I think the most useful part of this article is the last paragraph, where the author notes that none of The Pirate Bay's lofty projects have come to fruition yet. Until there is something that I can download and install on my computer, this is just vaporware.
Their projects only succeed if they use illegal content to push them.
This was also the reason youtube succeeded and right now they are only selling ads for about 4% of all content, because for the rest of the content, the original post doesn't own the copyright.
John Gilmore (employee number 5 at Sun, and co-founder of the EFF and cypherpunks) tried something similar in the late 1990s: http://www.toad.com/swan.html
Just to add to the above - FreeS/WAN project was too heavy on an ideology and it had a major impact on their technical decisions. They refused to support DES, because it was weak, nor did they want to include X.509 (PKI) support, because it was a hierarchy "controlled by selected ones". This limited practical usefulness of the project and it was one of the primary causes of its demise. Others were the fact that it was hard to install (required patching and rebuilding the kernel from the source) and tricky to configure.
There is a fork called strongSwan, which is still being developed even though it's Linux-specific and there is a native IPsec implementation in Linux 2.6.
From the article, it seems his goal was not met yet. However he is very proactive in fighting the illegal net spying. Maybe the Pirate Bay guys should consult him too.
Well, we have to do something with all those extra cores we're getting courtesy of the hardware guys. We might as well encrypt/decrypt the data that we're using with the extra cores that we're going to have.
Unlike in IPv4, encryption and authentication is a
mandatory feature of IPv6. Those features are normally
implemented using IPsec (which can be also used by IPv4).
Seems to me that if support for encryption is mandatory for IPv6, even if it's not on by default, using that somehow would be a better route than writing your own New Thing.
The essence of IPSec is a mandatory feature of IPv6. I don't see how this could mean anything other than "packets don't have to be encrypted by default, but encrypted packets must be allowed to be sent and received at any time". That makes IPv6 very much not orthogonal to the topic at hand.
What I'm trying to get to is this: why are they trying to figure out how to encrypt the internet when they can just use IPv6 (presuming that existing implementations support all the "mandatory" features)? The only rational answer I can come up with off the top of my head is if they think governments will have backdoors for transparently observing the traffic unencrypted, but that wasn't mentioned in the article as a rationale for anything.
Too bad TorPark (aka xBrowser and a lot of other things) swayed too far from the Mozilla/Open Source path and into the hands of a semi shady privacy company.
Onion routing solves a different problem: anonymizing traffic. This "project" doesn't aim to provide anonymity, just resistance against sniffing and some MITMs. Onion routing provides encryption as a side effect of anonymity, but it's massively inefficient compared to just encryption.
Yeah, onion routing will grind your data rates way, way down. It also very much works against the inertia of the internet as it stands, which it turns out to not be all that useful.
