***“Mistrust-based” DRM could very well reduce the number of people who upload a given file to peer-to-peer networks. But unless you can reduce that number to zero, it’s not going to do any good.***
Here is your absolutist reasoning at work again Tim. It falls inline with how you argue that unless DRM or the DMCA stop *all* piracy, there is no point in them, and how you argue that unless patents have 100% correlation with innovation, that they are no good, and how you argue that if DMCA 1201 affects *any* reverse engineering that it stifles innovation.
A more nuanced approach would be to state that mistrust based DRM is valuable if it reduces piracy to levels where its economical for creators to invest in online distribution.
Doug Lay
· 2 years ago
Noel, your lack of technical background is showing here. It only takes one uploaded copy of a file for the file to become globally available. It's a binomial distribution, not a multinomial distribution. Either a file is available or it's not. A single copy is as good as ten thousand copies.
Techniques that can reduce the number of downloaders can be effective incrementally. Techniques to reduce the number of uploaders, however, are only effective at all if they reduce available copies of a file to zero.
Noel
· 2 years ago
Doug, you adopt Tim's logic. Imitation really is the best form of flattery, but is it wise:)
Doug Lay
· 2 years ago
Noel, you can run around spouting 2=2=7 all you want. It's a free country. I just hope no content holders seriously think you're going to be able to save their bacon taking that approach.
Ed Felten
· 2 years ago
As Tim points out in the post, Peter Eckersley found a signature field in the Apple files. So I corrected the post that Tim quoted. For the record, his quotes capture the original version of the post correctly.
Noel
· 2 years ago
Doug, please you're getting ahead of yourself.
You claim that there are no numbers between 0-100% (by claiming that DRM is not useful unless it entirely stops piracy), then argue I'm bad at math:)
But in any case, even if I did run around yelling 2+2=7, its better than ranting about some worthless revolution that never and will never happen.
Doug Lay
· 2 years ago
In the case of uploading to file sharing networks, it only takes one copy to make the item available for downloading around the world. So in this case, yes, there are no numbers besides zero and one. This is what I mean by "binomial". Whether or not you call this a "revolution" is up to you, but it has happened. To claim otherwise is willful blindness.
Noel
· 2 years ago
DRM critics say it does not stop piracy, but then criticize law enforcement efforts in the name of a revolution. Basically, they're 1) making an argument, 2) trying to stop others from disproving the argument, 3) trying to generalize their cause. In some cases, DRM critics argue that peer-production obviates the need for professional creators to leverage DRM- an argument that seems intended to entirely moot steps 1, 2, 3 in case any of them are wrong.
eric
· 2 years ago
Noel, reality matters.
In the case of the recent release of the keys to high definition DVD encryption, one person uploaded the hex number to the internet. Shortly afterward, it was on hundreds of thousands of web pages around the world, and I'm sure on p2p networks as well. The same would be true for a single copy of a new album. This has also happened. The last U2 album was stolen (one copy) prior to release and uploaded (once) to the internet, and from there spread worldwide.
Tim's statement isn't just an argument, it is what actually happens in real life. To deny it is to reject reality.
Tim did admit there would be a marginal effect. Fewer copies uploaded in the beginning might slow the spread of the file worldwide by a short amount of time. Obviously there is a finite effect. Finite and infinitesimal. If that's the basis of your objection, OK, but it is irrelevant in practical terms. Between 0% and 100% it is so close to zero, I don't understand the quibble.
Steve R.
· 2 years ago
Noel: The piracy argument is a red-herring to hide the fact that the content producers are extorting property rights by depriving the consumer of property rights that they have traditionally enjoyed.
For example, I can read a book in the US or I can read it in England. The content industry has imposed region codes so if I buy a DVD here, I can not watch it in England.
I can buy the book today and read it five years from now. The content industry is claiming that if I "miss" watching a program that I paid for, I can not use technology to save if for future viewing.
From my point of view, the pirates are the content producers who seize every means to restrict the rights of consumers.
Noel
· 2 years ago
Eric, Tim doesn't seem to consider DRM alongside the DMCA nor law enforcement, and consequently does not consider that piracy can be deterred even when it takes only one uploader to make a file ubiquitous.
I"m not sure why Tim does not consider DRM *and* the DMCA (or other copyright enforcement mechanisms), since neither alone is meant to deter piracy. They're supposed to work together. Rather Tim seems to focus on one at a time, such as here, and argues that DRM alone does not stop piracy.
Tim also analyzes one patent at a time and asks how it contributes to innovation rather than a dataset of patents, so I'm not surprised to see him take a similar approach with DRM.
Doug Lay
· 2 years ago
Noel:
The world is bigger than your obsession with Tim. You will be a more effective advocate for content ownders or whatever it is you're advocating for if you get over this obsession and focus on how you can protect your constituency's interests. Content owners could care less whether Noel thinks Tim has an inflated reputation (no disrespect intended, Tim!). They care about knowing what works to protect their interests. Picker's proposal won't work, no matter what direction Tim's career takes.
Noel Le
· 2 years ago
Doug I dont care about Tims career as much as you do:):):)
eee_eff
· 2 years ago
DRM critics say it does not stop piracy, but then criticize law enforcement efforts in the name of a revolution. Basically, they're 1) making an argument, 2) trying to stop others from disproving the argument, 3) trying to generalize their cause. In some cases, DRM critics argue that peer-production obviates the need for professional creators to leverage DRM- an argument that seems intended to entirely moot steps 1, 2, 3 in case any of them are wrong.
Noel:
If someone comes up with DRM that does NOT require special legal protection which tramples on the First Amendment (e.g., DMCA) I would be more neutral about DRM.
On the other hand Noel neither you nor Solveig or the other IPCentralians have ever tried to explain how DRM could be compatible with the First Amendment, or to address the very real concerns that those who are concerned about the erosion of our basic freedoms have with the DMCA...
V
· 2 years ago
We're assuming that the entire market will be flooded with 256kbps songs from iTunes. As long as CD's are still for sale, there will be so many seeds that what iTunes does or does not do as a deterrent will not affect the availability of illegal downloads.
What it can do, however, is draw people to its store and away from P2P, which isn't entirely free if you try to put a price on viruses and spyware.
At best, the identifier might convince some quantity of users to uncheck the "share my music" box in their P2P client.
At worst, it will be used by the RIAA as evidence that not only the person hosting the songs is at fault, but so is everyone who's name is imbedded in those files. This isn't quite as easy as it sounds, because in order to get those names it would have to DOWNLOAD all the files. Many of those files would contain no information, some of them may be altered, which would require some effort to determine, and some would only provide evidence of one act of infringement. The RIAA tends to go after the high quantity offenders, so this all seems fairly impractical.
David McElroy
· 2 years ago
This is just a minor thing (and doesn't deal with the main arguments here), but I want to point out that Apple doesn't sell MP3s of any kind. The iTunes store sells Advanced Audio Coding (AAC) format files. AAC is the audio component of the MPEG-4 spec. Not all digital audio files are MP3s, contrary to common assummptions and usage. :-)
eric
· 2 years ago
Noel, I do see your theoretical point about DRM and DMCA working hand in glove and they need to be considered together. I'm sure that was the intent. However, in practice, we have DRM and we have DMCA right now, together. Are they deterring unauthorized copying? Clearly DRM can be easily cracked, from FairPlay to deCSS to HD-DVD. According to you, DMCA must work with DRM to deter. Why is DMCA not doing so, then? Is it possible for DMCA to do so, in any practical way? Well yes, it is marginally deterring some copying. Yet clearly it is not having the intended effect. On the zero to 100% spectrum, where are we? Closer to the zero end, I would judge.
Also, in the case at hand, the non-DRM AAC files (not MP3s, thank you David) are, well, non-DRM. So DMCA can't work with DRM, since there is none to begin with! I don't believe that stripping the identifying data from the AAC files -- in about three seconds someone will create a program to do this, if it doesn't already exist -- violates DMCA anyway, since there is no DRM. So how does your DRM + DMCA argument work here? That dog won't hunt.
All Comments
Here is your absolutist reasoning at work again Tim. It falls inline with how you argue that unless DRM or the DMCA stop *all* piracy, there is no point in them, and how you argue that unless patents have 100% correlation with innovation, that they are no good, and how you argue that if DMCA 1201 affects *any* reverse engineering that it stifles innovation.
A more nuanced approach would be to state that mistrust based DRM is valuable if it reduces piracy to levels where its economical for creators to invest in online distribution.
Techniques that can reduce the number of downloaders can be effective incrementally. Techniques to reduce the number of uploaders, however, are only effective at all if they reduce available copies of a file to zero.
You claim that there are no numbers between 0-100% (by claiming that DRM is not useful unless it entirely stops piracy), then argue I'm bad at math:)
But in any case, even if I did run around yelling 2+2=7, its better than ranting about some worthless revolution that never and will never happen.
In the case of the recent release of the keys to high definition DVD encryption, one person uploaded the hex number to the internet. Shortly afterward, it was on hundreds of thousands of web pages around the world, and I'm sure on p2p networks as well. The same would be true for a single copy of a new album. This has also happened. The last U2 album was stolen (one copy) prior to release and uploaded (once) to the internet, and from there spread worldwide.
Tim's statement isn't just an argument, it is what actually happens in real life. To deny it is to reject reality.
Tim did admit there would be a marginal effect. Fewer copies uploaded in the beginning might slow the spread of the file worldwide by a short amount of time. Obviously there is a finite effect. Finite and infinitesimal. If that's the basis of your objection, OK, but it is irrelevant in practical terms. Between 0% and 100% it is so close to zero, I don't understand the quibble.
For example, I can read a book in the US or I can read it in England. The content industry has imposed region codes so if I buy a DVD here, I can not watch it in England.
I can buy the book today and read it five years from now. The content industry is claiming that if I "miss" watching a program that I paid for, I can not use technology to save if for future viewing.
From my point of view, the pirates are the content producers who seize every means to restrict the rights of consumers.
I"m not sure why Tim does not consider DRM *and* the DMCA (or other copyright enforcement mechanisms), since neither alone is meant to deter piracy. They're supposed to work together. Rather Tim seems to focus on one at a time, such as here, and argues that DRM alone does not stop piracy.
Tim also analyzes one patent at a time and asks how it contributes to innovation rather than a dataset of patents, so I'm not surprised to see him take a similar approach with DRM.
The world is bigger than your obsession with Tim. You will be a more effective advocate for content ownders or whatever it is you're advocating for if you get over this obsession and focus on how you can protect your constituency's interests. Content owners could care less whether Noel thinks Tim has an inflated reputation (no disrespect intended, Tim!). They care about knowing what works to protect their interests. Picker's proposal won't work, no matter what direction Tim's career takes.
Noel:
If someone comes up with DRM that does NOT require special legal protection which tramples on the First Amendment (e.g., DMCA) I would be more neutral about DRM.
On the other hand Noel neither you nor Solveig or the other IPCentralians have ever tried to explain how DRM could be compatible with the First Amendment, or to address the very real concerns that those who are concerned about the erosion of our basic freedoms have with the DMCA...
What it can do, however, is draw people to its store and away from P2P, which isn't entirely free if you try to put a price on viruses and spyware.
At best, the identifier might convince some quantity of users to uncheck the "share my music" box in their P2P client.
At worst, it will be used by the RIAA as evidence that not only the person hosting the songs is at fault, but so is everyone who's name is imbedded in those files. This isn't quite as easy as it sounds, because in order to get those names it would have to DOWNLOAD all the files. Many of those files would contain no information, some of them may be altered, which would require some effort to determine, and some would only provide evidence of one act of infringement. The RIAA tends to go after the high quantity offenders, so this all seems fairly impractical.
Also, in the case at hand, the non-DRM AAC files (not MP3s, thank you David) are, well, non-DRM. So DMCA can't work with DRM, since there is none to begin with! I don't believe that stripping the identifying data from the AAC files -- in about three seconds someone will create a program to do this, if it doesn't already exist -- violates DMCA anyway, since there is no DRM. So how does your DRM + DMCA argument work here? That dog won't hunt.