Slashdot Mirror
Using Watermarks to Combat Piracy
TheEvilOverlord writes to tell us PC Advisor is reporting that researchers at the Fraunhofer Integrated Publication and Information Systems Institute have developed a new watermarking system to help track and combat piracy. From the article: "The system lets content providers, such as music studios, embed a watermark in their downloadable MP3 files. Watermark technology makes slight changes to data in sound and image files. For instance, the change could be a higher volume intensity in a tiny part of a song or a brighter colour in a minuscule part of a picture. Even the best-trained human eyes and ears, according to Kip, can't detect the change."
Even the best-trained human eyes and ears, according to Kip, can't detect the change.
Who says anything about using human senses to detect the watermark? If these watermarks are embedded by machine, I'm sure it won't be long until Watermark Bob creates a "cleanser" program to detect anything unusual, and maybe even remove it.
Virtual Betting on Facebook for non-geeks.
The system lets content providers, such as music studios, embed a watermark in their downloadable MP3 files
For whom was this intended again?
I'd be happy if there actually was plenty of music studios providing downloadable mp3's though.
Beware: In C++, your friends can see your privates!
through md5, sha1 hashes. nevermind the collisions though ;)
This is great, because.. oh, wait, I don't and never will buy DRMed music.
Never mind.
How well will this stand up to a lower bitrate/encoding setting?
x86, oh yes, I'm pro.
...and in order to defeat such a wonderful scheme, all you have to do is re-watermark the image/music/video.
I've yet to see a scheme that reliably survived that test unless it was specifically designed just for that test (like embedding high power signal in several random places), and upon detection, looking for that signal in those random places (hope is that 2nd watermarking didn't wipe out -some- signal data).
In any case, Watermarking doesn't work! Even Microsoft's researchers said so (damn, can't find link).
Maybe not, but I bet outguess can, along with a million other stego tools.
All's true that is mistrusted
And it will include checksums to ensure that one hasn't been applied along the way.
OK, so you tag downloads. Now what?
Assuming a "de-tag" program doesn't pop up an hour later, what do you do with this wonderful invention? Instead of passing around a "normal" mp3 of Metallica, they're now sharing a "watermarked" version that allegedly can't be discerned by mere humans. How does this help?
Cheers,
if the watermarks are imperceptible to humans, than what's to say that the matrices used by audio and video codecs won't remove them from the source because they're undetectible?
It should go through. Coders in general are not required to be deterministic, so some pattern recognition would have to go into identifying the watermark.
so what about the majority of pirated music: the mp3s ripped from cd?
me and my thinkpad, sittin' in a tree, c-o-d-i-n-g...
"If, for instance, you purchase and download a CD, burn a copy and give it to a friend and that person puts it on a filesharing network, our system will trace that music back to you and, depending on the legal system of the country you're in, you could be [hit] with an expensive fine," Kip said.
... what gives?
How, exactly? Supposing I went out and purchased a music CD (a radical idea, I know) with cash, how could they possibly trace that particular CD back to me should it somehow be made widely available to download? I mean, I wouldn't have provided any personal information to the store during the purchase so
Seems like an easily defeatable mechanism. To track you must be able to extract the creator's watermark from the customer files. If the customer has done further modification to the file, it could obliterate the ability to detect the original watermark. Watermark Bob needs to simply add his own slight data modifications to accomplish that goal. In fact, wouldn't that make his version of the song unique and not a copy? Would the creator's version, plus watermark, not be considered a copy? It's modified, however slightly, so which version is then copyrighted? The one with or without a watermark? Or both?
In other words, "Nothing to see/hear. Please move along?"
More seriously - although it could be stripped out (relatively) easily, you could embed watermarking data in the metadata segments of downloadable MP3s. I'd accept this as a tradeoff for music studios offering downloadable MP3 files: If some_hit_song_i_downloaded.mp3 shows up on a P2P network and contains metadata whose MD5 could only be generated by, say, hashing my credit card number with some_riaa_private_key, that'd be pretty reasonable grounds for RIAA to believe that I'm the schmuck who (a) paid for the right to download it from a RIAA-authorized source, and (b) uploaded it to a non-RIAA-authorized filesharing network.
Make it impractical for Joe Sixpack (who will be unaware of this type of watermarking, and who probably will be unaware of the existence of tools to strip it) to upload his files without risking fines/prosecution, and you can offer DRM-free MP3s to Joe Sixpack.
By embedding a watermark in a .mp3 file and making it available on P2P networks, aren't record companies implicitly giving you permission to download their music?
Worked for me just now... ?
This is already somewhat in use.
band releases early copies of an album to reviewers. if the album leaks, the people who sent out the advances can find out who leaked it.
Bury me in mashed potatoes.
Ok, you embed a user-id into each file downloaded and look for it on p2p networks? But giving the number of insecure home PCs, the files out there will propably be stolen from people who bought them legally online and then shared everywhere.
\u262D = \u5350
Even the best-trained human eyes and ears, according to Kip, can't detect the change.
That's what they said with the last corporate watermarking scheme.
And then it turned out (1) yes, even relatively untrained ears could hear the difference and (2) it was possible to destroy the watermark just by detuning the song by a human-inaudible amount.
Considering that the goal of "you own it but it's still ours" technology like watermarks and DRM is never to work, but always only to provide a false sense of security to IP holders, I don't expect that we'll get much better results even with a competent group like Fraunhofer working on it.
Dude, I hacked your account. Consider yourself iwned.
-CL
All one has to defeat those watermark is run through photoshop
sharpening or gaussian or one of many Photoshop filter.
Pretty cool idea. Works, for now anyway. I like it.
A public account: log in as "Communal Account", password is "kFhthALQ".
There are some real classics out there, you know like those Charlie Chaplin films. How do they convert this technology towards downloaded movies, especially silent ones. Will they change the narrative text between scenes or something? If not can this be seen as a workout which can be used in more modern films because they are more "recent". Jaws isn't as good without the sound!
Jonathanjk.com
While this may be a good idea in concept. When something does leak, who is to blame? The person who leaked the song? what if the person has no knowledge that his copy of the media was leaked?
1. User downloads media
2. Media gets watermark for user
3. User gets infected into a botnet
4. media gets leaked
So the person that media was attached to gets the blame, instead of the people that really leaked it.
After the person gets the blame they are going to get screwed even if it isn't there fault.
While this may not seem to be practical I could see it as a possibilty. I thought it was an interesting possibility.
Pay Cash. Send Kip email that says, "Neener, neener, neener!"
"I'd rather be a lightning rod than a seismometer." -Ken Kesey
The only thing a digital watermark could be useful for is authenticating a media file as an unaltered official release. It will never be useful for combating privacy except for only the dumbest users who do not cleanse the media files. Cleansing image files is trivial assumming you have access to free hardware/software (I guess thats the kicker). Cleansing music files is not much harder, and one can always simply re-record the music too, thats one avenue of DRM circumvention that will NEVER go away.
I'm hoping these kinds of anti-piracy actions work, and work well.
Things like the DRM and DMCA were put into place to fight piracy, and wound up just hurting regular consumers while the pirates just snickered as they continued along their merry way.
With these kinds of things, regular users will still be able to do what they like with their own copy, be it back it up or transfer it to another medium for personal use. At the same time, it will allow those tracking piracy to find the source and press charges only against that person, and not the random multitude.
I'm sure the pirates will figure out some way to work around this (be it to randomly change the volume slightly throughout an entire MP3, or brightening/dithering an entire picture), as they have everything else, but if this kind of technology can prevail and advance, it will allow those of us legally using our own purchased goods to do so without worry, while punishing those who deserve it.
I hate to be the guy to come out and say it, but I don't really mind DRM as long as it doesn't interfere with my user experience. I paid for my songs on iTunes, and I've rarely encountered any DRM restrictions that affected me. I wish I could just give them all to friends who wanted them, but let's face it, that's pushing the limits of fair use. And if I do want to share it, there are easy workarounds.
This goes for downloaded files, not physical media. If I buy a CD, I want to be able to do whatever I want to it, which includes sharing with friends. I've never made a habit out of sharing files, even back in the Napster days (Sorry, but I was a leech). Most of my file sharing is between me and my friends, and while I admit that it certainly pushes the limits of legality, it's the only "responsible" way to do things.
This watermarking idea just reeks of being absolutely unnecessary. People just need to learn to be more "responsible" about how they rip off music. I hate the record labels as much as the next guy, but I'm willing to work within the confines of a happy medium, and do most of my sharing via less (or is it more?) traditional means.
I don't see anything wrong with sharing TV shows that are freely broadcast over the airwaves, however. For most things, however, if you don't own the copyright, it's usually not yours to distribute.
What's my point? I really don't know. Try this: Steal all you want, just don't get caught, and don't let them force more of these silly things on all of us.
Slashdot: 24 hours behind every other site or your money back!
Guess I'll be keeping all my old ripping software that doesn't include watermarking technology, then!
You must think in Russian.
Call of Duty put in there as well. A lot of the battles are mostly or at least partly on real life history, and it would be an educational experience about the hell of war and price paid to secure freedoms. I remember for one of my history classes we had to watch Glory, a movie about african americans in the Civil War. It was quite an experience, and think this sort of medium can be an instructive teaching tool.
We're all hypocrites. We all have hidden parts, it's the contrast between them that make us more a hypocrite than others
If **AA prosecutes the original buyer of illegally distributed watermarked copies, then pirate distributors will create malware to steal originals from unsuspecting copy owners. Computer owners that don't secure their machines will find that someone has surreptitiously copied their media files, sold or traded them on the open market and made the owner of the infected machine liable for criminal act.
Two wrongs don't make a right, but three lefts do.
Are they planning to
The first is basically worse than DRM, the second is essentially an aid to enforcing existing copyright laws. I suspect that if the Content Cartel would finally accept that their business models need to change and go for the second approach, most of us could accept it.
Lacking <sarcasm> tags,
What about if I re-encode the MP3 to OGG or WMA at a higher bitrate?
Resident of Skara Brae since 1985
The actual term is called traitor tracing
Arash Partow's Philosophy: Be a person who knows what they don't know, and not a person who doesn't know.
I don't think you did. If you changed his/her password, it will just be resent to his email address, and he can update the comments. Also, if you change his/her email account, he can just re-set that too.
I think it was you, sir, who were pwn3d.
It's a bit like those secret government documents who have several purposely-placed typoes, different for each distributed copies, with which you can deduce who leaked the document according to the typoes that surface in the unauthorized copies...
So, what can prevent anyone from shaving-off the least significant bits and putting garbage instead? This way, you cannot tell who "pirated" the stuff...
..they're doing something that won't affect anyone other than the technologically inept pirates, who are already their easiest targets for prosecution. The resulting rise in lawsuit numbers and word of this scary techno-voodoo might put more newbies off using the fileshares, or it might get them to learn to use whatever unwatermarking tools pop up.
Slashdot Burying Stories About Slashdot Media Owned
Could something like this be used to gain privileged access? While I realize that it is in effect making different content, could a bug in the watermark application potentially create content so different that it would crash a media application?
At the least, a bug in the watermark application could cause degradation of the media. In a worse case scenario, it could cause media players to crash.
From another point of view, targeting MP3 files seems a bit odd and targeting DRM'd files would be useless. This technology appears to be out of date before it is used.
Proof by very large bribes. QED.
You wanted the video game thread.
Genius! Everyone knows pirates are afraid of water!
"Using Watermarks to Combat Piracy"
I saw an episode of Frasier once where the brothers decided they wanted to open a restaraunt that catered only to special clientelle. "We won't put up a sign." "Yeah! And we'll get an unlisted phone number!" "Yeah!!" And Marty jumps in and says "Why stop there? I can get a buddy of mine to sit on the roof with a rifle and pick'em off as they come in!"
Every time I hear the phrase "...to combat piracy" I remember this line.
"I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)
Regardless of what humans can or can't see in it, there will be byte for byte differences and patterns to those differences. If there weren't, the watermarking couldn't be spotted in the files later found to be pirated.
At the very worst, a simple matter of re-encoding the file in memory from digital to analog and back would insert enough variation due to nothing more than the variance introduced by floating point math to make the process easily circumvented.
Aren't people transcoding iTunes stuff now by just letting it play and re-recording it? If done in memory, I don't know how lossy that is, but I'm sure its going to produce "good enough" sound for most people, no? Those who care enough to claim they can tell the difference are not the ones pirating music anyway, AFAIK.
The plastic distribution industry (formerly known as the record industry) is spending stupid amounts of capitol to preserve a business model that rewards the distributor of the plastic on which content resides over the artist who made it or the consumer who bought it. Ultimately, that model will fail simply because it no longer makes economic sense.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
How does it actually pevent people from copying the files.... Cant this file be uploaded....Oh wait they will know who's file it is...except I live in a country using a credit card where that isnt crime...Woops better luck next time.
The phrase "more better" is acceptable English. suck it grammar Nazis
Two unique versions defeats this easily.
Mix the two and you get a clean version with no loss in quality.
But a pretty dumb file compare program will have no problem. Compare two versions of the file to see where the changes are. Compare them to a third version to assess how different each watermark is. Then fiddle bits to create your own version that they cannot no longer trace back to you.
It will cost more to deploy the embedding software and panoply of infringement detectors than defeating this mechanism, which leads me to wonder about its cost effectiveness. It will only catch the dumb crooks, and not likely even scare the smart ones.
Reminds me of how multi-thousand dollar traffic enforcement cameras are defeated by a low tech can of spray paint.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
Rent one or two copies to use as a reference Problem solved
Slow Down, Cowboy! It's been 60 minutes since you last successfully posted a comment.
just ask him.
Wanted: Clever sig, top $ paid, all offers considered.
Ok...so I'm not exactly sure how they are going to get their watermarks onto mp3s that were ripped from CD, since most of the downloadable music already has DRM in it. I'm not entirely sure how they are going to connect that CD to a person, assuming they can get the watermarks to work like that. I'm also not entirely sure how they are going to track unique watermarks for every song/artist out there. But as long as they want to throw their money at something like this, and continue to pass on the savings to the consumer, I think its a great idea...
The only change I can believe in is what I find in my couch cushions.
*whoosh*
-CL
Not only because that would be supporting DRM peddling assholes, but you have to be even more careful with these files than with DRM'ed files. You are now liable AND tracable if an mp3 you bought somehow gets shared online. It doesn't take a whole lot for that to happen.
Maybe you happened to leave a windows(Linux/BSD/Mac) share open on a hostile (college/company) LAN? Maybe you lost your iPod with those tunes you bought, and somebody else is happily sharing "your" MP3's on p2p networks? Maybe the PC repairman decides that he'd like a few of your mp3's for personal use when you brought your pc in (it's not like *he* could get in trouble, there's no way to trace it back to him)?
You have to keep an eye on your files, and your system constantly to make sure none of your mp3's gets away, or otherwise you can expect a huge fine in your mail when you least expect it. Buying those MP3's is even more risky than just downloading and sharing them online like many do now.
This was rolled out years ago, and plotzed with a mighty thud when it happened, due in no small part to the http://www.cs.princeton.edu/sip/sdmi/faq.html">wor k of Felten and his grad students at Princeton.
Basically, the Powers That Be came up with a very good watermarking system, but even the best system can be defeated by a very determined adversary -- especially since the watermarks can't be updated once the CDs are shipped.
Another problem that I've always had with these systems is the proof issue. If the RIAA tries to prosecute you for having watermarked files, they have to demonstrate the watermark. I can't imagine how they could show that without revealing exactly how the watermark is detected -- and once they do that, you should be off to the races.
Anyway -- this has been tried, and it has failed. The SDMI system was really quite sophisticated, and it failed almost immediately.
Thad Beier
I love Mondays. On a Monday, anything is possible.
It's a nice thought, but I'm still not buying music online because it doesn't give a significant incentive. The cost of a full album is nearly identical to a real CD, and for that you get regular 128-bitrate encoded files. I'd rather pay a little more and get the full-quality, which I can then re-rip according to whether it's going on a portable or another CD afterward. I'd be more skeptical about data management than the shortcomings of the actual watermark. Will unlimited-music providers like Rhapsody generate a watermark for each file streamed? That'd be an insane amount of data to collect and track, and it'll only make a difference if my particular rip becomes one of Kazaa's top 100-most-common to matter.
They said no one would notice the brown dots they add to movies to combat camming either, but I see them in theaters all the time without really trying.
It would be so easy to get 2 copies of the song, look at how each is different, there would be a volume jump at 1 point, different in each. Just edit it out.
For an image, again, just compare the two, and you will find the difference, the watermark. Edit it out.
I thought most of the content on the internet was ripped from CDs (which they can't watermark--or at least it would be a real bitch to trace it back to a given purchaser).
on top of that, they seem to be intent to charge as much to download a song as to buy a CD, sometimes more.
So how does this do anything at all?
Now, for p0rn it's another story altogether.
This whole data protection is crap anyway! Just deal with the fact that you are competing with piracy and approach it that way. Tighten your belts a little and charge $0.75 cents a song. Charge $5 for a movie and don't even bother with anything more than trivial copy protection.
Most people would be honest if given a REASONABLE choice, and if you can't afford the $0.75--well the record company has no business making money off your broke ass anyway!
One place where watermarks might be interesting--If all the songs I'd leagally purchased were "marked" with my personal code, I could scan through them and see which ones weren't mine, then I could pay for them in bulk, converting them to marked files yet retaining the flexible MP3 format.
Even the best-trained human eyes and ears, according to Kip, can't detect the change.
Just like we can't detect the imbedded dots in movies? That system annoys the crap out of me, being a gamer I'm used to react to small fast changes, I always get distracted by them.
It seems to me that the primary goal of watermarking system is simply to identify pirated content. Even if a pirate changes or removes a watermark, you can show that the mark was pirated or removed.
So, let's say you gave each legally sold copy of a song a unique randomly generated 64bit ID (that you record). The pirate could remove that ID. They might even put their own random ids in place of your id. The deal is, their IDs will not match those that you recorded, and you could make the the case that this is pirated music.
The thing that needs to happen is that publishers need to fight against the professional content copyright violators.
If done right, watermark technology would be sufficient to track and counter the extremely abusive copyright violations while allowing the use of open formats.
For that matter, I think "hidden" watermark technology is going in the wrong direction. The mark does not need to be hidden in the file. If you put a unique identifier on each thing downloaded. When you go to make the case that group A is pirating music you can either prove that there is a bunch of files with the same ID or with the wrong id. You don't even need to track back to the original buyer.
You code media players to detect the watermark (which would have to be in a standardized format) and refuse to play anything that does not contain the watermark.
So would independent recording artists be able to insert the watermark? If not, wouldn't that be grounds for an antitrust action? Or are they assuming that all possible songs are already copyrighted to a major multinational publisher, as hinted by this article and this article?
MP3's such a universally accepted format that i'd be able to purchase music online and be able to use it wherever i want - be it in the gym, on my ipod, on the tivo, and mac/pc/linux.
Watermarked MP3s would be a way that the music industry could say "look, we almost trust you!"
Ok in pictures I can see it beeing easily applied as it is now. Music however seems alot more difficult to keep a "watermark" applied. If it is only rasing the volume on one tiny piece couldn't I just convert my mp3 to some other format and turn on volume normalize? It seems so easy to bypass its not worth applying.
WTF?
The ide with watermark, is that they are normaly hard to copy.
And exactly how are these mp3s hard to copy?
Do most of the pirated MP3s out there come from people downloading from a store and then sharing them on P2P? Or buying a CD, ripping it, and then sharing it? I was under the impression that most of the "pirated" songs were put there by groups solely for the purpose of pirating music, if that makes any sense. Ie, if I paid to download a song, or bought a cd and ripped it to MP3s, I'm probably not going to go through the trouble of sharing said MP3s. Even if I'm feeling especially anti-RIAA that day, I doubt the majority of the teenybopper tio-40 kids are going to even think of something like that. It's the people who get the prerelease CDs and whatnot that are introducing them to the scene, and I'm guessing they're using sources other than downloading electronically. What is the point of this technology?
Real watermarks are for duplicating, not taking out. Absolutly nobody would want to take the watermark out of their 100 euro banknote. In an mp3 you would instead want to remove the mark.
Am I being anal? Well yeah but when it comes to security it is the only way to be. A banknote with the watermark removed just lost its value. A mp3 with its watermark has possibly just increased in value. It certainly has lost none.
So the type of attack they have to stand up against is totally different. A banknote watermark just has to be expensive to duplicate. Add enough expensive to duplicate elements to a banknote and you will make it unattractive to counterfitters. It is the reason you see so few attemps at counterfit cents. (Please do not post links to your favorite wooden nickel story okay?)
But all the 'counterfitter' has to do with the mp3 is to remove the watermark. Wich as others have already pointed out should not be too hard. This is totally different type of attack. Remember, the banknote is proud of its watermark and makes it very easy to 'see' it. It even forms a pretty picture to make it stand out more. The last thing you want in the mp3 version is for it to stand out. Adding a split second of mp3 codec that stands out shouting 'look at me I am a pretty watermark' is just asking for it to be edited out.
Oh well, will this work? Well only if they somehow manage to keep you from just removing the watermark. mis-Trusted computing anyone?
Funny thing, I own more LP's then any other medium. In fact as more and more anti copy protection is introduced, the less I own of it. LP (too many) -> CD (repectable) -> VHS (0 now but used have a okay collection) -> Mini disc (a couple)-> DVD (a few) -> iTunes (0)
Odd that.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
This sTorY is unsUbstantiaTEd. MovE alonG. MOve aLoNG.
Where be the ships and swabbies, the cannnons and cutlass, the parrots and wenches? Where be the sea chests full of plundered gold??
Arrr, what a poor excuse for piracy this be!
May ye be touched by his noodlely apendage
There is no "-1 offended" or "-1 you don't agree with me" mod options for a reason.
They don't need to prove you did anything. It doesn't even matter if you've done anything. All they have to do is accuse you of piracy, and you will settle out of court, because you cannot afford to pay the legal expense of defending yourself. Proof never gets the chance to enter the picture. That's how the system works. If this new technology allows them to accuse people of piracy, that's all they need.
not to mention, will it be able to hold up in court when the MPAA explains "ok, now we know you can't see the difference, but this machine says there is one"
-- lol pwned
Cleansing music files is not much harder
Are you sure? If the stego and the crypto behind it are good enough, you won't know that you even need to cleanse a given file, let alone which algorithmic steps to take to cleanse it.
and one can always simply re-record the music too
Recording a new performance of the music is called a cover version. RIAA isn't interested in covers, but Harry Fox Agency is.
I am not sure if the watermarking will survive a re-encoding to a different codec. ( The new codec might use normalization etc. to compress audio content ).
*ducks*
If every file bought in a store or downloaded has a watermark, why not just take 2,3,4 or more copies and overlay them to "wash out" the marks that can survive a codec pass? Once this is done, who is identified in that watermark? I think initial uploads will involve a 2-image wash, with later downloaded and merged copies going further. This erodes the capability of any tracking to a single market or consumer for the source of the watermark, without even knowing the crypto to reverse it.
Also, who gets the blame for the large current of stolen and fenced audio that is subsequently ripped onto the net? If my car's CD booklet was taken in a smash-n-grab from a parking lot (fairly typical), then all those songs end up on the P2P channels, there's no way I'm paying for it. Watermarking may narrow down the number of holes, but it doesn't remove them all, and as long as a single hole exists, sharing is unchanged.
FTA: The digital media watermark used in the Fraunhofer system also contains a 'hash value', which creates a link between the content provider and registered purchaser. "The hash value is like a fingerprint; it contains unique information about the user," Kip said. "The software we've developed can automatically search for fingerprints."
They're going after the wrong target. The big copyright pirate isn't someone who bought the song and shared it using Kazaa. As long as the *AAs keep going after the big fish, piracy won't be curbed and they'll just keep pissing off their (mostly) honest userbase and blame that for their losses.
As previous posts have mentioned, as long as you can hear it, you can rip it.
Maybe they should spend the money on promoting new bands instead.
My bad, I guess this wasn't really the same thing. http://www.wired.com/news/technology/0,1282,52665, 00.html
These watermarks are being developed not to catch the users whom download them, but to encode each copy for instance Universal hand out, and when they appear on the internet are able to see which source copy this came from. I have no idea how this will help for cams or telesyncs, but in the case of stollen screeners etc. if not removed before release the groups will be putting them and their sources at a major risk. There is an interesting video all about the new watermarks here: http://www.usvo.com/smartmark.htm didnt read the link in the article, so im sorry if this link is in there. or already posted. -Ap0c
If the song is so bad that a trained human ear cannot hear who is singing it, then there may be some merit in adding a water mark...
Fraunhofer is the company that owns the patent on mp3 technology, which is why your mp3 encoding software requires a paid license. This is why most Linux distributions don't come with an mp3 encoder installed.
My guess is that Fraunhofer sees the possibility of WMA and AAC files taking over the digital download scene, and they are just trying to make mp3s seem like a viable format for labels and their outlets to consider for commercial downloads. Because if everyone moves to WMA and AAC files, Fraunhofer stops getting paid for their mp3 technology.
You realize if you just analyze the wave content you will most likely be able to find this. Any user with Cool Edit (Adobe Audition I believe now) or Sound Forge can open up the file and find it.
So using the watermark, files traded on P2P networks can get traced back to their source?
OK, say the source is a CD I bought in a certain branch of Tower Records, and they know this was sent to that branch and sold on a certain day and time. Am I mistaken in thinking that by paying in cash there is no paper trail? (No credit card details, etc?)
Because it seems to me at least that I'm then free to batten down some hatches and if I've time to spare start bucking those bloody swashs I never seem to get around to...
Wouldn't modifying the original content of the song be considered copyright infringement on the artist's property - and possibly even considered to be a mix?
$signature_views++;
Um, how could they prove something as a watermark as opposed to data corruption? You know data corruption does happen
Don't buy any music, online or off, just pirate it all.. that way if any of your music finds its way to p2p, it isn't "your" music and the RIAA won't go after you, just whoever you got it from..
Yep, karma as bad as an AC. Yep real cool, a communal account will never work, all it would take is for someone to change a few things and poof, no more communal account. to the moderators, mod this and any post by this user down as troll or flamebait.
A public account: log in as "Communal Account", password is "kFhthALQ".
So what happens when someone uses one of the many iTunes shared file-ripping programs to copy the watermarked files of a user who believes his or her sharing to be legal (note that I say "believes", because the actual legality of the act may be questionable)? Prettymuch everyone at my college uses the iTunes share feature, which attaches enough restrictions (mostly the inability to actually save the files without using third-party software) that most people don't think it's at all legally questionable. How useful is the watermarking tool if it captures all these people and not the more informed users who would strip the watermark?
A lot of video ripping involves altering the FPS, resolution, and/or encoding. A watermark would have to be able to survive such conversion to be useful, in which case it would likely be detectable, if not by humans than probably by software.
Even so, the labels might adopt something like this. But it would be in addition to their current copy restriction schemes, rather than a replacement for them. Consumers still lose as they'll still have to wrangle with FairPlay, WMA, or whatever copy restriction scheme the labels want to use.
Penny - plain text accounting
Sharing programs will now accept a plug-in which randomises the watermark on any media file, both when downloading AND uploading. Not only will the watermark on the file you downloaded not match your source (including RIAA-infiltrated sources), but anyone getting it from you (including RIAA moles) will be guaranteed a non-matching copy.
Well, back to the drawing board.
they're right, i cant hear any high-pitched noises, but my dog whimpers whenever i play pirated music over the speakers...
Is it the user's responsibility to keep their music secure? If I park my watermarked mp3s on a PC and there is a security breach, am I responsible if someone takes the music I've paid for and shares it on a p2p network?
If I'm not mistaken, the RIAA was OK with you sharing music with friends. They just didn't like you sharing it with your 5000 closest "friends". If I pay for a song and give it to a friend who puts it out on a p2p network, am I liable if it gets shared across the world?
Watermarking sounds good in theory, but what are they really going to do to someone who only shares one song. The RIAA really wants to get those people serving up 1000s of songs. Its not likely that someone is going to purchase 1000s of songs and share them with the world.
In then end, what does this accomplish? Knowing that Bob downloaded a song 2 years ago. That he put it on a mp3 player that he pawned, lost or was stolen. That whoever got the player shared the song and now we know that Bob was the original purchaser.
Ninjas don't carry tic tacs
...our data corrupting overlords. Thanks for the bitrot guys.
-"...bad old ideas look confusingly fresh when they are packaged as technology" - Jaron Lanier (Digital Maoism on Edge.o
Especially if you tell them about it.
Seeing as how someone has already stolen this account, I'd like to offer a tip to whoever creates another one of these. First, create a PGP signing key and upload it to your new Slashdot account. It'll be located at http://slashdot.org/~$USER/pubkey. Then sign your first message with this key. See sllort's journal entry for more info on comment signing. After the inevitable theft of the account, create a new one with the same signing key. Sign this new account's first message with the key. Next, reply to as many of the original account's comments with the new account name and keyhash. Continue as long as needed. If a dumbass decides to change the stored signing key in favor of his own, reply to each of his messages with the original keyhash in the subject line and a comment containing the chain of signed first messages. The downside to this is that someone needs to be the "keykeeper". If the private key was available to all, the thief could use the key to make an imposter account. Feel free to poke any additional holes in this method.
Now that some jerk actually did something bad, I'd just like to make it clear that it wasn't me.
-CL
Then you hear "MPAA r00lZ... MPAA roolZ....."
Engineering is the art of compromise.
De-tagging applications aside, this system could never practically work until all music is distributed electronically on the net. Sure, you can watermark physical CDs - the problem is linking a physical CD with a purchase. First of all, that would require all music stores collecting personal information on all customers and submitting that information to the RIAA - clearly a privacy issue. Credit cards, you object, but a determined pirate clears that hurdle by brandishing the cash in his pocket. Of course, we haven't even touched on the issue of giving that same physical CD away as a present (perhaps to a stranger, sounds like a good legal excuse). A million other issues exist, this is just an example. Conclusion? It'll never fly.
Just write a program that re-encodes files to a personal set of standards. Example rules: make sure all linear fades are in fact completely linear, parabolic curves fit the parabola exactly... and all quirks watermarks can hide in disappear.
No one is going to bother to pirate mp3's downloaded from itunes, or some other mp3 downloading site. The quality lacks tremendously, and it is much better to buy the cd and rip the tracks to your hard drive yourself. (My preference is FLAC) And I'm quite sure that pirates would do the same, the only people who this would be catching are people who really don't know what they're doing.
In some countries, making a copy for a friend is legal, provided that you have the original. Now, when this friend, who has a legal copy uploads it somewhere, your friend does something illegal, but the watermark points to you. I can see a lot of legal hassle here.
"If, for instance, you purchase and download a CD, burn a copy and give it to a friend and that person puts it on a filesharing network, our system will trace that music back to you and, depending on the legal system of the country you're in, you could be [hit] with an expensive fine," Kip said. "This could certainly help deter online music piracy."
Another possibility is that I'd be detered from BUYING the file in the first place.
to whoever put them on the internet, and then get medeival on them for breach of contract.
Well I'm sure that they would like to do this, in their fantasies. But in the real world it's a music producer vs. Clear Channel, not some individual D.J. And if Clear Channel decides that it is in their best interest that the music go out to the P2P file sharers before the record is released, then there isn't a whole lot that the record producer can do about it.
Clear Channel gets money from every music source except sales of disks and downloads. In other words, it is in their interest to have the 'product' on the file share networks; copyright laws be damned. They own the radio stations, the concert venues, the ticket companies, the artist management companies (in some cases), and the billboards.
Maybe someday the entertainment companies will complain about Clear Channel's cavalier approach to their 'intellectual property'. But don't hold your breath waiting for it to happen. Huge companies tend to avoid harrassing each other over relatively trival things like this unless one is trying to do a hostile takeover of another.
Time to ditch MP3! either go OGG or for those who need maximum quality, forget compression and go pure .wav like I have. (only for those who have the space)
So you know it was a 'copy' that you just downloaded.. So what? People doing this really dont care where they come from, as long as its what they wanted.
---- Booth was a patriot ----
...how do they mark the water, and how do such marks on the water keep pirate ships from attacking honest merchant ships?
I respectfully disagree.
Your problem is that you are accepting the recording industry's propaganda, i.e. "We oppose piracy because people will listen to pirated copies instead of buying CDs."
The *real* objection of the recording industry, and this goes double for clear-channel, is that P2P sidesteps their promotion monopolies and makes the music market harder to manage and control. Fragmentation of the market costs them their niche at the top of the foodchain.
The best example of this attitude was, a while back, movie industry executives noticed that some heavily promoted presumed-blockbuster (I forget which movie it was, The Island maybe) was getting far less than the guaranteed level of attendance given the advertising budget. Careful marketing research traced this phenomenon back to bad word of mouth, which was spreading faster than it had in the past, chiefly by cellphone.
The response of the movie industry was NOT "gee, we'd better stop making movies that even brain damaged 11 year-olds regard as intellectually insulting", but instead "is there any way we can make it illegal to badmouth our movies by text message? Libel law, maybe?" Fortunately, they concluded that was a non-starter.
That long tangent aside, look at clearchannel. Clearchannel's business model depends COMPLETELY on the willingness of the general public to agree-to-like whatever 30 songs they decide they want to play/promote in a single month. They also need to make sure that people keep listening to the radio and not to ipods. Alternate routes of distribution are just as much a threat to clearchannel as they are to the recording industry.
The good and new comes from no quarter where it is looked for, and is always something different from what is expected.
Even the best-trained human eyes and ears, according to Kip, can't detect the change.
Just like those fucking annoying orange dots. Once you spot them during an action sequence in a movie, they keep jumping out at you, breaking the immersion in the film over and over. Actually, if they replace the dots with anything even *slightly* better, I'd applaud the move.
Also it requires every purchaser of a copy to be a registered one, it requires the purchaser to be very careful not to have the copy stolen or lose it, and it might also lead a hacked watermark to accuse an innocent purchaser.
---- MISSING MISCELLANEOUS DATA SEGMENT --- [sigdash] trolololol
So all their fancy technique becomes useless unless they violate your privacy first.
This is not to prevent copying, just to help track down where copies come from. This may not be too useful in current environment, but as cinema turns to digital distribution and terrestial broadcasts of television and radio go digital, it would help piece together a case against suspected infringers. I'm not at all for this, but just explaining how it would be useful (and likely will be used) by content providers. By the way did you know that many laser printers already have such watermarking, to allow police to prove that a printout was made by a specific printer?
Dear RIAA,
Want to beat piracy? I'll tell you a little secret... here's all it takes:
Give up on all this DRM crap which just pisses off your customer base. How 'bout putting less "suck" in each album, paying artists fairly (e.g. no more "breakage/spoilage" fees based on shellac record trasportation, billing your musicians at absurd rates for manditory "promotional" services, failing to advertise new talent, hell -- failing to RECOGNISE and SIGN new talent, etc.)? Oh, and don't forget selling that music at a competitive price?
You bellyache that your industry is in trouble! Well, no one would ever know it by your business practice. Most companies, in order to attract new sales, put everything on discount and make it really easy and enjoyable for the consumer to buy your product. Not so in the record business. Your industry obviously wants to keep everything the same and, unbelievably, acts even MORE hostile toward the customer. Yet you wonder why year-on-year sales have taken a dive. Who was your macroeconomics professor, Seinfeld's Soup Nazi?
Anyway, at this point I don't really care if your industry lives or dies. I, and my family, have learned to do without your products -- we just stopped buying. Or in the rare event we want music we buy it, just not from you. See, that's significant. In a lot of cases, we can now buy directly from the artist or purchase a used CD. Judging from your sales figures, we're not alone.
But then what do I know, I'm only the customer.
Watermarking can be used for good or for evil. For instance, this book has a watermark at the bottom of every page explaining that the book is CC licensed, and available for free from a certain URL. I did this because someone had taken the pdf file, carefully removed the copyright page and licensing information, and was selling the book on a CD on ebay :-) It was ironic, because he could have *legally* sold it for free according to the license, but he wanted to mislead his customers so that they wouldn't know they could get it for free off of a web page rather than buying from him. Of course, it wouldn't be all that difficult to remove the watermark, but because it's on every page, it would be *more* work to do that than it was for the guy just to remove the copyright page. Similarly, I don't think any DRM scheme for music will ever be able to stand up against a determined attack, but they can make it enough of a hassle that it will cut down on illegal copying. Whether you think that's good or bad probably depends on whether you're a college student or a music industry executive...
Find free books.
Aren't approaches like this weak in that you could apply a filter to the image, changing it imperceptively and destroy the watermark? Wouldn't an image recognition system be better? It could be similar to the kinds of face recognition software that can pick you up even if you are in disguise...
This "solution" assumes that somehow, someone will be able to track WHO bought every song and what watermark was assigned to them. This is a great added cost, if it's even practical. The number of bits required to embed in one song, to make sure each watermark is unique, might be too great to fit in three minutes. (Movies are longer and more expensive AND sell fewer copies, so my arguments do not really apply to them.)
- Systems Curmudgeon, AKA Precison Blogger
I read the article as "Using Watermarks to Combat Privacy."
Although, that would make sense.
"I am the king of the Romans, and am superior to rules of grammar!"
-Sigismund, Holy Roman Emperor (1368-1437)
What I don't get is the CD watermarking to discourage people from uploading ripped music to P2P networks. Typically, the ripped song is MP3 encoded before that happens which everyone knows is a "lossy" compression algorithm. In other words, rip a cd into a wave file. Compress the wave file into an MP3, then decompress the MP3 back into the wave file. The wave file you get back is not the same wave file you put in. My question is can the watermark survive a lossy compression scheme? The same goes for images and movies.
The article is a little short on technical specifics, but it's hard to imagine how a watermarking system would work with a lossy compression format.
If the watermark is applied to the file after compression to mp3, then it is very easily defeated by decompressing and recompressing with a non-watermarking encoder, of which many exist for mp3. The act of decompressing the file will obliterate whatever bits were flipped for watermarking purposes. If the hidden information is subtle enough, the lossy compressor will simply throw it out. If it's obvious enough to not be obliterated by lossy compression, then I can hear it in the file, and the product is inferior. The only option would be for the encoder to recognize the watermark and purposefully retain the data, and then we're tied to a specific piece of software just like DRM.
However, if the watermark was applied before the compression (i.e., directly to the wav file on the CD), then the act of compressing the file will change the watermarking somewhat, and matching the "fingerprints," as they are called in the article, would be statistical in nature, not exacting like a hash is. The fingerprint would have to be considered "close enough" to be a match.
Also, every single watermark would have to be unique in order to match it to a specific source, which means creating a Big Database (tm) of customer info, which is easily defeated by paying with cash. On the other hand, if the record companies weren't interested in identifying a specific source, but the presence of a watermarked file in an upload directory is sufficient, then that's no different than the existence of any other file in an upload directory which contains copyrighted material, which is what they've been going after for quite some time now.
And the point is...?
For security, the MD5 hash of this message and sig is 09f911029d74e35bd84156c5635688c0.
Watermarks do not have to be 100% security though obscurity... It seems to me that a proper system would use a private key of sorts to unencode the information, then the algorithm could be free (as in speech or BSD) and like SSL only they key would need be obscure. If you're using just a simple 256 bit key to decode the watermark, chances are it would be nigh detectable (assuming of course that the actual technology can survive recompression / reencoding).
The Geek in Black
I know my BCD's (when I'm Sober)
Take two differently watermarked copies of the same item (pirated film, etc.) and average each respective byte together. The resulting file should be viewable and not visibly corrupt. Who shows up as the original purchaser now?
I think there's a difference between encoding reversibly, and one-way hashing.
This is pretty important stuff, so correct me if I'm wrong, and pay attention if you haven't heard of this idea before.
Compression algorithms, for example, give you a result that you can expand to get the original.
Hash algorithms will give you the same result each time for the same input, but there is no process that you can apply to hashes to get the original data back.
So, be not afraid.
However, because you are assured the same result for a given input, you have to be careful about what your input is. If it is only a credit card number, one could just take the whole list of possible numbers and generate the corresponding list of hashes. A translation table. Seek the hash you're attempting to "decrypt" in your new table and see which card number created it—you've got it.
You could use some combination of other vendor-known input and "salt" to confound an attacker's ability to generate the translation table and make use of pre-computed translation tables.
http://en.wikipedia.org/wiki/Salt_(cryptography)
I'm not a cryptographer/cryptologist, but I bet there are a variety of methods that add to the computation required for generating the hash, but multiply for the computation required for seeking the hash. Develop a costly process that takes 1 whole second to generate the hash and takes all the (current) world's computers a few heatdeathoftheuniverselifetimes to seek. There are probably (computationally) cheaper and cleverer ways to test whether a known CC# was involved in the transaction without giving away the card.
I bet your CC# is much more vulnerable to an array of other attacks than trying to reverse some cryptographic watermark.
You are exactly right. I worked for ClearChannel for a few months, helping to launch one of their syndicated music shows. P2P networks were often the source of on-air material, because we could get it from the Internet faster than we could get it from the labels.
[Commercial music radio stations] also need to make sure that people keep listening to the radio and not to ipods.
In that case, they have the under-18 market sewn up, as students in public K-12 school systems in the United States are generally forbidden to bring an iPod or other electronic music player on the school bus without the express written consent of school administration.
[Transcoding audio] would also suck plenty of quality from your file...
But does it suck more quality than the hypercompression mastering techniques of the loudness race do?
My rippers will not honor the watermarks.
Your CD drive has moving parts, which will wear out. Future CD drives will likely insert watermarks into audio tracks. If no company makes a residential-use-priced non-watermark CD drive that can be delivered to your country anymore, then what?
Do you BURY me when I'm gone?
Do you teach me WHILE I'm here?
Just as soon AS I belong,
Then it's TIME I disappear.
groovy. Can't wait.
Training monkeys for world domination since 1439
new versions of open source media players are
and new burning programs are released which burn watermarked files into standard audio CDs.
Using what drive, if all CD burner manufacturers cave to the four biggest record labels?
Even assuming you could get around all the hacks and workarounds (ya right) how does this help? The original source for most MP3's on any network is probably some CD someone ripped. Unless you started putting individualized watermarks on each CD there'd be no point, and this would be way too expensive to do.
That would be ... *drum roll please* ... the music producers!! Yes, it was the music producers who initially distributed the file to said user. Individual watermarks are pointless because Windows security is non-existent. You cannot take a watermarked file and use it as proof to sue Joe Average. His system is riddled with viruses, trojans, and rootkits... some of which were installed by the rights owners themselves (I'm looking in your direction Sony). In any case, it will not only be plausible, but entirely probable that the user's system was rooted, and the files transfered off the system by some ner-do-well. Case closed: RIAA you loose. Go home, take your ball with you, and don't let the door knob hit ya where the good lord split ya.
Good luck with my velocity-insensitive, 6-note polyphonic, state-of-the-art 2001 MIDI ringtones.
--
Change any notes on my Nokia
ringer and you'll have to pry
it from my cold, dead fingers
Things like the DRM and DMCA were put into place to fight piracy
I think you're missing the most wonderfully ironic part of this whole story :-) The watermarking only exists as a strategy *because of* the DMCA... You see, thanks to the DMCA, the RIAA members cannot legally reverse engineer Apple's Fairplay DRM! So they are left with only ONE option if they want their music to play on an iPod without going through Apple... You guessed it! Unprotected formats. GOD THAT FEELS GOOD!!! F'ing the RIAA! How can you not respect Apple just a little bit for doing that to those a'holes (Go ahead Mac Trolls... that's your que. Flame away with the gay Apple jokes...)
:-D
I've been in public K12 far the last twelve years and never heard of such a thing. Is that really true? It sounds ridiculous.
It's ridiculous, but no more so than, say, copyright term extension. From the Fort Wayne Community Schools Student Rights and Responsibilities and Behavior Code:
Level 2 penalties include denial of extracurricular activities, after-school or Saturday detention, or suspension of driver's license and/or work permit. Penalties for repeat violations of any rule may increase at the principal's discretion."That's a valid intput, but steganographers thought of that years ago already. "
Of course, but that doesn't stop the slashdot experts from posting some kind of solution, like they're the only smart people on the planet, and everyone else is too stupid to have seen it coming.
Watermarking doesn't meet either of those criteria.
Thanks for the info. I always had this sneaking suspision that many broadcasters just got their copies off of P2P, while still paying for the rights.
That no machine can even pick it up! Don't tell hollywood; let them waste their money on yet another anti-piracy scheme. If it were undetectable, then machines would be nascent of it, and it would therefore not exist. We have seen every other antipiracy scheme fail, why not this one?
Melancholy Elephants by Spider Robinson
It presents an interesting argument about copyrights.
Jesus Christ! RIAA this, MPAA that, Clear Channel is the devil. Nobody is forcing anyone to watch, buy or listen to anything. Now that we have this interweb thingy people can find whatever they want if they're motivated. Marketing by itself cannot sustain anyone. Otherwise nobody would still listen to Black Flag; they'd be spending too much time listening to Puddle of Mudd.
If it's a supposedly undetectable change, couldn't someone just write a program to perform a large number of "watermarks" in random places in the song? This would make it hard to find the one that the manufacturer placed there. Otherwise, a simple shift down of the entire song's brightness and whatever other factors are being used by a simple amount should change the watermark.
When you release watermark content, you watermark ALL of it, every copy to some degree. There is no way to take two copies and come up with a clean one, you will either end up with gibberish or you will have no effect.
Watermarking is designed to overcome the simplistic attacks that have been put forth here, do you honestly think that the technology owners did not think of these attacks as well?
Watermarking is not a general solution, it is a specific solution meant to weed out the bad apples. There are many ways of applying watermarking, even the same watermark to multiple clips that would make it both very difficult to identify and almost impossible to remove.
The entire business premise of watermarking is that in order to defeat it you have to either use uneconomical attacks (which nobody has put forth here) or make the content itself uneconomical for distribution. For video this means that the video has to be reduced to below 320x200 and at less than 80kbps (IOW heavily compressed), with a starting point of baseband (21MB/s uncompressed, invisible to the naked eye) marked video.
You can throw out the geometric attacks as well, they don't work. And the whole SDMI thing is a non-issue as it was put forth by a company who did not have proven technology who was looking for validation in the market. There are companies who have proven technology (and I am not saying that fraun is one) that have met all of these challenges and then some.
I'm just having such a fun time imagining trying to explain the details of zero knowledge proofs to a judge (and/or jury). I can just imagine the looks a witness would get for a seriously non-trivial discussion of graph isomorphisms. (Or elliptic curves if encryption came up.)
This watermarking scheme is yet another reason for me to continue downloading music via P2P (which is legal here in Canada) as opposed to buying it on CD or over the net. I personally think that tracking what someone does with the product that they have bought and paid for, so long as it's not hurting anyone (and don't give me the bull about it hurting the artists), is ridiculous. It's also economically unfeasable for the music industry; behemoth that they are, the RIAA will have to invest millions of dollars to pay for the manpower and technology to track the watermarks and make investing in the technology worthwhile. Even if the RIAA tracks a file-sharer down, then sues and wins, they'll probably never see the money. With the kind of fines the RIAA is imposing, most people will never be able to pay; they'll just go bankrupt.
Between watermarking and DRM, I have absolutely no inclination to buy the music that I listen to. That is, unless the music is from an independant artist who uses neither method, and honestly, they're the ones who need the money the most in the first place.
Forget all that DRM stuff, just embed identifying details and personal marks into the media file. If its posted on kaazza, the media corps can find out who did it.
This won't work either. Assuming the best possible case for the recording industry, Joe Sixpack shares the song on a P2P network and gets nailed by the RIAA - THE SONG IS STILL OUT THERE!
Its still being shared, over and obver and over again. This doesn't STOP anything. Theres a fundamental flaw in this kind of protection scheme, it assumes that each time you detect a misappropriated copy that it goes away, that you stop it somehoww. Watermarks don't stop anything.
Once the genie is out, you can't stuff him back in.
If somebody thinks you have an illegaly copied file, they can trace back to the original buyer, who spread the file.
It traced me back the HMV at wakitaki, orishima. Uh.. exactly who bought it I don't know. WTF?
Of course my eyes and ears can and will detect the change - it's my consciousness that will not, because the small differences are filtered out in the brain's pre-processing.
In a dark room, a human eye can detect a single photon. Our ears are somewhat less sensitive, but some audiophiles have a hearing you'd not believe. Are their ears different - or their brains?
The point is that many things you don't consciously notice still lower the quality of what you see or hear. Low-quality encoding (video more visibly, but audio as well) is an excellent example. The picture will look blurry or "bad", even though you can't pinpoint exactly what it is that gives you that impression.
I'm sure you won't notice consciously, and you certainly won't notice if you hear your music on $2 headphones. But have you ever played an MP3 in a club/disco with an excellent sound setup? Man do you hear the difference to uncompressed CD quality.
Assorted stuff I do sometimes: Lemuria.org
Here's the old
http://hardware.slashdot.org/article.pl?sid=03/08
HD Trailers
to removing the watermark in an MP3 file (for example) would be to decode it back into WAV format (and normalize it while doing so) and then re-encode it to MP3 (again, normalizing it)
On any modern system, it would take only a matter of minutes to decode and re-encode an entire CD...
You can close your eyes to reality but not to memories.
Sales will nosedive, for any 'CD' that demands ID and paperwork before 'buying'- quite impractical for high rent CD shops where prices have topped out. Then there is the issue if the media is resold, the right of first sale will not go away.
Prior searches on watermarks will show they are easilly spotted, and the trick is to silence academics who show math is real and effective.
Then the watermark must be intrusive , if it is to survive the trials of reproduction and broadcast. This press release shows they hope to talk up share prices, venture capital, when the reality is, it has been tried and failed. Rehashes of this theme have been discussed in slashdot before, as well as threatening university lecturers with heavyhanded law threats.
and give him a cigar.
Look, it's not just audio tracks, it's video too. TV shows are starting to be downloaded, and with torrents you can download a MASSIVE video track from blockbuster/netflix/etc and also insert a private audio track. The video is useless without the audio. (unless your a Charlie Chaplan fan)
Most people are not endowed with a great sense of generosity. Most torrents come from a few people, over and over again. They don't need to get them on the first try, getting them on the 10th or the 20th share is good enough. Killing off the big sharers in the network will slow it to a trickle.
Once they find the culprit, by process of of elimination over dozens of cylces, they can insert a "gotcha" watermark into the file, that proves beyond a doubt that they shared at least that file.
Of course their is a consequence for killing off the big pirates: Legal downloads. People are now accostomed to downloading media from p2p network. Often discovering new artists. From an independent artists perspective, pirate networks constitute noise. If the RIAA/MPAA lower the noise level, their going to have a whole other fight on their hands.
I would rather be ashes than dust!
Nobody uses Fraunhofer's codec when ripping anyway. Everyone I know uses the Lame codec.
The notable thing here is that all the companies are doing is transferring that trust-anxiety from them->you to you->your friends. Now you have to look every friend in the eye and say "Will this person not put me in jail?"
I think "big brother" was pretty appropriate, considering how much like Stalinist Russia and other oligarchies and tyrannies this sounds like. The particulars are different, but we arrive at a very similar reality--the 'governing body' (RIAA) is has vague but powerful means to dispose of troublemakers and the 'oppressed' have really no way of knowing when they are going to get 'the bullet in their head,' except to constantly profess their loyalty (by distrusting everyone and never even taking advantage of their legitimate uses).
In the end, it's hard to forget that it's not our friends that are the spies, it's the MP3 file itself. I don't know, but to me that makes that sort of 'gray oppression' quite a bit more dehumanizing.
Does this mean owning music files = owning guns ?
If someone steal the music files and upload, then the owner is responsible for the act? Maybe there also needs to be a DRM police site where we reported the mp3 stolen?
Doesn't this approach have similar consequences to 70 yr old grandma or 12 yr old kids whose files stolen but charged for distributing mp3? How about lost or stolen ipod?
Does it also mean we need to physically smack the old hard disk to bits and pieces so no one can 'steal' mp3?
If owning the mp3 is going to be so much hassle, I'd rather not buy them.
AFAIK all published image watermarking systems are not robust.
Ah, that's it.
Thanks!
The good and new comes from no quarter where it is looked for, and is always something different from what is expected.
Well - here is the deal with watermarking and lossy compression.
The task of the perfect lossy compression algorithm is to store
only the information we notice. If the watermarks are not noticable,
the lossy compresser should not store it. It is by definition why
the data is compressed so well. So the situation is like this;
Either the watermark needs to be noticable - or as better lossy
compression algorithms are developed, they will automatically
exclude the watermark.
... until music players stop working unless you provide a valid watermark of course...
In my spare time, I sometimes take photos for a non-profit. Most of them aren't anything anyone would want to pirate, but there have been times where my camera was the only one in the room for a speech by one VIP or another, and that's a different story. They use the watermarking plugin from Digimarc, which I imagine isn't terribly cheap, but they've got it watching the web for illicit copies of thousands of photos...
Village idiot in some extremely smart villages.
That is the one. It is not terribly hard to defeat that watermark. Simply scalee the image and it is scrambled.
Do not look at laser with remaining good eye.
Below is more information about how Digimarc is worthless.
The Robustness of Digimarc Watermarks
We have made an informal investigation of the robustness of the Digimarc
protection scheme in the face of common-place image operations
in Photoshop. We used the example Digimarc image available from
Digimarc's web site4, which measures 215 by 142 pixels.
Compression Average { watermark becomes unreadable when
image is saved as low quality JPEG, even though the image quality
is not noticeably reduced.
Color adjustment and quantization Good { watermark is
retained until the image is close to unrecognizable under various
contrast, equalization and quantization operations.
Cropping, Translation Good { much of the image must be
overwritten before the watermark disappears.
Rotation Poor { free rotations of 1 degree completely remove
the watermark.
Scaling Poor { scaling the image down by 15 pixels removes
the mark, as does down-sizing by only 1 pixel and then applying
high-quality JPEG compression.
Filtering Good { resistance to Gaussian blurring and sharpening
lters, and also to addition of noise, is high.
Rotation is a less common image operation, and thus Digimarc's
extreme vulnerability to thismay not be so important when considering
non-hostile users. However, if the watermark can be removed by a small
amount of rotation then many non-technical attackers will be able to
remove the watermark easily from within Photoshop. The Digimarc
method's vulnerability to scaling, on the other hand, is critical to nonhostile
users.
I quit using it specifically because of how easy it is to defeat. Standard copyright is good enough to protect you. Trying to find a automated way to scan the net for your violations will never ever work. Stop wasting your money on digimarc and simply keep tighter control over your images.
Do not look at laser with remaining good eye.
Interesting paper - thanks! Tighter control isn't really an option, as the photos are being put on a public web site, but I believe our web guy may be implementing something where only members (free) can see the full-res ones. If I'm putting up a 200x200 headshot of someone, I don't Digimarc it, of course... it's the 2000x2000 version that gets the special treatment. :)
;)
Ah well, I'll just note it as one more expensive bureaucratic thing that makes someone up there feel safer, and makes life a little more annoying for those of us on the bottom of the pile.
Village idiot in some extremely smart villages.
What will you do once the 'Communal Account' starts at -1, lower than 'anonymous coward'?