From DBWiki

The following is a debate started on August 11th, 2004 between Bill Roselblatt of DRM Watch and the CEO/President of Digital Bazaar, Manu Sporny, about the state of DRM and its effects on the various digital content industries (music, movies, eBooks, etc). The debate revolves around peer-to-peer networks, online music stores, digital rights management technology and social and community-based solutions to digital piracy.

The original DRM Watch article can be found here: DRM Watch Article About Bitmunk

Readers unfamiliar with Bitmunk may want to read an introduction to the service: An Introduction to Bitmunk

Contents 1 Manu Sporny: P2P and DRM Debate Opener 2 Bill Rosenblatt: Piracy, Big Media and the DMCA 3 Manu Sporny: The Failure of DRM and Bitmunk's Solution 4 Bill Rosenblatt: Sharing and Copyright Infringement Woes 5 Manu Sporny: The Effectiveness of DRM 6 Bill Rosenblatt: The Problem with Fan-to-fan Selling 7 Manu Sporny: Bitmunk and the Use of a Sane DRM Scheme 8 Bill Rosenblatt: Final Points 8.1 Closing words from Manu Sporny

Manu Sporny: P2P and DRM Debate Opener

Hi Bill, thanks for covering us (Bitmunk) on your site - we have read through the story you did on DRM Watch and have noticed several errors.

Bill Rosenblatt wrote:

Only Digital Bazaar can translate Bitmunk's user ID scheme to the actual 
identity of a user.

Simply not true, any artist may download software to detect the owner of a particular file. They do not have to file a complaint with us. The owner detection software is free for anybody to use that is registered as an artist/label/publisher/creator on our network. There was a help item on this, did you miss it?:

How do I check to see if somebody is violating my copyright?

Bill Rosenblatt wrote:

But is Bitmunk truly copyright-respecting? ... our opinion would fall 
somewhere between "sort of" and "not really."

You are confusing the issue of copyright-respecting and copyright-enforcement. We are 100% copyright respecting, you absolutely cannot under any circumstances sell anything on Bitmunk that has not been cleared by the artist for sale. When the sale occurs, it happens under the artist's terms. We are respecting the copyright of the artists.

What we oppose is shoving restrictive DRM technology down the throats of consumers merely to ensure that a corporation maintains marketshare via file format and device lock-outs.

You are right in raising the point that our technology does not enforce copyright - we are fundamentally opposed to such systems (it has never existed in past technologies - it works against the artist, if you lock content away, it makes it difficult for word-of-mouth advertising to occur).

DRM technology isn't going to fix the problem of piracy. I would think that you of all people would understand this - every DRM system to date has been hacked and bypassed. What is the point of using DRM?

Piracy is a social problem, not a technological one. The most powerful thing about our network (which you failed to mention at all in your story) is that it brings the consumer into the content distribution cycle. They get a stake in artists and labels doing well - something that the big labels are very interested in doing if it helps their bottom line (which it does).

While we support the small labels and indies, we equally support larger, more established, labels. We're not shooting for just the indies - this is a distribution system for every sale-worthy piece of digital content that has and will be created.

I'd like a call the next time you consider running a story on us, as a story such as the one you posted about us is mis-leading, and completely misses the point of what we're trying to do. If you would like to discuss this further, my number is: 540-961-4469

Bill Rosenblatt: Piracy, Big Media and the DMCA

Manu Sporny wrote:

Simply not true, any artist may download software to detect the owner of a 
particular file. They do not have to file a complaint with us. The owner 
detection software is free for anybody to use that is registered as an 
artist/label/publisher/creator on our network. There was a help item on this, 
did you miss it?:

How do I know if somebody is violating my copyright?

Fair enough, I missed that and will correct the story. Ironically, I put the above line in the story as a way of mentioning that Bitmunk does not publish identities of downloaders, thereby preserving their privacy, as your literature also claims. The feature alluded to above belies that assertion, along with your other assertion elsewhere that Bitmunk does not track usage. OK, it doesn't directly, but it allows others - unknown to the users in question - to do so.

Bill Rosenblatt wrote:

But is Bitmunk truly copyright-respecting? ... our opinion would fall 
somewhere between "sort of" and "not really"
Manu Sporny wrote:

You are confusing the issue of copyright-respecting and copyright-enforcement.
We are 100% copyright respecting, you absolutely cannot under any 
circumstances sell anything on Bitmunk that has not been cleared by the 
artist for sale. When the sale occurs, it happens under the artist's terms. 
We are respecting the copyright of the artists.

First of all, you miss *my* point. The point I was making was that your scheme has been used before and rejected by major record labels (as well as film studios) for general distribution purposes. The type of fingerprinting/watermarking scheme you use simply does not scale as a way of catching infringers. It works OK if you know the identities of the 200 or 2000 users to whom you have distributed your content, but not beyond that. Again, this is the opinion of "big media," not necessarily that I agree with it. Big media wants to avoid calling a lawyer and prosecuting. They want to prosecute the tool makers instead, because it's more scalable. Congress likes this idea because it's a more efficient use of the nation's prosecutorial resources than going after individuals.

With a scheme like yours, a media company person will feel that there's effectively no difference between Bitmunk and the original Napster. The latter people also told everyone that they were "copyright respecting" because they left decisions about what files to put up on the network, and how to use them, up to users. You are trusting users to put up files to which they have the rights - I think that's your most serious hole. If you filtered out otherwise-copyrighted works using a fingerprint-based scheme like Wippit's (using the Gracenote technology and database), then that might be a different story. FYI, this is the direction that the DCIA and its membership is taking.

I happen to think that DMCA 1201 is a dumb law - a blunt instrument that gives copyright owners an unfair benefit of doubt, but it is representative of how RIAA/MPAA types think. They want to close off the possibility of any chain of events that leads to misuse of copyright. So, even if someone copies content or even circumvents copy protection for purposes that end up being classified as fair use, once those copies are made, then they could be used for infringing purposes. That's the rationale behind DMCA 1201 and behind the opposition to Boucher's bill. I'm mentioning this to give you an idea of why the RIAA/MPAA would not be interested... or let's even say "interested in a negative way"... in a scheme like yours. I am just conveying attitudes I have observed, I am absolutely not making any threats or insinuations, nor am I about to call the RIAA and tell them anything, nor am I a lawyer.

Next point: your assertion that every DRM scheme has been hacked is just wrong. The 2-3 latest versions of Microsoft Windows Media DRM, which have the largest installed base among PCs and portable devices by a couple of orders of magnitude, have not been hacked.

Look, I've been through these discussions with everyone across the spectrum. You owe it to yourself to stop parrotting this tired argument, which not even that many of the Copy Left believe anymore these days. Anyone who studies cryptography, or reads one of Bruce Schneier's more recent books, will know that it's not so much about whether a scheme is hackproof or not as it is about what happens once the scheme is hacked - what are the effects. Jon Johansen hacked FairPlay; have the record labels suspended their licenses to Apple until Apple does a field upgrade of its installed base? I don't think so. Dmitry Sklyarov hacked the Adobe eBook reader - the major trade book publishers are still publishing ebooks in Adobe format. DRM is a deterrent, not an absolute preventative. Is it awkward technology? Yes, but getting less awkward. Is it being used stupidly? Sure. Can it be used in a more consumer-friendly way? Yes - see the most recent MusicMatch announcement for evidence. Can it be used to support a scheme in which consumers can redistribute content and get paid? Yes - Weed, FileCash, and another service that I helped with recently that should launch later this year. (Not that I think this is a great idea, BTW, but that's another story for another time.)

I also don't buy the "piracy is a social problem, not a technical one." Yes, it is a technical one - technology has made it easy to pirate. If it weren't easy, less people would do it. Similarly, if taking illegal drugs were easier, more people would do it. The issue of whether infringing copyright is as bad for you or bad for society as taking drugs, or engaging a prostitute, or speeding, etc., is a separate question - let's agree to not go there.

I happen to think that one of the positive things about the P2P providers is that they provide enough ballast in the market to keep the legitimate players from exerting too much control. I see what you guys are doing as another data point in the continuum from Kazaa at one end to the crappier download services (e.g. MusicNow or Wal-Mart) on the other. The line will continue to get populated with data points. I'm a technologist who believes that the market should decide what happens, even if in this case it leads to the dismantling of the media industry as we know it. I do hope that Congress isn't hoodwinked into passing a law that shuts P2P networks down just because they are P2P networks.

Manu Sporny wrote:

I'd like a call the next time you consider running a story on us, as a story 
such as the one you posted about us is mis-leading, and completely misses the 
point of what we're trying to do.

Please understand that I hear about new schemes like yours practically every week, and I've heard all the philosophies from Fred von Lohmann and Mike Godwin all the way to record company execs. Yours is one of the few of these new services that has actually made it as far as a public beta, which is why I wrote about it. Congratulations on being one of the 5% that actually get that far. I think of Bitmunk as an interesting experiment, and I have given my opinion on how the media industry is likely to view it.

Manu Sporny: The Failure of DRM and Bitmunk's Solution

Bill, wonderful to hear back from you - thank you for your well thought out response... it is obvious to me that you know your stuff. It was equally nice to read an e-mail from somebody that seems to be as passionate and knowledgeable about this area as we are. You make some good points... I agree with about half of what you had to say, however, I belive some of your points fall flat (due to false assumptions).

DRM, as we know it today, will fail... it is just a matter of time. I don't know if we agree on that, but that is our position. DRM paints everyone as a criminal, it's central tenet is that without restriction - everybody is going to pirate copyrighted material. That is simply not so - there are plenty of examples to the contrary: Newsprint, eBooks, DVDs, television.

We have gone through the entire DRM thought process from the artist, publisher/label, and consumer's perspective. Our solution is the proper balance, and the conclusion that will be reached by big media in one to three years time, and here is why:

DRM music sold via a single corporation is an incentive for people to pirate and share cracked music files via P2P networks. Whenever corporations try to control people, there is an incentive to try and break free from that control - DRM is an incentive to break free. Why? Because people feel that they're helping others out by freeing music for play on car mp3 players, or their generic mp3 player or computer. Sure, there are assholes out there that will do it just because they're assholes, but these are the minority and we deal effectively with them on Bitmunk.

There is no incentive to crack watermarking technology (barring the technological challenge) - it is not restricting the buyer in any way. Are you going to complain that you have a receipt of sale (which is what our watermarking is) for something that you purchased? Furthermore, are you going to put what you purchased on a P2P network that doesn't reimburse you for sharing (Kazaa/eDonkey/etc), or one that does (Bitmunk)? One where it is legal to share with friends, or one where it isn't?

We're incentivizing everybody to play nicely with each other. We realize that the RIAA/MPAA wants to have full control over everything - but wanting something, and getting it are two very different things.

Please read on, as I have attempted to address all of your concerns.

Bill Rosenblatt wrote:

fair enough, I missed that and will correct the story. Ironically, I put the 
above line in the story as a way of mentioning that Bitmunk does not publish 
identities of downloaders, thereby preserving their privacy, as your 
literature also claims. The feature alluded to above belies that assertion, 
along with your other assertion elsewhere that Bitmunk does not track usage. 
OK, it doesn't directly, but it allows others - unknown to the users in 
question - to do so.

Still not quite correct, Bill... the only people that get identified on P2P networks are those that have distributed their Bimunk-watermarked files. In short, only the copyright violators are identifiable. If you buy a file on Bitmunk, you own it - don't let anybody have it (unless it is via Bitmunk-aware P2P network). If they sell their files via Bitmunk, their watermarked personal ID information is cleared whenever the file is sold to somebody else (the new owner's information is encoded into the file).

This scheme:

1. Does not track usage
2. Does not treat fans as criminals
3. Only punishes the copyright violators

Bill Rosenblatt wrote:

First of all, you miss *my* point. The point I was making was that your 
scheme has been used before and rejected by major record labels (as well as 
film studios) for general distribution purposes.

I did partially miss your point - however, your clarification sheds some light on your thought process (which, while logical, still doesn't take into account what our system truly does). You are right, watermarking has been tried and it has failed - but our system does not depend on watermarking to ensure copyright enforcement (which is the mistake everybody has made to this point). We are not banking on our watermarking technology to prop up our business model.

The combination of watermarking technology, a controlled copyright database, and reimbursing the sellers as well as the artists and labels via a P2P distribution channel has not been tried before. I know at first glance it sounds like a mish-mash of technology and methods, but it is the proper combination to ensure that both the artists, labels, buyers and sellers play fairly with each other.

What incentive does a music fan have to buy something from Bitmunk and put it up on an illegal P2P network? No corner cases - give me a good motive that they have to cheat Bitmunk. If you can adequately answer that, I will concede defeat.

Bill Rosenblatt wrote:

You are trusting users to put up files to which they have the rights - I 
think that's your most serious hole. If you filtered out otherwise-copyrighted 
works using a fingerprint-based scheme like Wippit's (using the Gracenote 
technology and database), then that might be a different story. FYI, this is 
the direction that the DCIA and its membership is taking.

Ah! Here it is - you're assuming that we're trusting our users to put up copyrighted works for sale - this is not correct. The only creative works that we authorize for sale through the network are the ones that have been cleared by the artists. Here is how it works:

1. An artist comes to our site and registers themselves.
2. They must then fax identification information in for themselves and sign a document stating that they own the copyrighted works that they are about to claim. We have their drivers license, signature, address and govt. issued ID card. We also check their claim against the US Copyright Database.
3. They set the royalties for and upload a "pristine MP3" of their creative work.
4. A seller sees that a creator has cleared a work for sale on Bitmunk, and associates their MP3 file (a performance) with the creative work (the description of the performance - for example: The Beatles, Penny Lane).
5. A buyer does a search for "The Beatles, Penny Lane" and finds the seller, the buyer then purchases the work from the seller, the artist is paid, the seller is paid and everybody walks away happy.

Now take the case of where the seller tries to rip off the buyer: they associate a Metallica song with "The Beatles, Penny Lane". Every file sale on the network is based on a transaction contract - an agreement between the buyer and seller that a fair exchange is going to happen. The transaction contract (which is stored with us and digitally signed by both parties) states that the buyer wanted to buy "The Beatles, Penny Lane", but instead got Metallica. They give us the watermarked MP3 file as proof - so now we know that the seller associated the wrong file with the creative work on their sales server.

Whether this was done maliciously or not, the seller is fined and we give the buyer a free download from our internal sales servers for "The Beatles, Penny Lane". If the seller keeps doing this, they are ejected from Bitmunk.

So, you see... we are very careful that legal file sales happen on the network. If you paid for something, and you got something that wasn't what you paid for - you would complain about it, wouldn't you?

Bill Rosenblatt wrote:

I happen to think that DMCA 1201 is a dumb law - a blunt instrument that 
gives copyright owners an unfair benefit of doubt, but it is representative 
of how RIAA/MPAA types think.

Absolutely - I agree 100% with you on this. We are not fooling ourselves into thinking that the RIAA is going to welcome this new technology with open arms - but it will help the music industry in the long run. The MPAA/RIAA might hate it in the beginning, but if consumers/artists and fans are on board - in the long run, it will be good for everybody.

Bill Rosenblatt wrote:

Next point: your assertion that every DRM scheme has been hacked is just 
wrong. The 2-3 latest versions of Microsoft Windows Media DRM, which have the 
largest installed base among PCs and portable devices by a couple of orders of 
magnitude, have not been hacked.

You owe it to yourself to stop parrotting this tired argument. Is it awkward 
technology? Yes, but getting less awkward. Is it being used stupidly? Sure. 
Can it be used in a more consumer-friendly way?

The key point being "2-3 latest versions". Give it time, it'll be cracked just like all the other DRM schemes. I am not blindly "parroting this tired argument". I am very familiar with Schneier's work - we are no strangers to cryptography, steganography, nor security theory for that matter.

Our argument isn't "DRM is stupid 'cause it can be cracked", rather it is "DRM will look like our watermarking technology in two to three years time because that is ultimately the balancing point that consumers, artists and publishers are looking for."

Bill Rosenblatt wrote:

A Big Media representative would object that such a channel is not adequate; 
the whole idea of encryption-based DRM technology is to proactively inhibit 
consumers' ability to copy and send content anywhere at virtually no cost.

But how successful is DRM at doing this, Bill? There is no way for you, nor a Big Media representative, to know because the DRM has been stripped long before the song makes it onto a P2P network. DRM doesn't solve the problem, it just does a nice smoke and mirrors job to provide a false sense of security.

Our argument to the RIAA is: "Stop beating the DRM horse, it's already dead and its not doing what you wanted it to do - its just frustrating the people that would respect copyrights regardless if the DRM was there or not."

Bill Rosenblatt wrote:

I also don't buy the "piracy is a social problem, not a technical one." Yes, 
it is a technical one - technology has made it easy to pirate.

I do admit that it is a technological problem as well - but it is not 100% technological, just as it is not 100% social. This problem is not that cut and dry. My point was that technology is not going to solve the problem - just continually deter it, and upset consumers in the process. The solution to this problem needs to be a combination of technology and social incentives... and if the social incentives fail, the financial incentives will kick in (human greed is a difficult beast to ignore).

Bill Rosenblatt wrote:

I'm a technologist who believes that the market should decide what happens

I can definitely agree with that sentiment.

Bill Rosenblatt wrote:

I do hope that Congress isn't hoodwinked into passing a law that shuts P2P 
networks down just because they are P2P networks.

Same here... I've written all my reps pleading with them to do the right thing - luckily Boucher is one of them.

Bill Rosenblatt wrote:

I think of Bitmunk as an interesting experiment, and I have given my opinion 
on how the media industry is likely to view it.

And I thank you for your opinion, I do see your logic, but wanted to clarify that some of it might be based on a mis-understanding of how our system works. We're being very responsible with how the system is created (which you really cannot say for most of the other P2P systems out there)... my objection to the story was that it only looked at the watermarking aspect of the system and no further - when it is the other parts of the system that make the whole thing work (you didn't mention a buyer's ability to resell at all in your article - which is a cornerstone of the system).

Would you be opposed to us posting our exchange (in debate format) and putting a link to DRM Watch on our site? All the content would be cleared with you before we would post it... you have raised some very good concerns that I feel we address with our system.

Bill Rosenblatt: Sharing and Copyright Infringement Woes

Manu,

Yes, you have my permission to post the thread on your site with a link to DRM Watch, as long as you link to the original article and don't reproduce it (I would have to clear rights to do this with Jupiter). However, I'm not done with this delicious argument yet... see below.

Manu Sporny wrote:

DRM, as we know it today, will fail... it is just a matter of time. I don't 
know if we agree on that, but that is our position. DRM paints everyone as a 
criminal, it's central tenet is that without restriction - everybody is going 
to pirate copyrighted material. That is simply not so - there are plenty of 
examples to the contrary: Newsprint, eBooks, DVDs, television.

What is your point? Every type of copyrighted material has been pirated since the term began to mean something in the context of copyright law. The book industry is actually more worried right now about physical book piracy (e.g. in China) than it is about ebook piracy... although that is likely because the ebook market is minuscule compared to the print book market.

Manu Sporny wrote:

DRM music sold via a single corporation is an incentive for people to pirate 
and share cracked music files via P2P networks. Whenever corporations try to 
control people, there is an incentive to try and break free from that control 
- DRM is an incentive to break free. Why? Because people feel that they're 
helping others out by freeing music for play on car mp3 players, or their 
generic mp3 player or computer.

I don't agree with that, beyond the fact that DRM is relatively novel technology and represents a juicy challenge to hackers. The fact that people "share" files (more on that term later) on P2P networks has nothing to do with whether they are DRM protected. That's a bad assumption. P2P "sharing" began when people uploaded music from CDs, which as you know have no copy protection (with rare and unfortunate exceptions).

Manu Sporny wrote:

Sure, there are assholes out there that will do it just because they're 
assholes, but these are the minority and we deal effectively with them on 
Bitmunk.

No, you don't, not really. See below.

Manu Sporny wrote:

There is no incentive to crack watermarking technology (barring the 
technological challenge) - it is not restricting the buyer in any way.

All you are saying is that no one cares about cracking watermarking because there's no need to crack it. You should replace "incentive" in your reasoning with "need," and see where you come out.

Manu Sporny wrote:

Are you going to complain that you have a receipt of sale (which is what our 
watermarking is) for something that you purchased? Furthermore, are you going 
to put what you purchased on a P2P network that doesn't reimburse you for 
sharing (Kazaa/eDonkey/etc), or one that does (Bitmunk)? One where it is 
legal to share with friends, or one where it isn't?

Let's deal with those two issues separately.

a) This idea that you get paid for sharing, which is embodied in Weed, FileCash, and one or two others. I find that there are two fundamental problems with this model. I will explain one of them here and touch on the other below. My position - and time may prove me wrong - is that this is one of those ideas that will appeal to "literally hundreds of people." I file this idea in the same column as those schemes we heard about a couple of years ago wherein people will want to pay to be able to create their own remixes of tunes. The idea will appeal to a small group of fanatics but not beyond that. Look at it this way: how many people sell records/books/videotapes to used record/book/video stores now? Not too many. You may well argue that this is because reselling physical goods is a hassle, and networks like Bitmunk will make it so easy that many more people will want to do it. Time will tell.

I believe that what you're talking about is a fundamental change in consumer behavior, which is very hard to effect. Take both an example and a counterexample from Apple. John Sculley failed to understand this point, which is why the Newton failed. Steve Jobs does understand it, which is why iTunes is a success - iTunes is nothing more than a record store that fulfills digitally, unlike most of the other models of the time (pressplay, MusicNet, etc.) which involved models that were unfamiliar to consumers.

b) Sharing with friends. That's a really tough one. Where does one draw the line between "sharing" and "copyright infringement"? Media companies will never, never, never trust consumers to draw that line in what they consider to be the right place. That is why watermarking without encryption will never catch on among media companies. Copyright law is based on making copies (unfortunately; this is not very practical in the digital age). "Sharing" involves making copies, unless you have a scheme wherein your copy doesn't play if you "lend" a copy to someone else. That puts sharing within the realm of copyright law and thereby causes defenses of the "I'm only making copies to share with my friends" ilk to rely on fair use law - which as I assume you know, is not proscribed in law but is a matter for a judge & jury to decide. There is no case law precedent establishing "sharing with friends" as fair use. (There isn't even precedent establishing "making a backup" as fair use.) You'd have better luck in Europe, where most countries at least have a provision in their copyright laws allowing "private copies" to be made for oneself or one's family members.

You'll argue that there is a significant distinction between sharing with a few friends and putting a file up on Kazaa. I'll certainly agree with that. A media company might agree in principle but will argue that both actions result in revenue losses.

Manu Sporny wrote:

Still not quite correct, Bill... the only people that get identified on P2P 
networks are those that have distributed their Bimunk-watermarked files. In 
short, only the copyright violators are identifiable. If you buy a file on 
Bitmunk, you own it - don't let anybody have it (unless it is via 
Bitmunk-aware P2P network). If they sell their files via Bitmunk, their 
watermarked personal ID information is cleared whenever the file is sold to 
somebody else (the new owner's information is encoded into the file).

Ah-hah... see how your logic feeds back on itself. You suggest that "sharing with a few friends" is OK and is not infringing. A user (downloader) may agree with you. A distributor of a file on Bitmunk may not feel the same way.

Consider this scenario: O is an owner. D downloads O's file and sends a copy to his friend F. O somehow finds the file on a CD that F has burned. O examines the file, reveals F's identity, and alleges infringement. D and F will feel that O has violated F's privacy. (That may be fine as a built-in check & balance in your design, even a smart deterrent to users being too loose with the definition of "sharing with friends" - and in fact Microsoft uses a similar type of deterrent in its ebook technology - but even in that case you shouldn't represent the system as privacy protecting.)

Now consider another scenario: O is a major record company. O finds a file with D's Bitmunk watermark on Kazaa. O goes after D for infringement. In this case, I think we agree that D has done a no-no. From the record company's point of view, all you have done is make it somewhat easier for them to catch D than is possible with technology that the record company already uses to catch infringers on P2P networks. It doesn't scale.

Manu Sporny wrote:

I did partially miss your point - however, your clarification sheds some light 
on your thought process (which, while logical, still doesn't take into account 
what our system truly does). You are right, watermarking has been tried and it
has failed - but our system does not depend on watermarking to ensure 
copyright enforcement (which is the mistake everybody has made to this point). 
We are not banking on our watermarking technology to prop up our business 
model.

I agree that mistakes have been made by people who thought that watermarking (alone) works as enforcement technology. Digimarc found this out to its cost a few years ago.

Manu Sporny wrote:

The combination of watermarking technology, a controlled copyright database, 
and reimbursing the sellers as well as the artists and labels via a P2P 
distribution channel has not been tried before. I know at first glance it 
sounds like a mish-mash of technology and methods, but it is the proper 
combination to ensure that both the artists, labels, buyers and sellers play 
fairly with each other.

I definitely agree that your particular combination of technologies has not been tried before - which is why I decided to write about it in DRM Watch. Again, you might be surprised (or maybe not) how many times I hear of schemes that people who haven't done their homework think are new but really aren't.

However, if you remove the "reimburse the sellers" component (which I think is not all that interesting; see above), you get something which *has* been tried before, by Activated Content among others. These companies have tried to get record companies to use its technology for B2C distribution, and they probably have been telling their VC investors that their end game is to do so, but it's not gonna happen.

Manu Sporny wrote:

What incentive does a music fan have to buy something from Bitmunk and put it 
up on an illegal P2P network? No corner cases - give me a good motive that 
they have to cheat Bitmunk. If you can adequately answer that, I will concede 
defeat.

The motive is the same as it is for anyone to rip a CD and put it up on an illegal P2P network. Let's say an artist puts a great song up on Bitmunk - Bitmunk is the only place where it's available. In that case, someone who is an (your term) asshole will put it up on Kazaa.

Manu Sporny wrote:

Ah! Here it is - you're assuming that we're trusting our users to put up 
copyrighted works for sale - this is not correct. The only creative works 
that we authorize for sale through the network are the ones that have been 
cleared by the artists. Here is how it works:

1. An artist comes to our site and registers themselves.
2. They must then fax identification information in for themselves and sign a 
   document stating that they own the copyrighted works that they are about to 
   claim. We have their drivers license, signature, address and govt. issued 
   ID card.

OK, I grant that requiring a fax is a deterrent - I missed that.

Manu Sporny wrote:

We also check their claim against the US Copyright Database.

The vast majority of copyrighted works are not in the US copyright database. There is no requirement to register copyright unless you want to sue someone for infringement. Are you saying that you require copyright registration as well? If so, then why not extract a fingerprint and build a fingerprint database - or perhaps better yet, license Gracenote's.

Manu Sporny wrote:

5. A buyer does a search for "The Beatles, Penny Lane" and finds the seller, 
the buyer then purchases the work from the seller, the artist is paid, the 
seller is paid and everybody walks away happy.

Tell me realistically how many scenarios there will be in which the seller and artist are different entities. This is the second problem I have with the seller-reimbursement model. The artist will be the seller, in 99.99% of the cases. If there are additional sellers, all that will do is create immediate price competition, drive margins down to zero, and remove sellers' incentives.

Manu Sporny wrote:

Now take the case of where the seller tries to rip off the buyer: they 
associate a Metallica song with "The Beatles, Penny Lane". Every file sale on 
the network is based on a transaction contract - an agreement between the 
buyer and seller that a fair exchange is going to happen. The transaction 
contract (which is stored with us and digitally signed by both parties) 
states that the buyer wanted to buy "The Beatles, Penny Lane", but instead 
got Metallica. They give us the watermarked MP3 file as proof - so now we 
know that the seller associated the wrong file with the creative work on 
their sales server.

Whether this was done maliciously or not, the seller is fined and we give the buyer a free download from our internal sales servers for "The Beatles, Penny Lane". If the seller keeps doing this, they are ejected from Bitmunk.

So, you see... we are very careful that legal file sales happen on the network. If you paid for something, and you got something that wasn't what you paid for - you would complain about it, wouldn't you?

I'm thinking more of the scenario where the song is "Metallica, Master of Puppets" and is called such on the network because someone misrepresented himself as the copyright owner - faxed you some false documents with the name "Ulrich, Lars" on them, let's say. Of course this is a pathological example, and I will admit that you have taken some careful precautions with this aspect of it.

Manu Sporny wrote:

The key point being "2-3 latest versions". Give it time, it'll be cracked just 
like all the other DRM schemes. I am not blindly "parroting this tired 
argument". I am very familiar with Schneier's work - we are no strangers to 
cryptography, steganography, nor security theory for that matter.

Our argument isn't "DRM is stupid 'cause it can be cracked", rather it is 
"DRM will look like our watermarking technology in two to three years time 
because that is ultimately the balancing point that consumers, artists and 
publishers are looking for."

Well, what I can tell you now is that I have heard record company execs say things like "There is only one correct answer to the question of DRM, and that is Microsoft." They say this to would-be new service providers who come to them to get licenses to use their material. I happen to believe that DRM technology is flexible enough, and market forces powerful enough, that DRM will prevail and encompass many (though perhaps not all) consumer-oriented attributes.

Manu Sporny wrote:

But how successful is DRM at doing this, Bill? There is no way for you, nor a 
Big Media representative, to know because the DRM has been stripped long 
before the song makes it onto a P2P network. DRM doesn't solve the problem, 
it just does a nice smoke and mirrors job to provide a false sense of security.

It makes piracy enough harder. Just like The Club makes car theft enough harder.

Manu Sporny wrote:

I do admit that it is a technological problem as well - but it is not 100% 
technological, just as it is not 100% social. This problem is not that cut 
and dry. My point was that technology is not going to solve the problem - 
just continually deter it, and upset consumers in the process. The solution 
to this problem needs to be a combination of technology and social 
incentives... and if the social incentives fail, the financial incentives 
will kick in (human greed is a difficult beast to ignore).

OK, I agree with this. I also agree that the media industry has some homework to do when it comes to providing social incentives. (Hell, they can't even do a good job of explaining to people why they should be interested in novel models like Rhapsody's paid subscription model, which I happen to like, but then I'm not a typical music listener.)

Manu Sporny: The Effectiveness of DRM

Hi Bill,

I apologize for the long delay in my reply - I have been out of town and incredibly busy with the technical side of the company. I also wanted to give myself a bit of time to consider your viewpoints.

These are the only points of contention as far as I can gather:

• Larger labels and artists will not be interested in Bitmunk because we do not DRM content.
• The "include the music fans as distributors" aspect won't appeal to music fans - it will only appeal to music fanatics.

I address both points below:

Bill Rosenblatt wrote:

What is your point? Every type of copyrighted material has been pirated since 
the term began to mean something in the context of copyright law.

My point is that everybody seems to be treating DRM as the magic bullet, when in reality it provides absolutely no security what-so-ever to the content creators. You have stated several times that our watermarking doesn't scale, I'm not contesting that - however, I fail to see how DRM does or any other technology does scale.

I do see the argument that you are trying to make:

"If we DRM everything, that takes care of most of the casual infringers - the people that would just take the file and put it up on a P2P network without really thinking about it."

However, if DRM fixes this problem - then why are the file trading networks still going strong? Have you checked out the P2P usage stats lately?

ITIC Study on P2P Networks (June 2004)

The combination of DRM and lawsuits have done nothing to deter this behavior over the long-term. The various online music stores have been out for over 12 months - and P2P usage has increased during the same time frame. So, what type of music fan is using the service? Is it the P2P user, or the music fan that wouldn't file-trade the music even if it wasn't DRMed. Don't get me wrong, there is some overlap between the two.

My point is: DRM isn't working - it is still too restrictive. DRM doesn't scale either. The solution doesn't lie in technologies that lock content down - the solution lies in bringing the music fan into the music distribution process: fans selling to other fans.

By counting on DRM to fix the problem, the content and technology industries have gone down a rat-hole that will take them a long time to get out of. They're missing the main problem here - are you?

Bill Rosenblatt wrote:
Manu Sporny wrote:

DRM music sold via a single corporation is an incentive for people to pirate 
and share cracked music files via P2P networks. Whenever corporations try to 
control people, there is an incentive to try and break free from that control 
- DRM is an incentive to break free. Why? Because people feel that they're 
helping others out by freeing music for play on car mp3 players, or their 
generic mp3 player or computer.

I don't agree with that, beyond the fact that DRM is relatively novel 
technology and represents a juicy challenge to hackers. The fact that people 
"share" files (more on that term later) on P2P networks has nothing to do 
with whether they are DRM protected. That's a bad assumption. P2P "sharing" 
began when people uploaded music from CDs, which as you know have no copy 
protection (with rare and unfortunate exceptions).

Do you think DRM is going to stop that sort of behavior? Do you think that the content industry is going to be able to lock down content for large periods of time?

Bill Rosenblatt wrote:
Manu Sporny wrote:

There is no incentive to crack watermarking technology (barring the 
technological challenge) - it is not restricting the buyer in any way.

All you are saying is that no one cares about cracking watermarking because 
there's no need to crack it. You should replace "incentive" in your reasoning 
with "need," and see where you come out.

I see your argument - watermarking doesn't prevent piracy. I'm not making the argument that it does prevent piracy by determined individuals. Neither does DRM or any single "file protection" technology.

A P2P network can detect our watermark (no licensing fee for that technology) and reject sharing of a file that has been bought via Bitmunk. Whether the P2P app companies add filters is another issue (I doubt they will unless their hand is forced).

Bill Rosenblatt wrote:

Let's deal with those two issues separately.

a) This idea that you get paid for sharing, which is embodied in Weed, 
FileCash, and one or two others. I find that there are two fundamental 
problems with this model. I will explain one of them here and touch on the 
other below. My position - and time may prove me wrong - is that this is one 
of those ideas that will appeal to "literally hundreds of people." I file 
this idea in the same column as those schemes we heard about a couple of years 
ago wherein people will want to pay to be able to create their own remixes of 
tunes. The idea will appeal to a small group of fanatics but not beyond that. 
Look at it this way: how many people sell records/books/videotapes to used 
record/book/video stores now? Not too many. You may well argue that this is 
because reselling physical goods is a hassle, and networks like Bitmunk will 
make it so easy that many more people will want to do it. Time will tell.

Yes, time will tell - we're banking on us making the process so effortless that it will be simple for somebody to sell digital work via Bitmunk. Here is the process as we envision it:

1. A music fan buys something off of Bitmunk.
2. A window pops up and asks: Do you want to re-sell this file on Bitmunk?

If the user clicks "yes", the file is automatically up for sale on Bitmunk. Easy as that.

Bill Rosenblatt wrote:

I believe that what you're talking about is a fundamental change in consumer 
behavior, which is very hard to effect.

Yes it is... forgive us for trying something new and bold :)

I'm not making the argument that this is just like every other record store out there - its not. Bitmunk is very unique from all the other "online digital content stores" - we bring the content fan into the distribution process. This results in:

• Lower prices for fans
• Greater profits for content creators
• Free marketing and press for content creators (word-of-mouth)
• A method of supporting fan/news/community websites

Yes, we could fail because consumers don't understand what we're trying to do, or they do not latch on to the idea. At this junction, however, launching another cookie-cutter online music store would be fatal. We believe this is the better solution - markets work out better when everybody feels as if they're getting good value out of participating in the market. That is what we are doing.

Bill Rosenblatt wrote:

b) Sharing with friends. That's a really tough one. Where does one draw the 
line between "sharing" and "copyright infringement"? Media companies will 
never, never, never trust consumers to draw that line in what they consider 
to be the right place. That is why watermarking without encryption will never 
catch on among media companies. Copyright law is based on making copies 
(unfortunately; this is not very practical in the digital age). "Sharing" 
involves making copies, unless you have a scheme wherein your copy doesn't 
play if you "lend" a copy to someone else. That puts sharing within the realm 
of copyright law and thereby causes defenses of the "I'm only making copies 
to share with my friends" ilk to rely on fair use law - which as I assume you 
know, is not proscribed in law but is a matter for a judge & jury to decide. 
There is no case law precedent establishing "sharing with friends" as fair 
use. (There isn't even precedent establishing "making a backup" as fair use.) 
You'd have better luck in Europe, where most countries at least have a 
provision in their copyright laws allowing "private copies" to be made for 
oneself or one's family members.

You'll argue that there is a significant distinction between sharing with a 
few friends and putting a file up on Kazaa. I'll certainly agree with that. A 
media company might agree in principle but will argue that both actions 
result in revenue losses.

Yes, all good points. It is a very difficult line to walk. There are some artists/labels that support limited "sharing" - while others are completely against it. One thing that I have come to realize in our conversations is that we should not be in the position to force a "no DRM" policy on our artists/content creators. It should be their decision - so we're backing down from a strict "NO DRM" policy. We will provide as much or as little protection as the artist asks for ranging from: none, to non-invasive watermarking, to Microsoft DRM.

Our official position on this is: DRM does nothing but harm the artist-fan relationship. Thus we are very anti-DRM. But, we will provide DRM capabilities for those artists that want to use them. Note that this was not our position when we first contacted you, but it seems as if the DRM marketing engine has convinced a vast number of artists and labels that DRM will protect them from everything.

We built our system to be as flexible as possible - the goal is to work with everybody - the problem is that everybody is out for themselves and finding the correct balance is difficult.

Bill Rosenblatt wrote:

Ah-hah... see how your logic feeds back on itself. You suggest that "sharing 
with a few friends" is OK and is not infringing. A user (downloader) may 
agree with you. A distributor of a file on Bitmunk may not feel the same way.

I didn't mean to suggest "sharing with a few friends" is OK and not infringing. We're well versed in copyright law - and we do not support any sort of "copy and keep" behavior. What we do support is somebody burning a CD of MP3s, giving it to their friend for a week so that they may listen and then their friend giving the CD back and buying some of the songs off of their friend's sales server.

Here is how we hope to tackle the "sharing" scenario:

You have a close friend F, who has an album you said you liked. Since that person is your friend, he says: "Don't worry about paying my fees, buddy - I'll give you a 100% discount on my sales server". So you can then buy the album for $5.00 instead of the regular going price of $8.00.

Regardless, the artist gets a sale and a full royalty out of that process. Yes, there will be people that copy the files to their computers and keep them there - we do not support that.

So, for buyers on our network that give their files to somebody else - they are fully aware that:

• Their personal information is now out in the open
• If their friend isn't careful with the file, they could get sued

Using DRM would prevent this very specific behavior - but at the same time, they would just go to a P2P network and download the file if there weren't an alternative - such as Bitmunk.

Bill Rosenblatt wrote:

Consider this scenario: O is an owner. D downloads O's file and sends a copy 
to his friend F. O somehow finds the file on a CD that F has burned. O 
examines the file, reveals F's identity, and alleges infringement. D and F 
will feel that O has violated F's privacy. (That may be fine as a built-in 
check & balance in your design, even a smart deterrent to users being too 
loose with the definition of "sharing with friends" - and in fact Microsoft 
uses a similar type of deterrent in its ebook technology - but even in that 
case you shouldn't represent the system as privacy protecting.)

That doesn't work as a valid argument - F should have been more careful with D's file. But really, copyright infringement occurred at the point when D copied the file for F. If you do that, you're willingly giving up any privacy protection that we offer.

O would allege infringement on D (not F) since D is the one with the ownership receipt embedded in the file.

That argument is like saying: "If D gives his credit card bill to F, and F photo-copies it and leaves it in a public cafeteria for anybody to find, that is an invasion of privacy and the credit card company did nothing to prevent that!"

See if your argument works for the iTunes store: "If D burns a CD for F, and F distributes that CD image to the web... then O has absolutely no way of finding out where that CD image came from... not to mention that F doesn't feel bad at all for putting that CD image on the web".

Which is why our system 'does' scale for infringing copying. D would be willing to give a file to F... but F would know that D didn't want the file to go any further.

The industry needs to stop ignoring the fact that this goes on... or worse, thinking that DRM fixes this problem.

Bill Rosenblatt wrote:

Now consider another scenario: O is a major record company. O finds a file 
with D's Bitmunk watermark on Kazaa. O goes after D for infringement. In this 
case, I think we agree that D has done a no-no. From the record company's 
point of view, all you have done is make it somewhat easier for them to catch 
D than is possible with technology that the record company already uses to 
catch infringers on P2P networks. It doesn't scale.

Neither does DRM. There is no magic technological bullet to solve the P2P piracy issue. Stop trying to sell DRM as that.

Bill Rosenblatt wrote:

However, if you remove the "reimburse the sellers" component (which I think 
is not all that interesting; see above), you get something which *has* been 
tried before, by Activated Content among others.

And we both know how that went. Thus leaving us at this point: fan-to-fan selling is part of the solution to this whole "piracy epidemic" thing. I think we just have a difference of opinion at this stage - we think that this is a -very- attractive selling point of Bitmunk. I go into this in much more detail below (3 questions down).

Bill Rosenblatt wrote:

The motive is the same as it is for anyone to rip a CD and put it up on an 
illegal P2P network. Let's say an artist puts a great song up on Bitmunk - 
Bitmunk is the only place where it's available. In that case, someone who is 
an (your term) asshole will put it up on Kazaa.

We will never do an exclusive deal - its bad for the artist and completely self-serving for any online music service. You're taking away consumer choice when you do that...

Bill Rosenblatt wrote:

The vast majority of copyrighted works are not in the US copyright database. 
There is no requirement to register copyright unless you want to sue someone 
for infringement. Are you saying that you require copyright registration as 
well?

For the large bands and labels we do require copyright registration... for the smaller artists we don't - we do require a fax with all identifying information, however (which we do a background check on). If they claim a work that they don't legally own - they have just committed a crime - and we can identify them. It is no different than the forms that artists/labels fill out when distributing their work via any other online music service. Bill Rosenblatt wrote:

If so, then why not extract a fingerprint and build a fingerprint database - or perhaps better yet, license Gracenote's.

We are looking into licensing Gracenote's database to offer a simpler experience for our creators and customers.

Bill Rosenblatt wrote:

Tell me realistically how many scenarios there will be in which the seller 
and artist are different entities. This is the second problem I have with 
the seller-reimbursement model. The artist will be the seller, in 99.99% of 
the cases. If there are additional sellers, all that will do is create 
immediate price competition, drive margins down to zero, and remove sellers' 
incentives. There are other things I could say here but unfortunately I'm 
under NDA with a company that may compete with you (which I consulted to up 
until a couple of months ago) so I shouldn't go any further.

There will be tons of scenarios where the seller and the artist are different entities. Here are a couple:

• Internet Radio Stations
• Music review/news websites
• Movie preview websites
• Movie review/news websites
• Palm-top software review/news websites
• Fan websites
• Bloggers

You're assuming that we're using very simplistic economic market models. Our software ensures that everybody gets a chance to sell something. If everybody lists a song for $0.80 - then the sales servers will be listed in a round-robin order - everybody gets a chance to sell if the prices reach commodity levels.

Sellers are only allowed to perform a fixed number of transactions at any given time - which means that if you're selling for cheap, your sales server will become saturated with transactions and you will do less business than if you raised your prices a bit.

All of this is transparent to the buyers and sellers, but ensures that everybody gets a chance to buy and sell for the best prices on Bitmunk.

The fans will always be better at distributing content than the artists. Lets face it - the artists have more important things to worry about.

Bill Rosenblatt wrote:

I'm thinking more of the scenario where the song is "Metallica, Master of 
Puppets" and is called such on the network because someone misrepresented 
himself as the copyright owner - faxed you some false documents with the name 
"Ulrich, Lars" on them, let's say. Of course this is a pathological example, 
and I will admit that you have taken some careful precautions with this 
aspect of it.

Yes, pathological indeed. :)

Bill Rosenblatt wrote:

Well, what I can tell you now is that I have heard record company execs say 
things like "There is only one correct answer to the question of DRM, and 
that is Microsoft." They say this to would-be new service providers who come 
to them to get licenses to use their material. I happen to believe that DRM 
technology is flexible enough, and market forces powerful enough, that DRM 
will prevail and encompass many (though perhaps not all) consumer-oriented 
attributes.

That is not the impression we were left with after speaking with one of the big four. Infact, DRM didn't really come up in the conversation at all after we explained our watermarking and social incentives method.

The music industry realizes how much fans selling to other fans would help their business - its grass-roots advertising, which has been the main driver for discovering and promoting new talent since the first musical work was performed.

I'm not saying that DRM has absolutely no place in the future - but its going to have to go through quite a few changes to win the hearts and minds of music and movie fans.

DVD did an excellent job of this in its first several years because it gave the consumer better value and people tended to not care too much that the content was locked down because they could do almost everything that they wanted to with the DVD. If the equivalent came up for purely digital content, we would support it as long as it didn't strip away the fair use rights of the consumer.

Bill Rosenblatt wrote:

It makes piracy enough harder. Just like The Club makes car theft enough 
harder.

No it doesn't. If I wanted to, I could download any song I wanted to right now off of a P2P network. DRM isn't stopping that.

Bill Rosenblatt wrote:

OK, I agree with this. I also agree that the media industry has some homework 
to do when it comes to providing social incentives. (Hell, they can't even do 
a good job of explaining to people why they should be interested in novel 
models like Rhapsody's paid subscription model, which I happen to like, but 
then I'm not a typical music listener.)

Our position is that the proper social incentive is going to be making the fans a part of the process. Fan-to-fan selling... there are many people in the content industry that support this business model (although, I'm not at liberty to say who we've been talking to in the industry that supports this - they're one of the big 4).

This e-mail is getting out of hand - I apologize for that, but you're one of the few people that we've had an exchange with that actually knows a decent bit about how these industries operate... and one of the only ones to last past the first e-mail. Did this e-mail help clarify our position with regards to your perception of Bitmunk?

My major points of contention with you are:

• DRM is years away from being usable by both content creators and consumers - and in the mean time it is damaging all content industries and making consumers out to be criminals.
• Fan-to-fan selling has much more value in it that you think it does. Many people trust their friends far more than they trust a corporation - even one with a good brand name.

I think the rest of the issues can be attributed to your unfamiliarity with our service (which we are partly to blame - our website needs to explain these things with a bit more clarity). It is a complex problem, and it is a combination of technologies and incentives that is going to make it work. We feel that we've gotten most of these things down, but our system is flexible enough so that the bits that we have missed can be added in as they are needed.

Tag, you're it. :)

Bill Rosenblatt: The Problem with Fan-to-fan Selling

So many points, so little time. Here are a couple of main ones:

You single-minded, fanatical abhorrence of DRM leads you to the totally erroneous conclusion that the music industry is single-minded in thinking that it is the magic bullet that will solve all of its problems. That is just not the case. The industry is using "all means necessary", including other technological means such as spoofing, in addition to legal means. At the same time, I just don't think that they will look at a technology like yours and feel comfortable with it. BTW, look at my recent story on MusicMatch for an example of how DRM can be made to support "sharing". It's a step towards better DRM, which I agree needs to happen. DRM is a set of capabilities. It can be implemented stupidly or smartly.

Your examples of fans-selling-to-fans are unconvincing, as is your economic explanation.

a) In most of the cases you present (fan sites, internet radio stations, etc.), they would find an affiliate network such as Amazon's or the one that Apple just announced with LinkShare to be adequate - no muss, no fuss, but they get paid. Many web radio sites use Amazon already. The only thing that needs to happen is Amazon needs to fulfill digitally, and that will happen very soon. The alternative is to keep complete control of the user experience, which Bitmunk does not let you do. I know this from my own experience with higher ed content. We chose to keep the user experience on our site and therefore go the expensive route of building our own ecommerce capability. Plus, I reiterate that there is no historical precedent for fans selling to other fans except in extreme geekazoid cases that do not scale and/or do not involve major-label content.

b) as for the transaction limits, round-robin algorithm if several members sell a track at the same price, etc., who cares. The prices still get driven down in the same way. Plus, if you limit the number of transactions, you are limiting incentives to participate, which (once again!) limits scalability. Notice that the folks at eBay do exactly the opposite by rewarding high-volume participants in various ways.

Manu Sporny: Bitmunk and the Use of a Sane DRM Scheme

Bill Rosenblatt wrote:

So many points, so little time. Here are a couple of main ones:

Your single-minded, fanatical abhorrence of DRM leads you to the totally 
erroneous conclusion that the music industry is single-minded in thinking 
that it is the magic bullet that will solve all of its problems.

Thats a strange thing to say, Bill. In the last e-mail I sent you, I stated that we will be supporting DRM (but only if the artist wants it - more than likely it will be Microsoft's DRM). Does that sound like the action of some single-minded, anti-DRM fanatic? The point I was making is that *CURRENT* DRM (that is, DRM in use today) doesn't work and only punishes the music fan that is willing to put some hard-earned cash down for digital music.

Perhaps we need a better definition of DRM: When I say DRM, I mean "DRM that is currently in use today and for the next two years - a closed algorithm that aggressively locks content down.". Maybe digital watermarking is too "loose" for your definition of DRM, but there are many artists that don't think so... which is why we decided to make watermarked MP3s our first cut at a "sane DRM" standard.

If there was some sort of DRM-standard, such as CSS (and we all know how well that worked) - we would support it (again, only if the artist wanted their work to be DRMed). I have a feeling that our "solution" to the DRM mess is more similar to yours than you think. We want a fair, open standard... don't you?

Besides, you dodged my question: I'd like to hear your opinion as to why file-sharing networks are doing so well. If DRM, spoofing, suing customers, and writing new laws are so effective - why is file-sharing at an all-time high? This stuff isn't going to take off unless the content industries focus on why people are still file-sharing. Don't say "Because it's free and easy" - that's a cop-out that grossly simplifies the market forces at work here. Its equivalent to saying "Because music fans are criminals"... which couldn't be further from the truth.

Bill Rosenblatt wrote:

I just don't think that they will look at a technology like yours and feel 
comfortable with it.

That's fine, I agree that we will have to show the industry the benefits of not being as extreme as they are being in their DRM philosophies. The content industries won't feel comfortable with -anything- that deviates from the traditional model. The traditional models (as well as copyright law) are based off of the difficulty of copying a work - it just doesn't work well with technology today. Bill Rosenblatt wrote:

BTW, look at my recent story on MusicMatch for an example of how DRM can be made to support "sharing". It's a step towards better DRM, which I agree needs to happen. DRM is a set of capabilities. It can be implemented stupidly or smartly.

Yes, agreed. Don't paint us as some DRM-hating zealots, we're not. We're scientists at heart, when we see something not work - we don't try and cover it up - We analyze what is wrong... we look for different solutions.

DRM, as it is implemented right now, is a horrible thing (much like many technologies when they first appear)... Our position is that DRM needs to be much lighter/friendlier than what is currently available (like CSS on DVDs)... or like the watermarking technology that we have introduced.

Bill Rosenblatt wrote:

a) In most of the cases you present (fan sites, internet radio stations, 
etc.), they would find an affiliate network such as Amazon's or the one that 
Apple just announced with LinkShare to be adequate - no muss, no fuss, but 
they get paid.

But they don't get paid enough... we want to enable them to make more money. 5% of the sale is pretty paltry (which is what most of those services (Amazon/LinkShare) provide (on average)). The sellers/affiliates don't have any control over the money they make... I've witnessed the amount that large Internet Radio sites take down on a monthly basis - its barely enough to cover bandwidth costs.

Bill Rosenblatt wrote:

Many web radio sites use Amazon already.

And they complain about how little they make from those affiliate links... but, its better than nothing (this is based on talking to over 30 of the top Internet Radio station owners over the past 2-3 years). If they could make more money by fulfilling their catalog's digitally, they will prefer that service. We're offering between 15%-20% of the total sales price to go to the seller. That is three times more than what they can make with Amazon, Apple, et al.

Bill Rosenblatt wrote:

The alternative is to keep complete control of the user experience, which 
Bitmunk does not let you do.

I really wish you would stop jumping to these conclusions. Where did you get that from? We operate as both a front-end search service for digital media/sales processing AND a back-end content delivery/sales processing system for websites/Internet radio stations/etc.

We will support web-services-based searching/browsing/etc. This means that they can provide a complete, controlled user experience with Bitmunk features. The buyer will not need to leave the music website or Internet Radio stations to perform a purchase of any music from that site. The site that uses our software gets to control the entire user-experience, the software costs them nothing up-front and it pays through three times better than any service out there.

Hope that clears things up - we want to help small and large websites alike by providing a complete, non-invasive, content delivery/sales system.

Bill Rosenblatt wrote:

Plus, I reiterate that there is no historical precedent for fans selling to 
other fans except in extreme geekazoid cases that do not scale and/or do not 
involve major-label content.

Ugh, what an unimaginative argument. No historical precedent is hardly a reason to discount a solution. The Internet used to be a geekazoid case that didn't scale and didn't involve any major corporations - most technology that we take for granted today started out like this.

Bill Rosenblatt wrote:

b) as for the transaction limits, round-robin algorithm if several members 
sell a track at the same price, etc., who cares. The prices still get driven 
down in the same way. Plus, if you limit the number of transactions, you are 
limiting incentives to participate, which (once again!) limits scalability. 
Notice that the folks at eBay do exactly the opposite by rewarding 
high-volume participants in various ways.

So, you're making the argument that driving down prices is bad? That good music, at very competitive prices, where the artist gets paid what they want on their terms, and support from their fan-base is a bad thing? I'm confused Bill, whats the problem there?

Sure, the sellers will have to differentiate themselves or provide some incentives with popular digital media - but, thats what people have had to do since the dawn of open markets.

There is nothing to keep a website that is using Bitmunk services from showing the content (on their website) that they are selling off of their Sales Server. If the visitors of that site see a good value-add, they will buy off of that Sales Server/website instead of searching for a low-priced version (to save a couple of pennies) on the Bitmunk website.

Thanks for the debate, Bill... I'll let you know when we get this stuff up on our website. I think we've clarified quite a number of things during these exchanges. Time will let us know which arguments are valid and which ones are junk.

Bill Rosenblatt: Final Points

Additionally, I have a couple of final points.

First: It's nice to say that you're going to offer 15-20% commission, but the way the economics work out, you're going to end up with product that is available more cheaply elsewhere. I don't believe this model works. Part of the beauty of the Internet is that it cuts out the middleman. The current legitimate services make no margins to speak of at 99 cents a track (to say nothing of Wal-Mart's 88 cents a track).

You'd have to convince me that your economic model leaves enough room for the middleman's 15-20% commission and still offers product at prices that are competitive with all of the other services that are beating each other silly over pricing. Once again, you are left with material that is rare or only of interest to fanatics (i.e., the stuff that isn't going to be for sale at 99 cents on iTunes or Napster), for which you can charge enough and still leave room for your 15-20% commission.

Second: regarding why the P2P networks are so successful in terms of traffic, you say: "Don't say 'Because it's free and easy' - that's a cop-out that grossly simplifies the market forces at work here. Its equivalent to saying 'Because music fans are criminals'... which couldn't be further from the truth."

No, those statements are not equivalent, any more than "You're against Bush" is equivalent to "You're supporting the terrorists" are equivalent. Come off it, please. Of course 'Because it's free and easy' is the right answer (although 'easy' is debatable nowadays with all of the spoofs and poor quality content on the P2P networks). Are people who drive 59 mph on the freeway criminals? They do so because (in most cases) they believe they can get away with it, and they do not think of the potentially harmful effects that are the reasons for the 55 mph limit, such as the increased chance of accidents. Similarly, people download from P2P networks because it's free, they think they can get away with it, and if pressed to even think about it at all, they would say that it doesn't really hurt anyone anyway.

Once again, you are spouting stale ideology from a couple of years ago, and you owe it to yourself to stop. I wish I had Siva Vaidhyanathan's new book in front of me, because it contains a great quote about how everyone thought that people were going to use the P2P networks for "sharing" but that has turned out to almost entirely *not* be the case. Although it is true that media industry behavior suggests treatment of consumers like criminals, and it is probably true that some media industry execs think that way, don't confuse their attitudes with reality. Getting something for free instead of paying for something is just basic human nature.

Closing words from Manu Sporny

And on that note, we're going to end our discussion about the current state of P2P and DRM regarding digital media. I'd like to thank Bill Rosenblatt for writing about us as well as his insight throughout this debate. His website is DRM Watch - analyzing some of the most pressing issues at the intersection of media, digital content and technology.

It is our belief that honest debate and discussion will help push the digital content industries forward to better serve the artist, publisher, distributor and fan. We are always open to change, as long as it is fair for everyone involved and we feel that this debate was a good example of that. If you would like to contribute to this debate, please add your comments on the Talk section of this page.