[tor-talk] phantom protocol

[grarpamp]

On Mon, Dec 15, 2014 at 7:42 AM, str4d <str4d at i2pmail.org> wrote:
> As far as the design goes, it is most similar to I2P's protocol. I
> have only read (most of) the original whitepaper[0] so far, and AFAICT
> at a high level it is almost identical to I2P, except for a few
> specific points (e.g. using IPv4 for identifying anonymizing paths,
> and bidirectional circuit-switched tunnels).

Yes the IPv6 interface is a specific point and a major feature. Like
CJDNS and OnionCat [1]... that gives the user access to whatever
apps they want to use over it, right out of the box. TCP, UDP, ICMP,
all of it, no problem. No porting, no waiting for someone to write an
app for the network from scratch, no need for socksv5.

I like forcing new dedicated apps (ideas) to appear [2], but not at the
expense of a variety of use / network adoption. Unless of course
the network is specifically designed for one thing. And with all
networks, using them securely is still up to the user as always.

[1] Tor-HS-v2 will break onioncat one-to-one addressing and thus be
no more useful than garlicat at that point. I think we need to come
up with *cat-v2 for those to regain one-to-one somehow.

[2] For example: I2P-Bote, Pond.

> It is an interesting protocol. I am reviewing the whitepapers, and
> intend to publish a comparison page on the I2P website[1] once I have
> a good understanding of the differences.

I'll add CJDNS somewhere when I can.