Wednesday, June 25, 2014

No Warrantless Searches of Cell Phones

The US Supreme Court has just handed down its decision in the cases of Riley and Wurie, which raised a single central issue: If the police seize your cell phone in the course of an arrest, can they search it without a warrant? Warrantless searches of your person are certainly permissible under some circumstances---they can check your pockets to see if you have a gun, for example. And the radius of permissible warrantless searches has gradually expanded through a series of court decisions. But if they check your pocket for a gun and pull out a smartphone, can they look at, oh, your bank account, health records, photo gallery, and email, all of which are either on or are accessible through that device?

This might seem to be a pretty straightforward privacy rights case, but of course there really is no right to privacy in the Constitution. It's a question of how far the Fourth Amendment goes in protecting "persons, papers, and effects."

In a unanimous decision, the court decided that your smartphone is the modern equivalent of your papers---even though in important respects your bits and your papers are not equivalent. For example, you may be able to delete your bits, or lock them away, remotely. Nonetheless, the court was unequivocal.
Held: The police generally may not, without a warrant, search digital information on a cell phone seized from an individual who has been arrested. 

Sunday, June 22, 2014

Surveillance Works

The New York Times reports that workers are increasingly the objects of digital surveillance. Tracking devices are so simple and cheap now that it makes sense to track the movements of waiters in restaurants, and to analyze the data to improve the efficiency of service. Yes, this is creepy. I sure would not like Harvard monitoring my movements around campus. But I also understand why it is seen as a competitive advantage to watch what your workforce is doing.

The interesting case in my mind is Uber. I have been watching a passionate argument about Uber, waged on an email list to which I subscribe, reflecting the passionate taxi-vs.-Uber debates going on in the city of Cambridge. Lots is being said about whether Uber is taking money from working class taxi drivers, whether public safety demands that Uber drivers be licensed and regulated the way taxi drivers are, and so on.

The thing nobody seems to be mentioning is that the reason Uber works is because everyone involved is being watched. That's the reason you don't need to identify where you are when you request an Uber; Uber knows your phone location. That is the reason you don't need to pay at the end of the trip; Uber has your credit card information. It's also the reason service is good. You have to rate your driver before requesting your next trip; Uber drivers are polite because they know that Uber will drop them in a heartbeat if they get a few bad reviews. They won't take you the long way around (generally they seem to just follow the directions in their phone app), because they know you'll rate them down if you do, and you'll have a map showing the exact route taken as part of your receipt, incontrovertible proof you've been screwed. And your driver can do a job on you too -- if you creep out too many Uber drivers, Uber will probably close out your account. (Not that there would be much incentive to threaten an Uber driver -- because all payments are electronic, there is no reason for them to carry any cash.) The mutual rating, with big brother Uber knowing all about both you and your driver, reduces many of the risks of taxis, for both parties.

It seems to me that the advantages of Uber are too great for any except the oddest of municipalities (that might be Cambridge) to make its business model unworkable. But let's not kid ourselves. The reason it works so nicely is because it completely eliminates the mutual anonymity of the taxi economy. If the taxi industry is smart, it will figure that out and go increasingly digital, to try to gain some of the convenience and safety that make Uber attractive -- which will leave fewer options for invisible life in the big city.

Ivory Tower

Ivory Tower is a documentary about the affordability crisis in American higher education. Ty Burr in the Globe gives the movie a mixed review, finding that it tries to cover too many issues. I thought it was better, but then I am a junkie on this sort of higher ed material. Also, my institution comes out looking good, and that may have colored my impression of the film. It's playing at the Kendall Square Cinema. To judge from the size of the house when I saw it, it won't be playing there or anywhere for long (if you can't make it in Cambridge, …).

Andrew Delbanco of Columbia, author of College: What it Is, Was, and Should Be, serves as sort of an overall narrator. Delbanco's is a good book, and he has written a number of other smart pieces, so he's a good choice as the prototypical professor, idealistic and realistic at the same time. (Though there were rather too many mood shots of him shuffling across the Columbia campus with his book bag.)

Harvard, as I said, gets treated well. As the film notes, Harvard is one of a tiny number of American colleges that admit need-blind and provide financial aid up to need for all admitted students. I think the frame on that says that only 1.25% of American colleges do that. That is probably the number that say they do it; the number that don't fudge on either admission or the definition of "full need" is probably smaller. The Boston Globe has a piece today on how the median "net price" of college (tuition minus aid) has continued to rise faster than inflation at almost all colleges. Harvard is one of the few where net price has gone down over the past few years. If you sort the table by net price, Harvard is near the bottom at $15,079, $5K less than MIT and less than half the median net price of BU and even Northeastern.

The narrative wanders in and out through Harvard and various other institutions, representing different models for higher education; Harvard is the one place where money is never an issue. I wish the whole Harvard faculty would see the movie, because it serves as a reminder how how much of our high moral ground is based on two things most faculty don't appreciate nearly enough: One, that we can afford to pay the full demonstrated need of the students we admit; and two that we do admissions with an eye toward creating the future, not rewarding applicants' past achievements.

Clayton Christensen, of all people, makes an early appearance explaining that things in higher ed can't go on as they have. Harvard is in the "DNA" of every American college, he says, but they can't all be Harvard and many are wasting a lot of money trying.* This tendency to build rock walls and luxury dormitories (neither of which Harvard has, in fact), and the growth in nom-teaching employees, seem to be the conventional wisdom now about why costs keep going up. A more nuanced observation is made by one person interviewed for the film---that at state universities it's the need to draw out of state students to balance the budget (out of state tuition being higher than in-state) that forces universities to compete as though they were luxury hotels. It also creates irrational exchanges, with Texas universities trying to fill their beds with students from California and California universities trying to draw Texas students, rather than either state university trying to educate the maximum number of children of their own taxpayers at a lower price point.

Harvard president Drew Faust also appears early on, both in a freshman event and in an on-camera interview, where she seemed to me pitch-perfect about liberal education (or "liberal arts eduction," as it is now customary to call it, I suppose less to avoid the sciences than in an attempt to forestall paranoia from the political right). The film traces a computer science student, a guy from a very disadvantaged background, on his freshman-year journey through Computer Science 50, so a lot of the faces are familiar to me --- my brilliant colleague David Malan who teaches the course, and several of his undergraduate teaching assistants. I actually know the student as well, and the guy on screen is the same guy I know. This all looks very true to me.

It was when the filming moved away from Harvard that I started to cringe. Some significant educators agreed to be interviewed---John Hennessy of Stanford, Peter Thiel (funny how nobody who quotes him adoringly on higher education these days describes him as an author of The Diversity Myth), Arizona State president Michael Crow, Wesleyan president Michael Roth, and Cooper Union president (and sometime Tufts provost) Jamshed Bharucha. The cringing is because some of these folks didn't have good answers to obvious questions. Crow contemptuously dismisses ASU's reputation as a party school (having acknowledged it by completing the interviewer's question); then there is a fast cut to several minutes of ASU bacchanalia footage. Roth answers a prospective Wesleyan parent's question about whether his daughter will wind up employable by preaching the virtues of a liberal education, which was not the question (though to be fair, I know that the real answer might have been edited out). Bharucha comes out the worst, in part because he has a facial tic that does not film attractively. He is asked how he can justify making about the same amount as Faust in salary when her institution is so much bigger, and his has such financial woes that it is starting to charge tuition, against the vision of the founder. He replies, with great confidence, that she doesn't have half the problems he has---but that's the point: Cooper Union has financial problems due to debt it unwisely took on, and for the president to take a modest pay cut might have been a politic gesture of shared sacrifice.

About the only place, other than Harvard, that comes out looking good is Deep Springs, which has 26 students, all men. This may seem to be an irrelevant distraction; higher education's cost problems can't be solved by generalizing the Deep Springs model. But it drives home a crucial question, what are you really paying for when you buy a college education?

The big villain, in the narrative, is student debt, which has been getting lots of attention lately. I do wish that some president of a college that charges a lot of money and can't balance that with scholarship aid would say the simple truth to prospective students and their families: "I hope you will attend Bozo U. It's a great place and you would get a lot out of being here. We are looking for people like you because we are no better than our students are. And we will do the best we can to make Bozo U affordable for you and your family. But you should not attend this university if you are going to wind up more than $X in debt when you are done. In fact you should not attend any college that will leave you that much in debt."

But what would $X be? One problem is that it depends on what you learn, but talking about that is hard at the same time as you are preaching the importance of a liberal education. If you graduate with a degree in computer science, $50,000 of debt would not be too much. You could probably pay that back in a year --- entry level jobs in Silicon Valley are paying in the six figures these days, at least for graduates of good programs.

The film quotes the statistic that the average (or is it median?) student debt on completing college is now $25,000. That is said as though it is a shocking statistic, but I am not sure why it is so shocking. If I remember the national data correctly, that is less than the delta between the annual salaries of college graduates and non graduates.

The film, however, pivots to interview at some length an articulate, unemployed college graduate with $140,000 in debt. (I hope someone sees this clip and offers her a job!) Now it seems to me the real issue is, what predatory forces got her to agree to take on that much debt? She did not need to be a math major to figure out how long it would take her to pay that back. I'll bet she got some bad advice, probably from people who stood to profit by her woes. I have a hard time thinking that the enthusiasm of the national "college for all" forces are really most responsible for encouraging such bad decision-making. (The whole for-profit higher education sector isn't mentioned in this film at all.)

Overall, it's a well done film. It certainly homes in on some of the hypocrisies and snake-oil salespeople in higher education, and its bottom line about MOOCs, which seems to be that online materials can help but education almost always requires direct communication between teachers and students, seems about right. That is, in fact, the impression from which you come away from all the CS50 footage, and that is the way that course works. It is a highly social sort of intensive technology education.

* The spat between my two Harvard colleagues Professors Jill Lepore and Clayton Christensen continues to roil. (See the previous two posts on this blog. I wish I had noted, when trying to parse Christensen's belittling way of referring to Lepore by her first name, that in the same interview he refers to himself as "Clayton Christensen," in the third person!) Economics writer David Warsh provides some helpful background; I had forgotten that two years before Lepore's article, a "hagiographic" (as Warsh puts it) profile of Christensen had appeared, also in the New Yorker.

Friday, June 20, 2014

Clayton Christensen is Mad

In a brief interview in Business Week, Christensen starts out polite about Jill Lepore (see The Bogosity of Disruption Theory) but gets significantly less so. At first, he says, he thought she was trying to protect his trademark, and he was shocked to discover she had something else in mind.
I was delighted that somebody with her standing would join me in trying to bring discipline and understanding around a very useful theory. I’ve been trying to do it for 20 years. And then in a stunning reversal, she starts instead to try to discredit Clay Christensen, in a really mean way. … I hope you can understand why I am mad that a woman of her stature could perform such a criminal act of dishonesty ….
If, like me, you twinged when you read that---wondering whether he would have been less mad to have been taken down by a man---the answer seems to be yes.
 Fifty-two years, Jill. Just so you understand, disruption doesn’t happen overnight.…  So—Jill, tell me, what’s the truth? …  Do they make rod and angle iron, Jill? No. Do they make structural steel I-beams and H-beams that you use to make the massive skyscrapers downtown, does U.S. Steel make those beams? Come on, Jill, tell me! No! …
 [Interviewer] You keep referring to Lepore by her first name. Do you know her?
I’ve never met her in my life.
Then why do you keep calling her by her first name, when the interviewer never used it?

The reliably conservative American Enterprise Institute dismisses the Lepore article as a leftist plot to  preserve the higher education dinosaur, titling its takedown of Lepore "Why business guru Clayton Christensen has landed on the left's hit list," and complaining that she "picks at the the corporate case studies that Christensen uses." Which is, of course, exactly right, except for the belittling verb --- that is what scholars do to each other, point out when their data do not support their conclusions. And then the AEI writer says that all she is trying to do is protect her own cushy job.
Count Lepore as one of many college professors who doesn’t like the idea of online learning disrupting how brick-and-mortar universities operate — and how profs do their jobs.
Which makes about as much sense as saying that the people who objected to Lysenkoism in Soviet Russia were just trying to preserve an archaic style of farming.

Paul Krugman takes Lepore's side, which I suppose just confirms the theory that this is a lefty plot. The AEI piece makes one interesting observation about the teaching mission of HBS, which I have no way to judge.
There is a big ongoing debate – Christensen on side, Michael Porter on the other — about how Harvard’s business school should deal with online education. Lepore’s piece can been seen as a mission in the campaign against Christensen approach.
Seen by people other than Lepore, I imagine. But there may equally be something else at stake---what counts as research and scholarship at places like HBS. Christensen seems to insist that he is doing something like science, and his theory has become more nuanced and gained more explanatory force. As he says about the iPhone in a Harvard Magazine profile grandly titled Disruptive Genius, "First I was wrong, and then I was right." But of course the real test of a scientific theory is its predictive success, not its pliability to fit facts. Or the amount of money people will pay you to spout it.

Monday, June 16, 2014

The Bogosity of Disruption Theory

Writing in the New Yorker,  Jill Lepore devastates HBS professor Clayton Christensen's theory of disruptive innovation. It turns out his case studies were cherry picked, and even the prime cherries don't actually provide persuasive evidence for his theories. Lepore notes that a theory is best validated by its predictive capability, not its ability to fit patterns of carefully chosen past events.
The theory of disruption is meant to be predictive. On March 10, 2000, Christensen launched a $3.8-million Disruptive Growth Fund, which he managed with Neil Eisner, a broker in St. Louis. Christensen drew on his theory to select stocks. Less than a year later, the fund was quietly liquidated: during a stretch of time when the Nasdaq lost fifty per cent of its value, the Disruptive Growth Fund lost sixty-four per cent. In 2007, Christensen told Business Week that “the prediction of the theory would be that Apple won’t succeed with the iPhone,” adding, “History speaks pretty loudly on that.” In its first five years, the iPhone generated a hundred and fifty billion dollars of revenue. In the preface to the 2011 edition of “The Innovator’s Dilemma,” Christensen reports that, since the book’s publication, in 1997, “the theory of disruption continues to yield predictions that are quite accurate.” This is less because people have used his model to make accurate predictions about things that haven’t happened yet than because disruption has been sold as advice, and because much that happened between 1997 and 2011 looks, in retrospect, disruptive. Disruptive innovation can reliably be seen only after the fact. History speaks loudly, apparently, only when you can make it say what you want it to say. 
It's really a good read. I am sure there is another side to the story, but it is always fun to see a Harvard snake oil salesman debunked by another Harvard colleague.

I have a bit of a personal stake in this. Lepore spends some time on Christensen's book The Innovative University, written about Harvard and BYU-Idaho, which would seem to be an odd choice for a counterpoint to Harvard until you realize that Christensen's old boss, HBS dean Kim Clark, followed the diktat of the LDS church leader to become the head of that university. It is surely not coincidental that Christensen himself proclaims the importance of his LDS faith.

The Innovative University treats Excellence Without a Soul with some reverence, though as Lepore points out, the parallels between 17th century Harvard and today's BYU-Idaho (NOT the well known Utah university) are rather strained. Still, having read the manuscript, I agreed to blurb the book, in a rather restrained way, so I suppose I have to consider myself a sort of apologist for the disruption mania of which I am now acknowledging my skepticism. Mea culpa.

Computer Science would be so much easier if we could sell our theories by attaching our names to them and claiming they were revolutionary and explained everything that had happened and predicted everything that would. Happily, neither our science nor any other works that way, though sometimes my colleagues try. A fellow Harvard professor once said in a faculty discussion of a colleague elsewhere that "The unit of bogosity is the micro-Jones," except that "Jones" was not the actual name of the computer scientist whose millionth sub-part might have been used as the standard measure of bogus claims. Out of respect for all parties, I will not use the actual name. You probably wouldn't recognize it anyway.

I am glad that my colleague in the history department decided to do the painstaking research needed to establish the bogosity of my business school colleague. Isn't there a business school professor, here at Harvard or elsewhere, who might have done the job?

Wednesday, June 4, 2014

Tiananmen, 25 Years Later

I posted this photo once before, but I want to do it again on the 25th anniversary of the Tiananmen massacre. I took it five years ago today when I joined the remembrance in Victoria Park in Hong Kong. It was a bizarre day, which started in Shantou on the mainland, where one could not have guessed that the day was an anniversary of anything at all. Perhaps there was an abnormal police presence on the university campus where I had been speaking the day before; I had no way of knowing what would have been considered normal. But there were no posters, nothing, until I got off the plane in Hong Kong, where the energy was everywhere all day, ending with this massive protest in the favoring presence of Queen Victoria herself, who still is enthroned in a huge statue in the park.

I'd like to add a plug for Rowena He's just-published book Tiananmen Exiles, which traces what happened to the student protesters. (Rowena teaches a popular Harvard freshman seminar on those events, and has devoted herself to promoting their lessons.) A nice moral counterpoint to today's NYT story about the success enjoyed by the head of the student union at the University of Beijing at the time, who stuck with the authorities. Today there are not one but two Tiananmen protests in Hong Kong, signaling (according to the NYT) both a division about how aggressively to confront Beijing, and an increased awareness of the events of 25 years ago and their significance for democracy in Hong Kong, in spite of the unwavering insistence by mainland authorities that nothing worth remembering happened on June 4, 1989.

Tuesday, June 3, 2014

An Odd Sort of Honor

I've said quite a few times that I believe in honor and wish there were more of it in the academy. I just turned down an invitation to go to China during term time, something that would not be against the rules, or at least not against any rules that anyone enforces. It just seemed to me dishonorable not to show up in the classroom when it is my job to teach, and when I consider it a privilege to be able to do so. Whether there are any applicable penalties or even rules seemed to me beside the point.

So one of my concerns about the Harvard honor code discussion is how distant it seems to have gotten from matters of conscience. Instead we read about psychological studies which, allegedly, show that reciting an honor pledge on a certain reinforcement schedule will tend to alter behavior, as though B.F. Skinner had put those quaint notions of conscience and honor on a firm scientific footing and we no longer needed to talk about dishonor, only its symptomatology.

So I reacted with some despair and bemusement when a friend pointed out to me that edX's Honor Code is embedded in its Terms of Service, just like iTunes and the like, right down to that terms-of-service classic gotcha: We can change these terms any time we want without telling you; not our fault if you didn't realize we'd changed the rules on you.
Please note that we review and may make changes to this Honor Code from time to time.  Any changes to this Honor Code will be effective immediately upon posting on this page, with an updated effective date.  By accessing the Site after any changes have been made, you signify your agreement on a prospective basis to the modified Honor Code and any changes contained therein.  Be sure to return to this page periodically to ensure familiarity with the most current version of this Honor Code.
Now I am completely fine with an "honor code" that is really a contract between student and institution. That is the way I interpret Harvard's current rules; when you accept our offer, you agree to abide by our rules. (But then there is no need for repeating the honor pledge. Once is enough to sign a contract.)

So a contract, yes. But an honor code as a contract of adhesion, which can be altered by one party without notice? Yes, the MOOC world will generate some interesting experiments. And I know Harvard is now a bit player in edX, even though it was a founding member. But really, do the people who dream this stuff up have any idea what words like "honor" actually mean, and how odd it is -- dishonorable actually -- for an academy of higher learning to tell people, cynically, that they should go back to some web page full of legal language and see if it has changed, if they don't want to risk falling afoul of some new rule?

Sunday, June 1, 2014

Unsurprising NSA story of the week

The New York Times reports that the Snowden documents reveal that the NSA is assembling a vast database of face photos.

No one should be surprised. Storage is so cheap that even for a commercial enterprise, it's better to save data once you have it than the throw it away -- no telling how it might come in handy later on A spy agency can start with a pretty good database -- photos of everyone who has ever gotten a passport or a visa, for example. And then it can mine the web, and cooperate with friendly governments abroad, who have their own local databases.

Is this creepy? You bet, though it is not completely obvious why. After all, there is nothing more public than your face. You walk around showing it to lots of people every day. Ordinarily, it's a pleasant surprise when someone you knew long ago recognizes it.

Two things about that. First, remember the koan: More of the same can be a whole new thing. There has always been a chance that someone would recognize you in a cafe while you were traveling halfway around the world. That is very different from an expectation that surveillance cameras in Istanbul will, through some intergovernmental cooperation, trigger an alert to a US agency that you are there -- and by the way, who is that woman you are there with? But that may come to be the norm, if the technology goes unregulated.

The other thing that I have always found creepy about face databases is the opportunities they present to influence behavior. Way back in 2005 Jeremy Bailenson at Stanford showed that voters' choices could be shifted significantly if they were shown photos of the candidates with a subliminal level of the voter's own face morphed into one of them. People tend to vote for people who resemble themselves, even when they don't realize (none of Bailenson's subjects did) that the effect has been achieved by subtly altering a photo. Think of the commercial and propagandistic uses of a tagged database of the faces of most of the world's population.

The article notes that Congress has done nothing on this front. I'll bet they won't, either, but they should.