Comments on: ‘Guaranteed’ anti-file sharing

By: Reader's Write

Reader's Write — Thu, 21 Apr 2005 04:16:34 +0000

Sure, we’re moving more into trusted network only sharing but really who cares? The point is, they’re forcing everyone to make p2p progress in develop 1000x faster than it would have without the millions of dollars of tv advertising they’ve given by attempting to talk negatively on TV as much as possible… instead they’ve kept the p2p fires burning…

Block all traffic except to port 80. Charge an extra fee to access a website not registered with a registrar (no home ip ranges unless a registrar has pointed to that ip and digitally affirmed through their own review that the site linked to is not a p2p hosting site)…

If you are not allowed to access any ip except a white-listed ip without paying an extra fee (business line with business reasons) then you basically would have surf access only internet..

Now ask me if I think people would still pay for that - at broadband prices… um, not likely.. for plain ol web surfing I’ll pay $4.99/mo.

lol, anyway - it’s pointless to try and stop it now…… people are already setting up wireless undernet’s without monitoring BS around it….. the internet was only the beginning….

Just my 10 cents.
_-Jile-_

By: Reader's Write

Reader's Write — Thu, 21 Apr 2005 03:31:54 +0000

It won’t work though.

Bittorrent has hashes for all of the blocks of a file.

Unless they managed to create a file with each of the individual blocks also matching the checksum with each stil beingl garbage (which is nigh on impossible) I don’t see this working for BT. Not to mention the files would have to be tweaked depending on the torrent.

This may affect Kazaa but won’t affect anything that using a decent checksum algorithm.

By: Reader's Write

Reader's Write — Thu, 21 Apr 2005 02:18:09 +0000

The problem is that a hash is funadmentally about mapping a large space (the size of your file) to a smaller space (the size of the hash). If you have a file size ten thousand times the size of your hash, you’ll have 10k other files with the same hash (on average, depending on the algorithm). The idea, generally, is to create a hash large enough that it’s “computationally infeasible” to generate a file with a matching hash, and to use a well-designed algorithm, to ensure that the output truly is “one-way”, e.g. the algorithm actually uses the whole bitspace. The last part is usually the tough one — most hashing algorithms have weaknesses that mean an attacker has to try several orders of magnitude fewer files to find one that works.

The only constraint is that generating the hash shouldn’t take too long, though “too long” is of course very subjective. Both generation and cracking times are usually tied to the hash size, so it’s a good idea to increase hash size from time to time as computing power grows — your enemy needs more horsepower to crack it, and while you need more cycles to calculate it, you have them, thanks to Moore’s law.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 19:53:19 +0000

One interesting thing that I want to point out about swarming techniques is that those options cannot be used with BitTorrent. BitTorrent uses a .torrent file that has the checksums of all blocks in the file. Even if fake blocks were sent, they would be sumarrily dropped by the protocol.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 13:46:48 +0000

Smells like crap to me, the technology supposedly overwrites the hash codes which would depend entitely on sharing enough dodgy files which would mean hosting the files and enough of them to overwrite the existing files on the system - someone came up with this idea over 2 years ago - it was crap then - crap now!

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 07:05:04 +0000

Q: What is a collision attack and a preimage attack?

A: A preimage attack would enable someone to find an input message that causes a hash function to produce a particular output. In contrast, a collision attack finds two messages with the same hash, but the attacker can’t pick what the hash will be.

http://www.cryptography.com/cnews/hash.html

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 07:00:18 +0000

Good read on the subject:

http://www.ciphersbyritter.com/NEWS5/CRCHASH.HTM#1999Apr1616.01.06.26203@koobera.math.uic.edu

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 06:51:15 +0000

So, the way Viralg is doing it is by generating fake/garbage file segments that will result in the same hash code as the original file segments when hashed.

This means that the hashing algorithm currently used is not capable of guaranteeing a unique hash for any oparticular file. In other words, it is possible that two files will have the same hash ‘accidentally’.

Apparently Viralg is capable of taking a hash and reverse-generate a garbage file that will have thte same hash when the hashing algorithm is used on it. They figured out how to do intentionally something that we knew theortically could happen by accident.

The obvious step for us to take is to find out whether it is theoretically possible to create a hash that cannot be broken - in other words, an algorithm that guarantees a unique hash for a file. I’ll go google around on this subject…

If such a hashing algorithm is possible (and may even exist already) we will have to switch to it. If it is not possible, we will have to come up with a stronger hashing method that cannot be broken as easily (or at least not for a few years

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 06:18:59 +0000

Security thru obscurity it seems… Soon enough, it’ll be gone or dealt with, as with essentially every other technical measure.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 05:45:04 +0000

I can think of numerous avenues of attack this uses, but can’t figure out which… also their site has a distinct lack of technical data.. i can’t seem to find it.

To me this means one of two things, either theyre scamming record companies or theyre actually intelligent enough not to give potential p2p debuggers/crackers clues as to how to defeat it.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 05:28:53 +0000

Even if it works as advertised, i give it 1 week of effective use before the first countermeasures are adopted… 6 weeks at most for the vast majority of p2p apps to leave it in the dust.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 04:45:15 +0000

according to slyck, it hasn’t been tested in BT.
http://www.slyck.com/news.php?story=750

i mentioned in early march that the breaking of the hash-1 BT code wasn’t just an accidental experiment or w student prank.

i got a lot of lip from people on that.

now you can see what has been going on behind the scenes.

it is still the ultimate goal of the New World Order and th’r mainstream sheeple-herding news orgs to control how and where we receive our news and information.

it’s a safe bet that viralg will be starting BT testing soon. and once that begins - and if they’re successful - the BT community and developers will have to find new ways to filter/block fake/corrupted files from accessing trackers.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 04:18:55 +0000

One of the things that the RIAA has said over and over is that filters should be installed in p2p applications. What is left out here is the enormous little detail. No filter to date has shown that it can be effective in distinglishing one mp3 from the next. Part of the problem with this has been that the majors are extremely reluctant to set up any sort of database that could be used to say this is an owned work. To do so would also lead folks figuring out just how much music they do own. Very possibly it is not the 70-90% claimed owned. It is very easy to say “someone should” but without the necessary data to do so, no one is going to pull that off.

I would suspect that Loudeye would be one of those that should be held to these law standards for virus and trojan sources as they were the ones to first seed p2p with fake files and recently announced that they would again be seeding wmv and wma licensed files with ads and trojan setup. The idea was that when the license isn’t found that normally the media player is setup to go to a license site for you to purchase a license. Only this one goes to load you with ads and the trojan.

I believe that this will either lead to ip blockers filtering out the identified ranges of those wishing to distribute such or that the hash alogthrims (SP?) will be strengthened to the point that such interuptions won’t be possible. A third solution would be not to accept any bits or bytes from other computers than the source that held the file you selected. This would be hard for applications that use swarming techinques but easier for other applications to avoid.

Professional filesharers? Isn’t that those like iTunes and Realnetworks that have that unconstitutonal DRM in them? Why unconstitutional? Simply, copyright laws spell out how long such can be renewed. DRM has no provisions to turn off a any date. Therefore they at some point will violate the provisions.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 03:25:05 +0000

Interesting that. I wonder how they plan to be able to distiguish “illegal” files being shared as opposed to legitimate ones? If they plan to just blanket-bomb the P2P networks, then it strikes me that legal action would be more than warranted.

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 03:05:12 +0000

I found this:

“The virtual algorithm is capable of mixing together files in P2P networks in a way that the illegal downloader will end up downloading useless garbage instead of real music, movie or game content. Our DRP technology is capable of destroying already shared functional files from peer to peer networks. ”

http://www.kauppalehti.fi/4/i/eng/releases/press_release.jsp?selected=other&oid=20050401/11135655917470&lang=EN

I wonder if the ‘Destroying’ part means they can change a file on my HardDisk… I would so slap them with a lawsuit

By: Reader's Write

Reader's Write — Tue, 19 Apr 2005 02:55:14 +0000

Hmm, this is either a scam or potentially illegal as hell. The Viralg site reveals absolutely nothing about the technology itself. I am guessing here, but smells either like a DOS attack (illegal), distributing fakes (big deal), or planting things on the users’ computers (trespassing).

Any further deteails on how this Viralg thingie works?