Tech Talk

This week, we're taking a look at the enormous leveraged buyout of TXU Corp., the Texas utility giant, by a private investment group led by Wall Street powerhouse Kohlberg Kravis Roberts and Co. (see Tuesday's Tech Talk). Today, we'll review some of the early reaction from environmentalists, who, as one would expect, greeted the news with enthusiasm, characterizing the big takeover as a big deal.

The US $45 billion acquisition of TXU—a record in the equities market—provides up front that the new management will withdraw its proposals before Texas authorities to commission eight of 11 new coal-fired power plants over coming years. This concession immediately drew cautious praise from leaders in the environmental movement. Both the Natural Resources Defense Council and Environmental Defense, leading ecological advocacy organizations, were participants in negotiations that led to the transaction.

"This turnaround marks the beginning of a new, competitive focus on clean, efficient, renewable energy strategies to deliver the power we need while cutting global warming emissions," said Frances Beinecke, president of the NRDC. "It is a big step forward for the State of Texas and for the American energy economy as a whole."

In its own press release, the NRDC's chief climatologist, David G. Hawkins, said: "What we're witnessing is the beginning of the end of investments in old-fashioned coal plants. These are very big investors coming to the energy table with very big ideas about where the competitive market is heading. Strategies to fight global warming and save energy are crucial for anyone hoping to succeed in today's electricity industry."

The president of Environmental Defense, Fred Krupp, said his association viewed the compromise on carbon dioxide emissions reached in the deal as sufficient to settle a pending federal lawsuit against TXU. "The debate over this issue has been a top priority for Environmental Defense, and we plan to work just as hard with the new TXU to implement this agreement," he noted. Krupp added, in a separate statement, that he considered the occasion "a truly historic day in the fight against global warming."

Still, not all of those concerned with the quality of the environment were pleased by the deal. The Associated Press reported Monday that the Sierra Club related that it doesn't understand why TXU's new management would proceed with the three other pulverized-coal plants, including two near Dallas, in the face of strong opposition. That stance was echoed by Dallas Mayor Laura Miller, who leads a coalition of Texas cities that opposes TXU's coal facilities. She told the AP, "Environmental Defense blessed those two stacks when they don't have the authority to do that."

Tomorrow, we'll examine the question of just where the new TXU intends to find the resources to generate the electricity needed to meet demand that the old TXU wanted to produce with the eight coal power plants that now will no longer be built.

Senior Associate Editor Steven Cherry looks into a Web site that has something nice to say about nearly every integer from 0 to 9999.

Steven Cherry

Now this is what makes the Web special.

Erich Friedman, an Associate Professor of Mathematics at Stetson University, in DeLand, Florida, has a page called "What's Special About This Number?" It has a list of numbers from 0 to 9999 and states a unique property for each one.

Here are a few examples:

5 is the number of Platonic solids.

6 is the smallest perfect number.

7 is the smallest number of faces of a regular polygon that is not constructible by straightedge and compass.

8 is the largest cube in the Fibonacci sequence.

Don't know what a "perfect" number is? The site links to other pages around the Web that give detailed explanations. Mathematicians have tons of these technical terms (many of which use ordinary words, like "perfect").

I had a major in math, and while I knew what a perfect number was, there are plenty of terms and ideas here that I didn't learn in college, such as "hexomino."

Friedman has an unfortunate tendency to link to Wolfram's excellent MathWorld site, which gives extremely complicated, mathematically rigorous explanations ("perfect numbers are positive integers n such that n==s(n), where s(n) is the restricted divisor function (i.e., the sum of proper divisors of n), or equivalently sigma(n)==2n, where sigma(n) is the divisor function (i.e., the sum of divisors of n including n itself).")

But the Web is rife with simpler ones, and Google knows where they are. For practical purposes, a perfect number is any whose factors, not counting itself, add up to itself: 1+2+3=6. 1, 2, 4, 7, and 14 add up to 28, which is the second perfect number.

Anyway, Friedman's page is delightful. He doesn't find an interesting fact for every number between 0 and 9999, but he does for 2849 of them, which is remarkable. For example, 35 is the number of hexominoes, which are structures made up of six squares. The Wolfram explanation is more helpful here and it has a diagram:

Did you know that 38 is the last Roman numeral when written lexicographically? Or that 40 is the only number whose letters are in alphabetical order? That 727 has the property that its square is the concatenation of two consecutive numbers? I sure didn't.

The best thing about this site is that Friedman asks:

If you know a distinctive fact about a number not listed here, please e-mail me.

We have a suggestion for the number 2600: It's the number of nuclear power plants needed to equal the energy contained in 1 cubic mile of oil, which happens to be amount the world currently uses annually. And 200 is the number of Three Gorges Dams it would take, and 5200 the number of coal-fired plants. You see, Spectrum's editors are almost as obsessed with numbers as Friedman, as they showed in an article last month, "Joules, BTUs, Quads-Let's Call the Whole Thing Off.")

Even without our energy numbers, over time, each of Friedman's 7151 missing factoids will be filled in, some of the existing ones will be improved, and the list will expand into the five-digit numbers and beyond.

The Web can be the sum of all knowledge, assembled collaboratively, lovingly, in a way that's never been possible before. We've put together a remarkable fraction of the world's knowledge so far. Imagine what it will look like in 100 years (the smallest square which is also the sum of 4 consecutive cubes), if we don't get in its way.

As battlefield technology accelerates into the 21st century, the expertise needed to design and develop new systems that can provide the margin of victory in combat grows year by year. This cold reality has forced the U.S. Department of Defense (DoD) to question the way it procures complex weaponry to see if it is making the right technical decisions at the right time. It's answer to it's own question, put plainly, is no. In a report in the March issue of National Defense Magazine, a Pentagon insider argues that the DoD needs more technology-savvy people to judge the merits of new defense programs right from the start.

James F. O'Bryon, a former Defense Department deputy director of operational test and evaluation, writes that, despite overhauls to the procurement process over many years, problems continue to plague technically complex military systems. In speaking with the U.S. government's top weapons evaluator and other industry sources, O'Bryon concludes that the need for more involvement by engineers early in the development process of new weapons has become a pressing concern at the Pentagon.

"The cost of programs is generally driven by decisions that are made in the initial 10 to 15 percent of the work that's done," Charles McQueary, the DoD's director of operational test and evaluation told O'Bryon. "Developmental testing is the place to find problems. Operational testing should be the period of confirmation, not a period of discovery."

McQueary added that poor decisions are often made because the government simply lacks enough technical expertise to oversee complex programs. "You need top-notch engineering capability in the government. Unfortunately, the government has lost quite a bit of its systems engineering capabilities, and when you lose these capabilities, you tend to do too much designing without an adequate knowledge of the trade-offs," he said.

In a kind of brain drain that seems to have snuck up on military authorities, much of the government's engineering expertise in weapon design analysis has trickled away in recent times, according to McQueary. "There is a statute that requires [the undersecretary for acquisition] to certify technology readiness levels. However, the government has backed off in some of its [other] acquisition reforms ... resulting in some of the developmental test people going away. There was little work for them to do since their work was contracted out."

When asked what the Pentagon should do to remedy the problem of developmental testing vigilance, attendees at the annual National Defense Industrial Association test and evaluation conference suggested steps such as: giving testers more authority to halt systems that are unsuitable, recognizing system improvements that may not be large or significant, improving the way test funding is handled, making operational tests more mission-oriented, and placing more emphasis on operational testing during concept development.

Sounds like the U.S. military needs more of what armies and navies have needed since the advent of armed conflict: skilled engineers.

The Association for Computing Machinery (ACM) today announced that it will present the 2006 A.M. Turing Award to Frances E. Allen, an IBM Fellow Emerita. Allen is the first woman to be so honored in the 40-year history of the prize for contributions to the advancement of computer science. In its announcement, the ACM cited Allen's 'fundamental contributions to the theory and practice of program optimization, which translates the users' problem-solving language statements into more efficient sequences of computer instructions'.

Allen, who retired from IBM in 2002, is known as a software pioneer in the areas of compilers, code optimization, and parallelization. 'Her contributions also greatly extended earlier work in automatic program parallelization, which enables programs to use multiple processors simultaneously in order to obtain faster results,' the announcement continued. 'They have contributed to advances in the use of high-performance computers for solving problems such as weather forecasting, DNA matching, and national security functions.'

"Fran Allen's work has led to remarkable advances in compiler design and machine architecture that are at the foundation of modern high-performance computing," said Ruzena Bajcsy, Chair of ACM's Turing Award Committee, and professor of Electrical and Engineering and Computer Science at the University of California, Berkeley. "Her contributions have spanned most of the history of computer science, and have made possible computing techniques that we rely on today in business and technology."

"Over the years, this foundation has enabled the advance of programming productivity based on the co-evolution of higher level programming language and optimization technologies," said Intel Corp.'s Vice President of Research Andrew A. Chien. "It is particularly timely that this award comes as parallel computing is becoming an element of the most pervasive of computing platforms—laptop and desktop personal computers—and the opportunities for new and important contributions to parallel programming and efficient implementation abound."

Allen told the Associated Press that it was "high time for a woman" to win the Turing Award. She then, though, quickly added: "That's not why I got it."

Sometimes it seems the American public has been regressing in its understanding of science lately, especially with all the media coverage of debates about climate change and evolution, among a host of controversial issues. However, a researcher who spends his time tracking the public's knowledge of scientific concepts last week said that, in reality, just the opposite is taking place: U.S. citizens are more fluent in science now than in recent decades.

At the annual meeting of the American Association for the Advancement of Science in San Francisco, Jon D. Miller, a professor at Michigan State University in multidisciplinary studies, argued that Americans are more scientifically literate than they were 20 years ago, but that they still have a lot of room for improvement.

"A slightly higher proportion of American adults qualify as scientifically literate than European or Japanese adults, but the truth is that no major industrial nation in the world today has a sufficient number of scientifically literate adults," Miller said. "We should take no pride in a finding that 70 percent of Americans cannot read and understand the science section of the New York Times."

Miller noted that about 28 percent of American adults currently qualify as scientifically literate, an increase from around 10 percent in the late 1980s and early 1990s. He attributed much of the improvement in scientific understanding to policies at U.S. universities in recent years that require students to take science courses, as well as informal science education resources, such as science magazines, news magazines, science museums and the Internet.

"Although university science faculties have often viewed general education requirements with disdain, analyses indicate that the courses promote civic scientific literacy among U.S. adults despite the disappointing performance of American high school students in international testing," he said.

Miller emphasized that a good grounding in science is fundamental for the well being of the public. He listed several reasons, including the need for a more sophisticated work force and for more scientifically literate consumers, as well as an intelligent electorate who can help shape public policy.

"Over recent decades, the number of public policy controversies that require some scientific or technical knowledge for effective participation has been increasing," he stated. "Any number of issues, including the siting of nuclear power plants, nuclear waste disposal facilities, and the use of embryonic stem cells in biomedical research, point to the need for an informed citizenry in the formulation of public policy."

To be classified as "scientifically literate," Miller said one must be able to understand approximately 20 of 31 scientific concepts and terms similar to those that would be found in articles that appear in the New York Times weekly science section and in an episode of the PBS program "NOVA."

Miller's work points to an improvement in the public's understanding of technical matters in general, but a passing grade from only 3 in 10 in the U.S. is setting the bar rather low. As he noted in his presentation, while Americans are holding their own, they are not even close to where they should be.

Our cover this month features a "Mystery Man." However, he isn't the mystery. His job is to investigate mysteries. In our special issue on dream jobs, Senior Editor Harry Goldstein demystifies our first selectee in "Andrew Paris: Electric Detective" by explaining what makes the occupation of forensic engineer so fascinating.

IEEE Member Andrew Paris investigates electrical and electronic devices suspected of catastrophic malfunction for Anderson Engineering of New Prague Inc., outside of Minneapolis. When asked by strangers what he does for a living, Paris has a brief answer that cuts right to the chase: "Have you seen the show 'CSI'?" And that gets their immediate attention.

Goldstein writes that forensic engineering isn't quite as glamorous as the television show makes crime scene investigation out to be, but for someone who loves solving technically challenging puzzles, it's just as compelling. Picking apart burnt lighting ballasts from a house fire, photographing an accident scene, questioning witnesses, writing reports, and preparing cases for trial, a forensic engineer wears many hats. "There's something new coming at you every day," Paris told Goldstein.

After graduating from North Dakota State University, in Fargo, with a B.S. in electrical engineering in 2002, Paris was further schooled by a former professor of his. "You're basically a neophyte for two years until you can start doing things on your own," he said to us. "It takes that long to go to enough scenes to understand the process and the legal issues." The apprenticeship paid off. Today, at 26, Paris is a top investigator at Anderson.

His work can range from investigating possible arson fires to occupational injuries caused by faulty equipment. They have one thing in common, though. They start out as mysteries and must be solved through rigorous field study accompanied by thorough research back at the office, not unlike what is depicted in the hit TV show. It's a real-world comparison he obviously gets a big kick from, along with the opportunity to uncover the truth.

"It's satisfying to know that you've helped people who have been wronged and that you're part of ensuring that products are safe," Paris told Goldstein with pride. "If nobody did anything about it, what incentive would there be to make a safer product?"

That's enough of an explanation for us.

If you're old enough, you'll remember the laughable ads in comic books for eyeglasses that promised to give you X-ray vision, just like Superman. Now, scientists at Brown University, in Providence, R.I., are developing a technology that could really deliver on that far-fetched vow. According to a recent statement to the media, a multi-disciplinary team from the college is hard at work on creating a system that would, for example, enable doctors to study damaged bones and tissue in motion to plan the most effective surgical approach to a patient's treatment.

The researchers from Brown are calling the new technology CTX, as it combines techniques from computerized tomography (CT) and X-ray fluoroscopy. At present, short of exploratory surgery, biomedical scientists have mainly a single advanced approach to studying hidden anatomical features in action. This involves using a procedure known as cinefluoroscopy, in which a fluoroscope and camera record two-dimensional moving images of the interiors of subjects (and which has been exaggerated itself in sci-fi entertainment such as the movie "Total Recall"). CTX, on the other hand, offers the hope of much more robust imaging in three dimensions, with software capable of rendering precise details from multiple perspectives. The Brown scientists say CTX should deliver images that will be able to track 3D skeletal movements with 0.1 millimeter accuracy, offering the equivalent of 1000 CT images per second.

Today's guest blogger is Anders Frick, a technology journalist from Lund, Sweden, who is working at IEEE Spectrum as part of a program sponsored by Vinnova, a Swedish governmental agency that encourages innovation.

Anders Frick

Toys are for kids, but toy fairs are not. Strangely enough, the American International Toy Fair, the biggest of its kind in the Western Hemisphere, bars all kids. To get past the guards, even journalists must prove they are at least 18 years old. And so middle-aged men and women are all you see as you go up and down the aisles, here in New York City's Javits Center, a glass-and-steel compound large enough to park airliners in.