� posted by Dan Gillmor 06:21 AM
� permanent link to this item
I'm heading this morning to the Digital Rights Management Conference at the University of California-Berkeley. They've put together a sterling crew of folks to talk about a topic you're going to be dealing with more and more in coming years -- restrictions on copyrighted material that flout your rights and wound the public domain. (Running notes here.)
Other notes from the conference include:
Bryan Alexander Aaron Swarz Kevin Marks Mary Hodder
It's about money and control, says Carl Shapiro, a U-Cal economist. Not about fair use? Not about customers' rights? Not about the public interest? I guess not.
Lon Sobel, a lawyer and editor, says ISPs would pay royalties to copyright holders -- he uses the word "owners," demonstrating his position on this matter -- at rates set by copyright holders themselves. Everything would have watermarks. Says he's been "assured" by DRM companies they can put watermarks/fingerprints on everything out there. (He's been assured of something that can't really be done unless we ban, say, encryption -- but never mind that.)
He says this is great for technology and copyright holders. He expresses basic indifference to the fact that there's not a hint of maintaining privacy of what we read, watch and listen to in the world he describes. Just an incremental privacy invasion, he says with amazing inaccuracy.
His plan offers markups to ISPs if they choose. Given that in a broadband world there will be just a couple of ISPs -- namely the cable and phone companies, given the way the government is giving them absolute rights to control content on their pipes -- do you suppose your friendly cable and phone monopoly will not choose to charge you substantial extra money?
Intel's Don Whiteside works for a company that claims to be on the customers' side, but which is creating tools that enable copyright owners to exert the control they want -- a company with fundamental internal conflicts, in other words.
We should recognize this isn't just about business models, but a challenge on larger fronts. Disruptive technologies have created an inflection point, forcing a choice on which path to take.
Small, nimble companies will figure out the new business model, he says. But incumbents should be the ones moving first.
Technology should focus on "fringes" where there's infringement, such as redistribution, he says, but not on basic customer rights for personal uses. He's bragging about Intel's participation in cross-industry forums where our rights are being decided -- without any customers at the table -- by technology and entertainment companies. Gosh, thanks.
Cary Sherman of the Recording Industry Association of America (servers seem to be under attack again) admits "we can't just stop selling CDs." There are new formats coming along, he says, but the industry is not in agreement on ways to "lure consumers back into record stores."
Winning court cases and litigation doesn't solve the problem, he says. Burned CDs are adding to street piracy, he says. Now it becomes a cottage industry. Downloading and burning is "a real substitute for purchasing," he says. The "web of rights holders" makes it hard for the record companies to go online, he accurately notes.
Consumers should be able to make personal copies but shouldn't be able to put songs on the Internet or give copies to other people, he says. (The fact that you can't have one without the other has escaped him.)
The industry is making progress, however, he says. Meanwhile the industry is building an infrastructure for online delivery, and working on copy-protecting music CDs while making it possible to enable personal copies. Such technologies aren't good enough, he says.
ISPs can be "incentivized" to help, he says. Education and enforcement, natch.
Sarah Deutsch of Verizon wants to offer legitimate content in a cooperative way.
Verizon is fighting off the recording industry's attempts to get ISP users' personal records. Deutsch says the RIAA not only wants the customer's identity but also to disable access to the user's hard drive. She calls this a gross invasion, based only on an "assertion that someone, somewhere, is engaged in copyright infringement."
A mere assertion, without evidence, would require turning over such information to anyone -- including stalkers, for example -- in what amounts to "private search warrants." If RIAA wins, the case will have a "huge chilling effect" on privacy, she says, adding that due process is a better approach.
(Sherman says this isn't really a big deal. The court decision, he says, shows Verizon's reading of the statute is tortured. And he says the RIAA and others are forced to show that there's a good-faith case, etc.)
She likens the war on peer to peer to prohibition of alcohol in the early 20th century. Result: Millions of people become criminals. Compete on doing the right thing to customers, she says.
IBM's Bob Blakley asks what happens if copyright holders get what they want. He asks -- apart from privacy questions -- if DRM ends up discouraging resale. Right resale value encourages high initial prices.
DRM can raise inventory costs, he says. Example: controls by country. Mass customization isn't perfect, either, he says. Raises costs and can actually encourage piracy. It can add life-cycle costs when things fail, because users will call.
Privacy liability is another issue, apart from invading privacy (which he assumes copyright holders don't care about). Now you have all that information, and liability for protecting it.
Does it create niches for competitors? Maybe people will find other things to trade, not yours. There may also be a revenue curve where all rights have to be effectively given away.
David Reed from the cable-TV industry's technology trade association says his group has a strategy for covering existing business models, but future models is "a tough one." There has to be inter-industry cooperation, he says.
Battleground has been drawn on various specifications in TVs of the future. Big picture: Info comes to set-top box, which has decrypter that sends info to devices. More restrictions on customer rights in the works, with encoding rules that restrict what we can do with material we buy.
The FCC will likely be ruling on new business models, he suggests. What could be worse?
Allan Adler, representing print publishers, notes that copyright doesn't work well in a "one size fits all" world.
Traditional publishing limits copyright infringement. E-publishing is a new environment.
DRM is the key to e-books, he says. But focus on security has "succeeded too well" by frustrating customers, he says. Publishers don't make DRM, having to rely on software and hardware industry. Vendors haven't been doing a very good job of it.
Business models derive from customer needs, he says. Publishers are learning that current DRM system fall short of customer needs. But there's improvement.
Will publishers and users ever get on same page? Maybe not, he says. People want to do with e-books what they do with real books for the most part, at least they think so until they see what augmented books can do.
Why should consumers enjoy added value without accepting some change in copyright rules to accomodate the risks, he asks. (Give up fair use in order to have a hyperlink to a dictionary? No, thanks...)
Let the market decide, he says, rather than government fiat.
Sobel's plan is being thoroughly derided by wiser heads.
--------------------------------------------------------------------------------
Second session is how DRM affects innovation, competition and security.
Microsoft's John Manferdelli, GM of Windows Trusted Platform Technologies, says DRM has many uses including protecting personal information. Enterprise rights management, he says, is about protecting corporate data, where fair rights doesn't come into the picture and privacy rights are reduced.
All systems have things in common, he claims -- and one is not to restrict the way people use computers. It's about enabling "new stuff," he says.
He says there are misconceptions, including: no wish to censor or disable content without user permission; no lockout of vendors or formats; total user control; no one owns machine key; it won't be perfect; Darknet.
The Darknet paper, he says, is being particularly misconstrued. It's not about the irrelevance of DRM, but recognizing the limitations.
When he talks about a negotiated equilibrium, he's claiming that individual customers have the power to counter what companies like Microsoft can assert. Sure we do.
Lucky Green from Cypherpunks is dismantling Manferdelli's warm and fuzzy comments. Trusted computing, he says, precisely aimed at stifling competition.
The big companies are conspiring to keep the public in the dark, he says.
What does "trust" mean in context of trusted computing? Not just your trust in the computer. But also that third parties can trust that your computer will disobey you (at their choice). Also that they can trust inability of third parties to compete.
Trusted computing, he says, can prevent all kinds of user control. It can prevent use of unlicensed software. It can make PC core of home entertainment. Also new opportunities for selling to government.
Quotes Bill Gates' comment that DRM is more interesting in realms of e-mail and other documents than music. Strict controls on how others can use what you may send them. "If you are the CEO of Enron, you love this because there's no evidence left."
Stifling competition: App vendors will wrap files with DRM. Q: What does federal government call a third-party application that is compatible with a proprietary DRM-wrapped file format? A: An illegal circumvention device. Huge penalties under DMCA.
Great line: "Using gasoline in a car is an opt-in technology."
Dave Farber says technical community has suffered from religious war without articulating what technology is capable of doing and what limitations are in technically valid way. Too much black and white.
Yes, there's a lack of perfect security, he says. But an acceptable level is more and more important in the current environment. Things that increase level of security are hard to turn down.
A relatively secure system makes it hard to keep out rights management, Farber says. If you have anything other than boot privileges it's getting hard not to host rights management. But if you say in law in computer environment that you can't implement rights management you may not be able to build secure systems, either.
He's less interested in protecting media companies, but very interested in not having people gain access to personal information. Protection mechanisms have an important role in protecting personal information -- who's looked at it, who's made copies, etc. Ultimately these things will be decided by market, legislation and courts.
The unwillingness of media companies to pay for good protection, he says, gives us Mickey Mouse solutions. A serious problem. Fair use -- there's no reason we can't have a rational balance that includes rights-management systems. Maybe we could have better fair use than we have; needs more research.
Berkeley's Hal Varian says DRM is just one of many business models, including advertising, bundling content with other things, subscriptions, low prices for authentic versions, micropayments. Pluses and minuses in each.
In DRM, seller should want to maximize value, not protection. More you give customer, more valuable customer finds the product. Pick the right tradeoff. DVD and libraries add customers, for example.
Crippleware reduces value, and there's competition it's hard to enforce. But DRM can cripple innovation. Sometimes you want customers to use products in innovative ways to find out new users.
Alex Alben of RealNetworks says the past does not provide sensible expectations. We need both copy protection and personal use for a functioning marketplace.
Problem isn't so much whether there's a perfect DRM. "We have a pretty good one," he says. But what's the price point for a 30-day download usable on a single computer?
To build market, everyone must give a little, he says. Allow mass distribution of content, personal use, transparent DRM, no regulation and new intellectual frameworks for making distribution work. Example: licensing models need work.
"Consumers have to use this in ways that are consistent with personal use," he says. File sharing to 1,000 people they don't know isn't acceptable, isn't justifiable.
Challenges: create a fair use exception that doesn't swallow the rule. Limit DMCA to protecting valuable media, not printer cartridges. Establish broadcast flag that doesn't create regulation of home networks.
Green: Uses to which DMCA has been put were not in Congress' mind at the time law was passed. But corporate interests did anticipate it. Doesn't matter what DRM sides believe, but what courts believe -- and courts continue to ratify DMCA in every case.
Farber: Let person who's going to accept stuff make policy on what they'll accept. Truth in advertising. Also, get rid of word "consumer" --- yes.
Pam Samuelson asks about the DRM patents owned by Microsoft and Sony. How does an open source developer wanting to do DRM, or individual author wanting to self-publish, deal with a world like that?
Farber: More than just those, but it's a big problem.
Green: Patents are intended to prevent competitors. Compatible with business models of patent holders, though. Microsoft has "DRM Operating System" patent. Policy on licensing?
Manferdelli: Microsoft has not said what it will do on this patent.
--------------------------------------------------------------------------------
Afternoon panel 1: Impact of DRM on flow of information.
Hal Abelson from MIT, says to policy makers: "Watch out. You're surrounded by delusional communities, lawyers and computer scientists, who have the delusion you make things better by making them more precise."
You start with the public good, and layer laws and code on top, and then forbit its removal. What's missing from discussion is not fair use in this discussion; it's the "de minimus" principle -- "do no succumb to the illusion that the public good is best served by forcing the strict alignment of practices with policy."
Copyright in scientific publishing: Abelson says scholars give rights to the journals, which now own it. Then then allow the scientist author to retain some limited rights. University gets no rights. Forget the public.
American Chemical Society gives authors right to transmit paper to not more than 50 colleagues. New England Journal of Medicine gives nothing except same fair use rights everyone else has.
Stationers Hall -- not about creativity of authors, but rights going to "responsible parties" that will exert monopoly control over publishing.
Sophisticated research tools are being stymied. Or: new network effects leading to further concentration and monopolization. DRM will exacerbate both of these effects, he says.
Are we headed back to the world of legally sanctioned monopolies dominating the infrastructure, ushered in and cemented for a very long time? The answer seems yes.
HP's John Erickson says polices are code, but increasingly not built-in (a good thing). That's useful where system must make a choice.
Potential, he says, is that policy creation, management, etc. is put into users' hands. Scary part is it's still code, limited by language, choices of people who control the copyright. But you can aspire to have values included.
Regimes that don't factor in "fuzzy" attributes -- include people in the loop -- will be removed from reality. "We have to figure out escapes."
Joe Liu, from Boston College Law School, presents a paper (MS Word document) on how the DMCA has affected encryption research.
Paper's main points: While there's supposed to be an encryption exemption in DMCA, and researchers should be able to continue their work in some cases (but not all), the DMCA will have "non trivial impact on the conditions under which such research takes place."
He says there's a danger of self-censorship, first of all. But scientists will have conditions placed on their research. Examples: limits on who can conduct research (law prefers people with formal training); new hurdles before doing research (such as consulting lawyers and getting permission from copyright owner); limits free communication about research; limits on publishing and content of publishing (not allowing code itself to be published).
This kind of regulation is a long distance from copyright infringement, he says, because we could be losing other kinds of innovation.
Ed Felten says DMCA takes all kinds of devices and turns them into black boxes. It means you're not supposed to look inside. The boundaries of the black box tend to grow.
Big policy questions: "Bans on understanding technology tend to cripple the public debate about these issues."
Examples: Total Information Awareness, obviously an issue. Feds want to mine all commercial databases. Advocates say DRM can prevent rogue uses by rogue law enforcement personnel. True? We need to understand DRM to know, but it's a black box.
Or, porn blocking. Advocates say blacklist is accurate, doesn't block good stuff. True? You need to look in the black box. (There's a DMCA exception for this...)
Or, electronic voting machines. Totally opaque designs. Do they tell the truth? No one knows for sure.
In all three cases, he says, we need to answer black-box questions. We can't because of DMCA.
Larry Lessig says we once had an online architecture that enabled everyone to take everything out there and share it. Created complaints from copyright owners. Political response was to force shift in architecture, flipping default to supporting total control.
Default world of total control. We're solving for the extreme cases, he says, thinking about the world only in binary terms. Defeats the massive and more important category of people in the middle, who want to share and take content in the middle.
Strategies: Fighting in courts and Congress, looking for balance. Not going to work, at least not in a timely way, he says.
Instead we need strategy that expresses the middle way, e.g. Creative Commons, finding a way to express rights -- Digital Rights Expression.
Voluntary action. Use this to create something different.
If it's binary, the forces of absolute control will win, he says. The cartel owns Congress. The courts aren't upholding the balance the Constitution says we should have. "Ideals and principles" versus "all the money in the world" hasn't come up in a while.
So we're left with you and me, doing it ourselves. We can build a balance ourselves, believing in freedom and showing it.
http://weblog.siliconvalley.com/column/dangillmor/archives/000821.shtml#000821
