Tiny Device Stores Light

IBM researchers have fabricated a silicon device that's a significant advance in making practical optical interconnects.

By Prachi Patel-Predd
An SEM image of an optical-delay line that has up to 100 microrings, all connected to a common silicon nanowire. The optical buffer can store up to 10 optical bits.
Credit: IBM

By forcing light to circle multiple times through ring-shaped structures carved into silicon, researchers at IBM have been able to delay the flow of light on a microchip. Being able to delay light is crucial for high-performance, ultrafast optical computers of the future that will process information using light and electrical signals.

It's easy to store electronic data in computer memory; light is harder to control. The new silicon device, described in this week's issue of Nature Photonics, is ten times smaller than those made in the past. It also works much better at high data speeds. "This work is approximately a factor of ten over the best achieved with [ring-shaped devices] so far," says Keren Bergman, an electrical-engineering professor at Columbia University.

Storing light on silicon is key for electronic-optical hybrid computers that researchers believe will be available a decade from now. In these computers, devices will compute using electrons but will move data to other devices and components using light--avoiding the use of copper wires or interconnects that tend to heat up at high computer frequencies.

But the optical interconnects would have to be laid out in an intelligent network, just as the copper wires on today's chips are. To transfer data packets efficiently between devices, the copper network on a chip has nodes where many interconnects converge. If a processor is sending data to a logic circuit, the data travels from node to node until it gets to the logic circuit. Each node in the network reads and processes the data packet to route it correctly to the next node. While the node makes a routing decision, it temporarily stores the data in electronic memory. To process and route data at the nodes of an optical-interconnect network, one would need to store, or delay, light so the node can make the routing decision.

Yurii Vlasov and his colleagues at IBM's T.J. Watson Research Center delay light on a silicon microchip by circulating it 60 to 70 times through ring-shaped structures, called resonators. The researchers make these resonators on a thin silicon layer mounted on an insulting silicon-oxide layer. They etch parallel trenches into the silicon that reach down to the oxide. The raised portion between the trenches acts like a silicon wire that shuttles light.

The researchers employ the same silicon wafers and techniques that are used to fabricate microprocessors at IBM. This makes it easy to "think of combining optical circuitry with electrical circuitry on the same chip," Vlasov says.

By connecting many rings, the researchers can build up the delay. With 56 rings connected to a common silicon wire, they get the longest delay: about half a nanosecond--which amounts to storing 10 optical bits--at a data speed of 20 gigabits per second.

Other researchers have made resonators on silicon before. But the smallest resonators so far have been about 100 micrometers wide, and cascading tens of them yields a device that is a few millimeters long--too big to be integrated into an electronic circuit. The IBM researchers make rings that are 12 micrometers in diameter, and they can fit up to 100 ring resonators into an area that is less than one-tenth of a square millimeter.

The size of the device is a major advance, Bergman says: "It is very close to the kinds of densities you would like to have on chip for optical interconnects." Achieving a delay of 10 bits at gigabits-per-second speeds, which would be typical of the data speed that optical interconnects of the future would be handling, is a breakthrough, she says. "This is a major step towards making optical interconnects a reality."

The device loses more light than would be acceptable in practical circuits, and Vlasov says that he and his team are working to reduce these losses. Once they do that, he says, they could put thousands of resonators together to store even more optical bits. For practical optical interconnects, you would need to store hundreds and thousands of bits.

It might take another 10 years before we see optical interconnects in computers, but the IBM research shows that the technology is viable, says Risto Puhakka, president of market-research firm VLSI Research, in Santa Clara, CA. "There are legs on this technology, and it could eventually be integrated with current circuits into chips."

Source: http://www.technologyreview.com

Microsoft's Plan to Map the World in Real Time

Researchers are working on a system that allows sensors to track information and create up-to-date, searchable online maps.

By Kate Greene

Researchers at Microsoft are working on technology that they hope will someday enable people to browse online maps for up-to-the-minute information about local gas prices, traffic flows, restaurant wait times, and more. Eventually, says Suman Nath, a Microsoft researcher who works on the project, which is called SenseWeb, they would like to incorporate the technology into Windows Live Local (formerly Microsoft Virtual Earth), the company's online mapping platform.

By tracking real-life conditions, which are supplied directly by people or automated sensor equipment, and correlating that data with a searchable map, people could have a better idea of the activities going on in their local areas, says Nath, and make more informed decisions about, for instance, what driving route to take.

[For images from the SenseWeb application click here.]

"The value that you get out of [real-time data mapping] is huge," he says, and the applications can range from finding a parking spot in a cavernous parking garage to checking the traffic flow in different parts of a city.

Other research groups at the University of California at Berkeley, UCLA, Stanford, and MIT are working on similar projects for tracking environmental information. For instance, UCLA has a project in which sensors -- devices that measure physical quantities such as temperature, pressure, and sound -- are integrated with Google Earth, the company's downloadable mapping software. In addition, companies such as Agent Logic and Corda process real-time data and can correlate it with a location, mostly for businesses and governmental organizations.

Moreover, within the past year, Microsoft, Google, and Yahoo have been vying with each other to generate the most useful electronic maps (see "Killer Maps"). For the most part, though, the local information offered by Web-based mapping applications is updated only infrequently. And sites that offer real-time, local updates (about the status of public transportation, for instance), while useful, are designed for a single purpose.

What makes Microsoft's experimental project different from others that track information, Nath says, is that it would allow people to search for different types of real-time data within a user-specified area on a map, and progressively narrow that search. For instance, a person could highlight a region of a city and search for restaurants. SenseWeb would gather information provided by restaurants about their wait times and display it in various ways: the wait at specific establishments, the average wait for all restaurants in the region, or the minimum and maximum waits. If you needed to find a place to eat quickly, says Nath, but you learn that the minimum wait is 30 minutes in a certain part of town, you'd know to look in a different area. "You don't have to take the time to look at each individual restaurant," Nath says.

Additionally, a person could zoom into an area and see newly calculated information, such as maximum, minimum, and average wait times, according to the newly defined geography.

Searching for these types of real-time statistics within different areas on a map is a new take on displaying data on maps, says Phillip Levis, professor of computer science at Stanford University. "It's very different to give the average wait time in the city than it is to scan around the city and see each restaurant's wait time," he says.

SenseWeb is composed of three basic parts: sensors (or data-collecting units), Microsoft's database indexing scheme that sorts through the information, and the online map that lets users interact with the data. The sensors used in the project can vary in form and function, and can include thermometers, light sensors, cameras, and restaurant computers. SenseWeb puts baseline sensor information, such as location and function, into a database that's searchable by location and type of sensor information.

Then, if someone wants to check traffic conditions along a stretch of highway, for instance, the database will direct queries to cameras ("Web cams") located along the route -- and an image of traffic shows up on the map.

In order for people with sensors -- from researchers at universities to a private citizen with a Web cam -- to participate in SenseWeb, Nath says, they would have to be able to upload data to the Internet and provide information to the Microsoft group about their sensor, such as latitude, longitude, and the type of data it provides (for example, gas prices, temperature, or video).

One challenge for the SenseWeb project will be making the different types of information pulled into its database consistent enough to analyze and sort, says Samuel Madden, professor of computer science at MIT. For instance, there would need to be standard units for temperatures. "As soon as you start integrating all this data, you can imagine that weird things will happen," he says. "It's really a challenge to build tools that work with generic data and to come up with a way that anyone can publish their information."

Another, more fundamental hurdle for the SenseWeb project, Nath says, is getting people to register their sensors and sign on to the free program. Gas stations or restaurants may not even know about the project, or may not have an efficient way to pass along their data.

Therefore, in coming months, the Microsoft group will extend SenseWeb to universities that have already deployed sensors for other projects. In addition, the team is talking to a company that has sensors on parking spots, which, if integrated into Live Local, could help people find available parking more easily, he says.

For now, though, SenseWeb and Live Local are separate projects, according to Nath. The Live Local team "really loves this technology," he says, but right now "what's missing is the actual data."

Saving Power in Handhelds

Taking advantage of human error tolerance could make cell phones more energy efficient.

By Larry Hardesty
Credit: Sándor Kelemen, Istockphoto.com

With the advent of the Apple iPhone and its big, clear screen, the idea of using the morning commute to catch up on missed episodes of Lost became a lot more attractive. But video chews through a handheld's battery much faster than, say, playing MP3s does. In the most recent issue of the Association for Computing Machinery's Transactions on Embedded Computing Systems, researchers at the University of Maryland describe a simple way for multimedia devices to save power. In simulations, the researchers applied their technique to several common digital-signal-processing chores and found that, on average, it would cut power consumption by about two-thirds.

The premise of the technique, says Gang Qu, one of its developers, is that in multimedia applications, "the end user can tolerate some execution failure." Much digital video, for example, plays at a rate of 30 frames per second. But "in the old movie theaters, they played at 24 frames per second," Qu says. "That's about 80 percent. If you can get 80 percent of the frames consistently correct, human beings will not be able to tell you've made mistakes."

Unlike the movies in the old theaters, a digital video isn't stored on reels of wound plastic; it's stored as a sequence of 1s and 0s. That sequence is decoded as the video plays, and the decoding time can vary from one frame to the next. So digital media systems are designed to work rapidly enough that even the hardest-to-decode frames will be ready to be displayed on time.

Qu thinks that's a waste of processing power. If the viewer won't miss the extra six frames of video per second, there's no reason to decode them. Lower decoding standards would mean less work for the video player's processor, and thus lower power consumption.

The straightforward way to ensure a decoding rate of 80 percent would be to decode, say, eight frames in a row and ignore the next two. That approach--which Qu calls the "naive approach"--did introduce power savings in the Maryland researchers' simulations. The problem is that such a system doesn't distinguish frames that are hard to decode from those that are easy: if frame five is the hardest, the decoder will still plow through it; if frame nine is the easiest, the decoder will still skip it.

Qu and his colleagues wrote an algorithm that imposes a series of time limits on the decoding process; if any of the limits is exceeded, the decoding is aborted. "You set certain milestones," Qu says, "and you say, 'Okay, after this time I still haven't reached that first milestone, so it seems this is a hard task. Let me drop this one.'" Using statistics on the durations of particular tasks, the researchers can tune the algorithm to guarantee any desired completion rate.

Raj Rajkumar, director of the Real-Time and Multimedia Systems Laboratory at Carnegie Mellon University, mentions that his colleague John Lehoczky and the University of Wisconsin's Parmesh Ramanathan have investigated approaches similar to Qu's. But he says that Qu's work is "the logical extension of earlier work. I think that what Gang did is very useful." Ramanathan adds that with Qu's approach, "my guess is that there will be considerable savings in power consumption. I think one can save quite a bit."

Indeed, the Maryland researchers' algorithm fared well in simulations, offering a 54 percent energy savings over the naive approach. "If you are using the current approach, which is going to keep on decoding everything," Qu says, "we are going to probably consume only slightly more than one-third of that energy. That means you can probably extend the battery life by three times."

Qu is quick to point out that the researchers' simulations involved signals similar, but not identical, to video signals; real video decoding might not produce such dramatic results. On the other hand, Qu says that more-recent video-coding standards call for frame rates higher than 30 frames per second. That means the decoding rate could drop below 80 percent, saving even more power.

And the tested algorithms do accurately model cell-phone voice decoding. In some handheld devices--notably the iPhone--voice communication is almost as big a battery drain as video playback. Without the handy reference of a near-century of analog movies, however, user tolerance for error in voice is harder to gauge.

Qu says his and his colleagues' power-saving scheme could be implemented in either hardware or software, although in the near term, software would certainly be the cheaper option. He adds that the work has drawn some corporate interest, but that there are no plans to commercialize it at the moment. Nonetheless, "if we got some partners," Qu says, "if they have a top engineer trying to work with us, this could be done in half a year."

Source: http://www.technologyreview.com

"Personalized" Embryonic Stem Cells for Sale

A company offers to generate and store stem cells from leftover IVF embryos.

By Emily Singer
Stem-cell insurance: A company called StemLifeLine offers to generate embryonic stem cells (shown above) from leftover embryos created for in vitro fertilization. The cells could potentially be used for future medical treatments, although no embryonic-stem-cell-based treatments exist yet.
Credit: David Scharf, Science Photo Library

It's a new, rather dicey form of life insurance. A company in California called StemLifeLine has announced that it will offer a service to generate stem cells from excess frozen embryos stored after in vitro fertilization (IVF). The company promises a huge potential payoff: the cells could one day be used to treat disease in the buyers or in their families. But the service is already garnering criticism from some scientists and ethicists who say that without current medical uses for those cells, there's no point in people paying for them.

"I think the company's website overly hypes what may be possible," says Lawrence Goldstein, director of the stem-cell research program at the University of California, San Diego. "They are almost guaranteeing that therapies are around the corner, and now is the time to start banking stem cells, but that strikes me as premature for the field."

The new service is meant to take advantage of a growing interest in the field of regenerative medicine. Stem cells from adult blood or umbilical-cord blood are already used to treat some diseases, including sickle-cell anemia and several forms of leukemia. But these cells are largely limited to treating blood-related disorders and can't be grown in large numbers. Embryonic stem cells, on the other hand, can be coaxed to form virtually any type of cell in the body and can theoretically be replicated indefinitely. Scientists are developing ways to use them to replenish cells lost or damaged in ailments such as diabetes, Parkinson's disease, and heart disease. But as of now, those treatments are limited to the lab: no embryonic stem-cell-based therapies are approved for human use.

Couples who have had children via IVF are often left with extra embryos--and the rather difficult decision of what to do with them. As of 2003, an estimated 400,000 embryos remained in cryopreservation in the United States. Embryos can be donated to research or to other couples, destroyed, or left languishing in frozen storage. According to Ana Krtolica, StemLifeLine's CEO, the inspiration to form the company came from requests from clients at IVF clinics who were donating their embryos to research but wanted to know if they would have access to those cells if they were ever needed. (The answer is no.)

"We had a patient whose husband is a paraplegic," says Russell Foulk, a member of StemLifeLine's advisory board andmedical director of the Centers for Reproductive Medicine, a private clinic with offices in Nevada and Idaho. "They wanted to have a child and were excited about the possibility of creating neural cells from the extra embryos."

The technology to derive these cells is not new. Scientists at StemLifeLine use a similar procedure to that employed by research scientists for almost a decade, although the StemLifeLine scientists have refined it so that the resulting cells are fit for human use. For less than $10,000 (actual price depends on the collaborating IVF clinic), clients can send in their excess embryos and, in return, receive a line of stem cells that have been "quality assured," meaning they have been checked for the molecular markers that signify that the cells can be differentiated into multiple cell types. The company received certification as a tissue bank from the state of California last month, and it's in the process of generating cell lines for its first group of clients.

However, critics say that the service is premature. Extra embryos can remain in frozen storage for years. And in the case of the paraplegic man, no treatments using neural stem cells are yet available. "There is no reason to take your embryos out of cryopreservation and make a line of stem cells and then freeze them again until the technology is available to actually use them," says Eric Chiao, a stem-cell biologist at Stanford's Institute for Stem Cell Biology and Regenerative Medicine, in Palo Alto.

Chiao and others argue that by the time scientists have figured out how to use embryonic stem cells as therapies, they will likely have developed better ways of generating the stem cells themselves, possibly using cloning, in which scientists would generate perfectly matched stem cells from an adult cell of the patient to be treated. "My offspring would be better off if they used cloning to generate stem cells for themselves," says Arthur Caplan, an ethicist at the University of Pennsylvania. "In America, the best thing you can do is take the money you would have used and invest it in an insurance policy to maximize the likelihood that your kid will have health insurance someday."

Krtolica counters that because it takes two to three months to generate the cells, it's better to have them ready before an approved use in case a client needs them immediately.

Stem-cell scientists also say that StemLifeLine's description of its product as "personalized" stem cells is misleading. As with organ transplants, cell transplants require that the immune profile of the transplanted cells match the host as closely as possible. Scientists generally use the term personalized stem cells to refer to a type of stem cell not yet possible to create: those generated through cloning, making them a perfect genetic match to the donor. Cells made from discarded embryos would not be a perfect match to family members, says Doug Melton, codirector of the Harvard Stem Cell Institute, in Cambridge, MA. "This would be like having stem cells from a sibling, so immunosuppression is still an issue."

The prospect of generating stem-cell lines from embryos is likely to ignite new ethical arguments over embryonic stem cells. Critics of embryonic-stem-cell research oppose generating stem cells from embryos for any reason. But this service could spark growth of a practice that some find even more problematic: the creation of embryos solely as a source of cells. For example, some people might want to undergo IVF expressly for the stem cells, not to have a child. Krtolica says that she hasn't yet fielded any such requests but that ultimately, it would be up to the fertility clinics. Foulk, for one, says he would perform IVF under these circumstances.

Source: http://www.technologyreview.com

Intel's New Strategy: Power Efficiency

Spurred by competitor AMD's rapid success, Intel is shifting its strategy toward more power-efficient microprocessors.

By Kate Greene

Amid increasing competition from Advanced Micro Devices (AMD), Intel is changing its chip-making philosophy: it's paying more attention to the power requirements of its microprocessors.

In July 2006, the chip-making giant will release a new microprocessor, called Core 2 Duo, designed for laptops and desktops. The new chip is based on Intel's current chip architecture, which replaced traditional single-core processing with two processing centers on a single chip. The company says that the Core 2 Duo will perform better than its current dual-core chip, and will be more energy-efficient, which could make laptop batteries last longer and desktop towers run cooler.

Paying attention to power consumption in microprocessors is a relatively new concept for the company, says Steve Pawlowski, a senior fellow at Intel, adding that the move may help Intel regain market share from its rival AMD. Historically, the most important metric in the industry has been processor performance -- the speed at which a processor can complete a task, such as calculating a spreadsheet. "We've always focused on performance at the expense of power [use]," Pawlowski says.

But basic changes have occurred in the PC market, which first led AMD, and now Intel, to rethink microprocessor designs. First, mobile devices have become the primary PC for many consumers -- who don't want a device that quickly drains a battery or gets too hot. Furthermore, as the size of transistors shrink, they're more likely to waste electricity through a physical process called "leakage," says Kevin McGrath, an AMD fellow -- and the more transistors on a chip, the more electricity is wasted.

AMD has been working on more-efficient microprocessors for several years, and now Intel is trying to level the playing field. Both Intel and AMD have tackled part of the problem by converting their chip line-ups to dual-core processors (see "Multicore Mania," December 2005), which turns out to be one way to increase efficiency. "Interestingly, going to multiple cores can be a very power-efficient way of computation," says Milo Martin, professor in the computer and information sciences department at the University of Pennsylvania.

Three aspects of multicore chips make them more efficient. First, when a chip has more than one core, the speed at which each core computes can be slowed down without impeding the speed of the entire chip. By slowing down the clock speed, explains Martin, engineers can decrease the computational rate of a single core by a factor of five, from one gigahertz to 200 megahertz, and the core consumes only one-30th of the power. Then, he says, even if five of those cores are assembled onto a single chip, only one-sixth of the power is consumed, yet the total computational rate of one gigahertz is maintained.

Second, smaller processor sizes reduce power consumption. The number of transistors each core has and the amount of silicon real-estate they take up determines the amount of power the core uses -- smaller processors have fewer transistors and thus use less power than larger processors. In a dual-core chip, the total number of transistors is greater than it is in a single-core chip, but each core has fewer transistors, making it more power efficient.

Third, some of the processor functions, such as controlling memory, can be shared between cores, so that each core consumes less energy by not performing a redundant task.

So transitioning to a multicore architecture is an obvious way to save power, and both Intel and AMD have done so. But they're looking at other ways to create efficiency. As Pawlowski explains, managing processors at the circuit and individual transistor level can also save power. For instance, specific circuits on a transistor are designated to control the manipulation of a photo or to play a DVD. When that circuit needs to be used, the transistors that comprise the circuit are turned on with a certain voltage. In a perfectly efficient chip, those transistors would turn on and off only when they're needed. However, even when a circuit is idle, its transistors are using a small voltage that slowly leaks out of the transistor, says Pawlowski. This leakage produces heat and wastes electricity.

While there is much overlap in the ways that AMD and Intel are approaching this problem of waste and leakage at the circuit level, their solutions are different. Intel is working to solve the problem by designating "sleep transistors" on a chip to micromanage the circuits in each core. These transistors completely turn off the voltage to transistors in circuits that are dormant.

AMD also puts portions of the processor to sleep, explains McGrath; but it does so by having an algorithm instruct the processor to go into various levels of sleep, by shutting down its clock speed so that standby computations aren't carried out as quickly. The algorithm "can ask a part to go into its lowest power state," he says, "there are five or six of these power states that are used depending on the load of the processor."

Intel has announced prices for its new energy-efficient chips -- they're less expensive than AMD's current offerings, which will put pressure on its rival. For Intel, though, the test of whether its power-saving chips can compete well against AMD's offerings won't come until its new processors hit the market.

Frozen Bacteria Repair Own DNA for Millennia

Mason Inman
from National Geographic News

Bacteria can survive in deep freeze for hundreds of thousands of years by staying just alive enough to keep their DNA in good repair, a new study says.

In earlier work, researchers had found ancient bacteria in permafrost and in deep ice cores from Antarctica.

These bacteria, despite being trapped for millennia, were able to be revived and grown in the lab.

Some researchers had thought that bacteria would have to turn into dormant spores to survive for so long.

But if bacteria merely went dormant, metabolism would stop and various environmental factors would begin damaging their DNA.

Like an ancient scroll that's crumbling apart, the DNA becomes so damaged that it's indecipherable after about a hundred thousand years. Then the cells can't ever reproduce and the bacteria are effectively dead.

"Our results show that the best way to survive for a long time is to keep up metabolic activity," said Eske Willerslev, lead study author and a researcher at the University of Copenhagen in Denmark.

Doing this "allows for continuous DNA repair," Willerslev added.

The work suggests that if bacterial life existed on Mars or on Jupiter's moon Europa, it might still survive locked in icy soils.

The new study appears this week in the online advance edition of the Proceedings of the National Academy of Sciences.

Living, Just Barely

The new study examined DNA from bacteria found in permafrost from Siberia in Russia and Canada. The permafrost dated back to about a half-million years ago.

What the scientists found is that the bacteria appear to have kept up their metabolism.

These barely living bacteria did not seem to be reproducing, but they were still taking in nutrients and giving off carbon dioxide, like humans do when they breathe.

The bacteria were using some of these resources to keep their DNA in good shape, the study authors said.

But the researchers found that bacteria couldn't keep chugging along like this forever.

"You see a large diversity [of bacteria] in the modern samples, and as you get older and older, the diversity declines," Willerslev said.

The amount of carbon dioxide the bacteria gave off also dropped with age.

The limit for life in the permafrost is somewhere around 600,000 years old, the researchers say.

In older permafrost, the team couldn't detect any carbon dioxide emissions or any large pieces of DNA indicative of living bacteria.

By about 750,000 years old, the bacteria trapped in the permafrost seemed to be completely dead.

Soil vs. Ice

Some scientists have claimed to be able to revive far older bacteria preserved in amber or salts, but Willerslev has doubts about these results.

"I've been extremely skeptical about these previous results," Willerslev said.

But in the much colder environments of Mars or Europa, life might be able to survive while frozen for much longer, Willerslev said.

At those lower temperatures, DNA damage would accumulate more slowly.

So the new results "could suggest that if you had similar life on Mars, it could exist for much longer," he said.

Brent Christner of Louisiana State University welcomes the new results, which he finds convincing.

Christner and others have been studying ancient ice from deep in the Antarctic ice sheet and have found live bacteria there that have been frozen in place for perhaps one to two million years.

These ancient bacteria seemed to be repairing themselves, but the team didn't have direct evidence showing how the microbes were surviving so long.

"This study confirms and corroborates everything we've been finding with ancient glacial ice," Christner said.

Still, Willerslev is cautious about making this connection.

Glacial ice, he said, "is a completely different environment from permafrost, which is basically frozen soil" and contains lots of nutrients.

Supersonic "Hail" Seeds Star Systems With Water

John Roach
from National Geographic News

Evidence of water vapor "raining down" on a newly forming star system is offering the first direct look at how water likely gets incorporated into planets, NASA researchers announced.

(Related: "First Proof of Wet 'Hot Jupiter' Outside Solar System" [July 11, 2007].)

The water—enough to fill Earth's oceans five times over—falls at supersonic speeds in the form of a hail-like substance from the envelope of dust and gas that gave birth to the star.

The hail vaporizes when it smacks into the dusty disk around the embryonic star where planets are thought to take shape, according to models that best explain the observed data.

"This is the first time we've ever seen the process by which the surrounding envelope's material arrives at the disk," said Dan Watson, an astrophysicist at the University of Rochester in New York.

Watson is lead author of a paper describing the discovery in tomorrow's issue of the journal Nature.

"Since the disk is what's eventually going to give rise to the planetary system around the star, what we are seeing is the process by which that disk formed and therefore the initial conditions of planetary formation."

Star Development

The new work is based on observations of an embryonic star system taken with NASA's Spitzer Space Telescope (see images of stellar nurseries captured by Spitzer).

Astronomers observe such protostar systems in the infrared spectrum, because visible light is easily absorbed by the systems' dusty environments, making them invisible to the naked eye.

Water vapor emits a distinctive spectrum in infrared light.

The protostar lies about a thousand light-years from Earth in a cloud gas and dust. The whole system is called NGC 1333-IRAS 4B, or IRAS 4B for short.

The star is a warm, dense blob of material at the core of the cloud. A disk of planet-forming material is believed to circle the blob.

The radius of the disk is just larger than the distance between Pluto and the sun: about 3.6 billion miles (5.8 billion kilometers).

Based on their data, Watson and his colleagues say that the surface of the disk is -153 degrees Fahrenheit (-103 degrees Celsius).

While this seems frigid by Earth standards, Watson explained, the properties of water are different at the atmospheric pressure of the protostar, which is about a billionth of the pressure at sea level on Earth.

In addition, material equal to 23 times the mass of Earth arrives at the disk each year, Watson said.

"That's the material that's heating on arrival and then gradually cooling as it joins the lower parts of the disk," he said.

"This is very wet stuff. The original state is very wet," he added. "There's plenty of water to make a solar system out of."

Right Angle

Of the 30 embryonic star systems observed with Spitzer, only IRAS 4B showed signs of water vapor.

According to Watson, this is most likely because the protostar's axis points almost directly at Earth.

"The other 29 could very well have just as much water emission as IRAS 4B, but they are turned the wrong way and you can't see them," he said.

The team has already identified hundreds more protostar systems like IRAS 4B and plans to observe them with the Spitzer telescope, including more stars that exhibit this rare orientation.

Higher Games

It's been 10 years since IBM's Deep Blue beat Garry Kasparov in chess. A prominent philosopher asks what the match meant.

By Daniel C. Dennett

World chess champion Garry Kasparov during his sixth and final game against IBM’s Deep Blue in 1997. He lost in 19 moves.
Credit: Stan Honda/AFP/Getty Images

In the popular imagination, chess isn't like a spelling bee or Trivial Pursuit, a competition to see who can hold the most facts in memory and consult them quickly. In chess, as in the arts and sciences, there is plenty of room for beauty, subtlety, and deep originality. Chess requires brilliant thinking, supposedly the one feat that would be--forever--beyond the reach of any computer. But for a decade, human beings have had to live with the fact that one of our species' most celebrated intellectual summits--the title of world chess champion--has to be shared with a machine, Deep Blue, which beat Garry Kasparov in a highly publicized match in 1997. How could this be? What lessons could be gleaned from this shocking upset? Did we learn that machines could actually think as well as the smartest of us, or had chess been exposed as not such a deep game after all?

The following years saw two other human-machine chess matches that stand out: a hard-fought draw between Vladimir Kramnik and Deep Fritz in Bahrain in 2002 and a draw between Kasparov and Deep Junior in New York in 2003, in a series of games that the New York City Sports Commission called "the first World Chess Championship sanctioned by both the Fédération Internationale des Échecs (FIDE), the international governing body of chess, and the International Computer Game Association (ICGA)."

The verdict that computers are the equal of human beings in chess could hardly be more official, which makes the caviling all the more pathetic. The excuses sometimes take this form: "Yes, but machines don't play chess the way human beings play chess!" Or sometimes this: "What the machines do isn't really playing chess at all." Well, then, what would be really playing chess?

This is not a trivial question. The best computer chess is well nigh indistinguishable from the best human chess, except for one thing: computers don't know when to accept a draw. Computers--at least currently existing computers--can't be bored or embarrassed, or anxious about losing the respect of the other players, and these are aspects of life that human competitors always have to contend with, and sometimes even exploit, in their games. Offering or accepting a draw, or resigning, is the one decision that opens the hermetically sealed world of chess to the real world, in which life is short and there are things more important than chess to think about. This boundary crossing can be simulated with an arbitrary rule, or by allowing the computer's handlers to step in. Human players often try to intimidate or embarrass their human opponents, but this is like the covert pushing and shoving that goes on in soccer matches. The imperviousness of computers to this sort of gamesmanship means that if you beat them at all, you have to beat them fair and square--and isn't that just what ­Kasparov and Kramnik were unable to do?

Yes, but so what? Silicon machines can now play chess better than any protein machines can. Big deal. This calm and reasonable reaction, however, is hard for most people to sustain. They don't like the idea that their brains are protein machines. When Deep Blue beat Kasparov in 1997, many commentators were tempted to insist that its brute-force search methods were entirely unlike the exploratory processes that Kasparov used when he conjured up his chess moves. But that is simply not so. Kasparov's brain is made of organic materials and has an architecture notably unlike that of Deep Blue, but it is still, so far as we know, a massively parallel search engine that has an outstanding array of heuristic pruning techniques that keep it from wasting time on unlikely branches.

True, there's no doubt that investment in research and development has a different profile in the two cases; Kasparov has methods of extracting good design principles from past games, so that he can recognize, and decide to ignore, huge portions of the branching tree of possible game continuations that Deep Blue had to canvass seriatim. Kasparov's reliance on this "insight" meant that the shape of his search trees--all the nodes explicitly evaluated--no doubt differed dramatically from the shape of Deep Blue's, but this did not constitute an entirely different means of choosing a move. Whenever Deep Blue's exhaustive searches closed off a type of avenue that it had some means of recognizing, it could reuse that research whenever appropriate, just like Kasparov. Much of this analytical work had been done for Deep Blue by its designers, but Kasparov had likewise benefited from hundreds of thousands of person-years of chess exploration transmitted to him by players, coaches, and books.

It is interesting in this regard to contemplate the suggestion made by Bobby Fischer, who has proposed to restore the game of chess to its intended rational purity by requiring that the major pieces be randomly placed in the back row at the start of each game (randomly, but in mirror image for black and white, with a white-square bishop and a black-square bishop, and the king between the rooks). Fischer ­Random Chess would render the mountain of memorized openings almost entirely obsolete, for humans and machines alike, since they would come into play much less than 1 percent of the time. The chess player would be thrown back onto fundamental principles; one would have to do more of the hard design work in real time. It is far from clear whether this change in rules would benefit human beings or computers more. It depends on which type of chess player is relying most heavily on what is, in effect, rote memory.

The fact is that the search space for chess is too big for even Deep Blue to explore exhaustively in real time, so like Kasparov, it prunes its search trees by taking calculated risks, and like Kasparov, it often gets these risks precalculated. Both the man and the computer presumably do massive amounts of "brute force" computation on their very different architectures. After all, what do neurons know about chess? Any work they do must use brute force of one sort or another.

It may seem that I am begging the question by describing the work done by Kasparov's brain in this way, but the work has to be done somehow, and no way of getting it done other than this computational approach has ever been articulated. It won't do to say that Kasparov uses "insight" or "intuition," since that just means that ­Kasparov himself has no understanding of how the good results come to him. So since nobody knows how Kasparov's brain does it--least of all Kasparov himself--there is not yet any evidence at all that Kasparov's means are so very unlike the means exploited by Deep Blue.

People should remember this when they are tempted to insist that "of course" Kasparov plays chess in a way entirely different from how a computer plays the game. What on earth could provoke someone to go out on a limb like that? Wishful thinking? Fear?

In an editorial written at the time of the Deep Blue match, "Mind over Matter" (May 10, 1997), the New York Times opined:

The real significance of this over-hyped chess match is that it is forcing us to ponder just what, if anything, is uniquely human. We prefer to believe that something sets us apart from the machines we devise. Perhaps it is found in such concepts as creativity, intuition, consciousness, esthetic or moral judgment, courage or even the ability to be intimidated by Deep Blue.

The ability to be intimidated? Is that really one of our prized qualities? Yes, according to the Times:

Nobody knows enough about such characteristics to know if they are truly beyond machines in the very long run, but it is nice to think that they are.

Why is it nice to think this? Why isn't it just as nice--or nicer--to think that we human beings might succeed in designing and building brain­children that are even more wonderful than our biologically begotten children? The match between Kasparov and Deep Blue didn't settle any great metaphysical issue, but it certainly exposed the weakness in some widespread opinions. Many people still cling, white-­knuckled, to a brittle vision of our minds as mysterious immaterial souls, or--just as romantic--as the products of brains composed of ­wonder tissue engaged in irreducible non­computational (perhaps alchemical?) processes. They often seem to think that if our brains were in fact just protein machines, we couldn't be responsible, lovable, valuable persons.

Finding that conclusion attractive doesn't show a deep understanding of responsibility, love, and value; it shows a shallow appreciation of the powers of machines with trillions of moving parts.

Daniel Dennett is the codirector of the Center for Cognitive Studies at Tufts University, where he is also a professor of philosophy.

Uninspiring Vista

How Microsoft's long-awaited operating system disappointed a stubborn fan.

By Erika Jonietz

Vista's Aero visual environment includes the flip 3-D feature, which allows a user to cycle through a stack of open windows to find the desired application, shown above, and translucent window borders. Vista also offers "Gadgets," small programs that recall Mac "Widgets" (far right of screen above).

For most of the last two decades, I have been a Microsoft apologist. I mean, not merely a contented user of the company's operating systems and software, not just a fan, but a champion. I have insisted that MS-DOS wasn't hard to use (once you got used to it), that Windows 3.1 was the greatest innovation in desktop operating systems, that Word was in fact superior to WordPerfect, and that Windows XP was, quite simply, "it."

When I was forced to use Apple's Mac OS (versions 7.6 through 9.2) for a series of jobs, I grumbled, griped, and insisted that Windows was better. Even as I slowly acclimated at work, I bought only Windows PCs for myself and avoided my roommate's recherché new iBook as if it were fugu. I admitted it was pretty, but I just knew that you got more computing power for your buck from an Intel-based Windows machine, and of course there was far more software available for PCs. Yet my adoration wasn't entirely logical; I knew from experience, for example, that Mac crashes were easier to recover from than the infamous Blue Screen of Death. At the heart of it all, I was simply more used to Windows. Even when I finally bought a Mac three years ago, it was solely to meet the computing requirements of some of the publications I worked with. I turned it on only when I had to, sticking to my Windows computer for everyday tasks.

So you might think I would be predisposed to love Vista, Microsoft's newest version of Windows, which was scheduled to be released to consumers at the end of January. And indeed, I leaped at the opportunity to review it. I couldn't wait to finally see and use the long-delayed operating system that I had been reading and writing about for more than three years. Regardless of widespread skepticism, I was confident that Vista would dazzle me, and I looked forward to saying so in print.

Ironically, playing around with Vista for more than a month has done what years of experience and exhortations from Mac-loving friends could not: it has converted me into a Mac fan.

A little context and a caveat: in order to meet print deadlines, I had to review the "RC1" version of Vista Ultimate, which Microsoft released in order to gather feedback from over-eager early adopters. Such post-beta, prerelease testing reveals bugs and deficits that in-house testing misses; debuggers cannot mimic all the various configurations of hardware, software, and peripherals that users will assemble. And Vista RC1 was maddeningly buggy. Although I reminded myself repeatedly that most of the problems I encountered would be fixed in the final version, my opinions about Vista are probably colored by my frustrations.

Still, my very first impression of Vista was positive. Quite simply, it's beautiful. The Aero visual interface provides some cool effects, such as translucent window borders and a way to scroll through a 3-D "stack" of your open windows to find the one you want. Networking computers is virtually automatic, as it was supposed to be but never quite has been with Windows XP. The Photo Gallery is the best built-in organizer I've used to manage digital pictures; it even includes basic photo correction tools.

But many of Vista's "new" features seemed terribly familiar to me--as they will to any user of Apple's OS X Tiger operating system. Live thumbnails that display petite versions of minimized windows, search boxes integrated into every Explorer window, and especially the Sidebar--which contains "Gadgets" such as a weather updater and a headline reader--all mimic OS X features introduced in 2005. The Windows versions are outstanding--they're just not really innovative.

Unfortunately, Vista RC1 contained bugs that rendered some promising features, such as the new version of Windows Media Center, unusable for me (an acquaintance who acquired a final copy of Vista ahead of release assures me that all that has been fixed).

My efforts to get Media Center working highlighted two big problems with Vista. First, it's a memory hog. The hundreds of new features jammed into it have made it a prime example of software bloat, perhaps the quintessence of programmer Niklaus Wirth's law that software gets slower faster than hardware gets faster (for more on the problems with software design that lead to bloat, see "Anything You Can Do, I Can Do Meta"). Although my computer meets the minimum requirements of a "Vista Premium Ready PC," with one gigabyte of RAM, I could run only a few ­simple programs, such as a Web browser and word processor, without running out of memory. I couldn't even watch a movie: Windows Media Player could read the contents of the DVD, but there wasn't enough memory to actually play it. In short, you need a hell of a computer just to run this OS.

Second, users choosing to install the 64-bit version of Vista on computers they already own will have a hard time finding drivers, the software needed to control hardware sub­systems and peripherals such as video cards, modems, or printers. Microsoft's Windows Vista Upgrade Advisor program, which I ran before installing Vista, assured me that my laptop was fully compatible with the 64-bit version. But once I installed it, my speakers would not work. It seems that none of the companies concerned had written a driver for my sound card; it took more than 10 hours of effort to find a workaround. Nor do drivers exist for my modem, printer, or several other things I rely on. For some of the newer components, like the modem, manufacturers will probably have released 64-bit drivers by the time this review appears. But companies have no incentive to write complicated new drivers for older peripherals like my printer. And because rules written into the 64-bit version of Vista limit the installation of some independently written drivers, users will be virtually forced to buy new peripherals if they want to run it.

Struggling to get my computer to do the most basic things reminded me forcefully of similar battles with previous versions of Windows--for instance, the time an MIT electrical engineer had to help me figure out how to get my computer to display anything on my monitor after I upgraded to Windows 98. Playing with OS X Tiger in order to make accurate comparisons for this review, I had a personal epiphany: Windows is complicated. Macs are simple.

This may seem extraordinarily obvious; after all, Apple has built an entire advertising campaign around the concept. But I am obstinate, and I have loved Windows for a long time. Now, however, simplicity is increasingly important to me. I just want things to work, and with my Mac, they do. Though my Mac barely exceeds the processor and memory requirements for OS X Tiger, every bundled program runs perfectly. The five-year-old printer that doesn't work at all with Vista performs beautifully with OS X, not because the manufacturer bothered to write a new Mac driver for my aging standby, but because Apple included a third-party, open-source driver designed to support older printers in Tiger. Instead of facing the planned obsolescence of my printer, I can stick with it as long as I like.

And my deepest-seated reasons for preferring Windows PCs--more computing power for the money and greater software availability--have evaporated in the last year. Apple's decision to use the same Intel chips found in Windows machines has changed everything. Users can now run OS X and Windows on the same computer; with third-party software such as Parallels Desktop, you don't even need to reboot to switch back and forth. The chip swap also makes it possible to compare prices directly. I recently used the Apple and Dell websites to price comparable desktops and laptops; they were $100 apart or less in each case. The difference is that Apple doesn't offer any lower-end processors, so its cheapest computers cost quite a bit more than the least-expensive PCs. As Vista penetrates the market, however, the slower processors are likely to become obsolete--minimizing any cost differences between PCs and Macs.

I may need Windows for a long time to come; many electronic gadgets such as PDAs and MP3 players can only be synched with a computer running Windows, and some software is still not available for Macs. But the long-­predicted migration of software from the desktop to the Internet is finally happening. Organizations now routinely access crucial programs from commercial Web servers, and consumers use Google's services to compose, edit, and store their e-mail, calendars, and even documents and spreadsheets (see "Homo Conexus," July/August 2006). As this shift accelerates, finding software that works with a particular operating system will be less of a concern. People will be able to base decisions about which OS to use strictly on merit, and on personal preference. For me, if the choice is between struggling to configure every feature and being able to boot up and get to work, at long last I choose the Mac.

Erika Jonietz is a Technology Review senior ­editor.

WINDOWS VISTA operating system
$99.95-$399.00
www.microsoft.com/windowsvista

Electric Fields Kill Tumors

A promising device uses electric fields to destroy cancer cells in the brain.

By Katherine Bourzac

Zapping tumors: Brain-cancer patients in a trial for a portable device that sends a weak electric field into the brain must wear electrodes almost constantly. One patient in a pilot clinical trial for the device, who still had cancer after radiation, chemotherapy, and surgery, experienced a complete recovery. The MRI at top shows a tumor on the left side of this patient’s brain before treatment. The MRI at bottom, taken after eight months of treatment, shows no tumor.
Credit: Yoram Palti, NovoCure (top image); Proceedings of the National Academy of Sciences (bottom MRIs)

An Israeli company is conducting human tests for a device that uses weak electric fields to kill cancer cells but has no effect on normal cells. The device is in late-stage clinical trials in the United States and Europe for glioblastoma, a deadly brain cancer. It is also being tested in Europe for its effectiveness against breast cancer. In the lab and in animal testing, treatment with electric fields has killed cancer cells of every type tested.

The electric-field therapy was developed by Yoram Palti, a physiologist at the Technion-Israel Institute of Technology, in Haifa, who founded the company NovoCure to commercialize the treatment. Palti's electric fields cause dividing cancer cells to explode while having no significant impact on normal tissues. The range of electric fields generated by the device harms only dividing cells. And since normal cells divide at a much slower rate than cancer cells, the electric fields target cancer cells. "An Achilles' heel of cancer cells is that they have to divide," says Herbert Engelhard, chief of neuro-oncology in the department of neurosurgery at the University of Illinois, Chicago.

Even after chemotherapy, radiation therapy, and surgery, about 85 to 90 percent of glioblastoma patients' cancer still progresses, and their survival rates are low, says Engelhard. He has about 10 glioblastoma patients enrolled in the trial, which is testing the unusual treatment in patients for whom all other approaches have failed. Engelhard says that the results are encouraging but that it's too early to comment on the treatment's efficacy.

The electric fields' different effects on normal and dividing cells mostly have to do with geometry. A dividing cell has what Palti calls "an hourglass shape rather than a round shape." The electric field generated by the NovoCure device passes around and through round cells in a uniform fashion. But the narrow neck that pinches in at the center of a dividing cell acts like a lens, concentrating the electric field at this point. This non-uniform electric field wreaks havoc on dividing cells. The electric field tears apart important biological molecules, such as DNA and the structural proteins that pull the chromosomes into place during cell division. Dividing cells simply "disintegrate," says Palti.

Palti, who for years has been studying the effect of electric fields on cancer and normal cells, says that he has verified this mechanism in computer models and experiments in the lab. "The physics are solid," says David Cohen, associate professor of radiology at Harvard Medical School.

Patients in the glioblastoma clinical trial wear the device almost constantly, carrying necessary components in a briefcase. A wire emerging from the briefcase connects to adhesive electrodes covering the skull. Alternating electric fields pass through the scalp, into the skull, and on to the brain. The Food and Drug Administration approved the device for late-stage clinical trials for glioblastoma following promising results from a pilot study in 10 patients, one of whom had a complete recovery.

One exciting result from his studies, says Palti, is that there is "excellent synergy between electric-field treatment and chemotherapy." In an unpublished lab study of several types of cancer, he says, adding electric-field treatment makes several chemotherapeutics more effective at lower doses. NovoCure is now conducting a pilot trial in Europe in which patients begin electric-field treatment in conjunction with chemotherapy when they are first diagnosed with glioblastoma. The results are preliminary, but, Palti says, "I strongly believe that the combination treatment will ... enable one to reduce the chemo doses to levels where their side effects will be significantly reduced."

Palti says that after more than 200 cumulative months of electric-field treatment in several patients, there have been no side effects beyond irritation of the scalp. "So far, toxicity seems to be low," says Engelhard. This stands in stark contrast to chemotherapy and radiation, which cause many side effects, including nausea, hair loss, and fatigue.

One worry is that the electric-field treatment could affect healthy cells that are dividing. The electric fields emerging from the electrodes can't be focused, says Cohen, and although they are primarily concentrated in the brain in the glioblastoma trial, they may also reach other parts of the body where cells are dividing. Cells in the bone marrow, for example, multiply at a great rate to create red blood cells and immune cells. But Palti says that the electric fields have no effect on blood-cell counts. The bone and muscle surrounding the marrow appear to protect the cells..

It's unclear how long patients will need to wear the device. "We're hesitant to stop treatment, because the consequences could be severe," says Palti, although one patient whose cancer has disappeared has stopped wearing the device. Patients must go to the clinic twice a week to have their heads shaved so that their hair doesn't interrupt contact between the scalp and the electrodes. The device itself costs only about $1,000 to manufacture, but replacing the electrodes twice a week is expensive.

Engelhard says that he got involved with the NovoCure clinical trial because the electric-field treatment is "radically different" from all existing cancer treatments. For patients with recurrent glioblastoma and other deadly forms of cancer, there are few options. "Researching and testing new therapies for this type of patient is very important," says Engelhard.