Decentralize BitTorrent with Kenosis

UnderScan writes "Eric Ries, writer/programmer/CTO, authored an article 'Kenosis and the World Free Web' at Freshmeat [Owned by Slashdot's Parent OSTG]. Kenosis is described as a 'fully-distributed peer-to-peer RPC system built on top of XMLRPC.' He has combined his Kenosis with BitTorrent & removed the need for a centralized tracker. He states: 'To demonstrate Kenosis's suitability for these new applications, we have used it to improve upon another peer-to-peer filesharing application that Just Works: BitTorrent. BitTorrent does one thing incredibly well. Using a centralized "tracker," BitTorrent manages efficient distribution of data that is in high demand. We have extended BitTorrent, using Kenosis, to eliminate this dependence on a centralized tracker.' See also the Kenosis README for details on using Kenosis-enabled BitTorrent."

Comments

This will be considered a troll, but...

[SilentChris]

Outside piracy, how useful is this?

Re:This will be considered a troll, but...

[Dachannien]

Decentralization is generally useful for any application where failure of some critical node results in failure of the entire transaction. Distribution of any data via bittorrent will benefit - regardless of content - if there is a possibility that a tracker host could fail.

Re:This will be considered a troll, but...

[bigbadwlf]

This will probably considered a troll also, but I think the obvious answer is porn.

Re:This will be considered a troll, but...

[athakur999]

One of the problems with BitTorrent is that the trackers themselves can get overloaded with too many clients. If this system can eliminate something like that happening then that'd definately be a good thing.

That being said, the busiest torrents I've seen are for copyright-infringing material, so I guess it's still a boon for piracy. ;)

Re:This will be considered a troll, but...

[nadadogg]

I'm thinking that your comment is a combo of both insightful and redundant :)

Re:This will be considered a troll, but...

[Anonymous Coward]

This will probably considered a troll also, but I think the obvious answer is porn.

Troll porn? That's just sick.

Re:This will be considered a troll, but...

[discord5]

Actually, even for piracy it's still not very useful. It takes the legal battle to the next logical step: the user

Re:This will be considered a troll, but...

[vhold]

Allow oppressed people to anonymously distribute large incriminating videos of their corrupt government?

Re:This will be considered a troll, but...

[Iamthewalrus]

Piracy isn't the only activity one might want to keep private and decentralized. Political dissent in totalitarian states, for example, is another.

Re:This will be considered a troll, but...

[Trolling4Columbine]

Why would someone need to hide a Linux ISO? Why wouldn't the old tried-and-true BitTorrent suffice?

Re:This will be considered a troll, but...

[jo42]

Obligatory: We welcome our decentralized BitTorrent Overlords!

Re:This will be considered a troll, but...

[bman08]

It's true. As long as there are tracker sites to shut down, and webmasters to sue, johnny-12-year-old can download the latest snoop without fear of losing his college fund.

Re:This will be considered a troll, but...

[WormholeFiend]

Still makes me wonder about network pollution though...

Re:This will be considered a troll, but...

[ceeam]

Remember, with enough lawyers _everything_ you do on your computer or download from the Internet may be considered illegal.

OTOH - if - for example - you crack a proprietary video codec so that it is suited for full-length movies distribution, add a cracked proprietary sound codec, name it all after a proprietary technology, then take some others' source, tweak it a bit, urge other peoples to contribute promising that it will be "free for ever", then demand money for it - it is still ok if you form a company! You can even put your certification on hardware players and stuff. Voila. (Yes, I do troll, "mod me down". But better yet - reply).

Re:This will be considered a troll, but...

[Trolling4Columbine]

And when we talk about "the user" on Slashdot, we only mean "grannies" and "12-year-old girls".

Re:This will be considered a troll, but...

[Ayaress]

Really, if you think about it, wouldn't anything that improves the ability to distribute files period will prove a boon for piracy?

Re:This will be considered a troll, but...

[Anonymous Coward]

A couple more:

Piracy == A crime committed on high seas;
Copyright infringement != Stealing;

Re:This will be considered a troll, but...

[glassjaw rocks]

"It pisses me off that even a /.'er refuses to make that distinction."

Sir, that isn't saying much. The /.'er is not an elevated race in terms of intellegence or maturity.

Re:This will be considered a troll, but...

[Rei]

I see several weaknesses in this, however. If I wanted to attack this, the route is pretty obvious to me. Several lines of attack stand out:

1: Start serving falsely-labelled file data that is correctly represented internally. There appears to be no moderation system built in, so bogus file data will pollute the system.

2: Start serving any file data that is inaccurately represented internally. For example, make all of your hash entries but one accurate, but make that one hash entry inaccurate. Users end up downloading most of the bad file before it errors out. Depending on the setup of the server and client, they may continue trying to get the data from elsewhere, in which case you could serve larger amounts of corrupted data, possibly by using bad clients working in conjunction.

3: Hash cracking: Brute force hash cracking could allow fake data to pass as real, hash-matching data; only a single cracked piece per file is needed. This would probably be economically inefficient, however, compared to #1 and #2 in terms of the ability to disrupt network usage.

4: Mass peer suits. If BT is the download manager here, getting the people who have the file being shared is laughably simple.

There's probably also some risks for their proposed change to allow multiple seeders, but I'd need to think about it for a while.

1, 2, and 3 require an "intelligent" client. In real life, we inherently weed out those who give bad data simply by our experiences and the experiences of those we know. The more we trust a person, the more we trust what they tell us about others. This sort of system tends to inherently isolate out the bad apples, even if they work together. Even if, working together, they manage to convince a good client that they're right and others are wrong, that good client too will simply be viewed as a liar and its data shunned. Overall, the system will remain intact. It's no easy programming task, however; yet, it is doable, as evidenced by the fact that we, as humans, do it every day.

#4 has a simple solution: Involuntary mirroring. If this system would automatically force the mirroring of data into a cache on the destination machine, and serve it from there, there would be no way to know whether a person was actually uploading copyrighted material or simply acting as a "router". Since our law has finally started to catch on to the fact that it is unreasonable to sue those whose computers pass through illegal data that they had no realistic way of knowing about, it would effectively anonymize *all* data on the network.

Re:This will be considered a troll, but...

[Omniscientist]

I was thinking that it was more funny than insightful, but then I remembered that porn is a very serious thing not to be taken lightly here at /.

Re:This will be considered a troll, but...

[Anonym0us Cow Herd]

Outside piracy, how useful is this?

Think how handy this would have been when I was downloading those hundreds of megabytes of Tsunami homemade camcorder videos, mpegs, etc. to play in Kaffeine.

Or what about if I want to distribute a customized CD of something that is non copyright infringing? A custom Knoppix CD? A custom OpenOffice.org cd? etc. Why are steps in the evolution of software that eventually will make it easy for everyone to share, and to download huge files a bad thing?

Why does everyone think p2p is only about piracy?

Re:This will be considered a troll, but...

[AndroidCat]

There's already a problem with torrents using stale dynamic IP addresses and never giving up banging on the next person to be assigned the address. I hope this doesn't increase it.

Re:This will be considered a troll, but...

[k98sven]

1: Start serving falsely-labelled file data that is correctly represented internally. There appears to be no moderation system built in, so bogus file data will pollute the system.

Can't be done. There is a moderation system built-in to BT. The SHA1 hashes which identify the file chunks simply cannot be 'correctly represented internally'. If you know a way of doing this without changing the chunk size by several gigabytes, I think some crypto researchers would like to talk to you.
Any reasonably coded client will start ignoring a peer who delivers enough bad chunks.

For example, make all of your hash entries but one accurate, but make that one hash entry inaccurate. Users end up downloading most of the bad file before it errors out.

Which will leave you with no 'seeds' but the original one. People would wise to this quickly.

Hash cracking: Brute force hash cracking could allow fake data to pass as real, hash-matching data; only a single cracked piece per file is needed.

See above. This is nowhere near practically possible. You have to add gigabytes of data. A chunk is far smaller than that. It won't fly.

#4 has a simple solution: Involuntary mirroring. If this system would automatically force the mirroring of data into a cache on the destination machine, and serve it from there, there would be no way to know whether a person was actually uploading copyrighted material or simply acting as a "router".

This would not work. First, most net users have lower upload speeds than download speeds. Second, and more important is that this requires a lot of bandwith for helping someone else. Expect a hacked client to be produced which does not proxy anyone else, saving their bandwidth, but leeching others'. You could create a tit-for-tat scheme where someone will only proxy for you if you proxy for them. On the other hand, bandwidth is still the main constraint here.

Re:This will be considered a troll, but...

[squiggleslash]

Piracy == A crime committed on high seas;

Your definition of the word "piracy" is both incomplete and ignores regular English usage.

1. a. Robbery committed at sea.
   b. A similar act of robbery, as the hijacking of an airplane.
2. The unauthorized use or reproduction of copyrighted or patented material: software piracy.
3. The operation of an unlicensed, illegal radio or television station

I know that the freeloaders hate any word with negative connotations being used to describe what they do, but the word "piracy" has had this definition for a long, long time.

Re:This will be considered a troll, but...

[Cr3d3nd0]

A +5 Insightful comment about PORN?!?!

and behold the final sign of the apocalypse is manifest.

Re:This will be considered a troll, but...

[k98sven]

WRONG. I can put any file up online and call it "Metallica - Sanitarium.mp3". It'll get listed as "Metallica - Sanitarium.mp3", and people will download it thinkint that it's Sanitarium, when really it's an 80 year old man reading recipies for lentil soup. Hashes don't come into play.

Uh, I believe we were talking about P2P clients here? And more specifically BitTorrent? Hashes certainly do come into play on such a network. They do not validate the content, they do validate that "the file 100 people are distributing with this name is the same file".
As for moderation:

You can't just ignore files that have no seeds but the original - otherwise, no new files could ever be listed. PLUS, working with a friend lying client, you can both seed the malformed file.

So a few users get burned trying to download the file if it's fake, or continue to seed if it's not. An authentic file will rather quickly gain seeds. A fake will have fewer or no other seeds, and a file which doesn't check-out will never have any. It works just fine. People won't get turned off even if they get burned 10% of the time. And that's an overestimate, since you'd think the P2P system would have more than ten users per hosted file.

The benefit of doing this, however, is that if you do it with any sort of significant percentage at all, the concept of mass lawsuits become untenable, because a significant percentage of defendants could prove that their computers were just having data routed through them.

Two words: Contributory infringement.

Bittorrent works by *exchanging* packets. The more work you do for others, the more work they do for you. This would work no differently - by being a proxy, you would be rewarded on your downloads.

So X passes a chunk from A to B. Who rewards him? A or B? And with what? If A and B are in the same file-swarm as X, the entire thing is pointless since they're all distributing the same file anyway, just in an inefficient roundabout manner.

If they're not distributing the same file, you cannot guarantee A or B has the file you want. Unless you tell them. In which case you're not anonymous: Other users can find out who you are and what you're downloading. And the exersize is again pointless.

Or are you suggesting some kind of system where you get 'credit' for proxying? That would of course be compromised by leechers and not work.

Re:This will be considered a troll, but...

[Rei]

> Uh, I believe we were talking about P2P clients here?

Uh, I believe we were talking about ways to damage a P2P client here, and polluting the network with incorrectly labelled IS a way to pollute the network. The way to resolve this is called moderation, and this client doesn't do it - end of story. Hashes don't remotely affect this issue, so it was pure idiocy to bring them up.

> So a few users get burned trying to download
> the file if it's fake

And while they're downloading it not realizing it's a fake, they're seeding it to others. Yes. Now repeat times 100 million files.

> An authentic file will rather quickly gain seeds

No quicker. If you don't realize until you're almost done that the file is a fake, you're seeding the bad file almost as much as you would a good file.

> Two words: Contributory infrigement

One word: ISPs

ISPs are now essentially immune, except the case of gross violations that appear to stem from the ISP itself, from successful prosecution (that doesn't mean that IP holders still don't occasionally try - not nearly as often as they used to, however). Why? Because the ISPs have an incredible degree of plausible deniability: The data was routed *through* them, and it was completely unrealistic to expect that they even *could* watch everything that passed through them.

The exact same thing applies here. If the RIAA wanted to do mass suits, they couldn't charge a large group of people with distributing content, because clearly a significant percentage of the people that they'd name would not have been distributing it - it would have been routed through them without their knowlege. The RIAA, if they wanted a mass suit, would have to show that the client itself was somehow illegal, and then do a whole RICO suit against all users and designers of it - an almost laughable concept.

I'll repeat: They can't just say "These people were downloading files for which we own the IP", because many of them *were not*. Even if they were contributing to infringement, that is a completely different crime, and even despite that, it's a stretch. This prevents the RIAA from launching mass lawsuits - the only way that they have that is at all economical.

> So X passes a chunk from A to B. Who rewards
> him? A or B? And with what? (etc)

http://books.slashdot.org/comments.pl?sid=135160 &c id=11282490

Re:This will be considered a troll, but...

[Daengbo]

Are you talking about some company in particular?

Re:This will be considered a troll, but...

[me at werk]

And because some people won't get it, this Informative post will explain.

DivX ;-) 3.11 was a hacked version of the Microsoft MPEG 4 layer 2 codec used for .asf files. A french hacker, Jerome Rota, extracted this codec and made it work for .avi files.

Wikipedia seems to skip a bit of history here, not mentioning the cracking of the sound codec etc, but does conclude with the information on "DivXNetworks, Inc" creating a Clean room design version of DivX (similar, in my opinion, to editing someone elses paper enough to make it not technically plagarism) and releasing it as DivX 4.

Re:This will be considered a troll, but...

[k98sven]

And it's idiotic to assume the client can or should do moderation. There is no foolproof way to determine if the data it's transferring is 'authentic'.

As for 'seed', I'm using that word in the bittorrent context, where 'seed' means someone with the whole file. And this is something which both Bittorrent and this proposed protocal make easy to verify, since the clients respond to queries of which chunks they have, it's possible for the clients to determine how many users have the whole thing.

Your reference does not answer the question, either. But if you believe you really have a terrific solution to all problems: Write a protocol specification, or even better: an actual program.

I doubt you could. I can't. And the reason is simple: It's simply not possible to write an *anonymous* and *efficient* P2P program, especially not with the features you want. If it was, then someone smarter than you or I would already have done so.

Re:This will be considered a troll, but...

[Rei]

> There's no foolproof way to determine if the
> data it's transferring is 'authentic'

First off, there are many basic ways to determine if data is authentic; completely foolproof methods revolve around trusted authorities. These need not be physical, destroyable entities, such as servers, but can be a select group of individuals. It is even be possible to have raters that don't do anything illegal - simply rate music and videos, if you can get a music hashing algorithm that works independent of actual encoding (i.e., bulk file waveform characteristics, ignoring minor waveforms that might get lost), to store in an advanced CDDB-type database for perfectly legitimate uses (CDDB itself wouldn't work, as it doesn't hash waveforms, but track lengths).

However, a variety of "effectively foolproof" moderation schemes exist for no-authority servers, relying on a web of trust. Moderation is a widely discussed issue; I suggest you FGI.

> where 'seed' means someone with the whole file

Clearly you couldn't have meant the bittorrent context, because there is only one true seed in BT (this network proposes changing that). You can have multiple clients "seeding", but they are not the "seed" - the lone client which maintains an authoritative list of peers (yes, the terminology is overloaded). A very low percentage of BT clients continue seeding the network after they finish the download - the vast majority of packets are downloaded from unfinished peers, making your argument completely irrelevant.

Unless, of course, you're going to go back to your whole argument that clients with one seed should get ignored. Which I'm assuming that by now you've accepted is preposterous, since new content can't ever be listed if you do that (and it doesn't defeat multiple IPs working together (i.e., the RIAA, MPAA, spyware pushers, porn pushers, etc) anyways).

> If it was, then someone smarter than you or I
> would already have done so

You're right - everything that's possible has already been written. That's why no one ever writes software any more, right?

Don't argue from "it doesn't exist yet, so it must be impossible"; that's obviously a fallacy, especially on something as new as widespread p2p software. And if you don't like the proposed mechanism, either pose individual challenges or concede.

Re:This will be considered a troll, but...

[FrzrBrn]

Outside piracy, how useful is this?

Kenosis is really only a platform for forming peer meshes. Are you telling me you can't think of any use for that sort of a thing? How about distributed computing, for one. As more nodes join the mesh, the workload is extended to encompass the newly available resources. Serverless instant messaging also comes to mind.

I assume that what you are really asking about is tying this to BitTorrent and the peer-to-peer distribution of the .torrent tracker file. As others have already stated, the elimination of a single point of failure in a networked system is generally considered a good thing. Consider trying to download the latest Slackware release when the server hosting the tracker is suffering from a DDoS.

So, in answer to your question, even outside of "piracy" this is still very useful.

Re:This will be considered a troll, but...

[AoT]

That is the beauty of bittorrent, it is completely unfeasable to sue someone for sharing *one* file, and this is all you can really find someone sharing on bittorent; at least at one time.

Example: Sony finds me downloading the newest eminem album(or whatever). They sue me for the damages from my sharing one album. Repeat for 6000 people. They cannot get enough money to even cover their legal fees from the settlement at that point, all users have to do is fight the case and sony would cave pretty quick.

Re:This will be considered a troll, but...

[Rei]

Quite true - these sorts of things are not in the BitTorrent protocol, and shouldn't be. BitTorrent is a way to transfer files - it's simple, and does it well (although there are some flaws in it... this isn't the time or place, however). Regardless, putting such things in the file location mechanism sounds like a proper place for them :)

> would someone posting a valid pattern for a
> metalica song be liable for pirates that use
> that pattern to validate against

If their code was sold as a replacement for CDDB to stop colliding CDDB hashes (which do happen, although rarely, and are a known flaw in the system), and was used legitimately and widely as such, that'd be a pretty hard sell. Pirates use CDDB to automatically label their files when they rip them, but you don't see anyone going after CDDB for that.

> I have seen many clients with built in trust
> metering systems, so I don't see a problem here.

Exactly. That's what I'd like to see this client have. The person who had been arguing with me seems, for some strange reason, to think that this is impossible.

The more bad data you get from a person, the less you trust them, and the more you tell your peers. The more your peers trust you and the less they trust the person you're complaining about, the more they'll listen to you, but repeated claims don't help you. Even friend-of-a-friend-of-a-friend communications work in such a situation, so long as each step of the way cryptographically signs their message and that each recipient atrophies the trust level based on how much they trust the sender and forwarders.

ke.no.sis

[KillerDeathRobot]

n. Christianity

The relinquishment of the form of God by Jesus in becoming man and suffering death.

Re:ke.no.sis

[Narchie Troll]

Um, what the fuck are you talking about?

What does that have to do with the definition of 'kenosis'?

Re:ke.no.sis

[Corporate Drone]

I think you left out a little thing called "The Resurrection." It happens to be the most important day of remembrance to the Christians.

Although the resurrection is important, kenosis only refers to Christ's act of humbling himself by the "emptying" of himself and taking on human form; it doesn't refer to the totality of Christ's human life, passion, and resurrection...

Re:ke.no.sis

[KilobyteKnight]

I think you left out a little thing called "The Resurrection." It happens to be the most important day of remembrance to the Christians.

Which is why they take Friday off, hide colored eggs, and eat chocolate bunnies - it helps them remember... uh..

What is it exactly it helps them remember?

Re:ke.no.sis

[KillerDeathRobot]

The eating of chocolate bunnies symbolizes the eating of chocolate pagans. I think.

Re:ke.no.sis

[xSauronx]

that chocolate bunnies MUST DIE

Re:ke.no.sis

[stinerman]

"Look mommy, the goldfish put a lincoln log in me sock drawer. Thats the story of Jesus."

How I miss Bill Hicks.

Re:ke.no.sis

[felis_panthera]

all that symbolism helps to remind them that they stole the festival from the pagans... this is a long read, but a good read...

Re:ke.no.sis

[Dr. Spork]

Yeah, I can imagine that if I ever got resurrected I'd wake up with serious munchies for colored eggs and chocolate bunnies. Mmmm, bunnies!

Re:ke.no.sis

[Cardbox]

Also - more simply - the self-emptying involved in creating anything at all (esp. anything free-willed).

Re:ke.no.sis

[mav[LAG]]

And while you're being objective, you can go here for another long but good read...

Re:ke.no.sis

[Michael Hunt]

I must have listened to that Bill Hicks bit 1000 times by now, but I still have absolutely no idea wtf a Lincoln Log is.

Care to explain?

C'mon people, thing of the big picture!

[Anonymous Coward]

How is the RIAA and MPAA supposed to stamp out bittorrent if you guys keep improving it? Where's your compassion?

Re:C'mon people, thing of the big picture!

[AnalogDiehard]

How is the RIAA and MPAA supposed to stamp out bittorrent if you guys keep improving it? Where's your compassion?

Welcome to mole country. Here's your hammer. Happy whacking.

Re:C'mon people, thing of the big picture!

[TeleoMan]

"Thing" of the big picture. Um. Okay. Sounds like a bad 50s monster movie.

Exeem alternative

[Anonymous Coward]

Neat, an open source alternative to Exeem which sadly turned out to be a spyware-ridden disappointment according to an earlier slashdot post.

Still falls just a bit short.

[Nosf3ratu]

This doesn't anonymize the users, does it? Your IP address is still readily available, no?

Then this falls a bit short of the "killer p2p app" moniker that it *almost* deserves.

Re:Still falls just a bit short.

[xZAQx]

I'd like to see someone take this and add anonymity to it.

Re:Still falls just a bit short.

[diegocgteleline.es]

And how do you expecto t "anonymize" yourself in a P2P? You can anonymize the searchs (the README says something about proxys) but for downloading the data, how could you be "anonymous"

P2P basis is that everyone sends data to anyone in the net. I don't think it can be "anonymous" - you'll be always connecting to someone's computer to download part of a file. It you want to "anonymize" it you'd have to go trought a central server and that doesn't works because it's centralized. You could play some tricks like freenet does, but for real-world P2P I don't think you can be "anonymous".

Re:Still falls just a bit short.

[Everleet]

As soon as you enlighten us on how to send a directed IP packet without a destination address, I'm sure someone will write a few dozen p2p apps around it.

Re:Still falls just a bit short.

[Norgus]

Who would be kind/stupid enough to provide a few proxies with huge bandwidth that would allow good p2p use, also who would want to open themselves to _everyones_ legal wrongs?

Re:Still falls just a bit short.

[Abcd1234]

Well, Onion Routing is a step in the right direction...

Re:Still falls just a bit short.

[teebo]

Coming up.

http://www.i2p.net/
I2P is an anonymous network, exposing a simple layer that applications can use to anonymously and securely send messages to each other. The network itself is strictly message based (ala IP), but there is a library available to allow reliable streaming communication on top of it (ala TCP). All communication is end to end encrypted (in total there are four layers of encryption used when sending a message), and even the end points ("destinations") are cryptographic identifiers (essentially a pair of public keys).

Re:Still falls just a bit short.

[Not_Wiggins]

And how do you expecto t "anonymize" yourself in a P2P? You can anonymize the searchs (the README says something about proxys) but for downloading the data, how could you be "anonymous"

Someone came up with one way to get around this. Basically, you setup a P2P network where everyone is only allowed to download from peers.

Let's take a simple, single-path example. Computer A connects to computer B and requests TinyLinux.iso.
A doesn't know if B *owns* the file or not, or is getting it from computer C.

B doesn't know if the file is destined for computer A or not... A may be requesting the file for some other user connected to A.

In this way, one could track who is connected to each peer, but cannot establish with certainty if the B is the source of the file that is being sent or is merely a link. Meanwhile, one cannot establish with certainty that A is the destination of the file or merely a link.

While it doesn't ensure true privacy (your peers know your IP address), it does make the culpability harder to pin on particular users... if I never store any part of a file transfering through my system, did I ever fully "steal" it?

The downside to this is, of course, you may not get optimal path downloads for a sourcefile AND you're limited to the bandwidth of the weakest link in the chain.

But, like an extension of the "Kevin Bacon" game, if you tell everyone you know to tell Kevin, "Love the movies," and they tell everyone they know to pass the message along... eventually, it *will* get to Kevin, but the source won't be known anymore. Kevin just has to tell the person that told him, "Thanks," and so on back to the original source for the communication to happen effectively. This is also how this P2P would work with multi-path (to aid in download speeds).

Re:Still falls just a bit short.

[legirons]

"This doesn't anonymize the users, does it? Your IP address is still readily available, no?"

So use Konspire, which is an anonymized version of BitTorrent. Or specifically, "deniable steganographic distributed transmission optimised for large data"

Re:Still falls just a bit short.

[rainmayun]

I would think, legally speaking, it would be more important to hide the source address.

Re:Still falls just a bit short.

[Not_Wiggins]

Mute is cool (totally forgot the name of the project, but that's the one I was thinking of when writing the reply... thanks!).

But, I'd check out WASTE instead... seems to have some promise, but only for "private networks."

Quite useful

[drewzhrodague]

QUite useful, of course! We could distribute spatial-data, and Wi-Fi locations to PDAs and laptops in this way. There are metric tons of useful applications for BT and K.

Re:Quite useful

[Em Ellel]

Application is everything in this case. If most torrent traffic is for stealing mucis, movies, or software then there is a problem regardless of other uses it could have

So, since computers are used to pirate stuff, we should ban them too? That will really make the music and movie industries happy. Not so much for software industry though.

-Em

Re:Quite useful

[dnoyeb]

I think you misunderstood his question.

"Outside piracy, how useful is this?"

"this" not being bittorrent, but being bittorrent without a central server. I have to ask the same question.

I would not take the risk of a corrupted server just to gain what? anonymity?

Re:Quite useful

[Moloch666]

Wrong. Porn would be more accurate.

Re:Quite useful

[Zerth]

Even though the grunt work of uploading/downloading the content is offloaded to the peers, a sufficiently popular file can cause so many join/finish/etc messages that the tracker can't respond to tell them who is in the swarm

Re:Quite useful

[halaloszto]

You are close but not there.

Guns are useful for hunting and all kinds of different stuff, still a gatling maschine gun has no other use than killing people in a war.

BitTorrent is a great way of distributing stuff. The only need for decentralized tracking servers is to hide the source. Which is only desirable in case of unlawful content.

As in the above example, if wifimaps want to distribute maps using BT, they can set up their own tracking server. The only issue with that is that it wil be well known that the maps are coming from them.

vajk

Re:Quite useful

[NullPhi]

eliminating a single point of failure helps all bt traffic, not just the "unlawful content."

Re:Quite useful

[yasth]

Actually a Gatling gun can be used for missile defense, and high speed package delivery.

Decentralizing tracking makes it easier to set up files. It also elminates a single point of failure, and some bandwidth issues of the host (tracking traffic is relatively small, but it can still add up very quickly as peers go into the 10s of thousands.)

Imagine you had a video of a very recent disaster. You could put it on a normal tracker, but you would first have to find one that was willing, and then you would have to hope it could handle the swarm.

Or imagine you have a torrent server for the presidential debates, most of the time your tracker will be idle, but right after releasing your version of the debates it might well fall over, this system would allow your server to be used by other systems as the majority of the time it is not being used, and then lean on other systems during your flash crowd.

Re:Quite useful

[aichpvee]

Wow, BT can walk into a music or movie store and steal merchandise? I must have been using it the wrong way.

Re:Quite useful

[Marvelicious]

No, you've got it all wrong. Right now we just download to our hearts content. We don't have to come up with a legitimate use for it 'til they try to shut it down!

That was a joke (for the sake of humorless moderators)

Re:Quite useful

[TubeSteak]

high speed package delivery
Does UPS know about this?
Is how those DHL guys in the yellow vans became really successful overnight?

Re:Quite useful

[chaoaretasty]

It isn't illegal until you find the RPAA (Randy Pornographers Association of America) knocking your door down because you've been illegally downloading 13-doggie-hot-sex-judy392.jpg, and as they drag you away lecturing you on how you're stealing money from these poor artists.

Re:Quite useful

[IpalindromeI]

They became really successful overnight because an already-successful air delivery service (Airborne Express) decided to create a ground delivery service as well.

From the feature list...

[hourieh]

From the feature list...

Kenosis works in almost any networking environnment, including restrictive corporate firewalls, because it uses XMLRPC for its network communications. It can also work with an HTTP proxy.

This alone makes a worthwhile project, for those stuck behind firewalls/proxies.

Hi there...

[gowen]

And welcome to KBTR (formerly K/.), all Bit-Torrent stories, all the time.

Enough, already!

Re:Hi there...

[ultrabot]

And welcome to KBTR (formerly K/.), all Bit-Torrent stories, all the time.

There are enough non-bittorrent articles to fill your workday, so move along, nothing to see here.

Bittorrent and p2p arp the hot topics of today (given all the police raids). New projects are certainly worth discussing.

Forced Evolution

[Superfreaker]

We all knew this was coming, but would this app get this kind of exposure had the MPAA not cracked down on those BT tracker sites?

It is just like Scour net (web based/centralized), then napster (p2p/centralized), then kazaa (p2p/decentralized). Every time they go after a technology, they force it to evolve into the next phase. They will never win IMHO.

Re:Forced Evolution

[Larsiny]

It is just like Scour net (web based/centralized), then napster (p2p/centralized), then kazaa (p2p/decentralized). Every time they go after a technology, they force it to evolve into the next phase. They will never win IMHO.

Remember that the next time one of your relatives' or friends' car gets stolen and stripped. Sure the technology to bypass all of the alarms and security measure not to mention the chop-shop techniques keep improving to the point where they keep outpacing the police. To paraphrase you "The police will never win IMHO" Won't you be glad!

Seriously, this is technologically good idea on top of a good service. But what is technologically good may not be the right thing. (Wanna bring up cloning anyone? I'm not saying I'm against it just wanna point out the possible moral and ethical points that will be raised) For anyone to say this would be most beneficial for anything other than illegal activities would be lying. For all the people who honestly try to argue the positive merits of BT, stuff like this just ruins the validity of it. This will eventually cause it's end just like all those other "technologies" you mentioned.

Re:Forced Evolution

[burns210]

Ironically, BT was actually a step back from the p2p generation. It was Napster-esque, and thus, first generation p2p (centralized p2p). Kazaa(Gnutella, etc) were 2nd gen (decentralized, p2p). Then the newest apps like Mute, Waste, and even Freenet are 'third gen' p2p. Comprising decentralized structure, p2p connections, and anonymity/security functions for protection.

3rd gen was an obvious move from 2nd gen. 2nd gen(Gnutella) was a smart, but obvious move from 1st gen(napster).

4th Gen? I have no idea. A more efficient (experience over time) protocol and clever tricks, for sure. But there is no clear 'next generation' p2p client.

Re:Forced Evolution

[Linuxthess]

I know this is a bit OT, but it seems that by convention of the English language we have confused the meaning of the word "Revolution".

People use "revolutionary" in the sense of an idea or concept which is mind-blowing, and brings us in a direction we have never encountered prior. "Evolutionary" is used to describe a system which is usefully extended, modded, but altogether not an unforseeable feat from the perspective of the original object or concept's invention.

But I think we have them con-fused. Revolve means to "come around again", hence the idea or object is not a totally novel concept, but a recycled one. Evolve would simply mean the progression from one form to another, and could be utilized in either sense of the conventional "revolutionary", or the actual, stunted form of "evolve".

This was all called to my attention when reading Neal Stephenson's Quicksilver (Vol. 1 of the Baroque Cycle) and his character's explanation of what a revolution is and is not.

Re:Forced Evolution

[DAMartin]

Perhaps we need a "de-evolution" at this point: has anyone tried to propogate torrents via USENET, which at this point has gone pretty much under the radar of the MPAA and RIAA?

Re:Forced Evolution

[787style]

IIRC, you could distribute the actual .torrent file via newsgroups but the problem is the actual tracker needs to reside on a web server whose address is provided via the text inside the torrent file. Once the address of the tracker is shut down, the downloaded torrent file is rendered useless. The concept of flooding a nntp feed with all the nodes relevant information would make spam look immaterial.

Re:Forced Evolution

[pcmanjon]

"For anyone to say this would be most beneficial for anything other than illegal activities would be lying."

Well, I used BT to get the Rubyx Linux distribution. It was available through BitTorrent and then a proprietary P2P method devised by the developer of Rubyx Linux.

That was the only two ways to get it... the maintainer ran the rubyx website off his home DSL modem.

As you can see, people *DO* use BT for legitimate reasons and people like me utilize that.

Granted, the majority of people pirate using BT, but if you open your eyes you'll find millions of legitimate uses and scenarios.

It's just that when a good thing comes along, the people who can abuse it start moving in really fast.

Re:Forced Evolution

[adiposity]

Right, in terms of vulnerability, Bittorrent was a step back. It was created for efficiency's sake. It was a huge step forward in p2p efficiency. Now that it has been proven as an efficient p2p system, it is being extended to have other advantages that p2p has. Thus, there are several categories of improvement that need to be considered:

P2P transfer, P2P search, P2P speed, and P2P anonymity. Actually, Kazaa had all of these except anonymity (it was speedy in many ways, but not nearly as optimized as BT). Bittorrent only has two of them (transfer and speed), and will have three with the addition of Kenosis. Emule already has all except anonymity, and its transfer algorithm is very similar to BT (horde vs. swarm). Freenet has all except speed, and Napster had only one (P2P transfer). So, I agree; BT is only just arriving as a true 3rd-gen P2P app.

One advantage BT has over other centralized apps, however, is its ability to use many trackers for many purposes. Thus, although suprnova is down, I am not affected, as the files I am interested in were never hosted there in the first place. Of course, the files I am intersted in are not illegal, but would never be hosted normally thanks to bandwidth issues.

On the other hand, e-mule is basically a decentralized bittorrent, so I'm a little surprised that people get so excited about being able to search torrents through p2p. For me, the main advantage of torrents was confirming before download that you would get the file you sought. Once you start trusting p2p searches, you can be fooled at the search level before you even begin the download. E.g., every huge file on emule seems to be a french porn flick renamed.

Decentralized searching is only part of the problem. We also need a way to inject the confidence that getting the torrent from a trusted site gives you. A voting mechanism, that can't easily be spoofed, that verifies that files are what they say they are. Perhaps this is 4th-gen. 3rd gen (emule) is still heavily susceptible to false seeding, as will Kenosis+BT be.

-Dan

Re:Forced Evolution

[Martin Blank]

Police are reactive. Even on stings, they're reacting to a threat (or at least a perceived threat). Installation of alarms and other security technologies is a proactive process, and those who want to get around them are largely reactive (though there is some reactivity on the part of the security industry to the occasional proactivity of the opponent).

A few days ago, there was a paper mentioned on Slashdot about safecracking. I found it fascinating, and it was really very eye-opening, because it cut through a lot of the mumbo-jumbo that's out there for physical security. The same author wrote a similar paper on dual-key locks that also altered my thinking about physical security. As a result of both papers that contained potentially dangerous knowledge, I am now much better informed on the capabilities and vulnerabilities (or at least potential capabilities and vulnerabilities) of the technologies securing my valuables.

Re:Forced Evolution

[LiquidCoooled]

This technology DOES have serious impacts, and having a private means to send data without "the man" watching is VERY important in some parts of the world.
There are countries and regimes in place where merely surfing the web is a crime.

Granted, its getting presstime because of the songs and movies, but as we both know, there are many other practical uses for it and similar protocols.

I stand 100% behind the protocol and its implimentation over something like freenet.
At least with bittorrent, I know I am helping somebody download a piece of the same file as me.
I know nothing illegal will pass through my box .

Re:Forced Evolution

[Gaijin42]

Stephenson's idea, while fun and interesting, is not accurate.

Revolution gets its current (upheaval) meaning from Copernicus.

Prior to Copernicus, the dominant theory of planetary orbits was the earth-centric one. IE, the earth was in the middle, and everything else went around, in very complicated paths.

Copernicus proposed the solar orbits we are so familiar with now.

He did this in a book called "De Revolutionibis" (The Revolution) published in 1543, talking about how the earth revolves around the sun.

Later (1616) Galileo did some promoting of the book, and subsequently got arrested for heresy etc. He is commonly misattributed with coming up with the idea of the revolution.

Newton comes along even later (1687) and publishes Principa, which gives the math for why Copernicus was right.

The word gets its current meaning (upheaval, overthrow etc) from the fact that ideas were overthrown (earth centric, vs sun centric) and has nothing to do with the concept of "coming around again, old is new etc"

Stephonson's characters might have been right to term what was going on a revolution, but they did not coin the word, there were making allusions to Copernicus, who got there several centuries earlier.

Re:Forced Evolution

[maxpublic]

But what is technologically good may not be the right thing.

The argument of those who favor the status quo. You want some piece of the world to stay exactly the way it is, and whine and moan and wring your hands when it appears it's going to change. But of course! Anyone in favor of altering that piece of the world MUST be a criminal! Because, like, you say so! And anyone who disagrees with you is a liar!

Really, this argument has been used since the dawn of time in a futile effort to keep things exactly the way they are because some people either just happen to like them that way, or they profit from the situation. And lo! Despite all this, technology continues to march on, and things continue to change.

You're pissing into the wind, and predicting doom when all the historical evidence in the world says that YOU are the liar here, whether you admit it or not. These technologies won't "end"; they'll continue to evolve no matter what you or anyone else - RIAA and MPAA included - think about the matter, or wish to be true.

Max

Re:Forced Evolution

[Hard_Code]

So then does the term 'revolt' date after the new meaning? Or did 'revolt' mean at one time, "the act of spinning"?

Re:Forced Evolution

[Gaijin42]

revolve means to move around (not quite spin), revolt (as a noun) did not exist until this time.

Re:Forced Evolution

[killjoe]

I hate to tell you but the police will never win. Despite thousands of years of police work crime still exists. Sure sometimes it goes up and sometimes it goes down but they have not been able to eliminate it.

As long as crime pays people will commit crimes. As long as pirating pays people will pirate.

Re:Forced Evolution

[Feztaa]

I agree with you. To say that a new file sharing protocol can only be used for evil is tantamount to saying that all files are evil. Not all files out there are protected by overbearing copyrights! There's plenty of stuff that is either not protected by copyright, or has a license that permits distribution! Any GPL'd bit of software is a good example, but there's also stuff like this:

http://www.legaltorrents.com/

Re:Wait

[litghost]

3. Sell content

Python

[ultrabot]

From TFA:

Kenosis is built in 100% pure Python

[snip]

He is author of several Free Software projects, most recently the peer-to-peer RPC system Kenosis, and co-author of several books, including The Black Art of Java Game Programming and Mastering Java.

It's fun to see how book-writing hackers act. Sell Java books to the Unwashed Masses, develop own projects in Python. BTW, interestingly enough, one could almost guess from reading the first few paragraphs that the implementation is going to be in Python.

Now we are just waiting for a platform-dependent implementation in C++ and MFC that is supposed to be faster because it's "native code", which all the clueless kids with 8mbit internet connections are going to download...

Re:Python

[arudloff]

It's fun to see how book-writing hackers act.

What, you mean, by using the right tool for the job instead of language snobbery? ;)

Re:Python

[Anonymous Coward]

Now we are just waiting for a platform-dependent implementation in C++ and MFC that is supposed to be faster because it's "native code", which all the clueless kids with 8mbit internet connections are going to download...

It would be faster. Python is SLOW, even compared to other bytecoded languages like Java, even if you use Psyco. Of course that's irrelevant because P2P is IO-bound, but that's not the point. ;)

Re:Python

[tetromino]

Now we are just waiting for a platform-dependent implementation in C++ and MFC that is supposed to be faster because it's "native code", which all the clueless kids with 8mbit internet connections are going to download...

Even network-limited programs like a bt client still need to worry about GUI responsiveness and memory usage. It would be insane to write the first implementation of such a program in C/C++ -- Perl and Python were given to us by gods to prototype these sorts of projects -- but once the basic protocol and UI behavior has been figured out, I would try to rewrite the client in a statically compiled (or at least a JIT-capable) language.

Re:Python

[ultrabot]

Of course that's irrelevant because P2P is IO-bound, but that's not the point. ;)

But that's exactly the point. Kids who don't know anything about programming thinking that changing the language is going to make their network throughput faster, not realizing that the interpreter overhead is dwarfed by the time spent blocking ahd context-switching.

Not to mention that a select() loop based Python approach probably beats a threaded C++ solution in performance anyway.

Re:Python

[bbtom]

I can only speak hypothetically, but considering that the BitTorrent client, GUI and TheShad0w's experimental client (BitTornado) are all written in Python, writing Kenosis in Python makes sense, since if it becomes popular, plonking it straight in to the main client could be achieved far easier than if you'd written it in any other language. (Of course, I haven't really researched this - it's an unproved hypothesis...)

Problems with decentralization

[Henry V .009]

The problem with this approach is dealing with untrustworthy peer. Without substantial protections, one peer can ruin everybody's downloads.

Re:Problems with decentralization

[hourieh]

Not really, if the md5 hash of each segment of the file is computed and transfered to clients, a client can check whether the segment is valid or not.

Kinda like EDonkey2000, not sure if BT works in a similar way.

Yes you need some protections, but it's still possible to protect clients from malicious peers.

Re:Problems with decentralization

[Anonymous Coward]

That is true, but you only need to ensure that the torrent you have is legit. Since the .torrent holds the checksums of the chunks, if a peer sends you garbaged data a couple of times then you can safely ignore it and move on, lookin' for other ones with the real data. MPAA would have to dump alot of bandwith to make it annoying enough, wich wouldn't be feasible, and even in such circumstance, it'd have to deal with the problem of being ignored after the clients realized it was sending false data.

You still have the problem with the MPAA/RIAA grabbing your IP, but that's another story...

Re:Problems with decentralization

[Jerf]

BitTorrent has said substantial protections built in to the protocols.

Remarkably, you are not the first to think of this problem. (Sarcasm.)

No central server?

[Psychic Burrito]

From the Readme:

Details

Kenosis-BitTorrent uses torrent files which specify a tracker of the
form http://hash.bt.kenosisp2p.org, where "hash" is the hash of the
original file.

Kenosis-BitTorrent downloaders will notice that this is a kenosis url
and use kenosis to find the tracker that is handling this torrent
file. Standard BitTorrent downloaders try to resolve
hash.bt.kenosisp2p.org as normal. Our dns server will look up the hash
in kenosis and return to the client the ip address of the kenosis
node that is tracking that file. If that tracker becomes unreachable,
subsequent lookups for that hash will switch automatically to the next
available Kenosis-enabled tracker.

Well, since there is a central DNS server at bt.kenosisp2p.org, how can they sincerely declare this to have no central point of failure? Yeah, of course dns propagates, but turn off this central DNS server and in a few days everything is gone, right?

Re:No central server?

[sinclair44]

Continuing reading, you can see that it's possible to directly have a torrent reference the network. The kenosisp2p.org bit is only for legacy clients that wouldn't know what to do with a "new" tracker location.

Re:No central server?

[Narchie Troll]

Kenosis-BT downloaders don't use the central server. Regular BT downloaders do. It's intended for backwards compatibility, I assume.

Re:No central server?

[MooseGuy529]

...turn off this central DNS server and in a few days everything is gone, right?

Wrong! The DNS server is a hack. Normal bittorrent links lead directly to a tracker. Kenosis bittorrent links lead to HASH.their.server.name. BitTorrent-Kenosis clients will recognize this and use the network. The purpose of the DNS (and the reason it's not btkn://HASH or something) is that legacy clients going there will be given the IP of any Kenosis client that can act as a tracker for it. Killing that DNS would kill legacy clients but not the enhanced P2P ones.

Re:Wait

[kzinti]

3 is "Get sued by MPAA anyway". Step 4 refers to the studios' profits.

Look, so long as the MPAA goons can trace at least one packet of a movie download to your IP address, you have liability. And with BT, as long as you're downloading, you're also uploading. Use torrent, and your ass is exposed, regardless of whether the index is centralized or decentralized. Call me paranoid, but that's how I look at it.

Hello draconian DRM

[acomj]

These little apps simply make the case for draconian DRM on everything, screwing us legitimate users even more.

Thank you pirates.

Re:Hello draconian DRM

[bogado]

So you sujest drakonian rule and/or ethics so that the internet would seem like a nice place that don't need draconian DRM on everything to the RIAA,MPIAA, SHITIAA?

Re:Hello draconian DRM

[HiThere]

In case you haven't been noticing, the *AAs haven't been waiting for new excuses to impose draconian DRMs. Or to corrupt the legislators into passing draconian new laws.

If I thought the pirates did them any harm, I'd acclaim the pirates as social heros. As it is, I merely consider them slightly more moral than the management (and some employees) of the studios.

O, I also consider that they are foolishly reckless, risking quite real harm for at best transient pleasure. (At best because most of what the studios have that *could* be subject to illicit copying would be a pennance to listen to or to watch.)

The Leia Principle Holds True

[NardofDoom]

"The more you tighten your grip, Tarkin, the more star systems will slip through your fingers."

When will the Empire^H^H^H^H^H^H *AA ever learn?

Re:The Leia Principle Holds True

[FireIron]

"Help me, O-BitTorrent Kenosis, you're my only hope."

Not really decentralized?

[kikawala]

from the README:

Kenosis-BitTorrent uses torrent files which specify a tracker of the form http://hash.bt.kenosisp2p.org, where "hash" is the hash of the original file. Kenosis-BitTorrent downloaders will notice that this is a kenosis url and use kenosis to find the tracker that is handling this torrent file. Standard BitTorrent downloaders try to resolve hash.bt.kenosisp2p.org as normal. Our dns server will look up the hash in kenosis and return to the client the ip address of the kenosis node that is tracking that file. If that tracker becomes unreachable, subsequent lookups for that hash will switch automatically to the next available Kenosis-enabled tracker.

Re:Not really decentralized?

[ledow]

"Kenosis-BitTorrent downloaders will notice that this is a kenosis url and use kenosis" - the other way is a compatibility interface for non-Kenosis BT clients.

Great Step, but still no safe haven for fileshares

[314m678]

This is an important step, but it still does not hide the user's IPs from the *AA.
From the Article:

It does not address problems of anonymity, privacy, or distributed data retention, although we hope to address these issues in future versions.

I don't think it's that useful for piracy

[tetromino]

because downloaders still need a way to find the content. So you switched from a central tracker to a distributed RPC system - but you still need someone to give you an IP address and a port to connect to the swarm. As long as Joe Sixpack can find the swarm and connect to it, so can RIAA/MPAA attack-bots, and then the fun begins...

Re:I don't think it's that useful for piracy

[BobPaul]

Well, seeing as since the MPAA has stated that they didn't want to sue end users and have gone after the Trackers and torrent hosts instead.

Since this decentralizes the trackers, what this most means is that the torrent hosting websites (places like suprnova was... hosted torrents but others were trackers) will be hit harder.

It's concievable that Kenosis could be used to host the torrents themselves. At that point, the MPAA could change their mind and start sueing end users ala the RIAA.

Nope

[EnglishTim]

If I understand this correctly, this doesn't affect communication between the bittorrent peers, just the client and the tracker. It still won't work through an HTTP proxy.

Re:Nope

[hourieh]

Hmm, really? I couldn't find any more info on the project's page, I think I'll test it and see if it's true.

Anyway, if your comment is correct, then it doesn't offer anything new over BT in this regard, as BT already uses HTTP to connect to the tracker.

Zero Defect Software?

[bperkins]

I think I found a defect.

This thing doesn't make any fucking sense.

I was really excited by this slashdot story, because I think something like this could be very very useful. I have to say that I was disappointed a bit by the download.

No docs or pointers at the top of the tarball.

One of the READMEs on the site says try "test.py" for an example, which seems to just hang.

Elsewhere it says to fire up bittorrent
trackers and clients.

There clearly is a lot of work that has gone into this, and the idea sounds really promising, but it looks like it needs a better end-user documentation before it's ready for primetime.

Re:Zero Defect Software?

[gosand]

There clearly is a lot of work that has gone into this, and the idea sounds really promising, but it looks like it needs a better end-user documentation before it's ready for primetime.

Whaaa?!? On Freshmeat? That's unpossible.

;-)

Re:Zero Defect Software?

[cs02rm0]

A brief look inside test.py suggested you needed to open ports for it.

At that point I removed it.

Re:Zero Defect Software?

[eries]

We'd love to have your help making the software ready for primetime. Join our mailing list if you're interested.

Re:Zero Defect Software?

[eries]

test.py is designed to show the API working in various real-world situations. If you'd like to see a comprehensive set of tests that do not require any ports, please see:

kenosis/nodetest.py

Re:Zero Defect Software?

[hey]

Still, instead of hanging it should give and error message or say what its doing. (If it was without defects.)

Why?

[jbb999]

Why would you want to remove the central tracker? Bittorrent works very well as it is.

Re:Why?

[Doctor Crumb]

Bittorrent works great. The problem is that bittorrent trackers are easily overwhelmed, lost, or become otherwise inaccessible. Since the rest of bittorrent is decentralised, it makes perfect sense to decentralise the trackers as well to further spread the load.

Re:Why?

[Fallen_Knight]

you forgot sued and/or hard to find.

Re:Wait

[Glog]

Umm, yes, sir, the solution is quite easy - it's called a proxy + encryption. In an offshore country. One without MPAA goons. With packets that look like garbage to anyone looking at them.

Are you an MBA or something?

[WebCowboy]

Because you have a fixation on money like some I've been acquainted with.

Seriously... I don't see how to make money off this...

Seriously...why is that important? Did you even read the article? The author of this BitTorrent enhancement does not even use the word "money"--it is WAY down the list of motivations for its creation, not does it seem to be about getting pr0n and warez. This guy sounds like an idealist in a very true sense--it's about decentralisation of control--making content available without being reliant on central servers.

I think this would be immensely useful. The reliance on central BT trackers has been shown to be BTs primary weak point--once a torrent is located and transfer is initiated it is incredibly robust.

Besides the fact that the admins of BT trackers are being harassed into submission by MPAA and RIAA, the more popular trackers seemed to be quite unreliable. If this innovation (open sourced to boot) addresses the reliablity issues in LOCATING the content that BT is so good at DISTRIBUTING then it could be start a dramatic shift in how we use the Internet, much like the WWW was.

It doesn't even have to be about piracy. Used within a VPN or on a corporate WAN it would make distribution of a large number of big applications much easier to distribute. I make VMWare and ghost images of machines that are many gigabytes and this solution would be a great way of distrubuting them to a large customer with global sites (keep in mind that these clients are legally permitted to use these images--my employer is a stickler for that).

A small operator could distribute software this way and save on the costs and time associated with maintaining a critical server with big pipe to the 'net. Security patches could be distributed this way very effectively without reliance on a single entity for distribution. The possibilities are endless. It might not be a money making machine, but it is the kind of thing that (if it works well) could change the face of computing.

Re:Are you an MBA or something?

[krack]

Blizzard used BitTorrent for the World of Warcraft Open Beta downloader.

Re:Are you an MBA or something?

[alan_dershowitz]

I'd like to second this. I'm working for an international company that shall remain nameless. We have hundreds of independent "sites" across the world that periodically need to update their client software. With a recent build, the size of the files transferred has jumped from about 15 megabytes to over 200 (this includes a lot of metadata.) So we started looking at other ways of reducing the download from the main server (and maybe dump some back on the client.) We're currently looking at Microsoft BITS and Bitorrent protocols to build our solution on top of.

Bittorrent and co. has several completely legitimate uses. I wonder how many companies are using it and haven't told anyone. It's not like we go around advertizing how we build our internal apps.

Do we really need...

[ltwally]

I just read about Kenosis from its homepage. And, I'm forced to ask:

Do we really need yet another bloated python p2p app? I can feel the flamebait and troll mods comming.. but seriously: Python sucks at gui work. It has to use generic wrappeers, like wxPython, that are extremely inefficient. Sure, like Pearl or Java, you can write gui apps using Python... but they always come out slow and over-weight.

Consider the BitTorrent client. Just running the application, without an actual torrent being transfered, consumes 23 MB of memory (on Windows) -- for that cheesy, very simplistic little GUI. When you actually start running a torrent through it, it'll easily chew 40 MB's and gobble considerably more CPU time than a comparable program written in C/C++.

I'm not saying Python isn't a useful language... But it was not designed to run P2P apps.

Just because a programming language can be extended to creating GUI applications does not mean it's a good idea. Python's strengths are elsewhere, and I for one am tired of the BitTorrent community using it to write p2p clients in.

Now go ahead and mod me down for having a modicum of common sense.

Re:Do we really need...

[cpghost]

Sure, like Pearl or Java,

Pearl? You mean Perl, right?

Just running the application, without an actual torrent being transfered, consumes 23 MB of memory (on Windows) -- for that cheesy, very simplistic little GUI.

Python itself is awesome when it comes to rapid prototyping. But nothing forces you to use it. ctorrent is a nice BT CLI client written in C, which won't use that much memory. Remember: it's about the protocol itself. As soon as it stabilizes, the apps could be recoded in C.

Re:Do we really need...

[Rakshasa Taisab]

It's a good strategy to prototype in a language that is quick to code in, and easy to change. It's easier to make a stable app in python than C++. You can let other people implement a more efficient client later.

Shameless plug: http://libtorrent.rakshasa.no/

Re:Do we really need...

[yerM)M]

bzzzt, nice try. You do know that you can run bittorrent without the GUI don't you? Check out btlaunchmany.py. This has multiple benefits including that you can download several torrents in the same process plus the memory footprint is much, much, smaller. There is even a command to download all torrents in a directory, very useful stuff.

The size of the application comes from one thing, wxWidgets which, you guessed it, is written in C++ and not python. The MacOS version runs directly on ObjectC and is much smaller. There might be a direct windows port, python can interface with the win32 api, check out venster. So now you have two options for reducing the size of bittorrent, get coding!

Obligatory rant: If you are tired of the BitTorrent community using python, then write it in another language and stop complaining, nothing in life is really for free.

Re:Do we really need...

[pcmanjon]

"Do we really need yet another bloated python p2p app? I can feel the flamebait and troll mods comming.. but seriously: Python sucks at gui work. It has to use generic wrappeers, like wxPython, that are extremely inefficient. Sure, like Pearl or Java, you can write gui apps using Python... but they always come out slow and over-weight. "

Mod parent above insightful not flamebait. WTF is up with the mods? Were the mods VisualBasic fans or something who downmodded this post? He has a point, python and java **ARE** slower than C/C++ applications, have crappy memory management and are usually more of a kludge.

I think parent deserves +3 insightful if anything. It's about time developers trashed these 'compilable scripting languages' and went to C/C++

I code a partitioning tool completley PHP and then compile it with Zends optimizer, but would it go fast?

Re:Do we really need...

[Hatta]

I'm not saying Python isn't a useful language... But it was not designed to run P2P apps.

Just because a programming language can be extended to creating GUI applications does not mean it's a good idea.

I don't follow, what does P2P have to do with a GUI?

Re:Do we really need...

[Valar]

You know... I don't recall it having a footprint that big. I'm without a windows box right now, but on by iBook, the BT client with a transfer running is using 8 megs. Sounds like the problem is in the specific widget library used for the windows version (which happens to be written in C++). Of course, considering the possiblity that something is wrong with a specific library or program is the rational thing to do. Do you think anyone designs an interpreter (especially one that implements garbage collection!) that is just going to allocate 23 megs of memory and do nothing with it?

Another thing to consider is this: given the choice between getting something done and having it be slightly inefficient and having a great unimplemented specification, I'll take the program over the wishware. If it is more likely to get done with python, then that might be the right choice.

Re:Do we really need...

[shish]

btlaunchmanycurses, downloading 15 files in small chunks: ~30MB
btdownloadcurses, downloading a single file in 4MB chunks: ~300MB

Based off numbers like that, and other random probing, I'd think that number of currently downloading chunks * chunk size is much more of a consideration than the GUI

Re:Do we really need...

[runderwo]

thanks for increasing the Signal to Noise Ratio on Slashdot.

Um, he'd be decreasing it. Or are trolls considered the "signal" now that the rest of the site is so humdrum and useless?

This isn't the MPAA's worst nightmare (yet)

[RebornData]

If you read the article carefully (or not so carefully), you'll note that this product does NOT include a fully distributed / decentralized tracker... an web server tracker is still necessary for the initial torrent retrieval. If that tracker becomes overloaded / unavailable this system will have real value, but there's still an originating central tracker for the MPAA to go after.

However, it's only a very short matter of time. The author explains that such a thing could be easily created with this framework. Clearly he could have done it if he wanted, so I'm guessing this is a purposeful strategy on his part to avoid any potential direct or indirect personal liability or legal issues down the road...

-R

Re:This isn't the MPAA's worst nightmare (yet)

[barawn]

but there's still an originating central tracker for the MPAA to go after.

Absolutely no way could the MPAA go after the "kenosisp2p" DNS server. Otherwise all DNS servers are committing copyright infringement. When you feed "ftp://www.moviewarezsite.com/pub/MPAA_Movie_Here. avi" into a program, it does the exact same thing as this.

The MPAA could not go after the DNS server that resolves "www.moviewarezsite.com" (yah, good luck!) - but they could go after the holder of the IP address that it resolves to.

The problem now is that the MPAA still can't go after kenosisp2p, but the holder of the IP address that it resolves to changes quite often. They might end up going after those people, but they could get someone who's in another country, or has no money, or has only traded one movie ever, or has a hacked computer. Not a good legal strategy.

I'm not saying that they wouldn't try going after kenosisp2p. They might. But if they do, they'll almost definitely lose. They probably will lose going after BitTorrent trackers as well, but there's some argument there as there's hash information being passed around.

Re:This isn't the MPAA's worst nightmare (yet)

[pcmanjon]

"Absolutely no way could the MPAA go after the "kenosisp2p" DNS server. Otherwise all DNS servers are committing copyright infringement. When you feed "ftp://www.moviewarezsite.com/pub/MPAA_Movie_Here. avi" into a program, it does the exact same thing as this."

Maybe it would be illegal according to a judge, there's never been a court sueing based on this premise before has there? The situation is also a TAD bit different than your example.

In the end though www.your-favorite-pirate-tracker.com would still probably go bust, unless the owner was rich. Having to afford tens of thousands of dollars can kind of make you want to settle with the *IAA out of court AS FAST AS POSSIBLE

Re:This isn't the MPAA's worst nightmare (yet)

[sycomonkey]

I don't know. Once it's ironed out a little bit it could prove extremly useful even without the decentralized torrent retrieval. I imagine, for example, an IRC channel where you directly download torrents and then various Ops and such run a network of trackers such that they would all have to fail to stop the distribution. I'm very interested in this project, myself.

Re:This isn't the MPAA's worst nightmare (yet)

[barawn]

The situation is also a TAD bit different than your example.

Howso? The DNS server has no way of knowing about the content of the website that it returns the IP address for.

Likewise, the kenosisp2p server has no way of knowing about the content of the node that it returns the IP address for.

In the end though www.your-favorite-pirate-tracker.com would still probably go bust, unless the owner was rich.

Yup. But the DNS server that provided you with the IP address to www.your-favorite-pirate-tracker.com wouldn't be touched.

Re:You /.'ers are horrible people!!!

[wernercd]

Well with better P2P software, we can pirate more stuff thus have more money to donate. Sheesh. Look at the bigger picture.

Not meaning to sound mean or anything myself here, as I do feel for those in the path of the devestation.

Did Miss something? Since when does an accident stop the world from turning? It doesn't! In this case the world actually turns faster!

There's alot of horrible things in the world today. Tsunami is just one of them. Twin Towers was another. Many Many others out there. The world moves on. So should YOU. Help if you want to. Support others that want to help. I donated. I'm also living my life.

Get a fucking clue, please.

Re:I don't get it.

[ledow]

RTFA and read it properly. The server is merely a pretty interface for older BT clients that will search the decentralised version and return the tracker address of the last known tracker for that item.

That's got nothing to do with the decentralised network itself.

Circumventing central DNS servers with spam!

[cpghost]

The problem with Kenosis is, of course, it's reliance upon a central DNS server to point to a list of distributed trackers. Many will undoubtely point out, that this DNS server could be taken off, and that's it.

Now how can we really circumvent this problem? One solution would be to advertize a list of DNS resolvers on USENET. A preconfigured list of newsgroups could be used to bootstrap this, and new usegroups (should the original newsgroups get closed) could be regularly advertized as well. A client would just go to those newsgroups, and fetch the updated list of DNS servers, newsgroups etc...

This system would be much more resilient to attacks by RIAA or MPAA because they won't have a single point to attack. Closing newsgroups is much more difficult than taking one DNS server from the upper zone.

Another way to advertize the DNS servers would be via spam! Yes, you didn't misread this. One can easily encode the location of DNS servers in spams and have clients read those spams, effectively extracting an updated list every now and then!

This is very important, because spam is already used as a covert channel to prevent traffic analysis. Specialy crafted spam checkers can extract useful information from spams. One such information would be the distributed location of trackers (or DNS servers that point to them).

Just because it's unethical (to piggy back useful data on top of spam), doesn't mean that it's not already used on a quite wide scale. There's no reason why it shouldn't work on a new generation of distributed BitTorrent trackers!

Re:Circumventing central DNS servers with spam!

[Anonymous Coward]

Perhaps you should see the other posts, or even better; RTFA.

The system does not rely on a single DNS server. Only for backwards-compatibility.

Re:Circumventing central DNS servers with spam!

[Kunnis]

From what I understand, you're missing the point of Kenosis... the entire DNS thing is for reverse compatability. I can use Kenosis to get the .torrent file, OR I can use DNS to resolve the ip to get the file, either way I can download the .torrent file. if the Kenosis DNS is taken down, we can still use Kenosis to get the torrent files, or at least that's what I understand.

Re:Circumventing central DNS servers with spam!

[slavemowgli]

Hey, that's a great idea - not so much one to advertise DNS servers (or whatever), that is, but to crack down on spam. Just use spam as a vehicle for transferring data that those in power don't want to be transferred, and spam will be dealt with in an effective way faster than you can say "spamassassin". :)

Re:Circumventing central DNS servers with spam!

[complete loony]

For legacy BT clients (and only for legacy clients) this process relies on a DNS hack to resolve the hash of the file to the current tracker. The parent domain is included in the torrent file. The torrent could point to any domain that included this DNS hack. The BT client does not need to obtain a dns server externally, only those publishing a torrent would. For futher armouring, the DNS server could reply based on a white/black/grey list to attempt to keep the MPAA and it's agents off the network.
On another note, this DNS hack could be implemented in your local DNS server allowing any .blah.com to resolve the current tracker, and allowing you to run you favourite BT client.

Re:Circumventing central DNS servers with spam!

[cpghost]

Hey, that's a great idea [...] to crack down on spam.

That was intended from the start! :-) Or let your representative know that terrorists [could] use the "spam network" to hide their tracks... and pronto! spam will be outlawed faster than you can say Rolex!

Re:YAP2PA

[Anonymous Coward]

You suffer from what is know as "lack of vision"

btw , who the hell are you to tell others what they should be doing?

just because you are too narrowminded to see the value in this, doesnt mean everyone else is.

you came up with some ideas, better guis. where is your work on the subject, where is your code. what have you done. ohhhh you expect everyone else to have the same priorities as you so you can just leech off their work all while being a dick about what people are doing with their time.

do something and stop bitching to others about what they choose to do.
i dotn see you doing jack.

Decentralized?

[MasTRE]

Lets simplify this. You are a program that doesn't know anything about the world, because you are a de-centralized program. You are started by your master ("user," in human speak). What do you then do? Who do you connect to? Surely if you had an address hardcoded somewhere you would no longer qualify as being decentralized. Do you start walking the IP space, trying to connect to 1.1.1.1, 1.1.1.2, and so on? Oh, so the IPs you have coded in your config are "only hints," huh? Okay, then you should be able to cope with all those "hints" having gone bad. When those hints are all bad, what do you do, Mr. D. Centralized Program?

Decentralized, my ass.

Re:Decentralized?

[garbletext]

So would you call Kazaa centralized? It's not, because there is no one server that you could shut down to destroy the network. In that loose sense, Bittorrent is already decentralized, because it requires a tracker for each package, which is easy to set up, and not by any means a central point of failure. No **AA crony can use a court order to stop all bittorrent activity, they can only shut down a single file at a time.
The IP addresses a decentralized P2P client connects to obviously are not hard-coded into the program, and can be arbitrary and manifold. In a centralized model, like napster or soulseek, the user has no choice of where his client connects to.

Re:Decentralized?

[Chexum]

Okay, then you should be able to cope with all those "hints" having gone bad. When those hints are all bad, what do you do, Mr. D. Centralized Program?

• Depend on my user to replace me with the latest version, which is available, since not every user of that is a copyright-limit-explorer.
• Pray that in those years multicast is finally implemented by the ISPs, and listen to the next periodic update on the hash-dependent randomized multicast address.
• Wonder who could shut down the whole Tor network (useful for many things citizens really need to do), and how could they find all of the tor-hidden 4914a6dfd3634e54f9f8457ca4cb6f39.onion hosts.

Just my two cents. I happen to be that bitter too sometimes, but I more and more believe technology *will* be advancing us. Not as in entrepeneurs, but as humanity.

Re:Decentralized?

[tepples]

The IP addresses a decentralized P2P client connects to obviously are not hard-coded into the program, and can be arbitrary and manifold.

Then watch the copyright holding company trade groups go after web sites that host a GWebCache (for Gnutella), a server.met (for eDonkey), or other protocols' equivalents.

Re:Decentralized?

[tepples]

Depend on my user to replace me with the latest version, which is available, since not every user of that is a copyright-limit-explorer. Pray that in those years

You mean pray that in those years legislatures have not implemented even more measures against recognition of substantial non-infringing uses.

Re:Decentralized?

[gibson042]

Well, this app uses BitTorrent, and thus relies on .torrent files in order to get anything. One could provide an IP address alongside every .torrent, to be passed as the --kenosis_bootstrap option. Is it messy? Yes. Will it work even if root.kenosisp2p.org:5005 goes down? Yes.

Of course, it would also be trivial to keep lists of long-lived and reliable Kenosis nodes. In that case, I believe a simple Google search would come to your rescue.

Decentralized does not mean omniscient; it means not reliant upon static centralized nodes. From TFA: any Kenosis node can effectively join the network ("bootstrap") from any connected node. I think few will object to trying to bootstrap using a known and likely active node before requiring user intervention.

Free Culture Good, Piracy Bad

[KrackHouse]

There is a ton of good legal content that will be created once the bandwidth issue is solved. It's sad that the default comment is "well this sucks because the **AA will still be able to track me down when I use it to break the law." Most of use see the cultural usefullness of these things but the handfull of anarchists among us are hurting the movement.

The fact that this can get through firewalls and that it won't fail under heavy load (as happens with bittorrent trackers) are the important things.

chocolate bunnies from hell

[sum.zero]

old band. i am getting old. soon i will be dead. sigh...

sum.zero

MORE centralization with Kenosis?

[DarkHand]

Correct me if I'm wrong, but wouldn't routing all tracker requests through a single domain create MORE of a bottleneck?

0 defects...Can see at least one

[flibuste]

From the article:

Kenosis is also "zero-defect software".

Ok Mr Kenosis, you just lost my vote. Considering that computer science theoricians (namely Goeddles) mathematically demonstrated the impossibility of a "zero-defect software", that makes the article quite impossible to trust or even consider. I hate when people think they are infaillible.

Re:0 defects...Can see at least one

[MikeBabcock]

The goal is nonetheless laudable. Dan Bernstein is quite willing to guarantee his software as being security defect free. I don't see why, although impractical in many cases, it wouldn't be possible to guarantee freedom from explicit defects in a piece of software.

Re:0 defects...Can see at least one

[flibuste]

It is definitely a nice goal to strive having as less defects as possible in a software. Now as for "explicit" defects, those can be fixed since you know what they are. The theory is very different when you come to consider the instable nature of a computer (you CANNOT build a defect-free software because of this. It actually boils down to the state of your computer at a given moment in time, which will never be the same the next time, hence your software may not behave the same way the next time you run it.) I reckon this is highly theoretical, but saying my piece of software is "completely defect free" is highly pedantinc!

Re:0 defects...Can see at least one

[eries]

Thank you Mike, you are quite correct. Our goal with Kenosis was to hold ourselves to a higher quality standard than is usual with the litany of "p2p app of the month" candidates that were neat ideas but completely unstable (and hence, unusable as a platform).

Of course we'll fall short of the goal, but this way we encourage everyone who finds a bug to call us out and make us feel bad :)

Re:0 defects...Can see at least one

[GileadGreene]

Without getting too bogged down in terminology, you are confusing (conflating really) "defects" and "failures". A failure is an observed behavior that doesn't match the specified behavior of a system. A software defect is a flaw in the design/implementation of a piece of software, where correctness is measured with respect to an assumed execution environment. A software defect may or may not result in a failure (or an error leading to a failure) depending on whether or not the code branch in question is executed. However, as you implied, it is possible to have system failures that result from something other than defects in the software (for example hardware defects). These failures are caused by variations in the software execution environment that fall outside the set of assumptions used to generate the software design. There's an entire field of research (software-implemented hardware fault-tolerance) dedicated to developing software that can mask execution environment defects (essentially by adding the potential defects to the assumed execution environment). This is distinct from software fault-tolerance, which seeks to mask defects in the software design (using e.g. recovery blocks)

Re:YAP2PA

[Anonym0us Cow Herd]

Please, give it a rest. You cannot change the bottom line. Spend your time writing better GUIs and promoting a 'standard'.

I agree. Joe Fourpack needs P2P applications with better GUI's; and a standard p2p is needed so that everyone's shared file collections are not scattered amongst different p2p protocols.

exeem anyone?

[Polysics]

this sounds great, but havent we been waiting a while now for exeem (which they're taking their time to develope to make sure it really works)?

how is it a project like this can come completely out from under the raidar. this sounds exactly like exeem, but out much quicker.

is this gonna be an app that winds up to be very disappointing because it was hyped up before it was thuroughly tested?

Re:exeem anyone?

[nagora]

this sounds great, but havent we been waiting a while now for exeem (which they're taking their time to develope to make sure it really works)?

No, they were taking the time to get advertising. Exeem is dead in the water, see the previous discussion. I can't imagine anyone wanting to use this pile of steaming dog crap now that the details are out.

TWW

Re:exeem anyone?

[LuSiDe]

No! We haven't waiting a while for Exeem. At least, i haven't.

Exeem is:
* Adware/commercial.
* Proprietary, Windows-only.
* It has nothing to do with Suprnova.org. The people behind Exeem just use Suprnova.org as part of their advertising by saying Exeem is a follow-up to Spurnova.org. It is not, however.

Give me equal or better software than this without adware, and with a honest developer behind it, and i'm over to that.

(Although i prefer Usenet.)

We are looking for help

[mcm]

(I am one of the authors of Kenosis.)

We are planning improvements to Kenosis in a number of areas such as better integration with BitTorrent, a more distributed BT tracker, simulation of larger Kenosis networks and making Kenosis work over NAT.

We'd love help with any of these or other areas.

Please join the mailing list to get involved.

Re:We are looking for help

[wdd1040]

What about adding encryption with a fall-back deniability feature? (Example, the Off The Record plugin for GAIM)

Re:We are looking for help

[eries]

That is a very good idea. Any other suggestions?

Want to join the development effort?

What about Azureus ?

[rpk]

Azureus is an open-source Java-based BitTorrent client with a built-in tracker.

Re:What about Azureus ?

[shish]

The point isn't that each node has it's own individual tracker optionally available, but each node is part of a distributed net of trackers

A good start

[nrlightfoot]

While this looks like a good start, this isn't likely to catch on until it can be installed from a single .exe file for windows users. Then it would have to have one GUI that provides a seemless interface for finding and downloading .torrent files distributed among Kenosis nodes, and then automatically starts downloading the files using the Kenosis distributed trackers.

P2P without destination IP addresses

[BobPaul]

As soon as you enlighten us on how to send a directed IP packet without a destination address, I'm sure someone will write a few dozen p2p apps around it.

1) Have your p2p app run in promiscious mode.
2) Sniff traffic and find an IP address on your local network that you can see traffic flowing to.
3) Instead of giving the other p2p nodes your address, give them one of the addresses you found through sniffing.
4) Your P2P app must now scan for all traffic going to the address you gave to the other P2P nodes looking for traffic destined for you. As I understand networking, the computer who's address you're using will be recieving lots of data not intended for it, but it should just end up ignoring all of that. The only time this will be a problem is if it's listening on the same port that the P2P packet is sending to.

Unfortunately, this probably won't work on the internet where the prevalence of NAT Routers etc to access the internet, and that I don't think you can put a cable/dsl modem into promiscious mode.

On a LAN environment free of switches, this has potential, but that's about it.

Re:P2P without destination IP addresses

[dossen]

> On a LAN environment free of switches, this has potential, but that's about it.
And how often would you expect to be on a switch-free LAN? Aren't most (ethernet) networks (even two host home networks) done with switches? The only exception I can think of is WiFi, and I'd say the odds of those having public routable IPs are low.

Re:P2P without destination IP addresses

[BobPaul]

And how often would you expect to be on a switch-free LAN? Aren't most (ethernet) networks (even two host home networks) done with switches? The only exception I can think of is WiFi, and I'd say the odds of those having public routable IPs are low.

That's exactly the problem. At the university I attend, we're on a 10MB ethernet network directly attached to smart hubs. The backend uses 100Mbit switches and routers, but the students (dorms, labs, etc) are still only on smart hubs. Something like this could be used on our network to avoid detection by the school (perhaps as a layer between WASTE and the network) but has very little application to most networks or even the internet as a whole where you cannot see your neighbors traffic, which is why it'll probably never be implimented.

To be fair...

[Dekks]

I was raised christian but don't practice, make your own preach joke :) Anyway, to be fair I don't think it was the church who started producing chocolate and making easter synonymous with gorging yourself with candy.

Re:To be fair...

[Dekks]

Thats a fair comment, I was meaning more I don't particularly think about it, don't go to church. I believe in a higher power but not nessecarily the bible etc, and in that regards you could say I lead a christian life, or a islamic life, or bhuddist or anything other religion that basically says don't be a dick, and try to help others if possible.

ob Eddie Izzard

[Drathos]

And kids eat chocolate eggs, because of the color of the chocolate, and the color of the... wood on the cross. Well, you tell me! It's got nothing to do with it, has it? You know, people going, "Remember, kids," the kids who're eating the chocolate eggs,

"Jesus died for your sins."

"Yeah, I know, it's great!"

"No, no no, it's bad, it's bad!"

" No, it's bad! It's very bad. It's terrible! Whatever you want, just keep giving me these eggs."

And the bunny rabbits! Where do they come into the crucifixion? There were no bunny rabbits up on the hill going, "Hey, what, are you going to put those crosses in our warrens? We live below this hill, all right?" Bunny rabbits are for shagging, eggs are for fertility. It's a festival - it's the spring festival!

Re:ob Eddie Izzard

[P0ldy]

Small note: Hey, what, are you going to put those crosses in our warrens? "in our forest."

Re:ob Eddie Izzard

[Drathos]

don't know where you got your version, but on both my dvd and the mp3s i've got (which are slightly different), it's 'warrens'

Re:ob Eddie Izzard

[P0ldy]

I suppose I misheard. Makes sense now that I think about it.

Congratulations.

[Kickasso]

You just have mathematically demonstrated that you have no clue whatsoever about Gödel's theoretical work. Way to establish your street creed.

Re:Congratulations.

[jayed_99]

Thank you!

I was looking at the parent post and thinking to myself "Goeddles? Goeddles? Who is this guy? Could it be Gödel? Nah, dude said this guy was a "computer science theoretician"; if he meant Gödel he would have said mathematician. Shit, I'm gonna have to google for this Goeddles guy -- I wonder what other cool stuff he said".

Re:Congratulations.

[Jonboy X]

Way to establish your street creed.

Way to expose your own deficiencies while nitpicking others'. I believe the word you were searching for was "cred", as short for "credibility".

BT traffic as a percentage of internet bandwidth

[runamok1]

This may be off topic...

I was wondering what the slashdot community thinks of the estimates that BT uses 1/3rd of available internet traffic.

Considering that it is a bit more private and "exclusive" than things like Kazaa and Edonkey does that number seem possible. You know. It takes the install of the basic BT binary as well as the GUI client of your desire. Then you need to find a decent BT website/ community.

If this figure IS true, wouldn't making BT even more viable eventually choke the internet?

Thoughts?

It would be ironic that something that was designed to "manage" bandwidth may end up hogging all that is available.

What About For WIndows?

[StratoChief66]

Do you know of a bit torrent client written thats faster for windows? cTorrent might be useful for Linux users, but is there an alternative for windows users. I've seen that bit torrent is slow and am interested to know that it is because it was written in Python.

Re:What About For WIndows?

[mushroom blue]

yeah. actually, Azureus is much faster. on my computers (windows and linux), it's barely a speck of CPU usage.

of course, the official clients under windows are about the same, as they're compiled into machine code, so it shouldn't be having any speed issues either.

woo!

[mattyrobinson69]

got an ebuild anyone?

Re:woo!

[justrob]

I committed an ebuild to portage about an hour ago ;)
net-p2p/kenosis

Re:woo!

[mattyrobinson69]

good work.

Re:writer/programmer/cto

[multiplexo]

Oh, man, you don't know how many times I see that on resumes today. Everyone who has a blog and wrote some little rinky dink peice of software under their little fake business puts that on their resume. Fact is, they are just another out of work programmer trying to fill the 3 year gap.

The one I love to see on resumes is "HTML programmer". "HTML programmer"? WTF is that? What did you do before that? WordPerfect and WordStar programming? When I have an embarassing gap in my resume I usually lie and say that I was a transvestite cabaret dancer. It might not be true but it does get HRs attention.

It's even easier than that

[shadowmatter]

DHTs work like this: Every node on the network has a 160 bit identifier. Given a key, through the DHT we can find the node whose identifier is 'closest' to the key. In Kademlia, the closeness of a node is quantified by treating its identifier XORed with the key as an unsigned integer. The node with the smallest such integer is the closest, and is therefore responsible for the key.

If you look at their readme file, they're just using the hash of the file kept in the .torrent as the key (extracted from the tracker URL, in the .torrent). So say you have a .torrent whose tracker you would like to eliminate. Just choose your node identifier, when you join the network, as either equal to the hash in the URL or close to it (such as by simply flipping one of the lower-order bits). That way, you will with near-certainty be the closest node to that hash, and thus be designated the tracker for that torrent. Now just ignore all requests from clients.

It can easily be done.

- sm

Re:It's even easier than that

[Rei]

Clever. I didn't even notice that ;) Allowing clients to basically pick the peers of another client is a big no-no in decentralized systems.

Re:It's even easier than that

[mrogers]

Looks like we'll finally get a large-scale demonstration of how well (or how badly) DHTs cope with Zipf-like query distributions.

Re:It's even easier than that

[mrogers]

In many DHTs (I don't know about Kenosis), your node ID is the hash of your IP address - you don't get to choose your ID. (If you have access to a large number of addresses you can try them all and pick the closest, but chances are there's someone else in the network who's closer still.)

Still

[TubeSteak]

None of this matters if the first seeder never manages to complete his upload.

The other issue i see is that when a DDOS is carried out, it'll quite possibly affect significant numbers of computers on residential connections... Your cable/dsl/sat company won't like this.

Does it work with Cygwin?

[tepples]

ctorrent is a nice BT CLI client written in C, which won't use that much memory.

For C and C++ programs using the POSIX API, you have to take into account the overhead of Cygwin as well.

Kneejerker suggests: "Then just dual boot!" Clients for file transfer networks are designed to run in the background while the user does something not bound by continuous network throughput in the foreground (such as word processing, image editing, surfing an HTML based web site, or playing a 2D video game) on the same machine. Rebooting between a Free operating system for the client and a more popular proprietary operating system for use of peripherals with poor Free driver support stops the transfer, defeating the purpose.

Kneejerker suggests: "Then just use a Free operating system to begin with!" Buy me a new scanner and I will.

What "good, legal content"?

[tepples]

There is a ton of good legal content that will be created once the bandwidth issue is solved.

Oh really? Watch the big copyright owners sue the independent producers of so-called "good legal content", alleging that the so-called "good legal content" is in fact subconsciously copied from an existing copyrighted work. It could happen, especially with music.

Publishing ease.

[believekevin]

Self-publishing content without the need to find a reliable tracker will help lawful BT users considerably.

Re:Cool, now we can have viruses that communicate

[Daengbo]

Right, so simply taking the original site the virus downloads from won't affect it. Cool!

An actual case: read it and weep.

[tepples]

Basically a bunch of speculation, that COULD happen. I COULD get hit by a bus ... Call us when it ACTUALLY happens.*

Had you taken the time to click the link, you'd see a bunch of cases leading up to one where it has happened. In Bright Tunes Music v. Harrisongs Music , George Harrison got sued and lost for inadvertently copying a song on his solo debut album.

Seems to be similar to Exeem

[Guspaz]

However it also seems to lack Exeem's kazaa-like simplicity.

Each has it's pros and cons. Exeem will very likely be a much easier and more streamlined solution, but BT/K, being opensourced, opens the possibility of support in other opensourced clients such as Azureus (Which seems to be the most popular client based on what I see in swarms).

Re:An actual case: read it and weep-Kleenix.

[tepples]

Also the allegation still has to be proven (there's a reason we have a legal system).

Most independent artists don't have the means to finance a legal defense. Many Slashdot users agree that the litigation process needs some sort of reform.

Courts also do take into mind "what did so and so do to comply with the law?" e.g. Compaq and the IBM bios.

So which steps would you suggest that a songwriter take to avoid subconsciously copying a song he happened to have heard 15 years ago on commercial radio?

torrents over gnutella or something like it? why

[jetpeach]

This is a question that I have been wondering about, why couldn't somebody code an application that used the gnutella network (or some varient similar to it) to share torrent files? For example, the application ignores all the non .torrent files, such that when you search, it works similar as gnutella network and displays the number of any .torrent files shared. Then you download the torrent file first through gnutella or something like it, which is tiny and will be fast, then the BT part kicks in and the speed of the bittorrent protocal kicks in. It seems very simple and would offer solutions to a lot of the problems people keep mentioning. Of course, if it was attacked it would need modification etc, but I am just wondering conception why this wouldn't work. jet

That's not how Betamax reads.

[cduffy]

The legal standard is substantial noninfringing use. Not "primarily noninfringing use", substantial. If you want to use your own standard, fine -- but I think the one encoded in current US case law works quite well, thank you.

I agree inasmuch as the fact that laws are being broken on a regular basis indicates that there is indeed a problem somewhere -- but it doesn't mean that the problem is with the technology used to break the laws. Whether it's related to the behaviour of the people who break the laws, the economic models which encourage individuals to break said laws, the laws themselves, or other factors is well beyond the scope of this discussion.

Finally, coming back to topic: Having distributed failover for BitTorrent trackers is a Good Thing! If I'm providing a massive download (say, a patch to my game) to a number of users and someone runs a DDOS on my server, I'd really quite prefer that the download still stays up.

Re:torrents over gnutella or something like it? wh

[Joel from Sydney]

You'd suffer from the Gnutella network's biggest problem as well, fake files. Using this method it would be very easy to start spoofing files. That's the beauty of a centralised web repository of .torrent files, they are guaranteed to point to known good copies of files.

Re:Do we really need... Another exploit!?

[Admiral Burrito]

Do we really need yet another bloated python p2p app? ... it'll easily chew 40 MB's and gobble considerably more CPU time than a comparable program written in C/C++.

Do we really need another network-heavy (client and server) C/C++ app with multiple buffer overflows waiting to be exploited?

I, for one, am glad this thing is written in Python.

Re:torrents over gnutella or something like it? wh

[bbtom]

Plus, it isn't the distribution of the .torrent's that is the problem Kenosis is trying to solve, but the role of the trackers.