Slashdot Mirror

Then, those organisations have even fewer targets to pick from, so it becomes easier for them to decide which users seem to be contributing most. So then they can stamp on them. And not only do even more people get scared off, but the ratio of contributors to leechers drops badly (it often tends to be that the minority contributes the majority) so those trying to download find much less, and many of the people left sharing stuff find they get even more traffic, probably putting them off further.

As I keep saying, unless you use a file sharing network that gives anonymity, this factor can become an achilles heel. Try GNUnet, if you're using *nix (and follow it's progress if you're using Windows, there'll be a version for you eventually), it's not yet finished, but it already gives you anonymous file sharing, and more is to come!

These points also really apply to Freenet, but IMO, GNUnet is ultimately superior, like for one thing, any user on dial-up or with a not-very-permanent connection and IP is basically a leech on Freenet, whereas GNUnet allows practically anybody to both benefit and contribute (not only in uploading files but also improving effectiveness of the network).

GNUnet has other advantages too, but is still in development- you can use it already, but not on Windows (*yet*), and AFAICT, the protocols aren't set in stone yet so you have to keep your GNUnet programs up-to-date to use it. And if you don't find it good enough for you, just keep coming back every few weeks, as work continues (or you can even help out).

*-I say "claim to be copyrighted", as just because Group X says Person Y is doing something, and has a log somewhere, it doesn't really follow that it is true- I could cheerfully tell the world that Hilary Rosen goes through my dustbin, and that I drew a picture of the alleged event. Wouldn't be true, but it'd be my word against hers (OK, that would be rather unlikely, and I'm in another country. Blah blah). Further, just because Person Y appears to be sharing a copy of There's Something About Hilary Rosen, it doesn't follow that that file is what it claims to be. Why wouldn't it be? Hell, I don't know. But it doesn't have to be. If you've used Gnutella (you say you still are. I for one stopped ages ago), surely you'll have encountered this sort of thing by now.

Thanks for the heads up regarding the differences in traffic analyzability between gnutella and freenet

Not gnutella - GNUnet. Gnutella is not anonymous in any way.

I do not believe they will feel that the evader was the user.

You may have something there.

They don't get prosecuted, because they have not done anything wrong.

Well, yes and no. If they can prove intent, posession of stolen property is a felony in most US jurisdictions. In actual practice, however, the only people who go to jail over this are those who have committed POP (pissing off the police), and were convicted on every conceivable charge because they tried to run over a cop or some such. In most cases, it's used as leverage to get people to identify and testify against the thief.

The trouble is, a screenshot is just bitmap data. It can easily be faked.

Even if the screenshots were real, it wouldn't even mean that they had that actual content, they could be random files with fake names- potentially even created with a virus.

If that sort of thing can be considered sufficient evidence to effectively extort money from people in a court, then you might as well cobble together some image of a program window wtih "I support Al Quaeda and give them money" written in it, and claim that it was gleamed from someone's machine (perhaps by hacking, or that Tempest thing).

In fact, maybe these people being put through the wringer for doing something so minor, should do that, say "They're trying to bring down the western world by prosecuting innocent members of the public!". OK, possibly that's tasteless (not very).

Anyway, ultimately this kind of thing just proves how important anonymous data-sharing networks like GNUnet * and Freenet are- people keep on saying things like "Oh, why would you want anonimity, you must be up to no good"- but as you can see, pretty much anything you might want to share with someone, it is likely that someone somewhere will want to stop you. Whoever it was that created DeCSS would probably be quite content and comfortable, not worrying about prosecution (OK, maybe he was let off, maybe not, I can't keep up with it all), if he'd distributed his initial code via GNUnet or Freenet. And now, so would these people in Denmark.

* - After a few times of trying to get GNUnet to work, I got pissed off at it seeming so impossible to setup properly (because the documentation was too unclear, or there were still bugs, I dunno), and a month ago I mouthed off about it on slashdot somewhere. Since then, I ended up having another try as they released 0.4.7, and tada, I got it to work. And once it was working? Pretty decent. Bit slow, but that's because of the anonimity measures. Not much content there, but the way to improve that is to join in and provide some! Anyone who decides to use it should realise that it's still work in progress, and could well give you a hard time. But, unlike Freenet, it doesn't need Java, and is capable of making use of "transient nodes" (fairly sure) - so if you use it on dial-up, you aren't just leaching -I believe this was why I always found it so hard to get anything on Freenet, not enough permanent nodes. So to sum up, I'm now backing GNUnet big style! Watch out for version 0.5, coming soon apparently.

No, they're not paying me ; )

. . no pun intended . .
: P

Just wondering, because it seems inevitable to me that Freenet and GNUnet are going to be the only place left in the world where independent idea publishing can be, a short few years from now.

How will their differences in design/engineering change their 'base', or
otherly phrased, what portions of our idea-world will live freely in each 'net'?

Also check out GNUnet which is similar to freenet but is searchable and written in C instead of Java. I think GNUnet has more long term potential than Freenet.

Crypto won't hide your IP address but relaying the information through a number of proxies sure will.

The only p2p model that has a remote chance to survive in this hostile world of RIAA omnipotence is the Direct Connect (DC)

What about the GNUnet model?

I've been thinking about this problem for quite some time. My thoughts aren't quite ready for publishing, but I did get quite a bit of information first.

Some links:
GNUNet
This is aimed at a being an anonymous backbone, of sorts, but is currently being developed more towards the P2P file sharing aspect. That said, the papers on their website are fascinating.

Freenet
This is more of an anonymous content publishing network. A partial solution.

Please reply to tell me if these are what you're looking for.

The only fight we have is in politics, there is no technical solution to this problem. As much as you would like to think you'll win this battle whipping up some code in C, you are going to find there is nothing you can code that will keep the handcuffs off of your hands.

I wouldn't be so sure. This is what systems such as trust-network, anonymous networks such as GNUnet are supposed to resolve. As long as we're allowed to have general-purpose computers, open networks, and good bandwidth, I think technical solutions can stand up. If we don't have general-purpose computers, or open networks, we've got other issues. Destroying our bandwidth is probably one of the few non-immoral attacks that can be effected, but an attempt at doing so likely won't succeed given the average persons's desire for it (for whatever reason).

I'm leaning towards the idea that the politics will change, not because we affect today's politicians, but because the up-and-coming persons of society are being conditioned differently (supporting things such as Napster). We might just have to out-live the current generation of politicians.

Keep in mind I'm not certain that a new cycle of politicians will help. Greed and powerlust is ageless.

Don't forget GNUnet -- read their docs. A different approach, but it looks nice. It is at version 0.4.5, but when I have time (hopefully soon) I will hook up a node here. Lots of mp3z,some moviez,some other cool stuff. GNUnet

There is a P2P network layer called GNUNet. I've studied the papers on it, and the design looks extremely solid and resilient.

Take a look at GNUNet. I've been studying it for a while now, and it looks extremely good. (Not perfect, but pretty damn close.)

Freenet is nice but, GNUnet is better suited for completely anonymous and distributed file sharing.