Leonard Slatkin did a great thing last night:  he made an entire hall of jaded, sophisticated concertgoers (bathing in the cream bath that is Carnegie Hall) really listen to Beethoven's 9th.  And it's all related to the current battle between innovation and intellectual property overreaching.

Slatkin turned around and talked to the hall about Mahler's re-orchestration of this symphony.  He pointed out that Beethoven died sixty years before Mahler's prime as a great opera conductor and composer, and that during those sixty years a lot changed:  orchestras got bigger, halls got bigger, and instruments evolved.  Notes that a flute couldn't play in B's time became accessible.  (There are still notes that the sopranos in the Kennedy Center chorus couldn't reach last night, but let's be charitable: pitch has gone up an awful lot since B's time, and human vocal cords haven't changed that much.)

Mahler, as a guy with dramatic flair and an urge to actually hear the melodies of the piece, liberally redrafted the orchestration of the 9th.  He used massed winds to emphasize points.  He changed dynamics and articulations.  He had the entire wind section raise their bells (a very Mahlerian move) to play what had been a lone piccolo's trill near the end of the last movement.  He marked up the famous recitative so it would sound meaningful instead of plodding.  Slatkin illustrated this for us, using the orchestra to provide musical examples.  Slatkin pointed out that lots of conductors had done similar things to the 9th -- the Szell 9th, the Toscanini 9th, the Walter 9th. 

He ended by saying that during the last 25 years or so we've adopted this prayerful, pure (my words, not his) approach to "classical" music.  We see and hear these works as unchanging and unchangeable.  But that's not what they are -- they're not frozen in amber, they're not things we're supposed to respect in the abstract.  They change with the times.

Some people near me arrived after this portion of the concert, before the NSO actually played the symphony all the way through.  When I told them about what they'd missed, one woman said: "Oh, well, I'm very familiar with the piece," as if she couldn't imagine that it might change.  And then she told her husband that the symphony was 70 minutes long.  Wellll, maybe it is and maybe it isn't! It depends what's being done to it.

Maybe (here's the tie-in to innovation and intellectual property) we're in an era in which we're beginning numbly to accept that "content" is just provided to us.  It's an atom, a thing that floats in space, unchanging.  We can hear or see it, as part of a mass content-absorption experience, but we are at a distance from it. 

But musical experiences are informational.  They're made of bits, not of atoms.  They should happen anew every single time, if things are going well.  Music isn't wallpaper, and you don't "acquire" concerts.  You experience them. 

Anyway, the performance was stunningly beautiful, full-hearted, and novel.  Maybe that's just me.  I thought it was. 

And I did get the sense that everyone in the hall who had been there for Slatkin's talk was listening intently with new ears. 

The truly talented Paul Marino came to talk this afternoon about Machinima, and I'm looking for volunteers to help me with "Property Law: The Video Game."  Seriously, folks, this is big.

Paul (and many others) are working on and using tools that make film-making within a real-time 3D environment easy for the rest of us.  They're using game engines (like Quake) as a way to produce movies.  This all started with teenagers recording what they themselves were doing in the game -- so they could show their friends -- and has amplified into a new way to make animated movies with a real director's touch. 

Paul (who is writing a book that will be coming out soon) says that Machinima was coined in Scotland and means "machine cinema."  At this point, people are deep into building post-production tools that provide all kinds of control.  There are some breathtaking films on this site.  Watch the one with the flower.

Basically, the Machinima tools take a game environment (many different game environments) as a given, and allow directors to choose camera angles, lay down "tracks" of different characters (so they appear to interact), add music -- you name it.  Game developers have mostly been quite receptive to Machinima, because it provides another channel for their engines.  So far, the licensing part of all this is a little ad hoc, but will likely become more standard in time.  This may be the way the next blockbuster animated film gets made.

Paul pointed us towards Red v. Blue, a group in Austin that's making very funny films using Halo.  Guys just talking and standing around in a multiplayer game, talking about standing around.  Take my word for it -- it's funny.

You can imagine all the gamer sorts of things that can happen in these films -- characters gathering together to be directed (people playing characters who are playing actors), avatars wandering on to the set, killing/maiming (but not really in what Paul showed us)..

He showed us some very realistic character work for which the interface for human directors would be sliders.  Raise a slider, raise an eyebrow!  And if you have a .wav file of someone speaking, that will drive the character's speech.  Really neat (words fail me at this point).

If you join the Machinima Academy, you can get a tool that allows creation of these things for a modest yearly fee.  As long as you don't make commercial films.  Paul did a demo for us in about 3 minutes -- very impressive.

So the law professors started asking questions:

can we load documents in and make this a course?

can we modify the environments?

can we use this to stage reenactments of accidents?

Paul pointed us to Adobe Atmosphere, a virtual world tool that allows for documents to be involved.  I'm most interested in the course applications.  What if you walked into a property course, and it was a game like one of these films?  (This is a lot of work for the professor.) 

But it might just be the future.

Part of the premise of the accountable net is that each individual will be setting up his/her own set of rules for who they connect to (this is a who rather than a what question -- trustworthy sources of bits).  We're still working on the paper, so there's nothing to blog yet, but the roadshow has been quite constructive. 

At the Berkman Center last week, we heard from several people who are concerned about making any changes at the ISP level.  The worry is that packets won't be guaranteed to get anywhere unless they find a route through congenial ISPs.  If any ISP can block anything, the reasoning goes, we'll end up with a net that doesn't actually interconnect.  No quality of service guarantees.

The response to this is that ISPs, to be valuable, will want to connect to networks that make adequate security guarantees (and enforce them).  So interconnection will be desirable and growing.  Backbones will want to make the same deals.

It's at the individual level that things get most interesting.  Let's assume that we flip the default setting and connect only to those we actually trust (or who are recommended to us).  In a sense, we will have set up an individual set of "laws" for online interactions.  We could even draw a picture of these laws so that we'd understand them for ourselves in a continuously updated way.  If we're not up to doing this for ourselves, we could go to vendors of rules and get the package that made sense.

Now, if laws can be bought and sold in this fashion, what does that mean for our respect for and understanding of "real" law?  If "real" law only governs atoms, is it more or less meaningful to us?  It may be that we end up with two complementary systems, with overlaps.  After all, the sovereign who has power over us, physically, could try to mandate particular sets of default rules -- a tricky task to enforce.  Things will be moving too quickly for the sovereign to keep up with our filters and connections.

But a new question does arise -- what happens to respect for "real" law when another system of rules exists that has legal effect in the online world?

Two recent (or upcoming) ICANN moments should cause concern. 

First, the sTLD beauty contest.  The application requirements for these sTLDs make it seem as if registries are applying for venture funding rather than a string.  Take a look, particularly at the financial and business plan requirements.  Headcounts down to the mailroom.  Travel plans.  It's as if ICANN has hired an investment banker to look into these plans.  ICANN has no special competency in any of these areas, and it would make much more sense -- and fit ICANN's limited role so much better -- if ICANN had a neutral third party develop minimum technical/financial standards.  ICANN could then then roll TLDs (not sponsored, not unsponsored, just TLDs) out as applications came in and were approved. 

Second, the suggestion that ICANN needs a Policy Development Process about changes to the "architecture and operations" (or "services and actions" -- pick your broad description) of a registry.  Although the staff says (and I believe them) that all they're doing is trying to develop a coherent process for responding to registry requests to amend their contracts, it's very clear that other constituencies view this PDP as an opportunity to ensure that no registry does anything without their permission.  This has gone far out of scope and needs to be hauled back.  Yes, we need a docket for the process and substance of actions ICANN is taking -- but we don't need one or another constituency blocking registry services for idiosyncratic reasons. 

The idea is that registries are free to innovate unless there is a consensus policy in existence to the contrary.  This PDP should not be viewed as changing that default setting, which is set forth in each of the contracts ICANN has signed with gTLDs.  Nor should the sTLD process be looked on as a model for future TLD activities.  It's time to move on to an ICANN that is more clearly in the business of technical coordination.

The MPAA comments in the broadcast flag proceeding are worth reading.

At p.9, the MPAA asserts that "the focus of attention on unauthorized redistribution should be on whether a proposed technology affirmatively and reasonably constrains unauthorized distribution beyond the local environment. . . " What the MPAA means by this is that it believes no content protection technology that allows transmission of content online could ever be added to Table A.

What does that mean in humanspeak?  Well, it means that if you're using a "compliant" TV and you see a news clip that you'd like to send to your parents (who, let's assume, don't live with you), you won't be able to.  Nope, not unless you find a way to make a hard copy and mail it to them in a box (and they have a compliant device in their home that they understand and can use to play the recording).  Sounds cumbersome, doesn't it?  Sort of puts things in a box. 

Focusing on a "tightly defined geographic area" for redistribution allowed by content protection technologies seems odd in the age of the internet.  But even stranger is the MPAA's contention that software demodulators (code that "tunes" TV signals so that people can see them) must be covered by the flag rule.  So this means that the FCC is now in the business of assuring the "compliance" and "robustness" of code.  Software demodulators, the MPAA claims, can only be sold on the market if they are incorporated in a compliant "Demodulation Product."  Again, the box:  software "tuners" will have to be sold in a box with approved hardware in order to be legal. And won't be available, separately, for use in PCs (unless the digital outputs of those PCs are adequately robust and protected).

This is shaping up to be quite a battle.  Stay tuned (but don't use software to do so).

It would be good to get mainstream computer enthusiasts interested in this proceeding.  What's the best way to do that?

There's reputable work going on with human cloning. A group of South Korean scientists has figured out how to clone a human embryo.

Immediately, we're into the circle of ethical handwringing that is familiar to us all. But a new intersection struck me as I listened to the stories on the radio.

There's a sense that we want to hold technology back, to prohibit certain kinds of copying that some view as immoral. I have a strong view on this when it comes to humans: so much about a pre-wired person is shaped by their environment that a clone could turn out entirely differently than its original. I refuse to worry about cloning, and I think the importance of creating useful stem cells far outweighs any concerns about creating embryo copies that are "human" in some sense.

Similarly, the copyright debate (particularly in the broadcast flag setting) is often focused on limiting technology so as to prohibit certain kinds of copying that some view as immoral.

I'm going to make a provocative suggestion:  Although it is true that too much copying produces cancer (or destructive infringement), the machines that make copying possible don't have moral content.  So there is no reason to punish copying machines when they do their job -- either when copying  embryos or content. 

Additionally, the act of copying is purely informational.  Too much of it, or too much public distribution of it, can be considered wrongful (cancerous or infringing), but small amounts of copying for personal purposes don't meet this test.

Don't punish technology.  Punish people who copy too much.  (And let stem cell research flourish.)

A good property/cyberlaw connection for this week.  Courts often strongly discourage self-help in landlord-tenant disputes because of the risk of violence, even when the landlord and tenant have agreed in a lease that the landlord may retake if there's a breach of the lease conditions.  (Tomorrow's case:  Berg v. Wiley.) 

We just don't like vigilantes.  And we point to things like eviction summary proceedings as the place landlords should go.

In the copyright context, though, we're encouraging self-help by content owners by saying that technical copy protections are in effect the law of the land.  We say that it's too cumbersome to go to court and fight about each individual infringement.  Instead, just fix it through technology.  (Looking around, I find that Julie Cohen figured this out a while ago.  Maybe I can provide a different spin.)

In the broadcast flag setting, we've gone even further.  Not only are we saying that technical copy protection is the law of the land, but the FCC has attempted to claim that the flag has nothing to do with copyright.  So there's no chance that anyone could say "wait a minute, go to court before you say I'm infringing, don't prompt violence through self-help."  It's not about copyright.  It's about I Love Lucy.  It's about the American Way.  It's completely unconnected from legal liability rules.  "We have to use self-help," say the studios, "because otherwise broadcast television as we know it will cease to exist."

There's no contract in place between broadcasters and the public, as there was in the Berg case (a lease that had pretty clearly been breached).  There's no claim that copyrights are being infringed by non-compliant demodulators (because the FCC's jurisdiction would be weakened).  But self-help is being written into regulations nonetheless.  

Surely that's doubly wrong.  Self help is wrong in the first place (as the Berg court reminds us), but self help that claims no legal background for justification must be really wrong. 

I've been thinking about how to present a suite of offerings to law students -- a basic cyberlaw course, a seminar for case studies, a group of people working on papers -- and I think something that's missing is actual technical expertise (both in me and in many students).  We have long textual introductions in articles that explain, over and over again, how the DNS works, for example.  Surely it's time to standardize and expect that that knowledge as well as more sophisticated understanding.

So I think I'm going to suggest a technology course (or set of course-lets) to run alongside the basic cyberlaw course.  The technology course would be pass/fail, so as to eliminate anxiety.  It might consist of a few long sessions, or one short session a week.  I've seen a sample tutorial that is given at Harvard, and I thought it was great.  I'd like to have someone do something like that at Cardozo.

The comments I got on "what is cyberlaw" were extremely helpful, and I'm going to adopt a more modular, case-studies approach next year.  This is another authentic plea for commentary:

What should law students (or business students, or any graduate student in a relevant field) know about the internet, networks generally, or the personal computer?  How is this best taught? 

Last night, John Palfrey, David Johnson, and I gave the first roadshow presentation of the Accountable Net, a paper that we're working on. We talked to a group of Yale and Harvard cyberscholars. 

David led off, summarizing the paper wisely; John cleaned up, sagely noting what points we weren't taking on (e.g., the longtime discussions about whether the internet can be governed at all); and in between I said some things about the strengths and weaknesses of the paper.  A very intense discussion ensued.  All involved were brilliant and insightful.

It's too early to talk about the paper in any specific way on this blog -- the three of us are still thinking through what its scope will be, and we go up and down in our various assessments of its possible impact. 

But it's not too early to talk about whom to trust.  From my perspective, choices of private labeling/filtering systems, choices of private DRM, choices of platforms, choices of private content -- all of this is good.  I assume that competition will produce lots of choices for the future of communication online.  I refuse to be worried about allowing private firms to make these kinds of decisions for us if we're too technically incompetent to make them for ourselves.

(Note yesterday's NYT front-page story about how incompetent people are at arranging for their own online security, and how important it is that they learn about this; note that Cardozo's own Zach Rubenstein made a point of talking about how idiotic law professors are in particular on this subject.  But I digress.)

I also think that we're at a crucial moment of growing knowledge and understanding of individual power to form communities online.  Social software, brutal and blunt though it is right now, is taking off.  We're all linking wildly.  We're learning how to include as well as exclude.  We will someday have much more nuanced understandings of our own online networks, and we'll be accountable to each other as individuals for what we send and receive.

So I trust private firms, and I trust individuals, to come up with a scenario that is a viable alternative to some sort of locked-down, government-controlled future internet.  That alternative may be more filtered and more limited than the internet we have today, but it's better than the vision that the copyright industry and the Department of Homeland Security have of online life.

But I heard last night from several people who are not as optimistic about human nature and who do not want to privilege private firms (of any size or description) in providing forms of "governance" through linking/filtering tools.  There's a sharp split between those who trust governments and those who cannot imagine trusting governments (much less groups of governments) in the creation of rules about non-physical-harm-causing bits.  There's a deep divide between the common-carrier, we're-all-in-this-together-and-it's-got-to-be-fixed view of the internet and the let's-fix-it-ourselves view.

Yes, this is a simple and obvious set of points.  But it may be that we're not as distant from the "can it be governed at all" set of arguments as we'd like to be.  We'd like to say, "oh, that's SO late 1996," but in fact we haven't progressed very far.  Some people think: governments have a role when it comes to atoms, and have physical power over ISPs in their countries, but most online problems aren't easily addressable in any mass way (much less fixable) by courts and legislators.  Other people think:  nothing much has changed here, of course governments should be fixing spam and content issues and everything else under the sun, bring in more treaties right away.

And the two camps aren't drawing more closely together -- at least not yet. 

Tomorrow morning, at 10 am in 2141 Rayburn, the Subcommittee on Courts, the Internet, and Intellectual Property is holding a hearing on "Internet Domain Name Fraud -- New Criminal and Civil Enforcement Tools."  At that hearing, the Subcommittee will be considering a new Whois bill creating new penalties for people who provide false data when registering a domain name.

We need to raise our collective eyebrows at this bill (which was suddenly dropped the evening before this hearing).  The title of the bill is the "Fraudulent Online Identity Sanctions Act." (FOISA)

First of all, it includes in the category of "willful" trademark infringements those of a "violator. . [who] knowingly provided material and misleading false contact information to a domain name registrar. . ."  This is significant, because monetary damages for trademark infringement can be increased up to three times if the infringement is willful -- and usually "willful" is a question left up to a judge.  It's also significant that there doesn't appear to be a necessary connection between what the violator has done and the whois data issue. 

Secondly, the bill amends the Copyright Act by adding to the "willful" language for damages in that Act the idea that an infringement shall be considered to be willful where "the court finds that the infringer. . . knowingly provided material and misleading false contact information to a domain name registrar."  Again, this is significant because willful copyright infringement can mean increased statutory damages awards of as much as $150,000.  And the bill ties this amendment to cases of "infringement occuring at or in connection with an online location" -- clearly broad language intended to get at P2P file trading.  Again, usually courts decide "willfulness" on a case-by-case basis.

This is like sentencing guidelines for intellectual property law.

And, in fact, it is.

The final section of the bill adds a sentencing mandate for "falsification relating to domain names in connection with offenses."  Maximum penalty:  increased by 7 years if, "in furtherance of that offense," the defendant provided material and misleading false contact information to a registrar.

This is outrageous, and here's why.  As a matter of ICANN policy, domain name registrants in gTLDs under contract with ICANN (with certain limited exceptions) are required to provide, and registrars are required to publish online, data about themselves -- including phone numbers and email addresses for technical contacts.  The vast majority of registrants in gTLDs are small businesses and individuals.  Small businesses and individuals worry about privacy and spam, and it is well-known that WHOIS information is regularly mined by spammers. 

It is already unfair to force registrants to provide all this data, and as a result many registrants do lie.  IP interests view the WHOIS database (an artifact of a gentler, academic age, and not mandated by any law) as their special red telephone information mother lode.  So they want to ensure its accuracy by mandating hugely enhanced damages if someone fails to tell the truth.  Law enforcement also has this interest.

There are ways to help both IP interests and law enforcement to data by providing special access to approved entities.  But it won't be accurate.  It costs far too much money for registrars to verify this data, and individuals will, predictably, continue to lie.

As Officer Short Shrift says in the Phantom Tollbooth, we're all guilty, guilty, guilty, and should be sent away for six million years.

From "Music, the Brain, and Ecstasy":

"[Composers] who lose their youthful rebelliousness are in grave danger of losing their talent as well.  Such was the destiny of Mendelssohn and Saint-Saens.  After a youth brimming in confidence and daring, Mendelssohn essentially worked himself to death in academic life, all the while becoming more and more conservative in his outlook and more and more detail-oriented in his composing -- a perfectionism he described late in life as his 'dread disease.'  Saint-Saens suffered a worse fate, becoming so reactionary late in life that he schemed to quash the careers of youthful free spirits like Debussy.  He once wrote in regret, 'I ran after the chimera of purity of style and perfection of form.'  The innovative Berlioz, who knew Saint-Saens as a glittering prodigy, was less charitable:  'He knows everything but lacks inexperience.'"

In our musings about the accountable net meme/essay/road show, I and my co-authors have thought that a positive view of accountability (friends pointing friends to great content) might be a breakthrough.  Rather than just shielding ourselves from spam/spyware/security problems by only accepting bits from sources we can verify (something that technology is now making possible), we could also be pointed by our friends to great, worthwhile stuff.  Like blogs, but bigger.

The problem is, as Cory Doctorow reminded us three years ago, that people are lazy, people lie, it's hard to describe things, and no schemes are neutral.  So it might be difficult to rely only on the individual choices of content of people/ISPs you trust to get you to things you want to see. 

Of course, Doctorow's comment was about metadata.  And it might be that you could create an aggregated set of links that was really the collection of useful stuff that people you trust had decided to link to -- rather than stuff that had been described in some consistent way.

But people are still lazy, and they still do lie, and it's hard to get them to join block associations.  Some people join, sure, but not everyone.  Is it too optimistic to hope that virtual block associations will arise to point us to worthwhile material, while protecting us from destructive bits?  Maybe.  And our machines are increasingly dark to us -- incomprehensible boxes that swallow useful text without warning.  (For a different view of computers, read Dream Machine by Waldrop.  I just tried to link to it and lost this post, twice, so you're on your own in finding it.)

Anyway, the question is whether the accountable net can also be the pathway to everything useful and worthwhile online, and if this new net, built by us, can avoid the problems that make governments want to regulate.  Stay tuned -- road show begins this week.