Slashdot Mirror
New Kazaa Lite Protects Identity
Denver_80203 writes "Found this story about the new Kazaa K++ 2.4.0 and it's new sister program which claim to protect your identity while sharing files. Any of you folk know how legit this could be? We all knew it wouldn't be long... is this the war or just another battle?"
I don't know if Kazaa K++ can hide your identity, but what I do know is this: Kazaa K++ is an excellent program. It is so much better than vanilla Kazaa. No ads, spyware, many cool features make it a great program.
#include "sig.h"
mldonkey is pretty good and has Fast Track (meaning Kazaa) support.
Please forgive me if I'm wrong, but UDPP2P does not seem to be "promising".
I've checked the web site. It basically says "we broadcast all the queries and if someone has the file we meet each other by using secret codes hidden in those queries".
A peer-to-peer network that does queries in terms of network-wide broadcast is always doomed to fail. Gnutalla failed (and was redesigned) the same way. Even Novell NetWare was unable to scale because of SAP (service advertising protocol).
Nevertheless, the web site says "peers will somehow know each other". This is also a big problem in P2P networks. -- No design only big words.
Anyways, if I were you, I'd use freenet. It's anonymous, and it works much better than the scheme explained on the web site.
May I point you to giFT-FastTrack?
Only law enforcement agencies can be accused of entrapment. There's no such thing for a non police corporation. They can entrap all they want. Remember, you're going to be going to civil, not criminal court.
--
RumorsDaily
It aint gonna work. The reason is simple : The rules have changed. Distribution of music is now much easier and cheaper than before and a large chunk of the old distribution network is *no longer necessary*. This is totally irrelavent as to weather or not this new distribution model is legal or not. It is happening. It probably cant be stopped(I mean the software industry tried and failed thru the 80s/early 90s)
So now the RIAA have several choice.
1. Try to roll back the technolgy that enables this new distribution channel. This is possible but not very likey.
2. Use more draconian law enforment techniques. Posibble but I mean whata ya gonna do... start sending colleage kids to prison ? For what stealing a Brittney track ? Is this what we want ?
3. Try to adapt to the new medium. Be creative and come up with new profit channels that take advantage of the medium.
Personally I dont think 3 is very likely either... I think RIAA is going to have to be dragged kicking and screaming into the 21st century.
This is a really good idea. You can extend it to make it even better though.
Part of the good thing about the erasure-correcting code approach is that if you use a big enough very low-rate code (although its quite tricky to do that with good CPU and memory efficiency) then you can have downloading from several servers concurrently without having to tell each server which parts of the files you want (just send random parts of the encoded data and theres a low chance of overlap from multiple servers).
Now, here's the clever part: you use IP Multicast with multiple sources spoofing the same sender address. This means that (a) you save quite a lot on bandwidth since many P2P clients will be downloading the same source file (this is important since a big reason many ISPs and Universities have banned P2P is the bandwidth); and (b) it is MUCH harder (not impossible, but hard enough if you are not an ISP or a router at the very end) to find out who either the source or the destination is.
I don't know if anyone has thought of this idea and tried to implement it. Someone should; maybe I'll give it a go when I have time.
PS. There is a sparser and more CPU-efficient solution than VanderMonde matrices, look for Low-Density Parity Check codes.