What Are Your Most Critical Test Problems?

Posted on by

Every test engineer, design engineer, and manager must address a unique mix of challenges. At what stage do you have the most difficulty? Why? How do you distribute the responsibilities? Can you get designers to create products that you can test? How easily can you get management to approve capital

TAM 8 interview: best way the world will end? | Bad Astronomy

Posted on by

While I was at The Amaz!ng Meeting 8, attendee Scott Carnegie (from the Winnipeg Skeptics) grabbed a moment with me to ask me which of the ways the world might end would look coolest from the ground. Here’s what I said:

I sometimes wonder what it would be like to be able to actually see a comet in the sky, and know that in some amount of time, months say, it will 100% for sure and for real hit the Earth. What would happen? Riots? Panic? Or acceptance? Probably disbelief until the last minute, and then panic.

Yikes. Good thing there’s nothing out there so far that can hit us. But that day may yet come… and I hope by then we have a space program in place that can take care of it.


The Ignored Disaster: Exploding Refrigerators

Posted on by

The planet seems to be consumed by disasters these days, natural and man-made. From massive oil spills in the Gulf of Mexico to earthquakes to erupting volcanoes. Yet there's another potential disaster looming in either the frozen food section of your local mega-mart, or even the residential model r

Love of Logic [Science Tattoo] | The Loom

Posted on by

Cantors proof tattoo600

Melissa writes,

I have a mathematical tattoo on my left forearm. It’s in Frege’s notation (from “Grundgesetze der Arithmetik”), which was one of the first modern logical notations. If it were written on a flat surface, it would start with the short vertical line, which is the assertion sign. What it asserts is: If {Cantor’s theorem} then {heart}.

Cantor’s Theorem says that the power set of any set is strictly larger than the set itself. (The power set of a set is the set of all its subsets.) For finite sets, this is pretty obvious; for example, the power set of {1,2} is {{}, {1}, {2}, {1,2}}. In general, if a finite set has n members, its power set has 2^n. But Cantor’s Theorem is also true for *infinite* sets, which is kind of unexpected. After all, the set of all even numbers is the same size as the set of all numbers — why does the power set of the set of all numbers have to be bigger?

That’s why the proof of the theorem is so cool. It proves it for finite sets and infinite sets, no matter how huge, at the same time. You start by assuming that some arbitrary set S has the same number of members as its power set P(S). That is, assume there’s a one-one function f which maps the members of S to the members of P(S). Now consider the set D, which consists of all and only the members of S that don’t get mapped to a set of which they’re a member. (So, for instance, if 7 is a member of S, and f(7) = {4, 5, 12}, then 7 is in D because it’s not a member of f(7).) D is a subset of S, so it’s a member of P(S). That means that f maps some member of S, call it d, to D. But: is d in D or not? If it is, then it’s a member of f(d), so by the definition of D, it’s not in D. If it’s not in D, then it’s not a member of f(d), so, again by the definition of D, it’s in D. Either way leads to a contradiction, and there’s only one way out: it’s not possible to have a one-one function from any set to its power set. QED! (Of course, you also have to prove that P(S) can’t be *smaller* than S, but that’s easy.)

When I saw how short and simple (and beautiful!) the proof of such a powerful theorem was, I knew I could spend the rest of my life doing set theory and logic. So last year, when I got my bachelor’s degree in philosophy and went on to grad school, I celebrated by getting the theorem tattooed on my arm. As for the tattoo itself, it’s easiest to read from the bottom. The stuff on the right-hand side of the ‘=’ means: for all a, if a is in r, then a is in u. (In other words, r is a subset of u.) The whole bottom line means: for all r, r is in v if and only if it’s a subset of u. (So v is the power set of u.) The bottom line and the one above it together mean: if v is the power set of u, then v is strictly bigger than u. So those two lines state Cantor’s Theorem, and the whole tattoo means: if Cantor’s Theorem, then {heart}. (Incidentally, I got that heart symbol from an illustration in “Alice in Wonderland”. It’s the top of the King of Hearts’s crown.)

Click here to go to the full Science Tattoo Emporium.


From the Vault: Clint Is Dead, Long Live Clint! | The Loom

Posted on by

[A post from 2005 I'm fond of]

Clint chimp.jpgClint, the chimpanzee in this picture, died several months ago at a relatively young age of 24. But part of him lives on. Scientists chose him–or rather, his DNA–as the subject of their first attempt to sequence a complete chimpanzee genome. In the new issue of Nature, they’ve unveiled their first complete draft, and already Clint’s legacy has offered some awesome insights into our own evolution.

The editors of Nature have dedicated a sprawling space in the journal to this scientific milestone. The main paper is 18 pages long, not to mention the supplementary information kept on Nature’s web site. In addition, the journal has published three other papers that take a closer look at particularly interesting (and thorny) aspects of the chimpanzee genome, such as what it says about the different fates of the Y chromosome (the male sex chromosome) in chimpanzees and humans. Other scientists offer a series of commentaries on topics ranging from brain evolution to chimpanzee culture. The journal Science has also gotten in on the action, with a paper comparing the expression of chimp and human genes as well as comments on the importance of chimpanzee conservation and research. (Thankfully, some of this material is going to be made available online for free.)

Why all the attention to the chimpanzee genome? One important reason is that it can tell us what parts of the human genome make us uniquely human–in other words, which parts that were produced by natural selection and other evolutionary processes over the past six million years or so, since our hominid ancestors diverged from the ancestors of our closest living relatives, chimpanzees. (Bonobos, sometimes known as pygmy chimpanzees, are also our first cousins, having split off from chimpanzees 2-5 million years ago.) Until now, scientists could only compare the human genome to the genomes of more distantly related species, such as mice, chickens, and fruit flies. They learned a lot from those comparisons, but it was impossible for them to say whether the differences between humans and the other species were unique to humans, or unique to apes, or to primates, or to some broader group. Now they can pin down the evolutinary sequence much more precisely. Until scientists rebuild the Neanderthal genome–if they ever do–this is going to be the best point of comparison we will ever get. (For more of the background on all this, please check out my new book on human evolution, which will be out in November.)

The analysis that’s being published today is pretty rudimentary. It’s akin to what you’d expect from a reporter who got to spend an hour flipping through 10,000 pages of declassified government documents. But it’s still fascinating, and I’d wager that it serves as a flight plan for research on the evolution of the human genome for the next decade.

First off, scientists can get a more precise figure of how different human and chimpanzee DNA is. In places where you can line up stretches of DNA precisely, there are 35 million spots where a single “letter” of the code (a nucleotide) is different. That comes to about 1.2% of all the DNA. The scientists also found millions of other spots in the genomes where a stretch of DNA had been accidentally deleted, or copied and inserted elsewhere. This accounts for about a 3% difference. Finally, the scientists found many genes that had been duplicated after the split between humans and chimps, corresponding to 2.7% of the genome.

By studying the human genome, scientists have also gotten a better picture of the history of the genomic parasites that we carry with us. About half of the human genome consists of DNA that does not produce proteins that are useful to our well-being. All they do is make copies of themselves and reinsert those copies at other spots in the genome. Other animals have these virus-like pieces of DNA, including chimpanzees. Some of the genomic parasites we carry are also carried by chimpanzees, which means that we inherited them from our common ancestor. Many of these parasites have suffered mutations that make them unable to copy themselves any longer. But in some cases, these parasites have been replicating (and evolving) much faster in one lineage than the other. One kind of parasite, called SINES, have spread three times faster in humans than in chimps. Some 7,000 genomic parasites known as Alu repeats exist in the human genome, compared to 2,300 in the chimp genome. While a lot of these parasites have no important effect on our genome, others have. They’ve helped delete 612 genes in humans, and they’ve combined pieces of some 200 other genes, producing new ones.

In some cases, the interesting evolution has occurred in the chimpanzee lineage, not in our own ancestry. Scientists have noted for a long time that the Y chromosome has been shrinking for hundreds of millions of years. Its decline has to do with how it is copied each generation. Out of the 23 pairs of our chromosomes, 22 have the same structure, and as a result they swap some genes as they are put into sperm or egg cells. Y chromosomes do not, because their counterpart, the X, is almost completely incompatible. My Y chromosome is thus a nearly perfect clone of my father’s. Mutations can spread faster when genes are cloned than when they get mixed together during recombination. As a result, many pieces of the Y chromosome have disappeared over time, and many Y genes that once worked no longer do.

Scientists have discovered that Clint and his fellow chimpanzee males have taken a bigger hit on the Y than humans have. In the human lineage, males with mutations to the Y chromosome have tended to produce less offspring than those without them. (This is a process known as purifying selection, because it strips out variations.) But the scientists found several broken versions of these genes on the chimpanzee Y chromosome.

Why are chimpanzees suffering more genetic damage? The authors of the study suggest that it has to do with their sex life. A chimpanzee female may mate with several males when she is in oestrus, and so mutations that give one male’s sperm an edge over other males are ben strongly favored by selection. If there are harmful mutations elsewhere on that male’s Y chromosome, they may hitchhike along. We humans are not so promiscuous, and the evidence is in our Y chromosome.

As for the mutations that make us uniquely human, the researchers point out some suspects but make no arrests. The researchers found that a vast number of the differences between the genomes are inconsquential. In other words, these mutations didn’t have any appreciable effect on the structure of proteins or on the general workings of the human cell. But the scientists did identify a number of regions of the genome, and even some individual genes, where natural selection seems to have had a major impact on our own lineage. A number of these candidates support earlier studies on smaller parts of the genome that I’ve blogged about here. Some of these genes appear to have helped in our own sexual arms race; others created defenses against malaria and other diseases.

When scientists first lobbied for the money (some twenty to thirty million dollars) for the chimp genome project, they argued that the effort would yield a lot of insight into human diseases. The early signs seem to be bearing them out. In their report on the draft sequence, they show some important genetic differences between humans and chimpanzees that might have bearing on important questions such as why we get Alzheimer’s disease and chimps don’t and why chimpanzees are more vulnerable to sleeping sickness than we are, and so on.

There is also a lot of variation within our own species when it comes to disease-related genes, and here too the chimpanzee genome project can shed light. The researchers show how some versions of these genes found in humans are the ancestral form also shared by chimpanzees. New mutations have arisen in humans and spread in the recent past, possibly favored by natural selection. The ancestral form of one gene called PRSS1, for example, causes pancreatitis, while the newer form does not.

But our genetic defenses and weaknesses to diseases aren’t really what we’d like to think make us truly, uniquely human. The most profound difference between the bodies of humans and chimpanzees is the brain. Much of the evolution that’s been going on in genes expressed in the brain has been purifying. There are a lot of ways to screw up a brain, in other words. But some genes appear to have undergone strong positive selection–in other words, new mutation sequences have been favored over others. It’s possible that relatively few genes played essential roles in producing the human brain.

You can feel the excitement of discovery thrumming through these papers, but it comes with a certain sadness as well. It doesn’t come just from the fact the chimpanzee whose DNA made this all possible died before he became famous. Lots of chimpanzees are dying–so many, in fact, that conservationists worry that they may become extinct from hunting, disease, and habitat destruction. And once a species is gone, it takes a vast amount of information about evolutionary history with it.

I was reminded of this fact when I read another chimpanzee paper that appears in the same issue of Nature, reporting on the first fossil of a chimpanzee ever discovered. It may be hard to believe that no one had found a chimp fossil before. A big part of the problem, scientists thought, was that chimpanzees were restricted to rain forests and other places where fossils don’t have good odds of surviving. The fossils that have now been discovered don’t amount to much–just a few teeth–and they raise far more questions than they answer. They date back about 500,000 years, to an open woodlands in Kenya where paleoanthropologists have also found fossils of tall, big-brained hominids that may have been the direct ancestors of Homo sapiens. So apparently chimpanzees once coexisted with hominids in the open woodlands that were once thought to be off-limits to them. More chimpanzee fossils will help address this puzzle, but they may never fully resolve it.

The chimpanzees of Kenya became extinct long ago, and now other populations teeter on the brink. To make sense of Clint’s genome, scientists need to document the variations both within and between chimpanzee populations–not just genetic variations, but variations in how they eat, how they organize their societies, how they use tools, and all the other aspects of the lives. If they don’t get that chance, the chimpanzee genome may become yet another puzzling fossil.


Vector Group

Posted on by

Sir

We are going to install 3 No of Alternators each of 7540 KVA and 6.3 KV.

Our project is that we have to sync. alternator on a 6.3 KV bus then used STEP Up by transformers ( 6 NO each 4000 KVA paralle connected ) to 11 KV bus.

Our problem is that what will be the Vector Group of Tr

Sarah Palin pulls into a tie with Obama for 2012

Posted on by

Stunning Turn-around!

Just one year ago, Obama led Sarah Palin by 12 points in a similar match-up. Now, Palin has pulled into a tie.

Breaking from Public Policy Polling (PPP):

With his approval numbers hitting new lows it's no surprise that Barack Obama's numbers in our monthly look ahead to the 2012 Presidential race are their worst ever this month. He trails Mitt Romney 46-43, Mike Huckabee 47-45, Newt Gingrich 46-45, and is even tied with Sarah Palin at 46. The only person tested he leads is Jan Brewer, who doesn't have particularly high name recognition on the national level at this point.

PPP surveyed 5 GOP likely opponents for Obama. All, except Arizona Gov. Jan Brewer were tied or ahead of Obama. Romney did best 46 to 43%. Gingrich and Huckabee also polled slightly ahead.