Slashdot Mirror
U of Wyoming Fingerprinting All P2P Traffic
mk2mk2 writes "News.com has an article on how they're preparing to shut down P2P sharing of copyrighted content: 'For months, the digital equivalent of a postal censor has been sorting through virtually all file-swapping traffic on the University of Wyoming's network, quietly noting every trade of an Eminem song or "Friends" episode.'" It's scary until one realizes that most P2P traffic isn't encrypted, like back when everyone still used telnet.
This new technology will last for about 1 day. That's how long it will be until Kazza, Gnutella, Limewire, et all will switch to an SSL encapsulated protocol. Suddenly all the "fingerprints" will be shot. Each and evey download of the exact same file will have a different, unidentifiable, "fingerprint".
Sounds to me like this company took a copy of Snort, set up a few rules for the "fingerprints" and sold it to the University of Wisconsin. What a waste of money!
Heh, nowadays everything (wired, at least) is microsegmented -- you won't be able to sniff anyone else's data. Now, insecurely encrypted wireless links which are cropping up in a lot of universities nowadays, is a whole another story.
It's pretty obvious you can't copyright a length 1 bit string, so how many bits do you need before you own it and I don't? 10? 100? 10,000? I know you can't trademark a number, can you coprright one?
Well, I'm sure this will appear in the large ISP's if it's proven to work on the small-scale...
Perhaps with this 'fingerprinting' technology the big boys can just charge us the ($.50/$1/whatever) a song they want from us anyways? Instant delivery system for them that they didn't even have to build!
This whole deal about copyrighted material somehow reminds me of the war-on-drugs... Making criminals of all the users didn't work there... Trying to stop the supplies at the street level didn't work either. The only thing that will work is legalizing the controlled substance... then taxing the hell out of it... hehee
This claim is interesting in a variety of ways.
If the notion of privacy in our communications is going to be utterly discarded, I rather wish the school had elected to eavesdrop on every phone call made on campus to help catch thieves, domestic abusers and other violent criminals, etc.
There are plenty of people who say what goes on the internet shouldn't be private; that there's no expectation of privacy there. I guess we'll get into this issue a bit on this topic. Just please don't forget to have a little imagination. This is all new. We're making the rules as we go along. Sometimes I think if the phone had been invented last year there wouldn't be an expectation of privacy on phone calls either.
Remember this is a "private" institution doing this, i.e. not a law enforcement agency. Remember that just because they can write a fancy terms of service that authorizes them to do whatever they want with the network, it doesn't make their actions legitimate, let alone moral.
Finally, most interestingly, remember that Fasttrack (i.e. Kazaa, etc) is encrypted over the wire (see this link). There's nothing saying that the whole thing won't be reverse-engineered and cracked sooner or later, but to my knowledge, that hasn't happened yet... of course, that could just be last I checked.
Want to Know How to Cheat the GPL? Read On!
So, ok these guys have essentially done what FastTrackMovies has done and hashed each file. Hunky dory. So, people implement this and think "no one can trade my files, cause we know what they look like (and have the hash), so we can block it."
.zips or .tars the music or movie.
.zipped asset from being traded? I know it won't compress the MP3, but it will change the fingerprint.
Now, Joe Pirate simply
Exactly how would they then block the
Methinks WinZip is the Sharpie for this expensive DRM.
"The pie shall be cut in half and each man shall receive.....death. I'll eat the pie."
Theyre looking to block copyrighted audio content. Sure, that's fine. But you can't "fingerprint" something as complicated as a DVD or somebody's home-ripped pr0n movies because each ripper/encoder works a little differently.
Youre going to wind up filtering everything but *porn*. I can't really see that being what they intended to do.
"But it's getting to be the only way to control our bandwidth."
In one 24-hour period, for example, the most popular file traded using the Gnutella network was an MP3 by rap artist "Big Tymers," which passed the network monitor 188 times.
The students should really set up their own, internal P2P network. This would put less tax on the University's external bandwidth, downloads would be quicker, and, assuming it's restricted to local users, the RIAA couldn't really prove any wrongdoing. (Although their FUD generally scares universities enough.)
Universities are generally big enough to support a network on their own. They should.
At our university you promise to not engage in criminal conduct on the University network. Sharing movies illegally (now that is unequivocally illegal) breaks the AUP and you have no expectation to privacy while committing a crime, do you? Does a burglar have the right to privacy when he discovers that he was caught with a surveillance camera in your house?
Comparing it to Windows will be a moot point, since El Dorado is going to have a 40% larger code base than XP.
Or we may find ourselves without the ability to enforce the GPL.
If I own a telephone set and an associated line, it would still be illegal for me to record my friend's conversation when he's using it, at least without asking his permission first. A company can't legally record its employees' conversations, either. Your argument does not apply.
It seems to me any easy way to bypass (or at least extend) quotas at the University level is good old sneakernet -- much like we got our music when I was in School back in the '80s. One would make friends and get to know who liked what -- you want Dead Kennedys talk to Cosmic John, need Billy Joel, talk to someone else. We would build our collections a cassette at a time.
Since CD burners are so common now, why not do the same thing? Pass around CD-Rs with .OGGs or .MP3s around the Dorm (or between classmates) -- instant portable 600MB of "bandwidth" per CD-R. Great way to build up a collection without worrying about sniffers or using up the bandwidth.
Beware of Sleestak
But I do own the nic card that the traffic is sent out on... (with normal hubbed networks, not switched, that's a different story.)
I mean think about it... the electrical pulses are going onto the copper connectors on your nic... so why can't you look at it? You own it now...
Sig rhymes with Fig
Before some of our fellow slashdotters come up again with "They own the network": Yes, they do. But that does not grant them the right to monitor it continuosly and in detail.
Someone always owns a piece of infrastructure, be it an ISP, a University, the interstate authority or your 'landlord'. But they don't have the right to invade your privacy if you are using rented, leased or subscribed equipment. Imagine the owner of your apartment trying to monitor your living habits, to make sure "nothing fishy is going on in your apartment".
Network and telephone lines can transmit very private and sensitive information, and it is a serious crime to snoop that out. If you thought that was the right way, you're had too much time on corporate americas way of life. They are your customers, your contractors, if you like, but not only that, but living feeling humans that deserve to have a private life, one that's none of your business. You can imagine a thousand situations like this:
- You rented my car, why don't I have the right to monitor where you're driving, who you take with you and what roads you drive on?
- You rented my house. I claim the right to visit you whenever I deem it's necessary. And just to ensure, that my property is taken good care of and you don't hoard drugs there, I will make a full seizure every time I come.
-
I rented you my video camera, you've got to give me a copy of each recorded tape, so that you cannot film underage porn. Think of the children, my god!
-
And finally: I've given you Internet Access. Now that you can browse the web and do spiffy emailing, you must be utterly thankful to me. And since you are a student, you don't have any rights to complain, we will treat you as a slave and you have no private life. Be thankful, you even got a 'net connection and understand, that we have to make sure you don't do illegal things with it. We don't count the bytes, we don't have per-user quotas, we do the nasty GESTAPO stuff piling through all your traffic. If you complain, well, try another University.
Opening some other's letters is the same and I hope finally someone will punish the university for doing this.Let it happen, that on one incident, some very private information about a student is obtained that way and told the public to embarrass him. One lawsuit later, the U has lost 10 Million US$ for a settlement and the bandwitdh savings of 5 years are worth exactly nothing compared to this. Go ahead, wait till someone reacts. I'd do that.
Now a public phone company doesn't have that right, because it's not in their TOS, and if they put it in their TOS, somebody would fight it as being an illegal invasion of privacy. The Internet, and a University network at large aren't seen as a common enough utility, that is necessary for living in the current society to warrent those kinds of protections yet. At some point the Internet might get that kind of protection. However, given the proliferation of networks, my guess is that it will be a market driven thing. Phone companies are monopolies, so they have a lot more regulation then a University network ever will, because you can always get network access from a dozen other places if you don't like the terms of service the University has. A University is also a lot like a place of business. My company has the right to monitor everything I do on their equipment. All their wires, all their harddrives are fair game for them to search. It's a term of my employment. They also own all of the things I do on their computers that's in my IP agreement. They also can restrict my free speech because I signed an NDA agreeing that as a term of my employment, I can't talk about certain areas of expertise I have to other companies.
Technically, you don't need his permission to do record his conversation, you just have to tell him you are doing it (it's subtle, but there's a difference, he doesn't have to concede it's okay, he merely has to hear you say it's the case). If he continues to use the phone, I don't believe there is anything illegal about it.
Kirby
The point is, it's THEIR network. It's not the student network, it's not the taxpayers network, it's not even the Alumni's network. It belongs to the University plain and simple. University is for research, not d/l pr0n or sharing eminem. Students are given access to the internet in their dorm rooms to assist them with their studies.
That certainly is an interesting point. Please allow me to offer a counter point.
Universities are there for learning and growth of their students and faculty. They are not all about books and studying and stuff like that. Universities sponsor football -- why? Student unions and governments -- why? Those are extracurricular activities that help the students grow as people, round them out, etc. Ever meet someone in real life who thought university was there for books and no socialization? I've met one, and let me tell you, communicating to get to the immense book-smarts was tough, and he was not prone to creative, reasonably practical ideas.
The university network is there primarily for learning, but there should be a reasonable amount of respect for personal growth and exploration. I'm not sure I want to argue that pirating friends episodes and pornography are aiding that pursuit, but maybe they are. The university should make a reasonable effort to allow the students to do explore their freedoms and help enforce the law when subpoenaed to do so. I think it can easily be argued that the downloading of friends episodes leads one to think about copyrights and what use they have in the real world. The exploration of pornogrpahy, it can be argued, helps educate the "consumer" what he (or she) thinks about the impact on the models as individuals.
My education was, believe it or not, furthered by playing with a little known Unix clone named "Linux". It wasn't supported on my campus network, and there were times when I used bandwidth for this side project that did not contribute directly to my studies, but I believe it was worthwhile. I played Doom over Kali, and ended up learning something about network latencies and bandwidths. Completely illegal on the campus network, I even ran a password logger for some time -- this turned out to be a very powerful lesson in cryptography and network security. I did not have the money to set up a legitimate private network to explore these issues, but this was education that helped me become the person that I am.
I believe that university and college dorms are there, not for the exclusive pursuits of scheduled academia, but for the students to explore their own educations, as they pertain or do not pertain to their class schedules.
Funny, ensuring network performance is kinda what university monitoring of traffic is about, isn't it? How do you think QoS or packet-shaping works?
The interesting question was when someone pointed out that it's not your network unless you laid the fiber yourself. I think there'll be some very interesting cases in the next few years with regards to setting up wireless access points. A wireless mesh network, in which 100, 1000, or 10,000 users allow their boxen to be used as access points, is indeed one in which the users "own the pipe".
At 100 users, odds are that "someone else" owns the pipe where stuff eventually goes through. (Like your University owning the pipes through which much of your dorm's P2P traffic eventually goes.)
At 10,000 users, that's not necessarily so. A mesh network composed of 10,000 Freenet nodes scattered throughout a city might be able to cache Titney Spears' "OopsYouGotFuckedbyRIAAAgain.mp3" within itself -- and thus the "pipes" through which the MP3z flow are indeed owned by the users doing the flowing.
Both cases are clearly copyright infringement - but the latter case would be much more interesting from a legal perspective - RIAA has the right to ask the University to sniff its traffic, but do they have the right to sniff your traffic?
(The Feds, of course, suffer from no such restriction, but that's because we've given them the authority to enforce the law and laid down rules that govern when/what/who they can sniff. But unlike the Feds, RIAA has no more authority to sniff than you or I do. Fuck 'em :-)
I don't think this is relevant. I haven't looked at any packets going down the wire, but I'm assuming when you request a file from another user, you have to ask for that file. Filename request goes down the wire. Once you know the format of file requests for a given P2P program, you can just scan them to see what kinds of files people are requesting. If not the file requests, what about when the client replies to search requests? What about direct connect complete listing queries?
:) So in an effort to make things better, once the P2P catches on it will be made worse again.
Some users have already brought this up, but the way around this is to encrypt/re-code the traffic. That is, all the requests, all the listings, all the control stuff, and the file transfer itself. This may lead to an increase in bandwidth consumption just to encrypt everything though
Just like after Napster. When Napster was popular, there was a gradual movement to shut down access to it. So other services started popping up, then completely distributed services such as Gnutella. Gnutella is a tremendous bandwidth hog, as opposed to something more centralized.
I respect the universities that just try to limit the bandwidth consumption of the offenders. But just shutting this stuff down cold turkey is only going to lead to P2P more difficult to detect and filter.
Of course, organizations such as the shitty Adelphia cable should not BY DEFAULT have a 15kps upstream. Assholes.
-- Having a Creationist Museum is like having an Atheist place of worship
You don't have to listen to all the traffic. Just enough to fingerprint it. Or watch the opening of all the traffic - file transfering protocols have to identify the filename somewhere. If it's a suspicious filename, store the traffic on that stream for later analysis.
I can't say that I don't give a fuck. I've just run out of fuck to give.
Freenet's a beautiful thing but the encryption and such bring everything to a crawl, i.e. Cable connection->28.8k.
If you like freenet you should probably check out mnet HERE!
I am in charge of the network/server department at our college.
.au files when I was in college thinking how cool it was that my box could play the james bond theme.
We have a limited connection to the internet, which is usually being eaten up by P2P traffic. Today, over an hour period, we had three students that used a total of 4G of traffic in an hour.
I don't care what the traffic is, but when legit work can't get done, such as our payroll system which uses SQL*Net across the WAN (bad idea to begin with, but that's a state bueracracy for you.) and their processes just aren't working, shit is gonna have to happen.
We blocked port 1214 (kaaza) and a week later the port switching version came out.
Right now we are facing the choice of either doing some severe draconian network policies or buyin a packeteer.
And how long will that work before the next fileswapping act runs with ssl over 443?
I feel for the students - it's something fun to do...hell, I remember downloading
Makes my life a pain in the ass - how to be nice and let legit stuff go on, allow some fun and experimenting to go on, at the same time "protect" the network and make sure it is available when need be.
We don't call the university U of Wyoming or UW(you double-you). It's U Dub (you dub) :P
Proud freshman flunkout!
Maybe I misunderstood, please correct me if I'm wrong, but your post seems to imply that you think that anytime someone/some company does something that has the effect of furthering someone else's goals, then they are really doing what they're doing in order to help the other person/company. That's pretty flawed logic.
Suppose that I am married and my wife doesn't like guns. Further assume that in my house, what I say goes (I know, I know...but it's a hypothetical situation!), and I don't want the guns in the house because, though I like guns, I think they're too dangerous to have since we have children. By your logic, what I am really doing is conceeding to my wife, rather than making a decision based on my own beliefs, simply because it furthered her goals. That would be a wrong conclusion.
Now, back to the bandwidth thing. I am a network engineer at a large financial institution. We just upgraded our Internet pipes to 22 meg, because we need the bandwidth. Though we have plenty of money to pay for it, it may not be a cost effective move if we could have elminiated, say, 25% of the traffic (5.5 meg) through any valid (meaning, more cost effective) means. For a university (yes, I am very familiar with university networks and funding issues) this is even more critical, as their funding is much lower than where I work. And, in fact, even we limit bandwidth used by using a web proxy and by restricting sites that employees can go to (which, admittedly, does serve another purpose as well).
My point is, that this type of activity is very common, especially in well structured networking departments, primarily because a dollar that is spent on a recurring charge is a dollar that may be better spent elsewhere. The recurring charges are the budget killers, though some are necessary.
Just my $0.02...
Wow.. UW on the Slashdot front page... Amazing. Unfortunately the article hardly says anything, so as a former IT employee and currently part of the staff that deals with all things related to student networking in the dorms, I'd like to try and fill in the details: Unfortunately, Laramie is NOT a large town (26k counting students) and the bandwidth coming in is very limited. The University only has a 30 Mbit upload capacity coming through Cheyenne, which (limitedly) comes from the huge hub in Denver, CO and (so we've been told) "there isn't enough capacity going into Cheyenne for us to purchase more". Up until a year and a half ago there weren't any problems here with bandwidth. Then all of a sudden everyone is using P2P in the dorms and leaving outside sharing on. It wasn't a problem of people downloading with P2P, it was the rest of the world downloading from us. There was so much traffic going out of the dorms that the entire university network was slowed to a crawl. Their solution at first was to just limit the dorm traffic to 10Mb which fixed the problem for the rest of the university but made it impossible for me to even read slashdot from my room. Naturally that was still a problem, as even legit HTTP traffic couldn't get through. They've been messing with packeteer for a long time but can't come up with a good solution. Right now HTTP packets have highest priority, followed by FTP (which wasn't allowed any priority at first until a lot of students complained) and just about anything else is like squeezing the entire population of China through a single revolving door. Speaking of telnet.. I can't telnet to anything off campus from my room unless I want to WATCH the packets arrive every 10 seconds or so. P2P traffic is about 20 times slower than a modem (but everyone still uses it.. as I sit here writing on my ex's computer next to her latest list of mp3s to download). So how do the geeks here survive? A lot of people are running local FTP servers, which is all I use any more. We can't play networked games off campus, so we have set up our own servers. But even that didn't work- Games like counterstrike which needed outside authentication would time out after 60 seconds. We managed to fix that problem with http tunnel. Almost anything can still be tunneled out and is unaffected by the packet shapers, provided you can find a good, reliable proxy on the outside. As far as getting busted for file sharing, we have shut off quite a few ports because of letters from the RIAA/MPAA, but for the first offense the students are only required to give us verbal confirmation that all of the illegal material has been removed before we enable their ports again. After that the ports to their rooms are shut off for the rest of the semester. Oh, and as far as an agreement? I sure don't remember signing anything related to the network usage. Personally, I don't see anything wrong with them snooping the files going through to help increase the legit bandwidth, as long as they aren't trying to crack through encryption and they don't snoop local traffic. I also think they should look into local file servers... you'd be amazed at what you CAN'T find on a 320 Gb ftp server filled by students... I never have to get anything from off campus anymore, unless its the latest source code for my Gentoo box (wget through HTTP works beautifully). At least the article picked the right person to interview as Brad is one of the few people over in the IT department with a clue. Sorry, couldn't let the article make our IT department look like they really know what they are doing. Really they are just being guinea pigs for this new software that the article is hyping up. IT is, however, doing a good job of walking the fine line on illegal P2P sharing. As Brad stated, they have a somewhat "don't know, don't care" policy while at the same time acting as MPAA/RIAA whores upon request (which I think is what this software is really for). Anyway, hope I could clear up a few things for you from someone who has been quite involved with all of this. Post questions, I'll be happy to answer. --An Anonymous Coward, even though most people from UW already know who I am now-- And uh.. mod this up/link it to the article