Category: Uncategorized

In a Los Angeles Times commentary, Margaret Wertheim writes about one of the growning problems in modern mathematics:

People are now claiming proofs for two of the most famous problems in mathematics — the Riemann Hypothesis and the Poincare Conjecture — yet it is far from easy to tell whether either claim is valid. In the first case the purported proof is so long and the mathematics so obscure no one wants to spend the time checking through its hundreds of pages for fear they may be wasting their time. In the second case, a small army of experts has spent the last two years poring over the equations and still doesn’t know whether they add up.

She claims that mathematics is evolving into a postmodern study where, according to Philip Davis, emeritus professor of mathematics at Brown University, mathematics is “a multi-semiotic enterprise” prone to ambiguity and definitional drift.

What I found interesting was this bit:

The [four-colour map] problem was first stated in 1853 and over the years a number of proofs have been given, all of which turned out to be wrong. In 1976, two mathematicians programmed a computer to exhaustively examine all the possible cases, determining that each case does indeed hold. Many mathematicians, however, have refused to accept this solution because it cannot be verified by hand. In 1996, another group came up with a different (more checkable) computer-assisted proof, and in December this new proof was verified by yet another program. Still, there are skeptics who hanker after a fully human proof.

I don’t know if I buy the postmodern angle, but I believe she’s right about the practice of mathematics changing. It seems to me that as the complexity of mathematical problems grows beyond the abilities of human minds, we are going to require computers to prove our theorems for us. My prediction: the job of mathematicians will eventually turn into proving that theorem-proving software is correct.

Bjorn Freeman-Benson has started blogging about the open source development of Eclipse. I’ve had the pleasure to sit through some talks that Bjorn’s given at the past couple EclipseCons, including the tutorial he gave with Darin Wright on the debug framework in February. To say that he’s a bright and articulate guy is a bit of an understatement. I’m looking forward to hearing what he has to say.

When I Men are from Mars, Women are from Venus, I was sorely unimpressed with its (lack of) scientific rigour. I came across a Rebuttal from Uranus, a passionate challenge to Gray’s assertions, but found it equally lacking in scientific study.

Happily, Edge has run several stories recently relating to gender differences including a debate between Pinker vs. Spelke on the results of recent gender difference studies, John Gottman’s Mathematics of Love, and Simon Baron-Cohen’s Assortative Mating Theory. Good stuff.

For an embarrassingly long time, I’ve been promising Mandy that we would buy a digital camera. Last night, we finally took the plunge.

Our main criteria was price. We wanted to spend less than $300. After some surfing around TigerDirect, I found four Kodak cameras that worked: CX6445, CX7530, DX7440, and LS753.

The reviews of the DX7440 and LS753 on Imaging Resource and the dearth of reviews on the CX cameras convinced me that we were looking at a shootout between the DX7440 and the LS753.

Mandy and I compared the two. The LS753 is a 5 megapixel camera, but only has a 2.8x optical zoom. The DX7440 only has 4 megapixels, but a 4.0x optical zoom. We don’t expect that we’ll be printing out to many large (>5×7) photos, so the extra megapixel of resolution didn’t seem all that important. I wanted more than the 2.8x optical zoom of the LS753.

So the DX7440, it was. It should be arriving this week. Expect to see some shots on Flickr by the weekend.

Apparently Eclipse 3.1M7 is out. What’s new?

Space is full of cosmic radiation. Computer chips that go into satellites and other space gear need special shielding to protect them from the single-event effects, or SEEs, of cosmic radiation, but here on Earth, chip designers haven’t had to worry about it too much because the atmosphere reflects most of it away.

According to an article in EDN, with the increased densities of modern computer chips, SEEs are becoming the dominant reliability-failure mechanism here on Earth.

Paul Dodd, another foremost expert on SEEs and acting manager for the radiation-effects department at Sandia National Laboratories (Albuquerque, NM), says that commercial designs are also more frequently encountering SEEs but that designers are commonly missing or misidentifying them as other failures. “It could be happening on everyone’s PC, but instead everyone curses Microsoft,” says Dodd. “Software bugs probably cause a lot of those blue-screen problems, but you can trace some of them back to radiation effects.” And designers cannot yet quantify the breadth of the problem because, as IC-design and EDA consultant Pallab Chatterjee points out, “It is something companies don’t brag about.”

As chip densities and clock speeds continue to increase, it seems chips will become more susceptible to SEEs. It will be interesting to see how this affects Moore’s Law.

According to this New Scientist story, the Marburg virus outbreak in Angola that started in March has now spread into all age groups. Prior to this the virus was mainly infecting children. The virus has killed 271 of 311 reported cases.

Should I be surprised that this story isn’t getting the same kind of coverage in the mainstream media that, say, the SARS outbreak received?

Robert Nagle blogs about John Taylor Gatto’s online book, An Underground History of American Education. From the prologue:

The new dumbness is particularly deadly to middle- and upper-middle-class kids already made shallow by multiple pressures to conform imposed by the outside world on their usually lightly rooted parents. When they come of age, they are certain they must know something because their degrees and licenses say they do. They remain so convinced until an unexpectedly brutal divorce, a corporate downsizing in midlife, or panic attacks of meaninglessness upset the precarious balance of their incomplete humanity, their stillborn adult lives. Alan Bullock, the English historian, said Evil was a state of incompetence. If true, our school adventure has filled the twentieth century with evil.

The way he rails against the public education system, you might never guess Gatto was once New York State Teacher of the Year.

New Scale Technologies makes miniature piezoelectric motors:

Piezoelectric actuators in the SQUIGGLE motor ultrasonically vibrate a threaded nut, producing an orbital motion. The nut vibration directly rotates a mating threaded screw to create precise linear motion — with no parasitic drag, no backlash, and very high stiffness. The motor holds its position with the power off.

This simple, robust piezo motor generates no magnetic fields, is vacuum compatible, and can be made from non-ferrous metals for use in MRI, scanning electron microscopy and focused ion microscopy applications.

Combine some of these with a miniature low-power DSP, and you have the makings of one tiny robot. What on earth you’d want a miniature robot for is beyond me, but it comforts me to know its possible.

Fulcrum Microsystems developed a clockless semiconductor design methodology. Which allows them to design processors that run much faster than clocked chips and with lower power.

Back in 1994, near the end of the digital design course I took in university, the professor introduced the class to asynchronous design. Our professor showed us some tricks for solving simple problem with asynchronous logic, but told us that the technique wasn’t feasible for large scale design at the time, because it was terribly complicated with race conditions. It seems Fulcrum has solved the problem, and even designed some rather complex chips with it.

They raised $20 million on Monday.