-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 13/05/13 22:42, Ricardo Lanziano wrote:
> On Mon, May 13, 2013 at 4:06 PM, Yussi <[email protected] 
> <mailto:[email protected]>> wrote:
> 
> Now, there are two concepts which we need to clarify, with some 
> similar aspects, namely a mesh vs. darknet.
> 
> A mesh network connects clients by "chaining" them in order from 
> source to destination, a wireless mesh does that over the 
> wireless.
> 
> suppose you have a fairly large mesh, and you run a service (a 
> website say) which is only accessible to people on that mesh 
> network, this is a local network service, it's also in a sense a 
> darknet, if you employ high encryption and VPN like routing in the
>  mesh level, it's a darknet proper. infact that's exactly how i2p 
> works  (worked), each client holds a few VPNs, and services were 
> encrypted on each hop (hop pair really).
> 
> 
> Actually, a darknet (and specially I2P) does not work like that: 
> http://pastebin.com/gWzFwBiw
I've seen this post ages ago, and it says exactly what i just did, but
let's not get hanged up semantics here :)
http://www.i2p2.de/how_intro

> 
> I2P, Tor, Freenet, all work as a darknet in the sense they provide
>  strong cryptography to ensure high anonymity and censorship 
> resistance.
> 
> CJDNS, OSLR, Batman, etc, they work by defining the best paths for 
> creating routes in a lower level.
cjdns/i2p does this by using tunnelling, Batman OSLR do this on the
ethernet level.

> 
> Netsukuku is somewhere in-between.
> 
yep

> Obviously I'm oversimplificating things, but my point is that 
> Netsukuku is unique.
> 
> Throwing up the old C code was a bad idea IMO, the python port 
> never lifted up and development stalled. I don't know what happened
> to the original developers (Alpt, etc) but the only person that was
> working on was Luca. He did get busy aswell and development of the
> Vala port stalled.

I agree, this is exactly the sort of thing you want to do in
c/c++. No one says we can't use the c code any more though.

A protocol is a protocol, the more implementations there are the
better, but the base one ideally should be in c and capable of being
compiled on embedded devices/arm.
There is some argument for why small parts of the more complex logic
can be done in vala, but even then it's definitely a c project.
> 
> I'd like to know if Alpt is around. Hopefuly we can get a roadmap 
> started?
Fuck it what the hell, let's kick this up a notch, who wants to set up
a swarm/node? let's do this thing!

The c implementation still exists somewhere on the open-wrt
repositories, let's have a few people running the vala and a few
running the old c and try and get them to talk.


I might even have somewhere a wrt flushed image with c-ntk and a deb
image with vala-ntk for running in a vm and some routing games/scripts
on it using brctl.

I'll start in 3 weeks.
whose up for it?

p.s. Luca are you still on here?

> 
> -- Ricardo Lanziano To iterate is human, to recurse, divine.
> 
> 
> _______________________________________________ Netsukuku mailing 
> list [email protected] 
> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/netsukuku

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJRkYO5AAoJELGGH8ff/6zfvI4IALfp48C0/SqiAcjT00KMFaGG
AcBYQNTzdAWhdF7RKYfmEO0nvGmzHhQp9HF3mPKhYmoswrqHBYDHgkWSypforfLV
c1f3406i+G+4v3vhizVe6TN0kvGB2Zw0xXEzT93oDcU0cIl3OEMcXTFL+X2UpoVu
ly22dppGzR0F+UxmhQvjvQmtdVL+W9hQ7bfwBPjBhpV8DekSsHPxtdAYy6udvfNY
+TreB3NEqHYVKDDeFX0idrAYPh/X4xs06GkpCyABuNiHMN+X6qhI4v4sh4qeTWyr
io7bZJpib4e6qwmpmvjZeeGh2uangEF2gLhVs2DzcW1Wa86VCuYE3CnWSwLp1tw=
=NZN2
-----END PGP SIGNATURE-----
_______________________________________________
Netsukuku mailing list
[email protected]
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/netsukuku

Reply via email to