 
(http://en.wikipedia.org/wiki/MapReduce), and Hadoop 
(http://en.wikipedia.org/wiki/Hadoop) and domain-specific knowledge 
about the contents of encrypted messages, is it possible to retrieve 
such DES-56 keys? If the keys are downloaded over the cable network, 
can you capture them? Can you construct a device with a programmable 
serial number that could implement a virtual copy of an existing DTA 
on the cable network in your own home? And could you do all of this 
in a way that would make the software relatively generic so that it 
would require integration by the end user and not necessarily be an 
obvious target to content owners and their agents?

These are all interesting questions to ask, hypothetically of course.

-- Mache