Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

Fling:Anonymous Protocol Suite

Posted by CmdrTaco on Mon Jul 10, 2000 09:23 AM
from the who-said-that? dept.
_endgame writes "Fling is a new suite of internet protocols that perform the function of DNS, TCP, and UDP in a manner that's both untraceable and untappable. Fling protects clients from servers, servers from clients, and both from an eavesdropper in-between. The result is that anyone can serve or retrieve any data, without fear of censure."
+ -
story
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • ...this project is less than a week old and consists of some theories bandied about by a developer and he's friend (who is providing the crypto knowledge).

    Wouldn't have been better to post this when there was actually news to report? Simply because someone has an idea and backs it up with a webpage does not a headline make.

    PS: That said, I wish them luck. :)

  • One of the things that always strikes me as interesting about things like this is the posiblities for abuse. No - I'm not talking about things like trading warez, porn, MP3, or whatever the hot semi-illegal commodity of the week is.

    I'm more interested in the possible effects for companies that keep wanting to do things like map out the Internet (see article last week here on /. about the group maping the 'net for advertising purposes) but don't want to really tick off admins who's machines they are adding to thier map. Same goes for script kiddies looking for machines (using nothing more than ping to see who responds) but want to keep from possibly alerting the admin at some company they are maping out.

    Just a thought - I could, of course, be completely wrong!

  • by / (33804) on Monday July 10 2000, @04:34AM (#945998)
    Ah, but an idea backed up with an open-source-oriented webpage (sourceforge.net [sourceforge.net]) has always a headline on slashdot made. Besides, since some of the planning has been done, it will soon be time for bandwagoners to start contributing code, and it's nice to have such a heads-up.
  • What better way to attract attention and get some serious development effort aimed at it? For those of us who don't want solutions handed to us on a silver platter, this is the best time to get involved.


    ...phil
  • by Signal 11 (7608) on Monday July 10 2000, @04:37AM (#946005)
    There's no way to prevent man-in-the-middle attacks with a truly anonymous protocol as there is no way to verify the authenticity of the server.

    In addition, crypto without a pre-arranged way to mutually verify both parties is trivial to crack. The NSA will certainly not mind you exporting this protocol overseas. :P But that is just a footnote to the above problem I mentioned. You can probably derive the encryption keys by monitoring the beginning of the conversation with the server and thus decrypt the contents of the packet(s). However, I am no expert in this, so I may be incorrect about being able to derive the keys - specifically, I know nothing about the duffie-hellmann(sp?) public key exchange stuff, beyond "it works", so YMMV.

    The other problem I can see is that you're sending up a big red flag saying "Here I am! Look at me, I'm up to no good!" to your network administrators. Net admins are notoriously paranoid, moreso now with the proliferation of scripts. This means that if you use it at work, you stand a good chance of having your network access monitored/revoked and/or you getting your ass canned. Yeah! Go crypto!

    The ideal protocol for this would be one where monitoring would a) do an attacker no good (which means you have to verify the authenticity of the server somehow before you communicate over the unsecured channel (the 'net)) and b) look like normal traffic. This is important - either you encrypt everything, even non-sensitive material, or you encrypt nothing and rely on stenography. I like stenography better myself.. and it'll become more important as governments crack down on conventional crypto - witness new zealand, I believe, which made it a law forcing you to divulge the keys of every encrypted thing on your system under penalty of jail.. even when they can't prove you ever had them!

    Imagine an HTTP request to www.someplace.com where the downloaded JPEG contains the information requested and the POST contents contained the key+query. E-commerce cookies can easily look like crypto keys. Rewrite a few doubleclick cookies and no one will be the wiser.

  • by grahamsz (150076) on Monday July 10 2000, @04:37AM (#946006) Homepage Journal
    But what it lacks are any suggestions of how the system would scale... will it be like gnutella which now has so many users that the average modem user is struggling just to connect to the network.

    Plus if my PC ends up routing mp3 files for other people using my 128k connection I wont exactly be pleased.

    Added to this I would expect that there will be quite a reasonable bandwidth overhead given all the layers of encryption.

    Certainly as a system for trading textual data it's reasonably sound but then usenet probaly works just as well for most people.

    Added to this for a user to keep information persistantly on the network they still must be permanantly connected... which isn't really an option for opressed tibetan monks is it..?
  • by phil reed (626) on Monday July 10 2000, @04:38AM (#946009) Homepage
    The author's justifications are very much anti-tax (he appears to be a serious Randian). One of the unstated reasons that the U.S. government was believed to be anti-crypto was exactly that the widespread distribution of unbreakable crypto would allow the development of an underground untaxable economy. It's interesting that this web site's author comes right out and says pretty much the same thing.


    ...phil
  • I'm going to paraphrase the movie about Larry Flynt's life starring Woody Harrelson here. He said that he was a scumbag, the lowest of the lowlifes, and if the law protected his right to say what he wanted to say, then you be certain that the law would also protect fine upstanding citizens like ourselves.

    Well, we're not really willing to simply *trust* that the law will protect us. We want to ensure that the scumbags can never be censored. If that happens, then we find upstanding citizens can also never be censored.
  • by NetCurl (54699) on Monday July 10 2000, @04:50AM (#946022)
    Sure this sounds great in theory, but considering the current state around the world, how would this be received?

    The economy is globalizing quickly, and daily interaction across the globe is paramount. So considering China just recently picked Linux over Windows95/98 because it can examine the source code to make sure there aren't any caveats that the US could use to sabotage them in a crisis, and on the other hand, the US is so paranoid about other countries being super-secretive that they delayed the release of Apple's G4 machine because it could perform well in encryption/decryption. Would the US allow China to have this Fling technology? Would it not try to stop certain countries (*cough* Iran, China, Lebanon, North Korea *cough*) from utilizing "super-secure" technology to transport data?

    This project may be doomed to the "oh-that-was-a-neat-trick-but-where-is-it-now?" hall of fame from the start.
  • Do we really need a way for people to pass around child pornography without having a way to find out who they are (so we can stop them)?

    In a word: Yes. We do. For the simple reason that there _is no way_ for any of us to exert our simple right to anonymity without having a way to pass round child porn too.

    This is one of those circumstances where people will have to choose between a greater evil and a lesser evil. At risk of making myself very unpopular, I would suggest the evils that can come from denial of freedom of speech could be an awful lot worse than the evils coming from the hampering of one of the ways the police use to track down a class of particularly unpleasant criminals.

    Put it this way: would you like every tiny piece of data about yourself in big government database, even though this would clearly help to catch many criminals, probably including some child pornographers? Supposing you didn't mind this. Now would you make it compulsory for _everyone_ to be in this database? That's what you're asking.

    Supposing the goverment could identify the profile of a child pornographer with 90% accuracy from this data. So they imprison all the people with these characteristics. This is another way the government could reduce child porn, but few would argue that the benefits outweighed the drawbacks.

  • I don't think anyone can define what a scumbag is.

    It's more difficult than some people think. It's just as difficult as defining pornography. There's some people that know it when they see it. Funnily enough, to those people, nipples and clitorae are pornographic. To me, when I see guns, violence, and Microsoft Windows, those things look like pornography.

    Without defining what a scumbag is, you cannot hope to censor them. If you misdefine what a scumbag is, then you'll certainly censor a person who doesn't deserve it.

    The only solution is to allow all people to transmit, without censorship. We don't live in a safe society. The world is dangerous. Boo Hoo! All in all, I'd rather use my intellect to avoid or combat messages that I don't like. Every other animal has to use their feet to avoid a wolf's teeth that they don't like. What chance do the sheep have of ever "censoring" the actions of the wolves? None at all. The choice is clear: We can act like humans, using our brains to fight ideas we disagree with in an absolutely free forum, or we can act like animals and hope the wolf doesn't like the taste of woolly fleece.

  • We now have the classic conflict generated by criminal thought on both sides of the issue.

    Protection from criminal actions by governments, and more specifically criminals in governments, big business, financial instituations, etc. who use and write the "law" to protect their own limited criminal interests is vitally important. Equally, protection from individuals who use such protection to justify and protect their own individual thievery and rape of the creative elements in the society is important as well.

    What we have is a war between the criminal elements that make up and contribute to the current internet and global culture. It is a war between criminal organisations who want to maintain their monopolies, and individuals who have been driven to criminal behavior by the rip offs in the world around them. It becomes a part of the culture. It is extraordinarily difficult to treat everyone you deal with with some sort of "code of ethics" or "code of honor" if you run into the argument that "only losers pay full price", as noted in a recent Salon Article; or you are trapped in the culture of "Net Slaves" [salon.com]

  • by hardaker (32597) on Monday July 10 2000, @05:22AM (#946060) Homepage
    Glancing through the web page quickly I note a few things:

    1. He's basically just adding a seperate data routing layer over the top of the standard IPv4 addressing space. Hence, data doesn't get routed only based on the IPv4 routing tables, but gets routed fairly randomly around above this. This has 2 problems:
      1. You still know the IPv4 address of the destination (regardless of weather or not DNS is protected) and hence can still trace the ownership of that address.
      2. Since data is no longer taking the shortest path, it'll get routed many times around the network and hence will increase the overall traffic level of the network at large (possibly sending the data over a given physical segment multiple times).
    2. He's assuming that by routing things around the network using different paths that it'll be harder to pick up all the traffic by way of a sniffer. This may be true if the physical internet truly had different physical routes. I suspect most sniffers you have to worry about are the ones at the end points, not the ones in the middle. It's the box next to mine thats more likely to be sniffing my traffic and hence that this protocol won't help. Now, it will encrypt it multiple times with possibly multiply different keys, but it won't prevent the majority of that traffic being sniffed.
    3. Root domain name ownership is not based on a pricing model. Hence I can:
      • for i in `cat /usr/dict/words`; do register $i; register $i.$i; done
      And the internet is hereby mine!!! Muhahahaa.
    4. Protocols designed by a few people quickly, possibly inexperienced in the world of security, will certainly run into security related implecations they hadn't thought of. I hope that something like this would go through a lot of peer review by cryptologists before being trusted.
  • Great! Not giving away your IP address is a fantastic idea! As long as we don't need to get information back from the server, it'll work for sure! Exclamation points can make the suckiest idea sound good if used right!

    Seriously, though, you need to reveal your IP address so the server can send back the information you requested. That's what servers do.

  • by scott@b (124781) on Monday July 10 2000, @05:36AM (#946070)
    IMO overhead isn't a large issue, bandwidth continues to get cheaper. Low power wireless communications will be an excepion to this; but that is generally shortrange stuff while unstoppable distribution is more of a wide area issue.

    Secure protocols will have more overhead because they need certain things beyond simply getting the data to the target. To avoid traffic pattern analysis you try to pad packets to fixed lengths, split streams up and send some junk so that bursts don't stand out, send dummy packets when traffic is low, and so on.

    You need secure low level protocols to give yourself a fighting chance at anonymous exchanges. Running such protocols at a higher level over something that is essentially an end-to-end protocol just points out the path used to route the `crypted data. At that point the unfriendly government steps in and has you blocked or arrested.

    The same technologies taht allow you to publish your anti-government newspapaer in a totalitarian state allow the distribution of porn and information on controlled substances. Sorry, information is information; differing states have declared diffeerent bits of information "bad" at times, the tools to supress one type can supress all types of information

    As for Fling specifically, I noticed that it uses IP4 addresses putting it behind current tech. I'd like it better if it's internal addresses were larger than IP6.

  • Wouldn't have been better to post this when there was actually news to report? Simply because someone has an idea and backs it up with a webpage does not a headline make.

    This is one of the weirder things about /. -- like the Ogg Vorbis open-source audio compression posts... They posted that a beta was scheduled, but submissions when the beta was actually released were declined. [http]

    I've seen this happen many times now, where a headline states that something cool is *going* to happen, but no posts when the thing *actually* happens.
  • I've got three unrelated problems with Fling.
    1. Technical The crypto as described seems to add very little apart from immense delays to transmission. Unless you use Fling for everything, it doesn't even begin to protect you, and using a multi-host route is going to probably multiply latency by the number of hosts you add. Frankly this looks like a half-baked MixMaster anonymous email scheme applied with broad strokes to all of low level networking.
    2. Adoption It won't be. You have to write this into any client you want to use it? Name service is a completely seperate entity? This would basically mean redoing the whole networking thing from square one (UDP), and the people who have to implement it are exactly the ones you're try to hide info from. Do you really think MS would ever put truly secure transmission protocols in their TCP/IP package?
    3. Philosophy The entire philosophy behind it is repugnant to me. The conclusions are somewhat accurate, in that there ought to be a means to be anonymous on-line, but the basis is in Objectivist Libertarianism, and implies a freedom from obligation to others and debt to forebearers. It's a selfish, twisted, flawed philosophy, evident of weak thinking and small souls.
      1. I for one doubt it will really go anywhere.

    Ushers will eat latecomers.

  • by PD (9577) <slashdotlinux@pdrap.org> on Monday July 10 2000, @06:27AM (#946105) Homepage Journal
    I agree that it is repugnant. How will you draw the line?

    Obviously kiddie porn inclus photos of 3 year olds involved in sex acts, but what about the other possible cases including:

    *a 17 year old 45 year old man
    * a 17 year old with an 18 year old man
    * two 6 year olds holding hands
    * a 4 year old swimming naked at the beach with his family
    * a 6 month old taking a bath
    * a 2 week old nursing at his mother's breast

    You get the point. I remember how surprised I was when my very own grandmother demonstrated a suprising amount of anger at seeing a baby nursing at his mother's breast in a parenting magazing. She was absolutely of the opinion that it was pornographic - kiddie porn even.

    So, how do you define those fringe cases? How can you reconcile your definition of kiddie porn with my grandmothers?

    When I said that censorship should be absolutely banished, I meant it knowing the consequences. It means that kiddie porn will be uncensorable, and to prosecute it you'll have to actually catch people with it on their computers, or in production. You won't be able to catch it in transmission.

    Freedom exacts a horrible price. The penalty in blood from wars and in cases like your example is very high. I am still of the opinion that the penalty of censorship is still higher.
  • A first readthrough of the documentation tells me that it needs better documentation :-) No surprise - new projects are often that way, but doing a good architectural description is a critical factor in making a project like this succeed, because people can understand where it needs work and where to help, and because security models need to address a lot of issues to be able to meet their goals.

    I had trouble telling what the technical goals of the project were - are they addressing traffic analysis, or only protecting content? They're describing a bunch of complex shuffling, but don't indicate why they chose those methods and what attacks they're trying to protect against. Some of the earlier projects like Pipenet and Onion Routing found that there are theoretical weaknesses if you only send traffic when you have real traffic, or if you do anything that makes it possible for an eavesdropper to tell what the boundaries between messages are, because the eavesdropper can do enough correlation to identify reasonably accurately where the traffic is going. The alternative is to build connections between sites that always have constant traffic levels, using filler traffic when there's no real traffic. This has a major cost/performance impact that affects the willingness of servers to support this kind of application. By contrast, IPSEC gives you all the privacy you need by encrypting, but doesn't try very hard to block the user identification.

    Privacy servers like this also depend on having lots of users - if there are only two people using it, it's easy to tell who's communicating with whom. It's nice to do technology, but you also need to work on a social or business model that encourages lots of people to run the client, and if it's got separate servers, to run servers as well. That's one of the cool things about Zero Knowledge [zks.net] - they've got a model that they hope will achieve this, though whether they succeed will depend on whether they implement it well enough for users to accept it and whether they can market it well enough to really take off. Some things are overnight successes - Hotmail, Napster - while others limp along at a low level for a long time, like the current remailer networks, mainly because they're annoying to administer and responding to complaints when they're abused is annoying. I wish the Fling folks good luck - but there's a lot of work they've got ahead of them to make it working and accepted.