[cryptography] patents and stuff (Re: NSA's position in the dominance stakes)

Adam Back adam at cypherspace.org
Sat Nov 20 05:03:16 EST 2010


Yep another xor-curosor or actually even worse.  Retarded.

A few years back, at Zero-Knowledge Systems (privacy tech company Montreal,
Canada - once home of the "freedom" network and a few cypherpunks etc) the
CEO Austin Hill wanted to patent some "stuff".  We were grumbling and saying
"please dont put my name on it", etc.

In an attempt to reduce this patent revolt among the technical people, they
shipped in a patent attorney expert guy to give us a talk on patents. 
Apparently the test for "novelty" is so poor as to make you cry.  So while I
dont think the legal people are morons, the system is moronic and like most
people they just aim to optimize their pay working within the system.

Many government related systems are broken, and often recognized as broken
by most participants, but they are hard to fix due to inertia and
bootstrapping problems.

The seasteading institute has some ideas about starting new countries to
experiment startup-style with new systems of government:
http://seasteading.org/

I thought IBM also tried some kind of patent amnesty kind of thing (a
license with a promise not to sue anyone who joined the club).

I was hearing that the pirate party was thinking of setting up a torrent
server on a sattelite :)

http://torrentfreak.com/pirate-parties-plan-to-shoot-torrent-site-into-orbit-101020/

I guess the limitation with patents is even if they cant sue you if you are
not operating in a patent recognizing country, they can probably try to sue
your customers.

Maybe freenet kind of idea would be better - p2p file sharing without any
central control and with enough anonymity that you cant identify who shared
what.  (I had one idea which goes like: the hoster never knows the contents
of files they host, and each act of download results in the file moving and
being re-encrypted to a new location.  So its contents are continually
shuffling and the protocol ensures that the sender doesnt know what he
served, hopefully in some provable way (eg the key is known only to the
downloader?) and it wont be served by the same host, nor under the same key
next time.


I find legal people are not nearly as formal/logical as technical people, so
they will just take a fuzzy view or creative interpretation or outright bend
the rules to get around loopholes.  So most loopholes wont work, or wont
last long.


Maybe your best bet is to somehow persuade many large companies that patents
are a bad idea.  Start another patent troll company and play it to the
max.  Use the spare funds to lobby for patent reform.


I guess this thread is somewhat removed from crypto - is this ok for the
charter of this list?  The old cypherpunks list seems kind of dead to me. 
(And the remaining node hosted on a watch-list bait "al-quaeda" related
domain name for some reason, so I'm not too interested in subscribing to
that one.)

Adam 

On Sat, Nov 20, 2010 at 06:51:27PM +1000, James A. Donald wrote:
>On 2010-11-20 3:21 PM, John Levine wrote:
>>I am increasingly getting the impression that you've never been
>>involved in actual patent litigation.  The process is messy, but the
>>assumption that judges, particularly Federal judges, are stupid and
>>are unable to understand and interpret expert testimony is, shall we
>>say, counterfactual.
>
>Let us consider the latest software patent in the news:  Oracle's 
>lawsuit against Google's android.
>
>Let us look at the first patent in the their pile of patents:
>
>http://www.google.com/patents/about?id=XTKpAAAAEBAJ&dq=7,362,331
>
>Would they bother trying it on if judges had a clue?



More information about the cryptography mailing list