Slashdot Mirror
Startup Tries Watermarking Instead of DRM
Loosehead Prop writes "A U.K. startup called Streamburst has a novel idea: selling downloadable video with watermarks instead of DRM. The system works by adding a 5-second intro to each download that shows the name of the person who bought the movie along with something like a watermark: 'it's not technically a watermark in the usual sense of that term, but the encoding process does strip out a unique series of bits from the file. The missing information is a minuscule portion of the overall file that does not affect video quality, according to Bjarnason, but does allow the company to discover who purchased a particular file.' The goal is to 'make people accountable for their actions without artificially restricting those actions.'"
Sounds reasonable. But then how does the copyright holder distinguish between the purchaser engaging in illegal distribution vs being the victim of theft? The article never covers that. I think I can guess how the **AA will react to any watermarked file floating around the net with Joe User's name/account reference embedded in it. They'll call a SWAT team and have Joe's house raided. No proof. Sorry, Joe, for the mess. We're on to harassing the next person we vaguely suspect of illegal distribution.
Then hex diff it, find the missing bits add them, and then.... profit!
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
This won't effect people putting up pirated movies at all. Those who are smart will edit out the first five seconds of the movie. Those who are stupid will just post it with their information.
I didn't RTFA, but how is this any different than a digital fingerprint? As far as the info at the beginning goes, anybody who cares to do so could simply chop off the first few seconds in any decent movie editor.
Solution: re-encode the movie, I prefer 2 pass xvid
Could the missing bits affect the movie and be detectable?
From TFA:
I'll assume the people working on Streamburst are clever; but I wonder how susceptible the ghost-stream is to translation and recompression: whether it's possible to corrupt the signature-stream while retaining watchable quality.
If the objective is to determine the purchaser of a file by means of the bits encoded in the file, would it not be possible to identify which bits and bit patterns are being removed and simply remove or replace them all? Or perhaps re-encode the file to a different format to totally change things?
There is no mod option "-1: Disagree" for a reason. "Overrated" is not an acceptable substitute. Post something instead.
In Soviet Russia, movie watermarks you.
-1 not first post
I do like this idea, we all say that we can be sensible and will pay for things so long as its in a form that is acceptable so we can use it (ie. without DRM). This would also give you your full fair use rights and would be able to fall into the public domain when the ownership had expired (another great benifit)...
In fact the only thing that I worry about is how much info they will keep on me to verify at a later point that it was me (or that it wasn't me) who put the file on Kazaa or torrent or whatever... will it be credit card info, linked to your address? will it just be a name and e-mail... and how secure are their systems it?
*''I can't believe it's not a hyperlink.''
So, people who pay for a movie from these guys won't be able to share it via Kazaa or bittorrent or whatever is popular right now. I don't think that many people who pay to download a movie really do so with the intent of putting it on a filesharing network. I mean, why the hell would you do that? The people I know who do the whole illegal filesharing thing, don't pay for media they can get for free, and the people I know who buy digital download media, don't use illegal filesharing sites. Buying something legally kinda defeats the purpose of using a filesharing site, amirite?
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0 is the magic number.
It takes a man to suffer ignorance and smile
Be yourself no matter what they say
This sounds perfect. As they say, it makes *me* responsible for the file; I can make millions of copies as backup. Of course I wont give it away, to do so is at my own risk.
The authentication will be a problem of course; it means I will not be able to make an anonymous purchase on the web - something that people are quite reasonably concerned about being able to do. What will it be signed with? My DNA? What about identity theft?
A heck, I give up. I was wrong. It's another stupid idea.
Open Source Drum Kit, LPLC deve board - mjhdesigns.com
Yeah, right. First, just because "my" copy of a movie ends up all over the internet, doesn't mean that I did anything wrong; maybe it was stolen from me. Second, if an evil-doer buys (or steals) a few copies with different watermarks, it's a good bet that he can merge them in a way that obliterates any evidence of where they originals came from. Do your homework, guys.
that really holds people accountable too. What if someone hacks into their computer and starts pirating? Now the movie is freely available and the person it came from gets bent over.
Confucius say: "Man who associates with smarter men than himself is smarter than the men he associates with."
-Eric
SJW: Someone who has run out of real oppression, and has to fake it.
This seems like a reasonable idea. It would certainly allow you any amount of fair use... but like any attempt at controlling access to something (presumably you still wouldn't be able to distribute it to a few thousand close friends via the internet...) it is probably doomed to failure.
Something similar to this was featured in a couple Tom Clancy books, the "Canary Trap" where a few key words were changed in versions of a document, without changing the meaning. Find an exact quote and you know who gave it. The problem is that once someone knows about that system, its pretty easy to not give an exact quote.
Obviously this system is a bit different. BUT... find three copies, and take a look at the string of 1 and 0 that follow. If two of the three match for one position but the third is different... go for the two that are the same in the version you distribute. Obviously there are probably some interesting things they could do to prevent this, I'm sure, but given time I'm equally sure they could be overcome.
Does a line appended to your comment give your post meaning in and of itself, or only in relation to those without?
Your computer is infected with a trojan and the malicious person behind it takes all your movie files and uploads them to [insert favorite p2p network].
This will be a goddamn mess.
Real pirates probably already have the originals anyway.
Besides, this appears aimed more to stop casual file swapping by scaring the non-tech-savvy than it is at real pirates.
In principle, I like this idea. I don't really see a problem with it.
However, they already do something similar in theaters. Every so often in theatrical movies you will notice a weird pattern of "cigarette burns" that appears for a brief moment. (Yes, to my eyes at least, they are visible and sort of distracting.) The pattern is different for each copy of the film shipped. The idea is that, if someone sneaks into a movie theater and makes a cam of a first-run movie, the producers of the movie can analyze the video and figure out which theater it came from. That helps them put more pressure on theater owners to enforce bans on video cameras, etc.
But does it seem like there are fewer cam bootlegs out there since they started doing this? They started it maybe five years ago.
Breakfast served all day!
well?
This scheme seems to cheerfully ignore the implications of legally selling on a copy.
I take it you didn't even read the summary, let alone the article. Hint: your solution won't work.
Adding an intro to the video will already alter the data in the rest of the file, especially if two-pass encoding is used and assigns slightly different bitrates to parts of the movie. Just keep a checksum of each 1MB block on file and you are good to go.
I suspect this would be fairly easy to circumvent, but I love the idea!
I have always thought that piracy should be solved through law enforcement, not technology. Much like traffic law enforcement.
DRM is the equivalent to putting a 70 mph speed cap on all cars. This watermarking is sort of like requiring cars to have a license plate.
If they can find a way to make this work I'd be overjoyed.
You thought this was something intended to defeat deliberate large scale pirates? Why would you think that? I mean none of the DRM crap stops them either, so why should this? :)
This sig has been temporarily disconnected or is no longer in service
http://www.broadcastpapers.com/whitepapers/Content %20Technology-05-2006-046-048.pdf
r ticle.html
The Thompson system for watermarking video and there's also a Fraunhofer Institute system:
http://www.pcworld.com/article/id,124676-page,1/a
These are all good ideas IMHO. As long as
1. The watermark isn't easy to remove
2. There is uncertainty as to whether the mark is removed
3. It isn't used to apply DRM
1 is obvious, 2 is there because the pirate has to be uncertain if their copy still has the idea, and 3. because the advantages of the system over DRM are lost if they use it for DRM!
Imagine you can freely buy and use the media you use however you like, but if it shows up on p2p, the ID can be pulled and traced back to you.
Since the DRM doesn't work, (not a single piece of media has successfully been locked up by DRM yet, a 100% failure rate). And since the DRM is already so restrictive that it puts off genuine sales, and is causing competition problems as inter operation is non existent. Then watermarking scheme will take over.
This one, I'm not so keen on, since the watermark is too easy to remove compared to the more mathematical approaches. The key point of any watermark approach is the mark must be difficult to remove and there must be uncertainty that the mark has been successfully removed.
My 2 cents.
Its just a shame many people have already thought of this idea and realized that it doesn't work (for all the reasons people have posted here...subtracting more bits to hide themselves, diffing multiple versions from different accounts and fixing it, etc...)
I see a future with millions of movie files on the P2P networks that are watermarked "Blockbuster Video".
Sweet informative mod.
In the light of Vista and God knows what tricks hidden in its media player, would it be possible to identify movies either by watermark, or with [ducks] image recognition-type software build into the player, and then send the report back to Microsoft/studios, who would subsequently bill you for the content? No DRM, nothing wrong or unethical, all filesharing networks operational - just market expansion with Big Brother kind methods, and huge revenues for studios.
You can do almost anything, including while (1) { fork(); } but it's logged, so the sysadmin can ask you not to do that ever again (;-))
davecb@spamcop.net
Please disregard previous post. In true /. form, I just now actually read the summary.
Sweet informative mod.
As far as I know, this is the strategy employed by TiVoToGo, which lets you take video off of a TiVo and watch it on your laptop. Here's one article discussing it. Personally, I'd take a watermark over restriction any day.
OK, so lets consider a few examples that prove the flaws in the watermarking method:
(1) a person sells their used computer/disk, and it has watermarked content.
(2) a person makes a CD/DVD/floppy/tape backup of their content, and it is lost/stolen.
(3) a hacker hacks their system and downloads it.
(4) a houseguest makes a duplicate without the owner's knowledge or approval.
(5) pirates watermark their content with a fake name, and/or the actual name of a
third party.
(6) etc, etc, etc. There are limitless ways it could happen.
I don't think it will be easy to use that watermark as evidence in court. Not unless their is alot of other corroberating evidence. I don't see the watermark, in and of itself, as sufficient to prove anything in court.
"Forgive us our trespasses, as we forgive those who trespass against us." -Jesus Christ The Lord's Prayer
First there is the obvious fact that has been pointed out. How do you prove that you were the one who leaked your file, and not someone else stealing it? Second, this doesn't solve anything with the pirate that made his or her own file of copyrighted material and then shared it with the world.
Let's face it, honestly copy protection gets us down. However, this is DEFINITELY a step in the right direction. This is far less intrusive into a viewing customer's life. Joe Schmoe won't have a problem with this (I hope), and it does give the Industry some sort of protection. Not a bad idea at all.
I realize there are several problems with it in practice- and that pirates taking the effort to do so can break this. However, this leaves us with a copyright protection scheme that: A. Isn't a hassle (it doesn't restrict the customer) B. Is at least as effective at discouraging piracy as anything else they've thought of. This means that it is the best Protection racket^H Scheme people have come up with yet. There is the danger of the MPAA sueing some innocent people, but I doubt they'll sue anymore innocents than they already do.
You are reading a copy of my copyrighted post.
1-Buy 2 or more files from them
2-do a bit comparison
3-modify a copy to reflect a random profile of all removed info
this would make any compairson hard.
"There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy."
I applaud the idea of giving people the freedom to do what they please with the media they have purchased. This idea has a great motivation. I wish it could work, however, as much as I like the idea, someone will do the following:
Purchase two copies under different names.
Compare the two bit-for bit. Anywhere the bits are different, set the bit to a random value.
Watermark destroyed. Video intact.
- For the complete works of Shakespeare: cat
TimeTrax tried this with satellite audio recordings. Copyright holders will still pissed off b/c they had to do the work in enforcing their copyright with every tom, dick, and harry who got a copy of the recording.
Watermarks piss people off.
from TFA-
"Because of its design, the watermark even survives most editing changes and format shifts"
As others have pointed out, this could be easy to defeat (and so is DRM) but the technique is fine by me. I have no interest in infringing others copyright and that is why I find DRM so offensive.
Pitch to Consumers: It's not DRM!
Reality: Polished turd.
Why not use this same concept but imbed ads at the beginning/end of the file, then allow free distribution.
Develop a format whereby the decryption of the video is stored in the ads. Any modification of the ads will result in the inability of the video to play.
You could even have on the fly ads in cases like this, actually targetted ads that people may actually want to see.
At some point I did a scetch of a somewhat similar idea in some net forum. Though I would not remove bits, rather I'd do an encoding with slightly increased quality in a few random places. (That way I would hope to prevent people bitching about reduced quality). And how much the watermarking costs in terms of extra space could be computed exactly. I haven't done any calculations on the extra space, but I would expect a few KB for a full movie.
To explain what my idea was I'll first give a short reminder of how jpeg works. Blocks of image data are transformed using something based on fourier transformations. The resulting coefficients are then rounded to different scales. For high frequency components a scale with larger steps can be used as errors in these components are not easilly noticed. There is a table of standard steps to be used for each combination of horisontal and vertical frequency. (I left out the part about how to handle colour components, which is not relevant for the following idea).
Making a minor change to one of the step sizes is not going to cause a major difference in the size of the compression or the quality. By picking some of the entries at random and reducing the step size you are going to increase the quality of random parts of the picture. Now what I want to do is to make a redundant encoding of a signature on the text from the watermark and use those bits to choose places to increase the quality. The signed text itself is included in the begining of the file.
First of all removing the signature would means you couldn't compute the step sizes, and thus you couldn't correctly decode the file. And if the file was reencoded, you might still be able to extract the watermark by comparing with the original uncompressed movie. You would just have to find enough of the places where quality was increased. (And enough is a lot less than all of them).
The signature used in the encoding should be performed using the buyer's private key. In addition to this, I would sign the entire encoded movie using the seller's private key to be able to detect if a file is corrupted (as a service for the users). The part about the user signing something could be replaced with just using a hash of the text, but that might weaken the proof of origin of a particular movie a bit.
Now all of this could be combined with features to prevent users from accidentially losing a copy to a cracker/pirate. Since this is not intended to prevent users from intentionally copying the file, it could be a lot better and less intrusive than DRM.
Do you care about the security of your wireless mouse?
... it is THEIR data. In Soviet Russia, DVDs own you...Although as many have posted it might be hard to demonstrate liability on the owner of the copy it would be easy to prove that someone in possession of a watermarked file are *not* the owner of the copy.
A good watermarking solution is resistant to many kinds of processing, including re-encoding.
Thats the whole challenge.
Off course the watermark might not be resistant to extremely destructive transformation such as downscaling from HD to QCIF, but then who cares about pirated QCIF video ? But certainly a very accurate transcoding would not affect the watermark.
Current watermarking technologies are very much dependant on proprietary algorithm.
I heard of a similar idea (don't know if its been used). You download a program from a website, and you get a unique "program key" to unlock your unique version of the program. The key is your name, credit card number, and expiration date used to purchase said program. You're welcome to share it with anyone you like ;-) This allow people fair use, with a big incentive for them to not share it with others (at least until the credit card expires ). The only downside I see is like a previous poster mentioned, that you can't buy something anonymously. If you mail order something its not really anonymous anyway (they have to have your address, etc), but it is a problem compared to buying a game with cash at Best Buy.
The truth is anybody can break into your house at anytime. They don't because there's some risk, however slight, that somebody will notice and they'll get caught. Same logic here. It's not going to prevent somebody from pirating but it will discourage the lesser crimes.
This sig has been temporarily disconnected or is no longer in service
I think that the point isnt that this is unbreakable or that they may not even look to go after the author, rather they are relying on a couple of points:
1. Its just too much work to crack/reencode etc... Not impossible, just a pain in the ass (probably more of a pain in the ass then running a disc through DVD Shrink)
2. If something has your name on it, they hope that you will be more likely to keep everything honest. They don't have to come after me, or even threaten to, because if I see a file with my name floating around I might be worried what others think etc... This doesn't mean that no one will distribute, but many would think about it twice. They are basically saying that if you are the only one using it then you have nothing to worry about
For example, have a scene where in the background you have a bakery with a sign above the door showing that the address is 2149, then later in the movie perhaps a phone number on a poster in the background of 329-1934. With a bit of digital magic you could create millions of combinations of those numbers, and they won't affect the movie one bit.
The differences could be any number of things changed in the movie, but it would be visual.
That way, even if you change the bit code, or even throw it up on the wall and copy it with a camera it'll still have the information in it.
As far as the problems of "what if someone steals it and distributes it..etc". What if someone steals your gun and commits a crime? Is that a valid argument against having trackable firearms?
Nuf said.
Arash Partow's Philosophy: Be a person who knows what they don't know, and not a person who doesn't know.
Nine Inch Nails did something similar with the underground video for Broken. I believe I read on their website in the access section (I'd provide a specific link, but their entire site is needlessly in Flash, and thus not indexed by Google, let alone searchable with the browsers find dialog), that the film had specific static signatures that identified each of the 14 (I think) original leaked copies. So, years later, they were able to tell who leaked the copies. Of course, it was supposed to be viral and they wanted it to be leaked. While static does detract from film quality (typically), in this instance, it probably added to the overall aesthetic.
I suspect we'll see MPAA opening thousands of cases against John Smith.
Do not. Touch. Down.
Seriously, how can you have an article like this and not bring up SDMI?
Any Crypto-based DRM can be bypassed. RIAA/MPAA give the person the DRM'ed data and give them the key to play it, and then they tell them they are not allowed to copy it. I have news, if you can play it you can read it. Period. Failure guaranteed. The problem is that by making it unusable by DRM'ing it they actually ensure that someone will be pissed-off enough to put it out on a p2p share.
Watermarking on the other hand addresses the Social issue and is only a deterrence to sharing the file, not to using it anywhere or anytime the purchaser chooses. The drawback is that one takes a chance of the media getting into the wrong hands and then getting blamed for willingly violating the copyright laws. Yes, you can easily remove or destroy a simple watermark, however the watermark can be done in such a way that when repeated with several variations of bit flips during encoding the water mark can still be recovered, much like using parity bits to correct a memory storage error. The question the p2p sharer will have to face is whether they have sufficiently removed the redundant copies well enough to prevent the recovery process from revealing their identity. Of course you can buy it under false pretenses/name and then its all a moot point. Just being a deterrence to keep the honest person from sharing without suffering undue problems during its use is definitely a step in the right direction.
The dark side of this is that DRM could be added in such a way that the player would refuse to play without a second watermark being present. If you destroy the first then the second won't allow the media to be used. Thats only a speed bump for a true geek. You can count on the MPAA/RIAA to jump on that band wagon before long, and we will see more of the same until they come to their senses. They don't learn very quickly after all.
EZTakes is already doing the same thing in the US. A brief intro before the film shows who purchased it. I've tried ripping the disc just using selected titles (as the "watermark" intro is just a title on the DVD) to remove my name, but they've anticipated that also. I haven't got a good rip yet without the watermarked intro.
Not that I wanted to rip them off or anything. I just took it as more of a technical challenge as to whether I could remove the watermark on the otherwise non-DRM'ed DVD.
Pretty cool service, although you won't find any blockbusters in their selections quite yet. But for my taste in '70s cult films, I've found a few that I've downloaded for the fair price of $2.99. Not bad, not bad at all.
:q!
The effort to uniquely watermark each copy individually is similar to encoding...high processor time. So I doubt this will be done per consumer. Perhaps per block of time or server. Even if it's per consumer, there is a hitch:
The encoding format is a known standard, so this the manipulation of the frames and core data is already possible. For example: decode to HD, perform some minor bit flipping on colors or clear out metadata areas, then encode (either MPEG4 again or to another format). Lossy compression muddies the water on the encode, so that it may not survive the pass. Only they have the tools to determine if the original owner is detectable from the end file.
It's a nice idea, but the content providers will never go for it. They want to use DRM to limit fair use so they can sell you the same content in different formats. They can make themselves sound very self-righteous banging on about preventing piracy, but they are at least as interested (if not much more so) in preventing our fair use.
"DRM is doomed to failure and addresses the wrong problem, but watermarking addresses the Social problem in making it less desirable to share with the Internet at large."
What makes you think that technological solutions can solve social problems. DRM didn't, and watermarking is an even weaker technology. The honest have no need for any solution, and the dishonest will always escape them. The solution is hard, not because they require an engineer to develop, but because they require changing the human heart. And humanism so far hasn't come up with a good way to do that.
But what if I REALLY don't want my name tacked onto "Dude, Where's My Car?".... can I use an alias?
Considering all of the DRM junk that I keep reading about (and enjoying mp3s to avoid the complications), I do not see anything wrong with watermarking. I don't feel that hacking is an issue (if someone hacks into my comp I want them to tell me how they got through Verizon's portblocking). And an article like this when I have mod points too....
I'm starting to think GNU is the problem with "GNU/Linux" these days.
I do love this idea. And if Real & Napster used water marking for MP3's or AAC's (or OGG) that would run on any device that can play those formats, they could sell music to iPod users. Hell, anyone could with this technology. It such a no-brainer you'd have to wonder why the labels didn't come up with this themselves.
Ryan-
- Get three copies.
- Binary diff files 1+2
- Use the diff index to repair file 1 using file 3.
- Strip first 5 seconds.
TFA doesn't give detail, but it can't be that simple, can it?boakes.org
Are my legal rights when it comes to First Sale doctrine? Once I own a copy, I can do with that ONE copy whatever I want (except copy it).
The watermarking system disallows my LEGAL right of selling that object to somebody else.
Now, what would be interesting would be an online database of all the media conglomerates coming together to create a Ownership Library, in which one can buy a copy right, so that downloading it would be legal. Simply verifying if requested downloader has a copy provided to them could potentially make users on p2p legal.
For example, I'd like to download a new album. I'd go to the ownership library, buy a copy "right", then download from any source I wish (legit provider, or piratebay..). To keep these shares legit, it would potentially request that I have a copy right to access that file share, and after checking that I can own it, allows download. It could keep the users AND sharers from turning into copyright violators.
Just duplicate some frames at random spots in the movie. This will likely realign all the compressed block boundaries and change all of the bits. It won't just be a matter of figuring out offsets like the text version of diff does. You'll need multiple copies of the movie and code which compares each frame in all of them to look for the duplicates. And even that might not be possible since the encoder will likely throw out different pieces of information giving each different encoding slightly different artifacts.
"Pirates of the Caribbean"
Purchased by: Dan Glickman
(insert jargon here)
----
Suddenly he is no longer president of the MPAA and is living on the streets. . .
But, as stated in TFA, DRM isn't about piracy, it's about squeezing every red cent out of the end-user. Want it as a ringtone? CHA-CHING! How about playing it on the PC? CHA-CHING! MP3 Player? CHA-CHING!
They're gonna lose an awful lot of CHA-CHINGs if they use a watermark instead.. Jesus, this is a step backwards... I'm waiting for the **AA to realize they can sell the cover art, lyrics, etc as separate items and charge even more...
Ok, so you bout the music track.. Wanna buy the lyrics now? How about the bass and guitar parts?
Seriously though, I would definitely be more interested in an unencumbered format with a simple watermark than I am in the DRM crap that exists now.. I used to have an Audible account, which I loved. Problem is, after I downloaded the DRM encrusted file, I had to spend hours running it through a convertor to get an OGG file that I could use on my iRiver.. Seems iRiver has no interest in the Audible DRM format.. No piracy interest at all, I just want to be able to listen to the damn thing!
XenoPhage
Technological Musings
While I did not read the article or any posts; this measure won't be successful either. It is similar to [essentially] movie houses who provide DVD Screeners to reviewers and/or rental stores with a unique id # to which try to identify the source of the leak - which are in turn obscured upon re-release and re encoding [and removal of the 5 sec intro] of these particular releases should obscure the identity of the original user.
Movie houses and the recording industry in general need to recognize the real underlying issue of cost to the end user. When then market is fair in regards to both cost and the removal of restrictions [DRM], piracy will decline accordingly. When their focus changes from protecting their broken business model to creating a fair and viable one, their true issue should cease to be a concern.
First you have to know where to filter.
That's easy: Obtain two or more copies and compare them. The watermarks MUST be different, so the bits that are different tell you where they are.
Assuming the watermarks are statistically similar to a fixed number of random bit-flips, two copies identify half of them, three identify 3/4ths, four identify 7/8ths, etc.
Of course with a few samples you might be able to crack the system. If the watermark is a set of redundant copies of something you can identify, from then on it only takes two (the second being to be sure they haven't changed the system or added another.)
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Even with a single copy of the file it can be cracked, just pass it to an equivalent process which remove some more of the file so when they compare they see no perfect match.
sign(c14n(envelop(this)), x509)
In a file the size of a video file, they can make hundreds of subtle changes. With lots and lots of redundancy, even if the re-encoding removes the vast majority of the watermarks, there is a good chance (it doesn't have to be a 100% chance, really) that the file can be traced back to the original purchaser.
Music files at 5MB are a bit tougher, but it's probably doable.
You assume that they start with a stock copy and then change them. That would be a mistake for exactly the reason that you mention. I find it hard to believe that they did not think about that.
I suspect that they will instead provide a varying data set. It may be that they have a standard set of copies (say 1000) that are slightly differing. Or they may chose to run it through a varying compression amount (99 is different that 99.0001). All in all, it should not be too hard to defeat what you are suggesting. Where they will lose is that they can not possibly defeat all codec transforms. But I would guess that a set signature might be impervious to DIVX, but be defeated by mpeg. So what? If you release one or two films by accident or to a small set of friends/relative, then you are not a threat to MPAA. If you continue to release them and copy them to your friends, now you are a threat. In my mind, then the MPAA has the right to come after you because you are in competition. If I were a pirate, I would be thinking about how to steal films every so often or how to have varying IDs. There is no such thing as security except by not selling it (and even that is not fool proof). Though I suspect that each copy will have various codecs that will wipe out the signature. IOW, they may have a signature on frames 10,12,14,...20. Then it turns out that converting to divx or compressing to tighter will leave the sig intact.
I prefer the "u" in honour as it seems to be missing these days.
First sale is an exception to the distribution right defined at 17 USC 106(3) as the copyright owner's right "to distribute copies or phonorecords of the copyrighted work to the public by sale or other transfer of ownership, or by rental, lease, or lending." Without that exception, sales, lending, rental, etc., would be infringing.
First sale is defined in 17 USC 109(a): "Notwithstanding the provisions of section 106 (3), the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord." There are some exceptions to 109(a) regarding rental of music and computer programs (basically, you can't do that, for the most part, although libraries and nonprofits can).
What's important, anyway, in the definition of first sale is that it applies to lawfully made copies. Copies in the statute don't mean what technical people think of when they use the word copy. We tend to talk about copies as instances of a particular sequence of bits. So, you might say I have two copies of a file on my desk, etc.
But this isn't how copies are defined by law. Copies, by 17 USC 101, are material objects in which a work is fixed. Physical things not sequences of bits. So, an instance of a movie on your disk is not a "copy." The disk is the copy. You can have two or 200 instances of the movie on your disk, but you still have only one copy, and that's the disk.
This is important when considering what it means to distribute a copy. If you want to sell someone your copy, under first sale, you have sell them your actual, physical disk. Chances are you won't want to do that
OTOH, if you sold your computer, and the copies of the programs you have on your computer are lawfully made, it's probably legal under first sale to sell the disk unformatted (even while keeping the originals). That's complete conjecture and I'd have to look at it more clearly, I'm sure someone will correct me
What's actually perverse about US copyright law is that 106(3) is very clear (ie, by the plain language) that distribution means distribution of (physical) copies. But the current interpretation of the law says that transmitting copies to others over the Internet constitutes distribution. So, even though you wouldn't qualify for the first sale exception if you emailed a copy of the movie to someone, you can still be found to infringe the distribution right. This is wrong but the law nevertheless. The EFF is making a good argument against this interpretation in one of Ray Beckerman's cases and it'd be great if they (and he) prevail.
In other countries, transmitting a copy would be a performance, and so it might infringe the public performance and display rights rather than the distribution right. Suffice to say, the guy on the other end is infringing the reproduction right by making his own copy on his own disk, so even if you could get away with selling the work to him by transmitting it over the Internet, he'd still be infringing and his copy would not be lawfully made.
is to 'make people accountable for their actions without artificially restricting those actions.'" emphasis mine
That will be absolutely impossible while IP law remains on the books. Its entire reason for existence is to artificially restrict action, access, etc.
What?
They'll call a SWAT team and have Joe's house raided. No proof. Sorry, Joe, for the mess. We're on to harassing the next person we vaguely suspect of illegal distribution.
Yes, but this is a different problem than the plague of DRM and other such things currently used on media. Watermarking is a lot nicer than invasive DRM (although I wonder about how it affects ownership exchange situation). The xxAA should not able to able to stage raids on public citizens like they do currently, but that's a different issue altogether and certainly shouldn't be considered a reason to look down on the use of watermarking.
Does this mean that if I walk into a movie store to buy a movie, I'll need to provide ID which will then be recorded along with the serial number of the movie I purchased?
What many seem to be forgetting here, is that it only takes one source file to populate the P2P networks. And that happens fast! It doesn't matter if that source comes from a rip, a cleaned version of a watermarked source, or a stolen source with the ID data still in it. When it comes to removing the watermark, the amount of processing that may be needed is trivial compared to the available resources from a single P4, let alone multi-core or clusters. We truly have very low-cost supercomputers and clusters of supercomputers sitting under our desks. Reminds me of those who rely on CAPTCHAs... easily, trivially defeated by OCR, whether silicon based or carbon based.
"Under this scheme if the thief breaks your car window and steals your iPod (and shares your music files), you are a criminal. Big difference." Nonsense. Your're not a criminal, you've done nothing wrong. IN BOTH cases you're a victim, only in the latter you have a much harder time demonstrating it. Your point is completely belied by your moronic assumtion in the fae of reality. I don't know where you learned logic, but you suck at it. And you mods, god damn people what the fuck are you idiots thinking modding nonsensical tripe like this up? You should be ashamed.
Does not affect video quality!
Whenever someone introduces a new DRM watermarking technique, they always say that, and it always turns out to be a lie.
I'd really like to believe this, but I just can't.
Whether or not people are "fundamentally" good or evil isn't an argument worth having, in a way, because it's impossible (or nearly so) to take a person completely out of their environment and away from the threat or fear of consequences. However, I suspect that if you gave the 'average Joe' a Ring of Gyges, that he wouldn't help himself to the contents of the local bank/liquor-store/etc. (at least until the novelty of being able to possess anything wore off).
While you, in fact, may be so constrained by morality -- and if that is the case, I salute you -- but to assume that most people are, seems a bit of a stretch. Most people don't commit crimes, because the perceived risk/reward doesn't work out in their favor. I could go out tonight and hold up the 7-11 on the corner, but I'm not going to; the few hundred bucks it might gain me (at best) wouldn't be worth the strong possibility of spending the next decade or so in prison. However, to someone who was poorer, or strung out on crack, that equation might come out differently; the possibility of a small amount of cash might be more than enough to make the risk worthwhile.
We can argue about the fundamental nature of humanity all day -- after all, if it was good enough a subject for Plato, it's good enough for me -- but in the end, what matters is whether your philosophy produces a model that predicts how people actually act, rather than how they wish they acted, or how they justify their own actions to themselves. The risk/reward model does this fairly well, at least with economic and property crimes, and therefore seems far more likely.
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
I'm not sure all the "Oh noes this is taking away my resale rights" have a point.
As distribution becomes entirely free of physical media it was going to be hard to resell your copies anyway. What did you want to do - have people that popped over to your garage sale stick a usb drive in your computer and mv the copy over??? On the one hand we want physical media to die so that we can time shift and format shift to our hearts content, and on the other hand we want to maintain resale rights. I'd say be reasonable.
Resale rights have been dying for a while. A lot of new computer games come with cd keys that get linked to online accounts ala steam. You could try to resell them, but the guy at the other end would be buying a limited copy. Try reselling your itunes downloads recently? http://news.com.com/2100-1027_3-5072842.html. Old record stores were my favourite way of getting old music because you couldn't waltz into Tower records and buy it. Then emusic came along and I switched.
Availability of older material won't be so much of an issue if you have distribution thats free of physical media. That itself reduced the value of your resale rights in a way. Digital distribution with watermarking will very effectively kill the resale market. This will probably lead to nasty pricing issues with older material. But the point is they were bound to die ever since I could make a copy of a CD and sell the original at a garage sale. Or borrow a dvd from blockbuster and burn a couple of thousand copies with dvd decrypter and resell them in paper envelopes. In the process though I get dirt simple format, place and time shifting.
Now digital watermarking is a much more consumer friendly approach than DRM. You get a copy, do what you like with it except distribute it and if that means you effectively can't resell it then c'est la vie. Nothing by the way prevents you from reselling it - just the risk of getting hauled to court. Sort of what you'd expect in a world where you can keep a copy and make an infinite number of resales anyway.
DRM controls you much more. You cannot format shift easily (and frequently not without loss). Worse how you could use your content were more strongly controlled. I can imagine a world where if you wanted your iTunes to play on your iPod and your mac you'd need a different a different version for both. Or one where you couldn't buy a copy and only lease one on a pay for play. If any company gets a monopoly on online content distribution this will likely happen.
Reality must take precedence over public relations, for nature cannot be fooled.
OK, first, this is not a new idea, and its been beaten to death in the security literature - it doesn't work. The DMI tried this back in the 90's with SDMI, and it was found be just an unworkable back then. Lets count the ways this new idea is still the same old unworkable problem:
1. All purchases would require ID. ID's are not only easy forged, but legitmate IDs can be purchased through corrupt government employees. So you create a barrier to purchasing the product (I need an ID to buy a $9 CD?), plus its not reliable. IDs can be faked or stolen. Just because it says Joe Smith, does not mean its Joe Smith.
2. IDing would require a national/international register. Just because it says Joe Smith, does not mean that its the Joe Smith from NY. So you need still more information about the person AND a means to confirm it. Its not enough to just have a credit card, identity theft scams abound, so Joe Smith from 1234 hartford st may not be the real Joe Smith. So we need their SSN, or some other national identifing scheme. (You need my SSN to buy a $9 CD?) Maybe even some biometric information to know that its Joe. Yikes! All this for a stinking CD?
3. You expose the customer to identity theft by collecting this information. Now you have a database of all sorts of useful information about the customer, controlled by people that might just want to steal it. What about that clerk at the music store? Can you trust her with your social security number, and other identifing information? (All for a $9 CD?)
4. Could you use cash anymore? Maybe, but you'd still have to produce ID, and maybe other information about yourself. What if I don't want you to know who I am? What business it of yours if I want to buy a Depeche Mode CD for my Niece. Fuck off.
5. What if the purchase was a gift? So I buy some media for my Niece, with my name attached to it, and she then gives it to her friend, who then puts it online. Guess who gets the call from the RIAA, and has to pay the expense of proving their innocence. (Which creates an incentive to remove these watermarks, and possibly even a business to do this)
6. As others have pointed out, what about theft? If my digital media is stolen, again the burden rests with me to prove my innocence. Also, now some of my personal information is in the hands of a thief! Yikes! Sounds like a really good reason to remove the watermark!
7. How do you know you can trust the watermark? Is it cryptographically secure? Could someone change the watermark to incriminate their neighbor? If its cryptographically strong, then it needs more bits, which makes it even easier to find and remove.
8. Will the watermark survive removal from simple lossy compression? Probably not. If its not supposed to effect the medium in such a way as to be perceptible by a human, then it will probably get lost in compression (afterall, those bits aren't needed, so out they go). So is it an effective system then? The science says no, you would lose the watermark, and probably without even thinking about. Rip CD to Vorbis, and the watermark is gone.
9. You can remove the watermark via other means. Steganography is hard, really really hard, and individually marking files makes it much easier to detect the "hidden data" by opening the file to comparison attacks. But there a few other attacks we could use:
a: Rosetta stone attack) If you have an unwatermarked file, you can use a rosetta stone attack. This happened with the ill fated SDMI approach from the 90s.
b: Oracle attack: If the attacker has access to an Oracle (a device/program for detecting the watermark), he/she can fiddle with the file with impunity to find the watermark, and again remove/obscure/modify it.
c: Lossy compression attack: Just use lossy compression, it will probably destroy the watermark.
10. This won't stop illegal copying. Its just a reactionary method, based on the assumpti
Python
What if you wanted to sell the movie to someone else? Or even just give it to them? Now it still has your name on it...
Why don't you just subtract the watermark from itself? If you have a watermark that tints a pixel in a certain direction, just apply the reverse operation.
...because "hacker" sounds way sexier than "code drone."
just wait till someone releases a tool to create the watermarks.
The best test environment is production. - Me
chrome://browser/content/browser.xul
Ok, so you just need two different owner copies. You look for differences and remove them and Voila!!
Nice idea but it will fail.
They have told you about one watermark.
Does that mean that there are not others in there.
Nothing is stoping them using other methods.
Can't watermarks in lossy format media data just be wiped out by shuffling the low bits in which the watermark is encoded?
--
make install -not war
They don't. But the person who possesses someone else's watermarked media files has some explaining to do. That explaining could take the form of a simple purchase receipt. You can legally resell copyrighted works, even watermarked ones, just as you can sell a portrait painting -- as long as you transfer or destroy all your copies.
The signed, portrait painting is a good analogy. It's pretty easy to figure out whether someone has proper title to the Mona Lisa. The original purchaser and any potential subsequent owner both have ample incentives to play by the rules.
"I think the various **AAs should learn that the problem isn't piracy, but that piracy is the symptom of a larger underlying problem, that their business model is outdated and self-defeating (may I add draconian?), and their prices are unfair."
Oh Lord no! It's NEVER piracy's fault. That would then mean it's OUR fault, and we can't be having any of that. It's always "your business model is bad" (So why doesn't the poster create a new one? It's called competition.). Your business model is "self-defeating" (Looks like slashdot wants it both ways again. Either piracy is hurting, or it's not. You can't flip-flop). Oh and my favourite because it's brought out every single time, "your prices are unfair", for some never specified degree of "unfair" (Once again there's always the opportunity for said posters to start their own company. Signup all those acts thare are always "on tour". And set "FAIR" prices). Think it's going to happen? My odds are on the lottery.
If they go out far enough to risk attaining originals from movie theatres, pre-screeners or whatever; what's stopping them from purchasing these online videos through a public unprotected wi-fi connection using a fake credit card & name?
Then who cares whom it's registered to... and no cracking required.
KISS
No trees were killed in the making of this post; however, many trillions of electrons were horribly inconvenienced.
Wouldn't this be "watermark" be lost as soon as you transcode it to a different format?? Or just running it through a software equalizer, so that it shifts the bits, rendering the "watermark" no longer existant...
Or if you're a CS Major and want a fun little project, go buy three or so of the same track under 3 different names. Then, analyze each of the tracks, and the bits that are different should be the "watermark". Now you've identified the location of the changed bits, and you can just fill in random bits to throw it off... However, this method would require you buy at least two copies of each track to find the different bits, but hopefully, after a few different songs, you should be able to find a pattern, especially if it is just a quick 5 second intro as the article suggests. From that point, you could write a program to automatically "strip" the watermark.
Although, if Napster-To-Go would switch to this, I would sign-up immediately. Imagine being able to download unlimited tracks that don't expire. However, I highly doubt that the RIAA would ever allow this to happen.
Rirelobql xabjf gung EBG-13 vf gur yrnfg frpher rapelcgvba rire, ohg jbhyq lbh jnfgr lbhe gvzr npghnyyl qrpelcgvat vg???
Interesting. Watermarking music is something that may happen. When there was some thought of watermarking bullets (so you know who purchased them, after retrieving them from bodies), the NRA clamored enough that the thing went nowhere. It gives you an interesting insight into the scale of values: watermarking music, yes (poor music companies!), watermarking bullets, no (just bury the guy).
Really, why not to sell personalized versions of music/video/etc?
Think about selling the real thing with titles/frontends not watermarqued, but clearly personally labelled. It will be great to have 'YOUR' own CD album / DVD movie signed from the real artist(s), with your own words, its easy to do and adds an inestimable value to the purchaser.
What's in a sig?
Humans are strange beings. Often its "Push them in one direction, they will run in the other".
So it is with DRM - part of the DRM hate originates from the "you force me to do (or rather not do) something" thought and the wish to resist whatever they are trying to do.
Watermarking, on the other hand, is more subtle, and I could actually support it; Share that file with your friends - no DRM or copyprotection will (in vain) try stop you. And even the company you downloaded the file from won't know or care - as long as you share it with just a few friends.
If you put it on the internet, however, you are in trouble - 10thousands of downloads do kinda catch the attention of the powers that be. Now if they only demanded realistic prices I'd be convinced.
+++ MELON MELON MELON +++ Out of Cheese Error +++ redo from start +++
All it takes is ten minutes alone in your house and all sorts of things can be done which can land you in a heap of trouble.
Drugs, child porn, identity theft, etc.
Society isn't collapsing because of it.
No sig today...
We seem to be approaching a critical moment in these things. First there is the copy protection provisions of Visa, which must provoke protest when they get going in the market. Second, there's the pending release of Leopard. Now, its hard to see how they avoid releasing retail x86 versions of Leopard, and when they do, its hard to believe people won't crack them within weeks and install it on any x86 they want. Third, there is Open Moko, which looks set to test the locked smart phone model to destruction. And now we have a solution to the piracy problem, which conveniently deprives Apple and MS of the excuse that they are only locking their music to their players because the music industry makes them.
We might hope for a number of things to happen. First for Open Moko to take off and eclipse iPhone. Second, for Leopard to emerge and force unlocking from Mac hardware. Third, for the nakedness of the iTunes and Zune lockins to become totally obvious to everyone, and for such lockins to become uncool and generally despicable. Fourth, for people to refuse to tolerate the remote degradation of their Vista hardware as a penalty for some alleged possibly imaginary misdemeanor in the content they try to play.
This could be the year the whole thing blows up with a bang!
is that the best way to compete with piracy - which is merely a business model - is to COMPETE.
Wow! Competition! A word most of the intellectual property morons have never heard of!
Typical chimpanzee behavior: "I can't compete - so I have to coerce somebody!"
Unfortunately for the monkeys, there's always somebody better able to coerce than THEM. So you end up with investment in coercion, which leads to a non-productive society consumed by paranoia and coercion.
Sound familiar?
There's only two answers:
1) Humans need to learn to stop being AFRAID OF EVERYTHING - especially death.
2) Get rid of humans.
I opt for 2 - it's more feasible than option 1.
Have a nice day, chimps.
Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!