¶ The Laboratorium (3d ser.)

I go back to this tweet a lot:

This election is like if your friends pick dinner and 3 vote pizza and 2 vote “kill and eat you”. Even if pizza wins, there’s a big problem.
— Andrew Shvarts (@Shvartacus)
August 9, 2016

Three months later, “kill and eat you” won the 2016 election. Three years on, “kill and eat you” is an established feature of the United States political system. To continue the metaphor:

• One person wants plain cheese pizza.
• One person wants vegan fig and mushroom pizza.
• One person wants to kill and eat you because the talking lizards living under his bed told him that you have delicious space aliens hiding under your skin.
• One person wants to kill and eat you because anything else would be better than pizza again.
• One person hasn’t been paying attention, isn’t really hungry, and doesn’t really believe in killing people or talking lizards, but says you have to admit that the talking-lizards guy makes some valid points about the aliens.

Living together in this house is impossible, and yet it goes on. You can’t make the talking-lizards guy move out; he has too many friends. (Even if he did, could you ever really trust your other friend who went along with him?) You can’t move out; you have nowhere else to go. The fact of the matter is, dinner – and your life – depend on the person who knows and cares the least about either.

The model is simple, but it explains much that is maddening about contemporary political life. One substantial segment of the electorate has suffered the political equivalent of a psychotic break; another has not, but cynically accepts that playing along is the best way to achieve its preferred policy goals. Together, they make up the modern Republican Party, and they have discovered that yelling insane things at the top of their lungs is a viable political strategy.

It also highlights the structural factors that make intra-Democratic debates over policy and strategy so deeply frustrating. “Say no to killing and eating people” is a point of undisputed agreement. It is a necessary minimum for coexistence in society. As against a party whose de facto platform includes killing and eating people, it ought to be politically sufficient, and yet it plainly is not.

It matters deeply what kind of pizza we get when the shouting ends and things go back to normal, but there is no assurance that they ever will. Maybe the perfect set of pizza toppings will capture the imagination of the muddled middle – or then again maybe actual pizza toppings are no match for the false promise of fried space alien. Offer to compromise on burgers, or is that just giving up any hope of ever making a majority for pizza rather than being killed and eaten? Start shouting too, or keep trying to be the reasonable one? Maybe there’s a right choice, or maybe there isn’t and all roads lead to ruin.

In conclusion, impeach Donald Trump.

Greg Lastowka, who taught law at Rutgers-Camden until his untimely death of cancer in 2015, was a thoughtful scholar of copyright and virtual worlds, a beloved teacher, and a profoundly decent human being. He was generous and welcoming to me as a law student whose interests paralleled his, and then a generous and welcoming colleague. He was always one of my favorite people to see at conferences, and his papers were like the man: intelligent, unpretentious, and empathetic, with a twinkle of poetry.

Greg colleagues at Rutgers-Camden have honored his memory by holding an annual memorial lecture, and I was privileged beyond words to be asked to deliver last fall’s fourth annual lecture. It was a snowy day, and I drove through a blizzard to be there. I felt I owed him nothing less. So did many of his colleagues and family, who turned out for the lecture and reception. It was a sad occasion, but also a happy one, to be able to remember with each other what made Greg so special.

I am very pleased to say that thanks to the efforts of many – our mutual friend Michael Carrier from the Rutgers faculty; Alexis Way, the Camden editor-in-chief of volume 70 of the Rutgers University Law Review; her successor Alaina Billingham from volume 71; and the diligent student editors of the RULR – my lecture in memory of Greg has now been published as Bone Crusher 2.0. (The title refers to one of my favorite examples from all of Greg’s scholarship, a stolen Bone Crusher mace from Ultima Online.) Here is the abstract:

The late and much-missed Greg Lastowka was a treasured colleague. Three themes from his groundbreaking scholarship on virtual worlds have enduring relevance. First, virtual worlds are real places. They may not exist in our physical world, but real people spend real time together in them. Second, communities need laws. People in these spaces do harmful things to each other, and we need some rules of conduct to guide them. Third, those laws cannot be the same as the ones we use for offline conduct. Laws must reflect reality, which in this case means virtual reality. Using these three themes as guideposts, we can draw a line from Greg’s work on virtual items in online games – Bone Crusher 1.0 – to modern controversies over virtual assets on blockchains – Bone Crusher 2.0.

This is not my most ambitious paper. But for obvious reasons, it is one of the most personally meaningful. I hope that it brings some readers fond memories of Greg, and introduces others to the work of this remarkable scholar.

I set out to write a longer post, but there is really no need. The readout of President Trump’s July 25 telephone call with Ukranian President Volodymyr Zelenskyy shows him committing an obviously impeachable offense. The President of the United States personally asked a foreign nation to investigate a political rival, and he used nearly $400 million in American aid as a bargaining chip. This is corruption of the most basic sort: using his office to serve his personal interests. Nothing more need be said.

I agree that there should be an immediate investigation. But the point of this investigation is not to dig at some further factual questions of what exact words Trump used or what he meant by them. Trump himself has admitted that the conversation happened as described, and thee meaning is the meaning is clear enough. If Trump is too confused to express his demands more clearly, or too amoral to understand why they are so deeply wrong, these facts make him more impeachable, not less. They amount to a defense that he can’t be impeached because he is unfit for office in the first place.

This is not the first obviously impeachable thing that Trump has done. The Mueller report lays out, with painstaking clarity, multiple instances of impeachable obstruction of justice. The only difference is that this new story broke all at once, rather than being dribbled out over the course of years, so that the political shock of seeing everything so clearly in focus landed with full force. The financial self-dealing also probably rises to the level of high crimes and misdemeanors, although the facts there have been a little better obscured. The articles of impeachment should include obstruction of justice, and the financial investigations should continue.

With an urgent official impeachment inquiry underway,, it is time for the House to use all of its powers to compel documents and testimony, and to ask the courts for the most expedited rulings they are capable of giving. The nation has no more important business than this.

This is a moment of clarity. Trump’s conduct here is fundamentally incompatible with democratic self-government. To stand aside – or worse, to defend it – is to give up on the republic. Donald Trump must be impeached, so that the United States can survive.

Rebecca Tushnet blogged a trainwreck of a copyright opinion in Cisco Systems, Inc. v. Beccela’s Etc. from the Northern District of California. The software-licensing caselaw was not good, but this is one of the most confused opinions I’ve seen.

In brief, Cisco sells networking devices through a network of authorized dealers. The defendants allegedly sell Cisco devices outside of these authorized channels. Cisco sued on a variety of theories, including copyright infringement. In response, the defendants claimed they were making legal first sales.

Ninth Circuit caselaw (see Vernor, Psystar, and Christenson) has held that first sale doesn’t apply to software distributed on CD-ROMs or DVDs which are “licensed” rather than “sold,” and use a messy multi-factor test to determine whether a given shiny plastic disc is licensed or sold. The defendants here argued that the result should be different where the software is “embedded in hardware,” but the court disagreed that this made a difference. “The Ninth Circuit in these cases did not distinguish the first sale doctrine’s application between software and hardware … .” As a result, “[T]he first sale doctrine does not apply to software licensees even when the software is embedded in lawfully purchased hardware … .”

To which I can say only, what does the court think that software IS?

“Software” could refer to the information in a program – the sequence of bits or characters – or it could refer to a specific physical instantiation of the program – a chip, printout, or other object encoding that information. In copyright terms, the former is a “work”; the latter is a “copy.” Cisco has a copyright in the work, and we can assume that the copyright has never been validly licensed to the defendants. But in first sale, that’s irrelevant. If I’m “the owner of a particular copy … lawfully made,” then I can distribute that copy regardless of whether I have any license from the copyright owner. That’s what first sale is. The reason that Vernor and other cases rejected the application of first sale is that the copy had been licensed rather than sold: that messy multi-factor test tries to figure out what rights the possessor has over a particular shiny plastic disc. For example, does the copyright owner have the right to demand the shiny plastic disc back? If so, then the possessor may not be an “owner” of that “particular copy” and so first sale may not apply.

This reasoning doesn’t on its face distinguish between shiny plastic discs and computer hardware. But that doesn’t mean the two cases are the same. It’s right there in the Beccela’s opinion. In fact, it’s right there in the same sentence where the court announces its conclusion. Cisco’s software isn’t just “embedded in hardware”; it’s “embedded in lawfully purchased hardware,” in the court’s own phrase. That ought to end the case. If the hardware is lawfully purchased (note: “purchased” and not “licensed”), then the defendants are owners of the copies of the software and have full first sale rights. Remember: “copies” are “material objects … in which a work is fixed,” a definition that includes both shiny plastic discs and dense arrays of transistors.

The court here honestly seems to believe that software can somehow be “embedded” in hardware without the hardware being a copy of the software, as though a file were in the computer but not of it. But there is no such thing. That is what it means to store digital information in a thing: the physical structure of the thing becomes an encoding of the thing. Or, in copyright terms, a copy is a physical thing “from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device.” That’s how software works.

To be fair, I don’t think that courts in previous first-sale and software-licensing cases have been terribly careful about the work/copy distinction or about what software is. The opinions cited in Beccela’s are full of sloppy language that seems to invite this result. But that language was unnecessary; you could come out the same way in a DVD software first sale case while being careful about your terminology. Beccela’s takes these unintelligible fictions about how software works and turns them into an actual holding that is essential to the outcome of the case. It is rare to see the confusion at the heart of modern software copyright licensing so plainly stated.

One evening, about the time when bananas were first being imported in Britain, Lord Leconfield was dining in his stately home with a friend. His guest observed that nobody really knew how good a banana could be unless he had tasted one straight off the tree.

Lord Leconfield said nothing at the time, but next morning he sent for his head gardener. “Go,” he told him tersely, “to Kew. Find out how to grow a banana. Come back here and grow one.”

Off went the head gardener. A special greenhouse was constructed. The banana tree was splendid. Lord Leconfield took a lively interest in in its progress until it fructified. “I will have the banana for dinner tonight,” he said as soon as the banana was ripe. And so he did – amid a deadly hush. The head gardener himself was there, concealed behind a screen.

The banana was brought in on a splendid dish. Lord Leconfield peeled it with a golden knife. He then cut a sliver off and, with a golden fork, put it in his mouth and carefully tasted it. Whereupon he flung dish, plate, knife, fork and banana on to the floor and shouted ‘Oh God, it tastes like any other damn banana!“ Banana tree and all were ordered to be destroyed.

–T.W. Körner, An Unofficial Guide to Part III, retelling a story from the autobiography of John Wyndham, the 6th Baron Leconfield, about his grandfather, Henry Wyndham, the 2nd Baron Leconfield

The Ninth Edition of Internet Law: Cases and Problems is now available. It includes a new section on platforms as marketplaces, a half-dozen new cases, and updated notes, questions, and problems throughout. As always, the book can be downloaded directly from Semaphore Press as a pay-what-you-want DRM-free PDF. The suggested price to students remains $30. It’s also available as a print-on-demand paperback (this year’s cover color: purple) for $65.50.

The road to Zork began in late May of 1977, when Dave Lebling put together a very simple parser and game engine quite similar to Adventure’s, from which Marc Blank and Tim Anderson built their first four-room game as a sort of proof of concept. … The name itself was a nonsense word floating around MIT that one might use in place of something, shall we say, stronger in stressful situation: “Zork the bloody thing!” when a piece of code just wouldn’t work correctly, etc. The file holding the game-in-progress got named “Zork” as a sort of placeholder until someone came up with something better. In the case of Zork, though, a proper name was slow in coming. And so Zork the game remained for the first six months of its existence. …

At some point around the fall of 1977, the DMG hackers had decided that their creation really, really needed a “proper” name. Lebling suggested Dungeon, which excited no one (Lebling included), but no one could come up with anything better. And so Dungeon it was. … Shortly after that, MIT heard legal rumblings from, of all places, TSR, publishers of Dungeons and Dragons – and of a dungeon-crawling board game called simply Dungeon! TSR was always overzealous with lawsuits, and the consensus amongst the MIT lawyers that the DMG hackers consulted was that they didn’t have a legal leg to stand on. However, rather than get sucked into a lengthy squabble over a name none of them much liked in the first place, they decided to just revert to the much more memorable Zork. And so by the beginning of 1978 Dungeon became Zork once more, and retained that name forevermore.

–Jimmy Maher, Zork on the PDP-10

It occurred to me today, while reading a story about a person targeted by anti-vaccination activists, that her ordeal was very much like that of the Sandy Hook parents targeted by InfoWars, which was very much like … you get the picture.

It’s a commonplace that the Internet is conducive to online mobs: people come together and temporarily find shared purpose swarming a stranger. Each participant individually metes out a small share of what seems like justice, but there are far too many of them, result misery.

But what if that’s not exactly right? Social media can bring millions of people together, but it takes a much smaller group to light the match and fan the flames. If that group has shared values, private networks for coordination and mutual reinforcement, a common vocabulary and rhetorical toolkit, a lot of free time, and an overriding sense of purpose, it will be that much more effective.

In other words, maybe the Internet isn’t optimized for mobs, so much as it’s optimized for cults.

I am pleased to say that I have joined the editorial board of the Communications of the ACM, the monthly journal of the world’s leading computer-science professional society, the Association for Computing Machinery. I am responsible for editing a three-times-annual column, “Viewpoints: Law and Technology.” The column was created in its modern form by the estimable Stefan Bechtold, and he has done a great job getting a a group of very smart people to write very smart columns. (The estimable Pamela Samuelson single-handedly writes a regular column for CACM as well.) I have big shoes to fill.

This is personally quite meaningful to me. As regular readers of this blog know, bringing law and computer science closer together is my life’s work. It’s hard to think of a more visible symbol of that intersection than the law-focused column of this venerable computer-science journal. I am humbled to have been asked to do this and I have high ambitions to present cutting-edge issues in law and policy to CACM’s readership in a nuanced but accessible way.

There are some great columns by scholars I deeply respect in the editorial pipeline, but to mark the transition, I thought I would take the pen myself to reflect on where Internet law stands today. My inaugural column is titled Continuity and Change in Internet Law, and here is an excerpt:

Everything old is also new again with cryptocurrencies. People have hoped or feared for years that strong cryptography and a global network would make it impossible for governments to control the flow of money. There is a direct line from 1990s-era cypherpunk crypto-anarchism and experiments with digital cash to Bitcoin and blockchains. The regulatory disputes are almost exactly the ones that technologists and lawyers anticipated two decades ago. They just took a little longer to arrive than expected.

In other ways, things look very different today. One dominant idea of the early days of Internet law was that the Internet was a genuinely new place free from government power. As John Perry Barlow wrote in his famous 1996 “Declaration of the Independence of Cyberspace”: “Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace, the new home of Mind. … You have no sovereignty where we gather. … Cyberspace does not lie within your borders.”

If there was a moment that this Matrix-esque vision was definitively unplugged, it was probably the 2003 decision in Intel v. Hamidi. Intel tried to argue that its email servers were a virtual, inviolate space—so that a disgruntled ex-employee who sent email messages to current employees was engaged in the equivalent of breaking into Intel buildings and hijacking its mail carts. The court had no interest in the cyber-spatial metaphor. Instead, it focused on more down-to-earth matters: Intel’s servers were not damaged or knocked offline.

“Cyberspace” turned out not to be a good description of how people use the Internet or what they want from it. Most Internet lawsuits involve familiar real-world problems—ugly divorces, workplace harassment, frauds and scams, and an endless parade of drug deals—that have spilled over onto cellphones, Facebook pages, and other digital platforms.

Whatever developments the years ahead may bring for law and computer science, I look forward to helping the legal and technical communities to understand them by helping CACM carry on its editorial tradition of excellence.

I have a new essay, Renvoi and the Barber, in The Green Bag 2d,¹. Here is the abstract:

The renvoi paradox in choice of law arises when two states’ laws each purport to select the other’s law. The barber paradox in the foundations of mathematics arises when a set is defined to contain all sets that do not contain themselves, or, more famously, when a barber shaves all men who do not shave themselves. Which state’s law applies? Does the set contain itself? Does the barber shave himself? Each answer implies its opposite.

Conflict of laws is not mathematics, but it could learn from how mathematicians escape the barber paradox: by modifying their theories to to exclude the kind of self-reference that can go so badly wrong. Renvoi too is a paradox of self-reference. Ordinary choice of law blows up into paradox not when one state’s laws refer to another’s, but when a single state’s laws refer back to themselves. The purpose of renvoi rules is to prevent this paradoxical self-reference from occurring; they work by ignoring some aspect of a state’s laws. When a choice-of-law rule selects a state’s law, it always necessarily selects something less than whole law.

The essay also features a dozen diagrams, extensive Sweeney Todd references, some subtle shade on your favorite choice-of-law methodology, and discussion of restricted and unrestricted versions of the Axiom of Comprehension.² It may be the nerdiest thing I have ever written, and I do not say that lightly.

The Green Bag, like this blog, is in its second series. ↩︎

Which explains some of my previous reading. ↩︎