OFF Demo Disk Released

[Lord_of_the_Dense]

We have got a Demo of OFF.

It does not have network capability yet but you can store files in it and retreive them out.

Our demo has a pile of free music on it (nothing illegal) and it shows the power of OFF.

Get a copy at our Sourceforge project here:http://offsystem.sourceforge.net/

For help please post in our forums here: http://thebighack.org/modules.php?op=modload&name=phpBB2&file=index

[aqlo]

So what do you guys think about this? Some "real" code from out of the "fictional" conspiracy to destroy digital copyright?

In the weird world of The Big Hack this system is supposed to be death to the music industry, or at least the incompetent parts of it whose revenues depend on their attempt to copyright numbers.

But at SourceForge it appears to be just another method of storing files. The people who made the demo have gone to a certain amount of trouble to make sure there is nothing copyrighted on it. (Not to say indies aren't great, and those who are willing to share the greatest.)

But is this everything the audience wanted it to be? Is truth ever really even better than fiction? Stranger maybe, but better? You tell me.

[cpugeniusmv]

interesting concept. it would be interesting to see if it stood up in court.

[Stellar]

So what do you guys think about this? Some "real" code from out of the "fictional" conspiracy to destroy digital copyright?

In the weird world of The Big Hack this system is supposed to be death to the music industry, or at least the incompetent parts of it whose revenues depend on their attempt to copyright numbers.

But at SourceForge it appears to be just another method of storing files. The people who made the demo have gone to a certain amount of trouble to make sure there is nothing copyrighted on it. (Not to say indies aren't great, and those who are willing to share the greatest.)

But is this everything the audience wanted it to be? Is truth ever really even better than fiction? Stranger maybe, but better? You tell me.

Why aren't you thrilled? This is a big step in a huge battle.

So it's a demo, and it works. If it's not everything it could be yet, it's the best chance I'm seeing to really protect our rights from those RIAA bastards who like to sue little kids.

I think this is really fabulous. WAY TO GO guys, bring us a final version soon.

[wonderboy2005]

ok... i read the info page, and it all seems promising. however, im not going to get the demo - there's not point for me. i do have a few questions though

1) say i have song A, which OFF has split up into numbers A1, A2, A3, A4, etc. Lets also say i have song B, made up of numbers B1, B2, B3, B4, etc. now, given some information from an outside source (likely on a network), will i be able to compile numbers A1, B3, B2, A4, etc. to make song C?

or is this just a clever distrobution method of splitting up a song so that the individual parts are meaningless?

2) is network/filesharing in OFF's future?

3) assuming a networked future for OFF, will download speeds be comparable to other FS apps? or will it be slower due to the complexity of the app?

4) since OFF relies on mathmatical equations, would it be possible to shrink the file size of a song, or will using equations increase the file size?
ex.

equation = 9^9 = 387420489 ; notice the equation uses only 3 characters, while the number the equation represents uses 9 characters

-or-

123456789+123456789 = 246913578 ; notice that the equation uses 19 characters, where the number it represents uses only 9 characters

hope that makes sense...

[Sephiroth]

The whole idea is based off a fictional movie which is why i think they ought to call it, The Big Joke. So their crazy scheme might sound good and they might make alot of big promises but that doesnt mean it will really work.

You can split files up however you want but if you can search a network then can anyone else because its whats being served not how its being served that is important.

[l33ts0n]

1) It is possible, yes.

2) Yes.

3) Presumably. Node speed is always a variable (if bandwidth were infinite...).

4) I'm starting to regret crackerjack ever brought up the numbers anology. Its a precise anaology of how OFF works, but... eh. Can someone else from TBH with more time on their hands field this one?

You can split files up however you want but if you can search a network then can anyone else because its whats being served not how its being served that is important.

That is exactly the issue The OFF System addresses, Sephiroth. If I give you a file that is both The Bible and a Britney Spears mp3, how can you claim to hold copyright over it, or even go so far as to claim that I'm going to use the "infringing" part?

[Otto42]

4) since OFF relies on mathmatical equations, would it be possible to shrink the file size of a song, or will using equations increase the file size?
ex.

equation = 9^9 = 387420489 ; notice the equation uses only 3 characters, while the number the equation represents uses 9 characters

-or-

123456789+123456789 = 246913578 ; notice that the equation uses 19 characters, where the number it represents uses only 9 characters

hope that makes sense...

Shrinking the file isn't possible given the scheme being used. And there's no point in trying to make it compress the file anyway. Firstly, MP3 and other song formats in use are already pretty darn compressed. Secondly, there's plenty of compression programs around already. If you want to compress something before inserting it into the network, then that's easy enough to do.

Increasing the size is quite likely, sort of. First, you need to understand that when you're getting a file off the network, for every block of data that you get, three blocks are required. So the maximum size is basically 3x the original size. However, the blocks are multi-use, so you may already have several of the blocks needed for the file and won't be downloading them, as well as the fact that multiple blocks in the file may use the same block in different sections.

Inserting something into the network *always* creates blocks of the same size as the file itself is, since it's using preexisting blocks to create the new ones (the "multi-use" part). Pulling something out of the network will require at least 4 blocks more than the original file size, and potentially up to 3 times more than the original file size. This is why you want the node on a box with a lot of bandwidth.

But the network portion isn't totally worked out, really. This CD demos the concept by showing that multi-use blocks can produce files and that you can store files in such a node. It's a demo of the concept, not of the networking side of things.

[method]

As long as the OFF system doesn't transmit a full mpeg video frame, you may be alright. - If it does you'll be breaching copyright even with just that! :/

It's an interesting idea though.

[MonkeyMadness]

If I understand the idea here, the problem will be with the search function. This is a problem FreeNet has. You have to know the "key" to retrieve the material. The schemes necessary to protect anonymity and deniability (distributed, encrypted, chopped up file bits, and encrypted retrieval keys, and distributed search functions) make searching impossible... at least without compromising the anonymity and deniability.

[shawners]

lol.. i would of named it "on".. It needs a cool name so 14 year olds would use it =).. Like OFFazza.

[FreakinWeasel]

lol.. i would of named it "on".. It needs a cool name so 14 year olds would use it =).. Like OFFazza.

LOL shawners, how about I get my music OFF the net? or OFFnet?

[Otto42]

If I understand the idea here, the problem will be with the search function. This is a problem FreeNet has. You have to know the "key" to retrieve the material. The schemes necessary to protect anonymity and deniability (distributed, encrypted, chopped up file bits, and encrypted retrieval keys, and distributed search functions) make searching impossible... at least without compromising the anonymity and deniability.
This is true, but at the same time, indexing sites, ala ShareReactor, are becoming more and more popular. With these, the key could be distributed and searching added on that way.

But you're essentially right, it won't have any search function. At no time does the network contain copyrighted data, or even the filenames, in fact. There's nothing in which to search for anything. Every block in the whole thing is essentially "random" garbage... it's just random garbage with interesting properties. :)

But hey, look at the popularity of BitTorrent. It has no search functions either. It doesn't have a constantly "up" network, and "file existence times", sort of thing, can be measured in days or weeks, at most. It's still pretty popular.

[Sephiroth]

That is exactly the issue The OFF System addresses, Sephiroth. If I give you a file that is both The Bible and a Britney Spears mp3, how can you claim to hold copyright over it, or even go so far as to claim that I'm going to use the "infringing" part?

There is numerous problems with that like how someone is suppose to get their completed file from two seperate files. Because if any one on the network can make the file they requested from any parts then it accomplishes nothing.

That it will waste a huge amount of bandwidth and resources wasting 3 times more bandwidth per file. Plus the actual content of the dummy file becuase i dont think some people will like that their file will get mixed up with porno and etc.

So i dont think that it offers any kind of protection and that the system is so wasteful that it will never work because why would anyone in their right mind waste three times more bandwidth and time to get the same file that is on other networks for supposed security that is unproven and was taken from a movie.

Which to others who want to know one of people who is behind this and has posted here is Captain Morgan.

[l33ts0n]

So i dont think that it offers any kind of protection and that the system is so wasteful that it will never work because why would anyone in their right mind waste three times more bandwidth and time to get the same file that is on other networks for supposed security that is unproven and was taken from a movie.

Because it is 100% legal to transfer blocks over OFF. No one but you, yourself, know what you are assembling those blocks to be. Yes, it is very possible that any given multi-use block you download could be potentially used to reconstruct child pornography. Or a digital sketch of The Mona Lisa. Or a videotape of your parents-in-law having sex.

This is the concept of a brightnet: to prove that nothing illegal is going on by transferring files through the network. The concept of a darknet--FreeNET, MUTE, for instance--is the exact opposite; they seek to disguise what's really going on in the network, because they are doing "illegal" things.

The primary objective of a brightnet is not to obfuscate, because by its inheritent nature everything on the network is 100% legal, as ownerless as a block of data produced by this pseudo-code:

for(int i = 0; 1024 * 128; i++)
OFFBlock[i] = rand(255);


Who do the results belong to? Nobody. Utterly random (well, hopefully).


To explain the concept of OFF in more simple terms, think of a bucket of building blocks: they are all the same size, shape, weight, and uniform in every sense of the word. They are ownerless, and by themselves, they are worthless.

Now, assume you were to build a "structure" with these building blocks. This structure now belongs to someone--consider it akin to a file on a computer, copyrighted. In this form it is easily identifable as being someone (your's) work.

However, assume we take a sledgehammer to the structure, and depreciate the contents back into their raw state--now they are nothing more than a pile of uniform blocks scattered all over the floor. The blocks are no longer a recognizable structure, a copyrighted file!

To take this another step, assume that I pick up these scattered blocks... and assemble the greatest, most fantastic, almost orgasmic recreation of the Battle of Buffalo. The blocks now have two meanings.

This is called "multi-use encoding." Every block I used now has another meaning: they can be used to construct the first structure (copyrighted file), or they can be used the Battle of Buffalo.

Why is this relevant? Because you are no longer able to definitively single out any block I used and say, "The only use for this block is to recreate a copyright infringing structure/file!" because the very duality of its nature it belongs to two seperate functions.

But what if I destroy the Battle of Buffalo and construct an entirely different structure? And then repeat myself, ad nasuem?

This is what OFF seeks to do--through the fascinating wonders of multi-use encoding every block stored in the cache will be given 2, 3, 4, 5, 6 (!!!) meanings. It will become impossible to point at any given block in the network and claim that its only use is for reconstructing a copyrighted file.

To complicate matters, there are no searching functions in OFF. It works similar to Freenet: you supply a key to your "private" OFF node, which will dissemiate the information and download the constitutent blocks without alerting other nodes what file it will be constructing.

Someone above mentioned the flaw of inflating the download size by 3 (three constituent blocks to recreate one original block). You're right! However, OFF has been designed so that it can be hosted on a webserver. Essentially, you host it on your "free" web account, provided by your ISP (aww, they're so nice). Their phat-bandwidth handles the brunt of downloading the inflated file, while the end-user gets the entire file--no more, no less. Crafty users will sign up for one of those $9.99 web hosts on the internet and rape their bandwidth for all its worth.

To the end user, they simply click a link, a few seconds later the web server is streaming back the entire file to them.


What happens if someone wants to abuse this system and start sharing copyrighted content?

Well, that's up to them.

[tMoD]

Because if any one on the network can make the file they requested from any parts then it accomplishes nothing.

So if you share a couple of different encodings of the U.S. Constitution, you commit copyright infringement because the files could be XOR'd to produce part of a Britney Spears song? To use CJ's analogy, if Britney owns 12 then she must own 5 & 7 (5+7=12), 35 & 23 (35-23=12), 6 & 2 (6x2=12), 36 & 3 (36/3=12), and literally every number in existence because it's possible to combine them in such a way as to produce 12. If that's the case, then all data transfer on the internet is copyright infringement.

And if it's only the assembly of the file that's illegal and not the trading of random numbers then how is the copyright industry going to do anything about it? How do they know what you are doing with the blocks that you've downloaded unless they illegally hack your computer?

[aqlo]

Why aren't you thrilled? This is a big step in a huge battle.

So it's a demo, and it works. If it's not everything it could be yet, it's the best chance I'm seeing to really protect our rights from those RIAA bastards who like to sue little kids.

I think this is really fabulous. WAY TO GO guys, bring us a final version soon.
It's not that I'm not pleased. And if these guys really took the gibberish that was being spoken by that Stark Ravin character and made a sensible usable program out of it, I'm stunned and a little amazed. And if they skipped that chore, as I suspect, and went back to the basic concept made familar here by Captain Morgan, I'm even more amazed (though hardly stunned.) But will it do any good?

I understand that it is just a demo now, once it has network capability we will be able to see how well it works in practice. But just in the demo stage, without even the bother of downloading the demo CD itself, we already have people dead set against it. How can something destroy digital copyright when it can't even get a friendly reception for free music from people who love free music?

Cpug sort of hit the nail on the head too, it's a great theory but how will it work in court? Admittedly it's much the same theory that Kazaa used to protect themselves in court, and that worked fine for them even though no one here believed it. But they were a big corporation, and we will just be ordinary webmasters hosting these random blocks. Sure, the DMCA is supposed to protect us as a "safe harbor" for things we host without our direct knowledge. And the ISPs are making it harder and harder for the Copyright Industries to bully the small user. But will it be enough?

I went ahead and did more research about the 3 times bigger thing, this isn't as bad as it sounds. Sure, the first few songs or whatever that you construct will take longer, but as you acquire and share more and more blocks there is an ever-increasing likelihood that the blocks for new "downloads" will already be available to you. This acts as sortof built-in incentive to share, to maintain a large node, but once you do your songs and movies (and bibles) will just appear magically from your own cache without even a lot of network activity.

All this seems to put the burden back on the downloader for whatever files they construct however. Will we end up in a situation where people really are getting sued for downloading? Or will the people sharing these "uberhashes" that make the reconstruction of the prohibited material possible be the ones held responsible? I don't know what the precedents are to protect downloaders and hash-hosts, but we do know people get in trouble for sharing to some extent, even if the whole process hasn't been adjudicated yet. FreeNet is a good example of what might be able to go wrong with a system like this, I have never trusted the idea that the host will be safe just because they don't necessarily know what they might be sharing.

I hope a lot of you quieter people are actually downloading this demo and making judgements based on what it really does. I have looked it over and I can't find anything wrong with it, but the fact that we already have a storm of emotion against it based on some sort of "fiction" label makes me pretty alert.

By the way what movie are you talking about Sephiroth? I thought this overall dealie was based on things like Opus Dei and Paranoia, I never heard of a film reference so far.

[Sephiroth]

Because it is 100% legal to transfer blocks over OFF. No one but you, yourself, know what you are assembling those blocks to be. Yes, it is very possible that any given multi-use block you download could be potentially used to reconstruct child pornography. Or a digital sketch of The Mona Lisa. Or a videotape of your parents-in-law having sex.

This is the concept of a brightnet: to prove that nothing illegal is going on by transferring files through the network. The concept of a darknet--FreeNET, MUTE, for instance--is the exact opposite; they seek to disguise what's really going on in the network, because they are doing "illegal" things.

The primary objective of a brightnet is not to obfuscate, because by its inheritent nature everything on the network is 100% legal, as ownerless as a block of data produced by this pseudo-code:

Who do the results belong to? Nobody. Utterly random (well, hopefully).

To explain the concept of OFF in more simple terms, think of a bucket of building blocks: they are all the same size, shape, weight, and uniform in every sense of the word. They are ownerless, and by themselves, they are worthless.

What happens if someone wants to abuse this system and start sharing copyrighted content?

Well, that's up to them.

P2p programs have been splitting files up and transferring them across networks for years its called multi-source. People can choose to abuse current systems and share copyrighted material and it is up to them.

So this doesnt provide any kind of security at all. And just because they can invent a few buzzwords doesnt mean that its no different than other network. Because this is disguising the actions of people on the network. Just not in the same way as others.

There has to be some way that someone else can determine what chunks to download in order to get a completely file. They dont appear out of thin air and no one wants random bits of random files or corrupted files. The overhead for this overglorified PAR system would be tremendous and would provide zero benefit because no one cares how the files are transferring and those who have the hashes to construct the files is no different than if they were hosting the file themselves.

Alqo as for where i found out this whole thing is based off a movie/book is from their offical FAQ located here. (http://www.thebighack.org/modules.php?op=modload&name=FAQ&file=index&myfaq=yes&id_cat=2)

[tMoD]

All this seems to put the burden back on the downloader for whatever files they construct however. Will we end up in a situation where people really are getting sued for downloading? Or will the people sharing these "uberhashes" that make the reconstruction of the prohibited material possible be the ones held responsible? I don't know what the precedents are to protect downloaders and hash-hosts, but we do know people get in trouble for sharing to some extent, even if the whole process hasn't been adjudicated yet. FreeNet is a good example of what might be able to go wrong with a system like this, I have never trusted the idea that the host will be safe just because they don't necessarily know what they might be sharing.

It's kind of hard to imagine someone being sued for downloading a block that could be part of any file. There is no way of knowing what the block being downloaded is being used for since, in practice, it will be part of many different files and, theoretically, part of any file. Since OFF can be used as a simple distributed file storage system, the block can easily be part of an e-mail I wrote that only I have the key to. It can also be part of a public domain file. Furthermore, not only do you not know what the blocks are used for but the blocks themselves have substantial non-infringing uses as they can be used in the encoding of private files and public domain works. This goes back to MGM v. Grokster and even The Betamax Case where the courts ruled that you can't be held liable for copyright infringing misuse of something that also has substantial non-infringing uses.

As far as those who host the hashes, text URLs at the very least appear to be safe. 2600 magazine was sued for posting links to the DeCSS code on the basis that a URL with an <a href> tag is a "device" as it causes your computer to execute an operation. However a simple line of text doesn't do that and is not considered a "device". 2600 has had no problems since changing their hyperlinks to text URLs.

[l33ts0n]

P2p programs have been splitting files up and transferring them across networks for years its called multi-source. People can choose to abuse current systems and share copyrighted material and it is up to them.

So this doesnt provide any kind of security at all. And just because they can invent a few buzzwords doesnt mean that its no different than other network. Because this is disguising the actions of people on the network. Just not in the same way as others.

There has to be some way that someone else can determine what chunks to download in order to get a completely file. They dont appear out of thin air and no one wants random bits of random files or corrupted files. The overhead for this overglorified PAR system would be tremendous and would provide zero benefit because no one cares how the files are transferring and those who have the hashes to construct the files is no different than if they were hosting the file themselves.

Alqo as for where i found out this whole thing is based off a movie/book is from their offical FAQ located here. (http://www.thebighack.org/modules.php?op=modload&name=FAQ&file=index&myfaq=yes&id_cat=2)

You seem to have misunderstood how OFF works, despite my elaborate and well-meaning explaination.

OFF != Multi-source.

OFF == Multi-use.

There is a difference between "splitting up files" and multi-use encoding.

But I'll try explaining it one last time...

You have a block. It is 128KB. It is ownerless--random noise, if you will.
Unowned, you are freely able to trade this block, delete it, or copy it at random. Now, assume there is an entire network of blocks like these.

Let's say you download several of these files, and someone has given you the key to assemble The Constitution of the United States of America (an ASCII-based copy. ;p). Since the key is only used for assembling blocks, it will not be propogated through the network. All the assembly occurs locally. With a click of the mouse you know how an ASCII filed names 'Constitution.txt'

That's great.

Time passes. One of your friends gives you another key. How interesting!

<extremely dumbed down version>
You click your mouse, and your OFF cache churns a bit in the background. It notices that although you have most of the blocks to assemble, you are missing a few. With a simple HTTP GET request, it canvasses the list of known OFF Caches and asks each one for the missing blocks. Blocks that could be used in anything--bible, redhat ISO, screensaver. And so, you download these precious few blocks, and then... using the blocks in your cache you have assembled 'Happy Hour at Hardcore Debbie's.'
</extremely dumbed down version>

That is multi-use encoding. Transforming one file to be used in more than one file. Legos, man. Building blocks. Ownerless. Not "partial file sharing" in the aspect that a file is just gutted into several smaller pieces.

Please reread my previous explaination(s) more carefully if you still cannot grasp the concept of multi-use (NOT MULTI-SOURCE) encoding, or read CaptainMorgan's explaination at http://www.thebighack.org/modules.php?op=modload&name=phpBB2&file=viewtopic&t=669 if you think you're tech-headed enough.

[Otto42]

P2p programs have been splitting files up and transferring them across networks for years its called multi-source. People can choose to abuse current systems and share copyrighted material and it is up to them.

So this doesnt provide any kind of security at all. And just because they can invent a few buzzwords doesnt mean that its no different than other network. Because this is disguising the actions of people on the network. Just not in the same way as others.

There has to be some way that someone else can determine what chunks to download in order to get a completely file. They dont appear out of thin air and no one wants random bits of random files or corrupted files. The overhead for this overglorified PAR system would be tremendous and would provide zero benefit because no one cares how the files are transferring and those who have the hashes to construct the files is no different than if they were hosting the file themselves. [/url]

Well, here's the thing.. Say you're running an OFF node on the network. All communication between that node and other nodes won't be transmitting actual copyrighted files, just blocks of essentially random data. So someone looking at which blocks any given node is transferring won't be able to reconstruct which files you're downloading. The problem rapidly becomes too large for analysis.

Now, this doesn't exempt the individual downloader from committing the act of copyright infringement if he downloads copyrighted files. The law doesn't much care about methodology in that sense, you get the file, you did the deed. However, they've been going after people sharing files, and it's not only impossible to say what files any given OFF node is actually sharing, it's quite possible to say that any given node isn't actually sharing any files at all. Just blocks of randomized data.

It's true that a theoretical XXAA could go after sites hosting links to find the content, but places that regularly do this (ShareReactor, etc) seem to still be up and running. And the keys themselves look like just chunks of text. An example would be "a1617ed0fff34c434e118a0ead73807e7e795aa6/
5f93aac64309573ae0753063c5f55cff52e99e6a/
8c3f6d936b380bf9418fbaa18d3bd9dae6e164f0/
63d243a7379fd82ad342f600d67abd0418c3a24f/
37bca874270f9f80fa3b98953aee29e8ea61fceb" which is the key to reassemble a specific file (in this case, one of the html files that is on the demo CD). In and of itself, it's pretty meaningless. It's possible to linkify this for conveinence reasons, or to use javascript code to create a link to your own OFF node at the time you click it (the preferred method for such an indexing site to use, for various reasons).

I mean, it'd be hard to argue that distributing a block of random looking text that is relatively small is copyright infringement. Especially when a similar form of the key with a few changed numbers produces some completely different and unrelated file. And take, for example, magnet: links used by all sorts of programs. It's a standard, of sorts, for creating links to specific files on P2P networks. Are magnet links illegal?

Really, it's an (unreliable) distributed storage mechanism. You can stick files in, you can pull files out. Files are broken up, made randomish, mixed with other files, and distributed throughout the network. Getting a file without the key is essentially a needle in a haystack problem, since even the key blocks needed are multi-use blocks.

Sure there's overhead. That overhead reduces in some cases, but the main idea is kinda like Freenet except without the encryption. Freenet attempts to accomplish anonymitiy by using hard encryption methods. OFF attempts to accomplish a similar goal without the hard encryption or the need to obscure anything. Copyrighted data cannot be transmitted over the network, because all data transmitted over the network is non-copyrightable. You can put a file in, and take a file back out elsewhere, but you have to have the key to do it. Transmitting the key to others to facilitate copyright infringement may be against the law, but there's a lot more evidence to suggest that doing that more defensible than transmitting the files themselves. I mean, you can use Freenet for copyright infringement too. Insert a file into the network, pull a file out of the network, but you have to pass the key along to do it, yeah?

In the end, you're right, methods don't matter. Copying the file is copying the file. But the legal system isn't so clear cut, and by establishing a way to transfer a file without actually transferring the file, it brings the issue to the forefront. Are they going to ban transmission of random uncopyrighted data? Are they going to ban transmission of text based keys? The legal aspects are not that well defined to be able to say what will happen with any form of certainty.

[cheapprick]

I'm wondering in what way this beats freenet. Speed?

The legality approach is very similar.

[l33ts0n]

I'm wondering in what way this beats freenet. Speed?

The legality approach is very similar.

Freenet's slow-ass proxy routing or downloading ~300% of the original file size with OFF. Take your pick: they're both the devil in disguise.

The issue of legality is actually quite different. Freenet seeks to obscure illegal activity with proxies and encryption (something it does very well). The OFF System makes all content transferred between nodes inheriently legal, shifting the issue of copyright infringement to anyone who constructs infringing content.

Unfortunately, its quite impossible to tell who reconstructs a file unless you are physically (or remotely, via trojan) watching their machine, due to the multi-use / many-to-one encoding methods. The demo CD 'Shock' has a map in which it graphs out the uses for every block in the cache for the extra curious.

[cheapprick]

They are similar in the sense that only the end user knows what has been downloaded. To every one else in the chain there is no indication what person x has downloaded.

Your answer was somewhat vague. Would this have any speed/time advantage over Frost on freenet?

[l33ts0n]

Your answer was somewhat vague. Would this have any speed/time advantage over Frost on freenet?

I don't know--I only used Frost once on Freent, a long, long time ago, before they introduced NGRouting, so I'm not qualified to make a direct comparison.

There is no need for proxying in OFF because it is a brightnet, so I suppose the typical 'you download as fast as they upload' axiom applies.

The Spiced Rum guy would know more about this than me, but the pussy never ventures out of his forums.

[Otto42]

They are similar in the sense that only the end user knows what has been downloaded. To every one else in the chain there is no indication what person x has downloaded.

Your answer was somewhat vague. Would this have any speed/time advantage over Frost on freenet?

Well, that's not an easily answered question. Freenet does a lot of things.

Hard encryption, which takes some processing time to do, isn't done in this system. XOR is much faster than real encryption, but there's the size disadvantage.

Proxy routing, meaning that a packet can travel between several machines before getting to where it's going, is probably not going to be in OFF, because there seems to be no need. We're not striving for anonymity on the network, as the data is just random looking chunks, so direct connections between nodes is preferable.

So, finding and getting a given block could be faster than Freenet, but then you have the possible potential of having to get 3 times as many blocks. And then you don't have the overhead that Frost has, sort of thing. Until the OFF network exists, there's no easy way to make predictions about it. Plus, next to none of the network code has been written yet, and the design is pretty vague. Finding blocks, for example, is a bit trickier than it would seem to be at first. And what if a block has been removed from the network due to loss of a node or some such? It'd be nice to be able to detect that, I'd think.

Anyway, all that isn't fleshed out yet, so answering questions about it is premature.

[aqlo]

Ok, I think I've got this, but feel free to jump in and correct me if I go wrong somewhere. First of all, these songs on this CD definitely appear faster than stuff downloaded from FreeNet. That doesn't prove anything of course, because they are already on the CD, but at least it isn't the other way around, i e if it was slower than FreeNet even on its own CD with no networking then that would be it, there wouldn't be any point.

Now I am betting that some of your initial downloads will be slower than FreeNet. Not necessarily slower than the old badly-routed too-redundant FreeNet that we are all familiar with, but with NGRouting I have managed to get some things (not everything by any means) faster than I had expected. But as you build your cache, and as other nodes near you build theirs, I expect that OFF will end up being much faster than FreeNet. There are only a finite number of blocks, and once those exist widely, the process of acquiring and reconstructing them should become lightning-fast. At that point the only disadvantage is in reconstructing the file, and as I said that is already faster than FreeNet's complex decryption process.

But the real advantage of OFF over FreeNet seems to be in the legal or moral sense. That's what this "brighnet" idea is all about. With FreeNet I definitely have illegal content stored on my computer. I have the complete file, it may well be something horrible, and the data is used for that file and only that file. When they come scan through my node, I have these terrible things on it, and my only excuse is that they were encrypted, I didn't necesssarily know they were there, and I hosted them because the concept might help some people in say China mouth off about their government.

With OFF on the other hand, you might be able to reconstruct something bad using some of my blocks. But it would also use some of your blocks! even though all you downloaded before that was a Bible or a Constitution. And I don't have those awful things on my computer at all. All the blocks you use to construct them with are being used for something else. Let's say we dump Project Gutenberg out into this system. Long before we are done, every possible block will exist, and be used more than once.

This is what is supposed to make digital copyright impossible. In the story at least. Why can't it work in the real world as well?

[Sephiroth]

But there is no legal/moral advantage. Its not the developers of Freenet or any network that puts the content on it. Which Freenet's goal is not to be a "mainstream p2p" program. The whole "bright/darknet" terms are just marketing buzzwords.

What is really being done here in a nutshell is that they are cutting files up into jigsaw puzzle pieces, combining the pieces from numerous other puzzles and randomly trying to reassemble the orginal picture with whatever fits from numerous puzzles.

Since this hasnt been proven legal at all, and no real expert on the law has looked at this the only arguement that its legal is numerous vague arguements that dont clearly answer the question. Which this is still unproven and too early to be promoting it.

The last group of people to make a bunch of big promises and not be able to prove their claims like this was ESV and until it can be proven otherwise this is no different IMHO and another post about how splitting files up and putting them back together with random parts isnt convinceing at all.

[l33ts0n]

The last group of people to make a bunch of big promises and not be able to prove their claims like this was ESV and until it can be proven otherwise this is no different IMHO and another post about how splitting files up and putting them back together with random parts isnt convinceing at all.

The reason it is explained so many times is that is that the multi-use encoding system is what makes all content in the OFF system inheriently legal: without the multi-use encoding system its just another flakey p2p app with a freaky name. Its not just a matter of "splitting files up"--blocks in OFF in no way-shape-or-form resemble copyrighted content, because: they are not the original file, and have been depreceated into a form that gives them multiple meanings. If block Whatchamacallit is a random series of 0s and 1s that are not copyrighted, but can be used to reconstruct five different files... that block has five different *known* interpretations.

The way OFF depreciates the contents of a file is vital as to how you make the system entirely legal. It is key, my man.

Why is it legal? Suppose some asshat copyrighted/patented/"ownz0rs" a particular pattern on a Rubik's Cube (it'll happen sooner or later, US patent system sucks). You're not allowed to "replicate" that pattern anywhere, in any shape, without paying some kind of royalty.

Does this mean you're not allowed to give other people Rubik's Cubes, simply because one pattern is copyrighted?

No, it does not. Why?

Because the Rubik's Cube has many different interpretations. You can make one entire side green, another side yellow. Sproadic fruity colors if you want. As long as you don't transform the Rubik's Cube into the patented/copyrighted/trademarked pattern, there is nothing illegal about possessing the Rubik's Cube. You bought it, its yours own, you can have as much freaky sex with it as you want. You can even leave the edges sticking out of a half-turned 'Cube, even though your mother told you not to.

There is nothing illegal with giving people the Rubik's Cube (blocks in OFF).

There is something illegal about making all the pretty colors line up so it looks like that copyrighted/patented/trademarked pattern above (assembling copyrighted file in OFF).

OFF will not protect you if someone is physically snooping your computer and watches you assemble a file. At that point the legal content is transformed into copyrighted content, a decision made solely by the end user, and it is all your fault. Bad user. Bad.

As I've said, understanding how OFF stores blocks its vital to understanding why it is legal to transfer these blocks.

If anyone out there still doesn't quite grasp the concept, feel free to PM me. No fear, no fear, into the light we go.
---

Not to get off topic, but the term 'darknet' has existed for a long time. Brightnet is simply a spin off darknet, because it sounds really awesome and accurately reflects the nature of OFF--out of the shadows, you've got nothing to hide! Brightnet is also a crappy ISP.

[Sephiroth]

The reason it is explained so many times is that is that the multi-use encoding system is what makes all content in the OFF system inheriently legal:

Are your a lawyer?

Your just assuming that because because your breaking files down to such a small part that it will be legal and unable to identify the file.

[l33ts0n]

Are your a lawyer?

Your just assuming that because because your breaking files down to such a small part that it will be legal and unable to identify the file.

No. Again, that is not what OFF is doing... read the above explaination(s) if you still don't understand. The files are not being broken down into "small parts."

[tMoD]

What the OFF System does is chop up files into 128k blocks and encode them using XOR encoding. XOR encoding works by comparing the individual bits in two binary numbers and outputting a result based on whether the bits are the same or different. If two corresponding bits are the same—0 and 0, or, 1 and 1—then the output will be 0. If two corresponding bits are different—0 and 1, or, 1 and 0—then the output will be 1. For example, let’s say we want to XOR 101 with 110. The leftmost bits, 1 and 1, are the same so the output for that bit will be 0:

101
110
0

The middle bits, 0 and 1, are different so the output for that bit will be 1:

101
110
01

The rightmost bits, 1 and 0, are different so the output for that bit will be 1:

101
110
011

As you can see, the end result of this particular XOR is 011. To retrieve 101, you XOR the output (which in this case is 011) with the key, the same number you XOR'd the original number with (which in this case is 110). The leftmost bits are different:

011
110
1

The middle bits are the same:

011
110
10

The rightmost bits are different:

011
110
101

In short, XOR is a reversible algorithm so A XOR B = C and C XOR B = A. The interesting thing is that, depending on what you XOR it with, any number can produce every other number. 011, for example, can be a part of the encoding of any other three-bit number:

011
011
000

011
010
001

011
001
010

011
000
011

011
111
100

011
110
101

011
101
110

011
100
111

Let's say 010 was part of a Britney Spears song. If I wanted to download it I would download 011 and 001 and XOR them together. Note that at no time did I download 010 off of anyone. Note also that the two numbers I downloaded can, and will, be used in the encoding of many other numbers as well-- 011 and 001 will not simply be used to retrieve a Britney Spears song. For instance, 011 XOR 111 will give you 100, part of, say, The Hulk perhaps; 011 XOR 101 will yield 110 which might be part of some public domain work like Shelley's "Ode to the West Wind"; etc. The same goes for 001. This is what multi-use encoding is. In practice the blocks will part of the encoding of many files and, theoretically, they can be part of any file. As the assembly of the file takes place on your computer, there is no way (short of illegally hacking your computer) for any outsider to know what you've done with a block you've downloaded, given the infinite number of potential uses to which it can be put.

[Otto42]

aqlo: Yes and no. You really can't have every possible block. The space is too big. And you'll get collisions long before that occurs. Currently, possible collisions are really being ignored, because it's so unlikely as to not be an issue. Freenet does much the same type of thing, as does every other file sharing system. Collisions in the hashspace are just unlikely enough not to matter much.

Every time someone inserts a file, new blocks are created. But it'd take more files than probably exist to create every possible block. And it'd take quite a bit less to fill every possible hash. But that's still 2^160 possible blocks. For all intents and purposes, consider the possible number of blocks infinite. It's large enough that that works for conceptual purposes.

You're correct on the rest though.

[Otto42]

Are your a lawyer?

Your just assuming that because because your breaking files down to such a small part that it will be legal and unable to identify the file.

You're not just breaking apart files into chunks.

If a song is represented by the number 12, and I'm sharing the numbers 5 and 7 (along with 3, 9, 10, 2, and 6), have I broken the law?

OFF breaks apart files into chunks and combines them with other chunks of data from other file or several other files. The result of all this is chunks of data which are essentially random. If I share these random chunks, then I cannot have broken any laws, because these chunks are not the files they started from. Each chunk could be used to produce hundreds or thousands of files.

Yes, trading songs is illegal. But trading the chunks isn't because *they're not songs*. They're not anything. They're randomized chunks of data. If someone told you how, you could combine them in various ways to produce a song, or a copy of the constitution. You'd have to get information on which chunks to use and how to use them. Passing that information on how to combine them to somebody else *may* be illegal, but passing the chunks around definitely is not. The chunks are worthless in and of themselves.

They may be derived from copyrighted content, or they may be derived from public domain material, or hell, they may be derived using a random number generator. It makes no difference, there's nothing inherent to these chunks of data that makes them anything at all. They're not ownable. They cannot be copyrighted. They are just random bits in no particular order. It's only how you combine the bits to produce something that produces the file, and in order to do that, someone has to pass you info on how to do that.

But the nodes don't have that info. They don't share that info. This means there's nothing the node does that could be considered illegal. And that's the whole goal, to make the network itself lawproof. The network doesn't pass around copyrighted material in any way. It doesn't pass around information on how to produce copyrighted material. It doesn't pass around links to copyrighted material. It just shares chunks of random data with other nodes on the network. That's it. That's all it does.

The CD contains links to produce the songs because it's a demo. A network of nodes wouldn't contain links at all. It would contain just the chunks of random data. You have to get your links elsewhere.

All files are just numbers. Long numbers, but numbers nonetheless. But numbers are mathematical concepts, and you can represent any number in an infinite number of ways. By breaking files down into other representations, and mixing them all together, you make it impossible to produce any reference to one file from one of the chunks. Or multiple of the chunks.

Let's say I'm running an OFF node. The RIAA seizes my node via legal methods. Nothing on my node can produce any songs. Nothing on my node has links to any songs. I may have files on there that are random chunks of data, but I have no songs, no links, no nothing. You have to use information from outside my node to get those songs. You may have to use blocks that are not even on my node to get those songs. How can my node be illegally sharing anything if it doesn't have anything illegal to share on it?