Saturday, May 14, 2011

Stanford Engineering Hero Lecture: Don Knuth

Yesterday, Stanford Prof. emeritus Don Knuth gave one of the "lectures" in the series devoted by Stanford to their "Engineering Heroes." Jim Plummer, Dean of the School of Engineering, rattled off the names of the eight initial Engineering Heroes so quickly that I had to find them later here. They are William Durand, a pioneer in aerodynamics; Fred Terman, who is credited with fostering the emergence of Silicon Valley; Bill Hewlett and Dave Packard; Charles Litton, one of their associates and vacuum tube developer; Ray Dolby, whom we all associate with noise reduction in sound reproduction; Vint Cerf, one of the "fathers of the Internet"; and Don Knuth.

Like Cerf, Knuth was one of my professors at Stanford, and I am deeply indebted to his teaching. He is as close to a Renaissance man as you can get these days: a mathematician and computer scientist who pioneered the analysis of algorithm performance, but also invented many key algorithms for data manipulation; an accomplished player of multiple musical instruments, especially the organ; the developer of the TEX typesetting language (a derivative, LATEX, is still used today to typeset mathematical texts, including on Wikipedia and Blogger); the author of the famous series of textbooks called "The Art of Computer Programming," which he has promised finishing soon... since Vol. 3 appeared in 1973; and someone who, in the middle of all that, took time off his scientific pursuits to study religious texts.

Knuth used to devote the last lecture of his courses at Stanford to fielding questions from his students. True to form, he decided that his "Engineering Hero lecture" would not be a talk, but a Q&A session. He fielded questions from the packed auditorium on campus, as well as from the Internet audience who were following the event in real time. Dan Boneh, a cryptography expert who teaches at Stanford, was the moderator.

Below are my notes, providing a potentially biased summary of the session. I took the time to research links so that the reader can understand what mathematical or computational problems are mentioned in Knuth's answers. It should be noted that Knuth is often facetious, and will not bother giving a serious or complete answer to a question if he can get away by making a quip about it. He also tends to get off on a tangent frequently, and doesn't always return to the point. You can almost imagine the thoughts colliding in his head faster than he can express them -- nothing much has changed in this respect since I sat in his CS155 course, "Analysis of Algorithms," in 1973.

Q: If you wanted to give an interesting open problem of mathematics or computer science to a smart new student, what would it be?

A: One of the biggest open problems, of course, is P vs. NP. But really, this question is like asking a parent which of their children they love most. (Knuth searched for a specific problem in his own book, fumbled a bit, and gave up).

Q: In the 1790s, the French tried to introduce a decimal time system. How would a computer scientist design a time system?

A: Well, a decimal system could be very convenient, for example you could have birthdays more often.

Q: A lof of computer science students today don't know who you are. Do you think that's really bad?

A: I find it tragic when people don't know who Martin Gardner, who died about a year ago, was. He did a lot to illustrate mathematics in a really interesting way. For me, the important thing is not that they remember me, but that they remember the things I put in my books, and don't reinvent the same things over again.

Q: You developed TEX when computers were slow and user interfaces were not WYSIWYG. How would you do things differently today?

A: Actually, I wanted to design something that could capture texts in an archival format, something that would endure through technology changes. So I was never interested in changing it to track the evolution of technologies.

Q: I learned Pascal, then C, C++, and Java. After 25 years, I still think it is hard to write software, debug it, etc. Why, and do you think we'll finally some day overcome this?

A: I put a lot of ideas about this in my book "Literate Programming." I think the key is to write programs so that they can be easily understood by humans. But there is no silver bullet. Some techniques will make things better, but programming will still be hard.

Q: What is your opinion of quantum computing? If it ends up working, will it change the notion we have of algorithms?

A: Quite possibly. Quantum computers make some things we consider very hard today much easier. So if this works, I might be able to finish my books much faster.

Q: Your Web site has Frequently Asked Questions, but also Infrequently Asked Questions. Do you know why those questions are infrequently asked?

A: I am not an activist, but I think people should be talking more about some important issues like those. And my points must not have been made strongly enough, because I didn't get any hate mail.
[Note: I think that it may simply be that this page is hidden away and has not been mentioned enough. I just discovered it myself as a result of this question.]

Q: Can you talk about some mistakes you've made?

A: I once wrote a paper about the mistakes I made in TEX, and there were over 1000. One of them is that I based TEX on binary units internally, but decimal units externally (in what the user can see). The rounding error that results when someone specifies, for example, a space of 0.1 points, can cause some anomalies to happen.

Q: If you were starting as a Ph.D. student today, what area would you go into?

A: I'd probably be most attracted by computer graphics, because I like to use my left and my right brains at the same time.

Q: Do you think we will ever create real artificial intelligence?

A: I am not sure, and if it happens, we may not know it. But regardless of the result, aiming for that goal has been one of the most fruitful generators of ideas and of useful computer science problems.

Q: What are the biggest bottlenecks against improving the quality of life?

A: The eighth and latest volume of my collected papers ("Selected Papers on Fun and Games") is the "dessert issue" -- things I couldn't think of leaving out. I mention this because quality of life is about building a bridge to the other side of the river so we can cross it and go have fun on the other side. But I really don't enjoy the kind of ranking we tend to do, like "what is the most this, what's the best that?"

Q: What is your general approach about solving a tough problem?

A: One of the biggest problems I solved, with several other people, was the "giant component problem" in graph theory. When you randomly connect vertices in a set, the largest connected subgraphs stay small for a while, and then there's this sort of big bang explosion of the size of the largest component. We had to invent a way to look at the time steps in this process differently so we could see what was happening in slow motion, so to speak.
    I tend to train my brain to understand the domain of the problem by taking baby steps during the first weeks.

Q: What do you think about open-access (freely downloadable) texts and journals vs. traditional commercial publications?

A: I like to do control quality myself on my work, but I'm also now thinking of doing e-books. I'm in favor of open-access journals because I don't like it when commercial publishers make money from the free work of the authors and reviewers.

Q: Do you prefer analyzing a problem or solving it?

A: If I can solve it, I certainly feel an adrenaline rush when I succeed. But I can sometimes find the journey more interesting than the destination.

Q: A long time ago, you had said that computer science did not have enough hard scientific results to really qualify as a science. Are we there yet?

A: Yes, we must have passed that point 25 years ago. What happened is someone told me that a science needed to have 500 theorems or more to really be a science. I replied "or 500 interesting algorithms." We've had that many for a while now.

Q: When you write your programs, what language do you use?"

I use CWEB. I write about 5 programs a week, mostly pretty small ones, and that's what I use.

Sunday, May 8, 2011

Busy Week-End: "The Magic Flute" and "Ann"

In a rare coincidence, I had found some time ago two attractive performing arts events on this week-end's Austin calendar, and I decided to attend both.

"The Magic Flute" is a ballet by Stephen Mills, based on a compressed version of Mozart's opera. In fact, the score was written specifically for this piece, cutting the length in half and incorporating some of the sung parts into the orchestral suite. This was rather well done — the Queen of the Night's famous airs, for instance, were "sung" rather well by a trumpet.

I wish I could be as positive about the staging and the choreography. Mr. Mills, who spoke with the audience after the performance, said he was inspired to use shadows projected on a backdrop by something he saw at the Biennale in Venice a few years ago. The idea is okay... but leaves you to imagine what he could have done, for example, if Tamino and the dragon had danced their fight, instead of the lame shadow dragon we saw instead. Or if the water and the fire, through which Tamino and Pamina have to pass as their final induction rites, had been dancers in appropriately colored costumes, evoking flames and waves. There were tons of creative ideas that were just reduced to nothing through the limited and repetitive gimmick of this screen projection.

During his interaction with the audience, which he kept disappointingly short, Mr. Mills said that he "wanted to depart from the Egyptian and Masonic settings imagined by Mozart." I didn't think he got away from it at all, and indeed how could he? The story itself, even without any representation of a temple, implies a search for truth and purification, and a set of rites of passage, that clearly represent the Masonic traditions more than those of a traditional church. This is reinforced by the costumes chosen for Sarastro and his priests. As for the Egyptian references, they are, in most productions of the Flute, already limited to the minimum demanded by the libretto ("O Isis und Osiris," etc.) and since this production is voiceless, there wasn't much else to avoid.

But in fact my main criticism of the piece was that the costumes were overdone and did not serve the dancers or the work itself well. Schikaneder's libretto says that Papageno is dressed in the plumage of birds. That's fine for an opera, where the key artistic exploits are going to be first and foremost the singing, and then the music and the acting. In fact, a "busy" costume like Papageno's will be very good if the singer, instead of being a thin young man, is a bit older and corpulent. But this is dance. We're supposed to be fascinated by the movement of human bodies (none of which is going to be old or fat), not by the movement of layers of feathers!

In fact, I was reminded of Trey McIntyre's Peter Pan, created in 2004 for the Houston Ballet, in which I had also found that there was too much emphasis on complicated costumes. Note to choreographers: we are in the 21st century, and your predecessors started showing off the human body, with the minimum amount of costume needed to create visual harmony (and pass whatever decency standard seems currently applicable). There are lots of ways to suggest that Papageno is a birdcatcher, Tamino a prince (and the dragon a dragon, if there had been a tangible one) without covering them in things that hide their movements.

When Mr. Mills explained how the ballet came about, it was clear that he had spent months identifying and working with the people who could create the backdrop shadow effects, and he only started on the ballet five weeks before the first performance. This speaks a lot to the skills of the dancers, but unfortunately it doesn't say much about the choreographer, and I'm afraid it showed in the results. It was entertaining, it felt like a good family evening that would amuse the kids a lot, the orchestra and the score did an excellent job of reminding the people who know the work of all its superb musical moments... but this work seriously needs to be redone by someone in the tradition of the great choreographers of the last decades... a Stephen Morris, for example, or a Paul Taylor.

The next evening, I went to see "Ann," subtitled "An Affectionate Portrait of Ann Richards," a monologue written and played by Holland Taylor, a mostly TV and screen actress (Two and a Half Men, The Truman Show, etc.). At two and a half hours, including a fifteen-minute intermission, this one-woman show is quite a tour de force. Taylor has the accent, the mannerisms, and renders the often acerbic humor of the Governor. She had access to a lot of friends and colleagues of her subject, which gave her tons of materials to work with in order to tell the life, the challenges, the victories, and ultimately the gracious defeats (to two versions of hell: George W. Bush, and cancer) that marked her remarkable life.

The play was actually performed in several other Texas locations before hitting Austin for five sold-out performances in the last few days. It will be heading north later this year — it runs for three weeks in Chicago from Nov. 13 to Dec. 4. If it happens to be playing near you — and I assume that it will be back in Texas some time next year, unless if moves to Broadway or off-Broadway and stays there — don't miss it!

Saturday, March 26, 2011

Humor and Dance in the Capital


Washington, D.C., is well-known as a place where important people dance around issues, and sometimes unintentionally provide humor, but this post is about real dance and subtle humor, as seen in a performance of the Paul Taylor Dance Company at the Kennedy Center on Thursday.

All three pieces on tap that night were relatively recent productions (2009-2010) of the prolific choreographer himself. Brief Encounters, the first piece, was notable from the start by its fluidity of movement. The minimalist, stark black costumes (tight-fitting briefs for both sexes, plus bras for the women) and the often bright lighting combined to emphasize the movement and the individual bodies. Some of the frequent visual jokes in Mr. Taylor’s oeuvre were already noticeable, as well as some of the sexual ambiguity he often introduces.

Three Dubious Memories went one step further (so to speak) in that direction. In it, a love triangle is seen in turn from the perspective of each of the three participants. In each of the first three “memories” in question, a couple is lovingly minding its own business when a jealous third intrudes. A fight ensues. The contrast between the three recollections creates humor in itself: the Man in Blue remembers finding the Woman in Red and the Man in Green together, whereupon he beats up Green, and takes Red away. Laughter ensues when the second scene unfolds, and one realizes that Green remembers that he found Blue and Red together, beat up Blue, and took the girl. The third permutation brings an uncommon symmetry, as Red comes up on the two men frolicking together, goes from shock to anger, slaps them both and stomps away. The rest of the company forms a sort of Greek chorus that punctuates the storytelling. I can’t tell if it was the explicit allusion to a same-sex relationship that scared away the two ladies seated next to me, but they didn’t return after the second intermission.

The final piece, Also Playing, was a joyous romp made up of 15 short vaudeville acts danced with a faked incompetence that was sometimes reminiscent of the Trockadero Ballet. Of course, pretending to dance badly takes great mastery, even if the audience may have been laughing too hard most of the time to notice.

Sunday, February 20, 2011

Stanford Leading Matters (the end): Lecture on Stem Cells

Please read the Preamble of the previous post to understand the context of this article. As always, I encourage comments and responses. Remember that you can find and listen to the actual lecture on Stanford on iTunes.

"Stem Cells and the Promise of New Cancer Therapies" by Dr. Henry Weissman

A stem cell has a unique property, which is that when it divides, one of the two resulting cells is also a stem cell. Dr. Weissman took pains, at this point and several times later in his lecture, to debunk the myth that a stem cell can produce any tissue. He said that there are several types of stem cells (blood-forming, bone-forming, muscle-forming, etc.) and that they are not interchangeable at all.

Blood-forming stem cells can be obtained in three ways:
  • from the bone marrow
  • from "mobilized blood": you can give someone a drug that flushes stem cells into the bloodstream for just a few hours; during that time you can collect as many stem cells as you would normally get from as many as 100 bone marrow taps.
  • from umbilical cord blood — but again, people who propose "cures" derived from this source for all sorts of ailments are charlatans.
California has a law (Prop. 71, adopted in 2005) that allows a university to perform clinical trials of treatments that a company might not want to fund. This is what the Stem Cell Research Center at Stanford does.

The speaker described the potential to help fight blood-related and brain-related diseases with stem cells. However, there is a complex interaction between graft rejection mechanisms and stem-cell transplants. You need to circumvent the rejection mechanism to successfully implant foreign stem cells into a person.

Moving to the importance of this research for cancer, Weissman said that when a chemotherapy drug kills ordinary tumor cells but not the tumor stem cells (TSCs), the cancer will regenerate. Leukemia cells carry a CD47 protein that serves as a "don't eat me" signal to macrophages that would otherwise eliminate them. Therefore, antibodies targeted at CD47 seem promising, but so far experiments have only been conducted in the form of xenografts from humans to mice. In those experiments, the antibodies have arrested the development of the cancer cells transplanted into the mice.

In conclusion, Dr. Weissman pointed out that this research has a lot of implications because of the passionate opinions for or against the use of fetal stem cells. He did not shy from the controversy.

An audience member asked about potential solutions that could come from transplants from mice to humans. Dr. Weissman said that unfortunately, mice have about 150 viruses they live with, but which could potentially cause leukemia in humans. There are current efforts to raise virus-free pigs, because their hearts are about the same size as human hearts, so they would be suitable for transplants.

In answer to another question, the speaker said that his research center is performing clinical trials in the UK because patients are covered by a single national health insurance system. In the U.S., he would have to negotiate about the coverage with a number of private insurance companies, and after some of them deny coverage, the remaining sample population for the trials would be small.

Feel free to comment on these last three posts — either about the subject matter or about the usefulness of these notes.

Stanford Leading Matters, continued: Lecture on Innovation

Please read the Preamble of the previous post to understand the context of this article. As always, I encourage comments and responses. Remember that you can find and listen to the actual lecture on Stanford on iTunes.

"Harnessing Collective Wisdom" by Hayagreeva Rao, Professor of Organizations at the Stanford Graduate School of Business.

"Darwin's ideas of variations and selection apply to how the C-suite handles innovation."

Key questions about innovation in an enterprise: how do we encourage ideas without creating clutter? And how do we select some ideas, therefore kill some of them, without discouraging people from continuing to submit?

A study by McKinsey found that the way many companies measure innovation is flawed. Metrics and measures and useless without a good system of idea generation, which requires creative people and a positive climate, which in turn is created by good leadership.

In most companies, when you ask people what hinders innovation, you get two very different answers depending whom you ask:
  • people at the top think that there aren't enough ideas generated by people below
  • people in the trenches say that it is the top echelons of the company that act as a bottleneck
There needs to be a climate of confidence, trust and fun, enabling people to share their ideas and to try new things.

A good predictor of an organization's ability to innovate is how it handles failure. You have to:
  1. reward people for trying hard enough that they will fail a certain percentage of the time;
  2. make sure that failures are recognized quickly and that projects are stopped when they have failed, freeing resources to pursue others.
Peter Kim, head of the Whitehead Institute in Cambridge, Mass., used to tell people: "if you kill a project, I will give you stock options." At another company, the CEO frequently visits employees and hands them a business card, the back of which looks like a Monopoly "Get out of jail" card. Then he says: "Try something new! If it fails, give your boss this card, and I promise you won't get in trouble." At Amazon, Jeff Bezos gives a monthly award to an employee who did, without asking for permission, something that improved customer satisfaction.

Rite Solutions, a Rhode Island company, has created an internal stock market for ideas, called "Mutual Fun" (a pun on "mutual fund"). This virtual stock market trades cost reduction ideas ("savings bonds"), mildly aggressive ideas ("Bow Jones") and far-out innovations ("SPAZDAQ"). Once this market determines what are the most popular ideas, actual funds are allocated to them. Ideas are never killed by a corporate review board -- they are by the community when the corresponding shares don't sell. This approach also circumvents a common flaw of management review processes: senior managers tend to mostly empathize with people who are a lot like them.

At the end of his talk, Rao was asked about Open Innovation. He said that one size doesn't fit all. "You may not use the same incentives or the same rules with different actors."

Stanford Lecture on Energy

Preamble

In the last six months, I attended two instances of a series of events entitled "Stanford Leading Matters." This is a touring half-day conference that is going through 28 cities, mostly in the U.S. but with some foreign locations too. The purpose is to re-acquaint university alumni with the challenges that Stanford University is addressing. Of course, as a by-product of this goodwill, the university hopes to successfully appeal to their generosity and raise more money for the "Stanford Challenge," a $10 billion fundraiser.

Leading Matters was a rather stellar production, complete with making the meeting hall look like a scale model of the university's inner quad — sandstone arches and all. At every stop, Stanford University President John Hennessy spoke of the university's vision, which is no less than helping solve the world's toughest challenges; incredibly gifted and involved students provided their views in a panel moderated by Hennessy; professors gave lectures on important issues in today's world; and very good food could be sampled.

The 28th and last stop on this tour will be in Portland, Ore., at the end of May 2011. Because I was at a conference in Boston the week immediately preceding the Leading Matters event there, in late September 2010, I added a day to my trip and went. There were about 550 Stanford alumni, spouses and sometimes their college-bound children in attendance. Then, when Leading Matters stopped in Houston a month ago in January, I went again — in part because the faculty lectures and the student panel involved different people each time, therefore it wasn't a complete repetition.

What I am proposing to do in three successive blog posts is to give a summary of the lectures I personally attended (they can all be found and downloaded for free at "Stanford on iTunes" but some readers will appreciate a more personal touch).

Fixing the Energy System — Why Is It So Hard?" by Prof. James Sweeney

Sweeney is the Director of the Precourt Energy Efficiency Center at Stanford, thus named because it was founded thanks to a $30 million donation by Jay Precourt.

He stated that there are three drivers for a public energy policy:
  • Global climate change
  • National security
  • Economics
Concerning the first one, the world releases about 30 billion metric tons of carbon dioxide each year. To address CO2 emissions, we should focus on three things:
  • "decarbonizing" the generation of electricity
  • reducing the use of oil in transportation
  • improving how we use electricity
"Energy efficiency" means the economically efficient improvement of energy use — i.e., something that we can be willing to do not only because it is the right thing to do, but also because it will save money.

The possible improvements in the production or use of energy fall into several categories, in terms of their feasibility:
  • those for which technology advances are needed
  • those for which the technology exists but is currently too expensive
  • those that require new regulations
  • those against which there are no obstacles, other than our behavioral inertia
Why are some of the energy cost reduction opportunities not being pursued? Sweeney gave several examples:
  • Owners or rental property have no incentive to make improvements, for example in insulation: they cost money, but do not allow them to charge more in rent.
  • Consumers are not receiving enough information to understand their electricity usage; for example, a cable TV decoder/DVR box uses 45W when it is on, and almost the same when it is off.
The Google Power Meter is a free energy monitoring tool for households. It checks your pattern of electricity consumption, and reports the results a day later. However, apart from the fact that either your utility company must be part of the program (and only a few companies are), or you must install a monitoring device otherwise, Sweeney said that this tool "requires too much analysis by people, and feedback should be more immediate" if we are to obtain real changes in consumer behavior.

Electric utility companies, of course, do not have incentives to reduce consumption. Usually it is the opposite (although he should have mentioned that an increase in peak consumption may cost a utility more than they can recover, because they may need to bring online their most inefficient plants, the one they would otherwise keep shut off, in order to meet the demand). Instead, we should find ways to decouple volume from profit, which California has done through regulation and financial incentives.

For businesses, there should be software packages for energy use management, just like there is to manage and optimize other company resources (people, equipment, etc.). "What you don't measure, you can't manage."

After the speaker ended his talk, someone in the audience, who worked at the Department of Energy for ten years, suggested that there is a rift between blue (Democrat-leaning) and red (Republican-leaning) states: people in red states are opposed to energy management measures because these sound anti-business. Sweeney replied that he did not see this issue as a clear-cut Democrat-vs.-Republican attitude, but he also admitted that as a Republican he could be biased about it.

Sunday, December 19, 2010

Men and Near-Men of the Year

Wow... nothing posted since May 27. It would seem that my life got super-exciting since then and took me away from the keyboard. In fact, all that happened was that I was quite busy with consulting work from August to early November, and that we got a puppy in July. Other that that, it's the usual reason: procrastination. And perhaps a lack of something really exciting or infuriating to write about.

I received the TIME Magazine "Person of the Year" edition (and yes, I wrote "Men" in this post's title because I'm only going to talk about men here) and I was thinking of the meaning of some of the choices.

I've already heard the Luddites say that it was preposterous to bestow such a high honor on Mark Zuckerberg, but I think the point is that regardless of his actual personal qualities or limitations (disclosure: I have not seen The Social Network) the phenomenon that he and his company created is of great importance. I consult and write professionally about the importance of social media in the enterprise, and about the fact that companies that try to create an internal social network fail to realize that their employees want to be connected to a mix of colleagues and non-colleagues. Since Facebook and LinkedIn are the primary places that meet this need, they will not become faithful to an internal-only site. My latest paper on this, "If You Build It, They Won't Come" will be published by the Cutter Consortium after the turn of the year.

Now don't take me wrong: I'm interested in this enough that next time I walk by Antonio's Nut House, which is a couple of blocks from where I lived in Palo Alto in 1985-88 and 1993-2000, I may drop in to see if Mr. Zuckerberg is there (he may have to change his unofficial headquarters now that TIME has disclosed this habit), because I would absolutely not mind an autograph. It would be a change from my usual ballet dancers and violinists... OK, violinist. But Facebook (and LinkedIn) are now much bigger phenomena than any boy genius, and will probably not depend on their respective founders to continue (the usual counter-example to this is Stevc Jobs, of course).

The Chilean miners, los trenta y tres, almost made it. Some of them were truly heroic, others were well inspired to follow their leaders' advice. They are important not so much as 33 individual men, but because it showed how a country and the victims could take charge of a rescue operation that some thought was doomed from the start. The "Heckuva job, Brownie" exclamation was a gross sign of incompetence at all levels in the U.S. in 2005. Next time there is a Katrina-like catastrophe in this country, I hope we ask some of these miners, and perhaps the President of Chile and his cabinet, to come here and coordinate the rescue, since they know a thing or two about actually succeeding at a rescue.

And now, the man whose near-designation may be controversial, but who deserves some discussion: Julian Assange. Once again, here is a person who seems to have some deep flaws, and who is alleged to be a repeat sex offender; but regardless of his personality, he is the visible face of a movement to expose diplomatic secrets. I am quite perturbed, frankly, by the attitude of governments and banks toward the Wikileaks organization and even toward Assange himself. First, it has not been proven that Wikileaks participants procured any information illegally. The people who gave them the information may have violated their terms of employment by various governments, so go after them if you have legal standing, but don't blame the messenger.

Secondly, I find the diplomatic niceties tiresome. We all know that foreign dignitaries, be it Karzai, Merkel, or Sarkozy, are not angels. I find it refreshing to hear about the incompetence of the first, the imperiousness of the second, and the vanity of the third. Instead of suing someone for revealing this, the U.S. diplomats should say "oops" and move on, and the big shots who have been described in unflattering terms should look at what they can do to govern more competently or more humbly.

Thirdly, and the TIME article on Assange points it clearly, the financial institutions clearly acquiesced to apply punitive measures to Wikileaks as soon as Ms. Clinton's organization cleared its powerful throat. Paypal suspended processing payments because of "illegal activity" even though it is an allegation that has not been proven. Visa and Mastercard stopped too, even though they still let the Ku Klux Klan have accounts with them. And a Swiss bank has frozen Mr. Assange's account because he is not a permanent resident of Switzerland -- a rule they constantly overlook in other cases. This is just spineless capitulation and, until and unless an actual illegal action has been proven, a sad day for freedom of speech.

Even the fact that he was arrested in London just after the outcry from the U.S. government seems rather suspicious. I can imagine the debates at Scotland Yard: "if we arrest him now, we'll be called lackeys of the U.S. If we don't arrest him, we may be derelict in our duty. Oh, what to do, what to do? Here's an idea: let's arrest him and then free him up on bail. Let's see how this plays out."

So this Person of the Year issue is more interesting than the usual crop. I was especially glad that the designation didn't go to some Tea Party politician, because I doubt that we will remember them much in a few years, and it would have been another U.S.-centric story. Instead, in all the cases I mentioned (Zuckerberg, the miners, and Assange) we have people whose actions have inspired or disturbed the entire world, and who evoke in us some fundamental values that are not only American: connectedness, the love of life, and the right to know what our governments would like to hide.