Slashdot Mirror
SDMI Technologist Talal Shamoon Interview
A reader writes "Salon has an interesting interview with one of the brains behind SDMI.
Watermarks in music? Talal Shamoon, a key technologist for the SDMI, says that he's found the key to protecting copyrighted tunes."
> 4.Fact #4: If you can identify unneccesary data, you can remove it.
Any decent compression will remove the watermark automaticly...
I'm guessing the idea here is to imbed the watermark on CDs etc...
When someone rips a CD to an MP3 the compression will automaticly remove the watermark. It probably appears as unheard (or byond our hearing) noise.
Example Sound is "Really loud" watermark is "extreamly quite poping" you never hear it but it's there.
Here comes compression. Unheard sound? Yank.. Not in this MP3.
Even simple filtering would yank that example...
If it's not accually in the audio stream but in the data then it'll never survive being converted to an MP3.
Say your protecting an audio stream. I tap in and record whats comming off my chip (kmix RecSource master volume). I MP3 it...
All I need is your trusted client.. a program to control the audio on my sound card (kmix or the mixer built in Windows will do) and any audio recoding tool.
Convert to an MP3 using reasonable compression.
The watermark is history...
If the watermark is fluxuations in the sound that are to minnor to notice. Again removed.
Ideally lossy compression removes everything you don't hear. In the real world it ends up having an audioable effect and removes a few things you DO hear.
But compression is vital if you are gona trade MP3s over Napster. And thats what they are trying to prevent right?
I don't actually exist.
It's disappointing to see such an otherwise brilliant man so completely taken in by the media companies' need to protect their works -- a need which has never been convincingly demonstrated; to protect works which, strictly speaking, aren't theirs to begin with, but the originating artist's.
However, digital watermarking does have an important use in the infinite abundance of the digital universe, and it's not what Mr. Shamoon has been led to believe. Watermarks have a compelling use not as a basis for copy protection/management (erroneously referred to as "rights management"), but rather for reputation management.
Think forward to an age where everything -- including physical objects -- can be copied infinitely and perfectly at zero cost. Attempting to control copying in such a world becomes utterly pointless, not to mention childishly foolish. However, being able to track down the original artist(s) behind a given work will become extremely important. One way to do this is with difficult-to-remove watermarks. By scanning the work and recovering the watermark, you are able to identify the original artist, possibly to negotiate with them to do additional, similar work, and you're able to make this identification no matter how many hands the work has passed through. Thus, the artist is assured that their reputation will be preserved, and the recipient of the work knows they can track back directly to the originator rather than a faceless publishing house.
If Mr. Shamoon were to re-think his strategy from identifying and protecting copies (again, a pointless exercise in the digital universe) to identifying and protecting artists, I think he would find a good deal more support from artists, technologists, and consumers.
As for this new SDMI stuff, be very alert for it, as the media corporations are arm-twisting high-tech companies to cram it into everything. For example, Intel is working very hard to incorporate SDMI-like "features" into IEEE-1394 (FireWire). Also, the new Digital Flat Panel signalling standards from the Digital Display Working Group have space in the specs reserved for similar copy protection measures.
Personally, I can't understand why the high-tech companies are giving these guys the time of day. They won't be buying these devices; the consumer will, and the consumer has already made it clear they don't want this stuff.
Schwab
Editor, A1-AAA AmeriCaptions
... Nuff said.
========
Stephen C. VanDahm
The author is not clueless; he's being very calculating. He knows that people are wary of the big, bad internet, and that these same people are the ones who can't help but open binary attachments in their mail. They just aren't smart / thoughtful / informed enough to figure out that Gnutella won't do anything on its own. If millions of people will believe some stupid hoax about a virus that erases your hard drive, they will also believe that you can get a virus through the internet with Gnutella. It may be FUD, but it's semi-believable FUD for a lot of people.
Walt
P.S. There is some small risk in other people knowing your IP, but it ain't safe to be connected to the internet if that's an actual risk for you.
While the SDMI seems like a good idea, unfortunately it's too little, too late. Because of the music industry's years of vehement opposition to anything resembling an online strategy -- first shutting down mp3 web sites, then Napster -- they've inadvertently created a market for systems like Gnutella and Freenet, which are virtually unstoppable because of their decentralized, distributed nature. This is the music industry's equivalent of the unstoppable "superbugs" that might result from the overuse of antibiotics -- and now that they're here, there's no way to put the genie back in the bottle.
No matter what ingenious encryption or watermarking system is used, it boils down to this: at some point, between the 0's and 1's and the analog output, the music must be decrypted/decompressed/whatever. There is nothing -- repeat, nothing -- that will ever stop people from capturing the signal somewhere in the chain, recording it as an mp3, and putting it up on Gnutella. Sure, there might be some quality loss -- but since that doesn't seem to bother people now, I doubt it will bother them then.
Cheers,
IT
Power corrupts. PowerPoint corrupts absolutely.
First off, this is insane ... how many people who own MP3 players are going to voluntarily cripple them by installing "phase 2" software? A show of hands, please?
...
Well, the genie's not really out of the bottle. This is one of my big problems with the way people analyze this market. If you see it as a war between pirates and content creators, then it's under a completely different light than how I think it should be seen. Really, these are market conditions that have caused a black market to emerge. People are copying music because they feel somewhat disenfranchised with the options they have at their disposal in the digital space. It's up to the content industry to create value in the digital arena and they've made phenomenal steps in that direction.
I must disagree. This is a war between people who want to share music, and people who want to prevent others from sharing music. Too bad for the RIAA that Congress legalized all non-commercial music copying in 1992!
The Court of Appeals, in reversing the Napster injunction, basically told the lower court that it was completely wrong in its interpretation of the law, and explained how the law should be interpreted.
So what is the RIAA going to do, when either this court, or the appeals court, hands down a ruling that the 1992 Audio Home Recording Act completely legalized music sharing, and that the monetary interests of the RIAA have been completely accounted for by the collection of royalties on blank media, as provided for in the AHRA, and clearly spelled out by Congress, both in the law itself, and in the legislative history?
People will stop feeling guilty about using Napster, when they realize that the industry has been lying to them, and actually has been collecting royalties for 8 years on blank digital audio media, and then the war -- the war for people's minds -- will be over.
SDMI is just another attempt to use technology to make it physically impossible for consumers to exercise their legal rights. I think that SDMI products will die in the marketplace.
Pay no attention to the man behind the curtain
A lot of these abbreviations seem to have simple pronounciations that come about. SCSI being pronounced as "scuzzy" being one. I've heard a lot of people pronounce SDMI as "sodomy" and others pronouncing it as "sid my." Is there a correct pronounciation?
Prevent email address forgery. Publish SPF records for y
I think history pretty much shows that the odds are stacked against this outcome. When people went from listening to nothing to listening to phonographs, it was a clear improvement. When people went from listening to phonographs to listening to CDs, it was a clear improvement.
But there is no reason to adopt SDMI formats over CDs/mp3s. There is no improvement in several areas:
I can see the conversation with a sales clerk: "Oh, you want me to buy the more expensive model which sounds the same, but is a complete pain in the ass to use. I hope you don't mind if I just buy the CD player"
The only way people will adopt SDMI is if they are extorted into it by the RIAA companies by raising the price of CDs and lowering the price of SDMI content. But as far as people voulentarily adopting it because of "value being added to the online SDMI experience" or whatever the hell this guy's talking about, I don't think it's going to happen.
Do the power analysis. It requires a minimum of kT energy to flip a bit, where k is Boltzmann's constant (1.38 * 10^-23 joules per Kelvin) and T is the temperature your computer is running at (the background temp of the universe is about 3.2 K).
It would require an optimal computer [*] about 250 megawatts of power [**] and a full year of time to break a 128-bit cipher.
[*] We don't have anything close to this.
[**] 250 Mw is a hell of a lot of power. Don't ask me how you'd keep your computer from melting down from the heat. Or how you'd keep it at 3.2 K, for that matter.
128-bit ciphers are secure for the indefinite future. I don't expect anything short of enormous advances in quantum computation to make a dent in them.
When I read the last paragraph of this interview, I nearly leapt out of my chair. I, as a consumer, DO NOT WANT CDs to go away! I still have and play old LPs (these are titles that will never be available in the CD format). If this guy thinks I'm going to start replacing CDs with some new digital format he's got another thing coming. And I'm sure he'll hear from thousands of other music enthusiasts as well.
I'm positive that the replacement format will require payment every time I download their watermarked MP3 format files. Whoops! My hard disk crashed. Whoops! The batteries died (or whatever). Guess I'll have to repurchase my music collection. I got the impression that an ulterior motive of the SDMI might be to corner the market on digital audio players just the same way that the CSS got a stranglehold on the licensing of DVD players. Next thing you know, you and I are paying royalties per listen.
Also, I'm already not any fan of a new format that is a step backward in terms of sound quality (MP3s are that sort of step as far as I can tell). I wonder how long it will be before people begin noticing that they can, indeed, hear a difference in the sound of watermarked audio files after all. Some of us consumers aren't half deaf.
Let's hope that the marketplace has the sense to make the SDMI watermarked audio format Dead On Arrival.
--
CUR ALLOC 20195.....5804M
So are they going to put a unique watermark on every CD pressed? And then are they going to keep a record of who I am when I buy it? What if I buy the CD at a record shop and pay cash? Are they going to make it illegal to buy CDs without showing identification? What's to prevent someone from buying a CD anonymously, ripping it, and then distributing the MP3s to their heart's content? Maybe we'll have to start licensing people to use CDs so we can keep a record of who has which watermarks. Or maybe, the idea is to get rid of physical media entirely. Never mind the billions of $$ people have invested in audio equipment. How many years is it since the introduction of CDs, and yet you can still buy brand new cassette tapes? I don't think CDs are going away any time soon, and as long as I can buy a CD anonymously with cash, I can rip the songs and distribute them, regardless of any watermarking scheme.
-Vercingetorix
-Vercingetorix
"Necessitas non habet legem." -St. Augustine
If music players all force you to only play music that is "signed" by this technology, guess who gets to decide what kind of music people listen to? Only music labels.
The promise of digital technology was that it could make things like producing and distributing things like books and music accessible to anyone with a computer and microphone. But I guess we can't have that if it threatens the music industry. If this kind of technology spreads from the music industry to other places like software and publishing, we're really going to be in trouble.
Music publishers want to have it both ways. The freedom of the internet, and the control of copyright enforcement. These two forces are opposed to each other. Something's gotta give, and I hope it isn't freedom.
Artists will make great music even if they aren't paid. I have the recordings to prove it.
-jlg
ps. use Debian! www.debian.org
According to this article the watermark is audible. I am not buying any music with an audible watermark, I might buy music with an inaudible watermark, but only if it is significantly better than the plain old CD I can buy now.
Q.
OK, the author here needs a HUGE whack on the head with a cluestick. With napster, you ask a central server for a song, and then you set up a direct connection. With gnutella, the only difference is that you have to do so before you log on. He's right so far. But then he goes on to say that this increases the risk of virii. Grade A prime bull. You can't possibly get a virus unless you actually transfer a song. No client is brain-damaged enough that it will accept code (vbs or otherwise) from its nearest node without asking for it. And the stuff about child molesters is pure garbage. Chat is chat, whether or not you're using a decentralized server. Knowing someone's ip address doesn't help you hunt down their house.
--
Friends don't let friends misuse the subjunctive.
All the specification says is, "if you download a song to a PC it should be protected; if you transfer it to a portable device, the wire along which it travels should be protected and the portable device itself should keep it protected."
Oh, gee, what realistic requirements! Sadly, it will be hard to transfer data to your PC when it's secure (ie, turned off inside a safe at the bottom of the ocean). They're actually expecting end-to-end hardware protection! This would required rebuilding everything inside the machine (if it's possible at all, which I doubt). Not to mention the fact they you still have to trust whoever you got it from (ie, trust that nobody tampered with it over the wire). OK, let's rebuild the internet too! But we'll stop those pirates! (haha, yeah right).
The only hard bit is identifying the unnecessary data. But, it's only a form of steganography. If you know the message is there, then all you have to do is find it. It may be hard, but given the past history of the companies involved with SDMI, it won't be *very* hard.
This is another example of the 'Trusted Client' problem There ain't no such puppy as a trusted client. There can't be.
The millions being invested in SDMI is a waste. I hope the people involved have a *very* good set of excuses ready for when the shareholders start asking where the money went.
In the meantime, I will pay for the music I listen to. I'll pay for the DVDs I want to watch. But I'll play them on the platform *I* choose.
Share and Enjoy.
I can't help but wonder if these people are in a deep, deep, deep case of denial. They need pills, or something. Can't they see that they're fscked? I'd guess terabytes of mp3's swap hands every single day. There's no putting the genie back in the bottle. For some unknown reason, sales still are up. What, people are maybe at their core honest?
There is no digital protection scheme short of implanting an RIAA chip in your head that will would because you need to hear the music. This guy has a degree from Cornell. He's still an idiot, he's just an idiot with a degree. If there's a watermark, you can bet there's a pissed off hacker out there who's better than you who's going to take care of your watermark real fast.
The format is too widespread, there's no control over players and the numbers of people make it impossible and possibly not legal to sue everyone collectively. (Civil disobediance, who?)
Find a model that doesn't rape consumers and makes people happy for once, find a model that makes the artists happy - no, not Lars and his happy gang, but the 99% that get fscked when they sign on with the labels. Or face a horrible, horrible obsolecence. You won't be missed.
Doing my part to end RIAA monopolistic practices since 1996. :)
..don't panic