Tech Talk

The instrument-maker’s quest for smaller, faster, more sensitive, more flexible sensors is never-ending. The hunt drives them to smaller and smaller scales, where novel phenomena offer new ways of thinking about—and measuring—nature. There may be Faustian bargains along the way, though: at very small scales, they sometimes lose the ability to capture data that is easily available at larger scales.

In the present instance, extraordinary optical transmission (EOT) is a highly efficient light transmission through arrays of nanoholes. It’s an order of magnitude more efficient than conventional transmission through apertures of equivalent size, thanks to (in the words of its discoverers) "the coupling of light with plasmons—electronic excitations—on the surface of the periodically patterned metal film.” But EOT is only possible with perforations on the order of tens of nanometers in diameter, arrayed in a grid whose period is significantly shorter than one wavelength of the light to be studied.

Diffraction gratings—the workhorses of spectrometry for centuries—only work if the slits are significantly farther than one wavelength apart.

Now, though, researchers at the University of Alabama at Huntsville have devised at least one way of having their cake and eating it too, via a “super nanograting.” Haisheng Leong and Junpeng Guo, members of Huntsville’s electrical and computer engineering department, have built a surface plasmon resonance spectrometer (SPRS): a 300-micrometer-square silicon chip coated with a 50 nm of gold and then electron-beam etched with about half a million 140-nm-diameter holes distributed in a 420 nm grid. The duo converted this straightforward EOT sieve into a novel dual-scale grating by the simple expedient of omitting every fifth row from the grid.  The result is equivalent to a diffraction grating with a 2100 nm pitch…but with the order-of-magnitude-better light-transmission and plasmon-sensitivity characteristics of a nanoarray.

Go to your local Best Buy store and look at the computers. What’s the most attractive offering? Surely it’s the Chromebooks, at $249 for Samsung's and $199 for Acer’s. And that—more than razor-thin margins due to no-name competitors; more than a PC market share that has slipped by a third in the past six years; more than the uncertain success of Windows 8; more than carrying $15 billion in debt—is the challenge faced by Dell Computer, the one-time leader in PCs, now a distant third, as it leaves NASDAQ for the comfy confines of private equity.

In Wall Street’s largest leveraged buyout in six years, Dell Computer is going private. According to the New York Times, Michael Dell, who has put up a healthy chunk of his personal fortune, and his partners at private equity firm Silver Lake are sanguine about taking on an additional $15 billion because “Dell’s cash from operations has held steady for four of the last five years, coming in at $5.5 billion for the most recent fiscal year.” Me, I would be looking over my shoulder at those Chromebooks and worrying about how long the milk will flow.

The market for Dell’s PCs and laptops was already hollowing. Apple owns the high-value (and high-margin) end of the market—even five years ago, Apple was the manufacturer of two of every three computers costing more than $1000. iPads and its Android ilk are now directly eating into the commodity PC market—in the fourth quarter of 2012, tablet sales were up 75 percent, while PC sales declined by 6.4 percent, according to IDC. Just last week, Bloomberg reported that at Acer, Chromebooks are already outselling Windows 8 PCs. And the Chromebook threat isn't just the hardware. Google is, after all, a software company.

In the past few years, Google has turned Google Docs into a viable competitor to Microsoft Windows. And when I say viable, I mean in terms of features, usability, and reliability. In terms of price—Google Docs is free to individuals—it blows Office out of the water. (Personally, I’m coming to agree with Charles Cooper over at CNet when he says “it's time to move on.” Have you used it lately? It has a better word processor than Word.) With the browser handling almost everything else, including e-mail, what more do you need? Oh, how about your favorite Android apps? As it happens, James Kendrick over at ZDNet posed the question just last week, “Are Android apps for Chromebooks on the way?”

The recent death of Andre Cassagnes, the electrical technician who came up with the idea of using the response of aluminum powder to an electrostatic charge to create a drawing machine, has kicked off a wave of Etch A Sketch nostalgia.

For me, Etch A Sketch wasn’t a drawing tablet—I had drawing tablets (that is, paper and ink), that were a lot less frustrating for an eight year old. Instead, it was my first game machine. I spent hours and hours twiddling the knobs to create complex labyrinths. Then, I would challenge myself or a friend to navigate the stylus through the labyrinth without hitting the sides. I always included a nasty section of tight hairpin turns designed to trip the player up.

When first introduced to a “real” video game, Pong, I was pretty good at twirling the knobs, thanks, I’m sure, to those hours with my fingers on my Etch A Sketch. (Like the creators of Pong, the manufacturers of Etch A Sketch went with dials for controls, though Cassagnes original version used a joystick.)

By the time my kids were ready for Etch A Sketch, the toy was available in a miniature version that came with transparent overlays that turned the gadget into, yes, a video game. Guess I wasn’t the only one who had discovered Etch A Sketch gaming. I had been holding back on introducing my kids to electronic games,and was quite happy to hand over a shiny new Etch A Sketch at the beginning of a long plane ride instead of a Game Boy.

So thanks, M. Cassagnes. And this weekend I’ll see if I can dig up an old Etch A Sketch and challenge my kids to a maze game. I bet I can still beat them.

Follow me on Twitter @TeklaPerry.

Photos: The Ohio Art Company, the Pong Museum, Klutz

Just a few years ago, 3-D television was the next big thing in home entertainment, and sports was the next big thing in 3-D. Now, 3-D is a minor feature in television sets, kind of like automatic dimming or a memory card slot. And 3D sports broadcasts, while catching on in some countries, aren’t common place in the United States. Today’s Super Bowl will not be televised in 3-D; the broadcaster, CBS, is not currently producing any original 3-D programming.

Keeping the Super Bowl a 2-D event makes a certain amount of sense. The Super Bowl is typically viewed in large groups—I’m going to a Super Bowl party today, you may be too. And it can be tough to find the decent 3-D viewing angles for a crowd in a typical living room, though 3-D broadcasts in pubs in the U.K. have worked out just fine.

Just because it's going to be done in 2-D doesn't mean the Super Bowl is a simple broadcast to produce. A regular season game has cameras at fewer than 30 different positions, for today's broadcast CBS will more than double that number, to 62, says Howard Postley, CTO of 3ality Technica, who wrote about the challenges of producing sports broadcasts in 3-D in the November issue of Spectrum.

Meanwhile, true 3-D sports broadcasts on traditional networks seem to be fading away in the United States. ESPN, the network that had pioneered some of the early 3-D sports broadcasts in the U.S., has moved on to what they call 5-D TV. It sounds great, but it’s more like two-and-a-half D; the 3-D camera rigs are used to capture both a 2-D image and a 3-D image, meaning the broadcast cannot necessarily be optimized for 3-D.

“I think 3D channels are destined to go Over the Top,” Postley says, that is, delivered via the Internet instead of through traditional broadcast channels.

Photo: Randy Sagar/ESPN

In 2011, a very exciting study came out in the medical journal, the Lancet, that claimed to use electroencephalography (EEG) to detect unexpected levels of brain activity in a small sample of vegetative patients, suggesting that they were still capable of voluntary, perhaps even conscious, thought. The research garnered attention from the New York Times as a potential bedside tool for determining the status of non-responsive patients. However, a new, independent analysis of the study is calling the results into question.

"We showed that their method is invalid for their patient data," says Andrew Goldfine, a rehabilitation neurologist at the Burke Rehabilitation Hospital in White Plains, NY and an author on the new study.

In the original research, cognitive scientists gave a verbal command to patients with traumatic brain injury, asking them to listen for a tone and then respond by imagining movement in either their right hand or the toes on their right foot. With EEG, they looked for brain activity that would indicate processing of motor imagery and compared these results to those collected in healthy subjects. Of the 16 patients they tried this with, three seemed capable of completing the task.

Goldfine and his colleagues, who have also used EEG to detect residual activity in non-responsive patients, were at first stunned by the results. "We were really impressed because here they had done something simlilar to us," says Goldfine. "Then we learned about what their task was. And their task was much harder than ours."

Funded by a grant from the James S. McDonnell Foundation, which supports data-sharing and collaboration in brain trauma rehabilitation research, Goldfine undertook a fresh analysis of the data, performing the same statistical analysis that he does on his own results. When given access to the raw data, he found that the researchers had run many trials in succession, analyzing them as independent data points in a way that was inappropriate according to Goldfine. When the trials were run again as blocks of data, the reponses that jumped out in the original results disappeared and the brain activity looked completely random. "In the EEG before the beep and after the beep there was no signal change at all," he says.

What is the blackest black this side of a black hole? So far, the earthly title goes to carbon nanotubes packed together on end. Mats of these vertically aligned multiwall carbon nanotube arrays (VANTAs) reflect less than 0.07 percent of incoming visible light (wavelengths 390 to 700 nanometers) and about 0.35 percent of radiation in the 5000-to-10 000 nm range. Imagine sifting sand over a field of up-ended drainpipes: almost nothing will bounce out.

Researchers at the U.S. National Institute of Standards and Technology’s Boulder, CO, lab have exploited VANTAs’ near-perfect absorption to demonstrate a faster, cheaper, and more sensitive cryogenic radiometer—the workhorse instrument for calibrating radiation sources, including the lasers that drive fiber-optic communications.

In Optics Letters, Nathan Tomlin and laser radiometry project leader John Lehman describe their single-chip carbon nanotube cryogenic radiometer (CNCR). Like comparative radiometers used for more than half a century, their device (maintained at about 4ºK to increase sensitivity and accuracy, hence the “cryogenic”) uses changes in resistance to gauge the temperature increase generated by an incoming energy beam. They then use a precisely controlled heater to find the current needed to maintain the same temperature after the light is shut off. Assuming near-hundred-percent efficiency in both absorber and heater, the electrical power will be the same as the optical power.

Unlike current cryogenic radiometers, the CNCR uses VANTA nanotube mats (about the size of the disks left behind by office paper-punches) for all of the key components: absorber, heater, and thermistor. The VANTA mats are grown in place on a 25-millimeter-square etched and prepared silicon wafers. One black spot does double duty as both absorber and heater, and a second serves as the thermistor to track temperature changes.

Thanks to the VANTAs’ nonlinear thermal resistance (R=R₀ exp(T₀/T)^0.5), “measured resistance …is very sensitive to the spatial heating profile.” As a result, even the first prototype achieved response times two orders of magnitude better than conventional cryogenic radiometers. The developers note that more work is needed, including refinements that would reduce temperature fluctuations in the cooling bath and decrease resistance on the chip. In the long run, say Tomlin and Lehman, they should be able to reduce noise by seven orders of magnitude. Indeed, the “ultimate accuracy of the CNCR should be limited only by our ability to measure the reflectance of the nanotube array.”

Image: N. Tomlin/NIST

OpenSite published an infographic yesterday that covers some of the same trends I discussed last week in a podcast interview with the head of the American Library Association, Maureen Sullivan. Some additional facts about U.S. libraries we didn't discuss include (the full graphic is below):

• Library usage is way up (over the past six years). For example, Boston saw library visits rise 44 percent, Chicago 36 percent, and Brooklyn, N.Y., 31 percent.
• 70 percent of libraries report rising use of public access computers. Yet,
• 76 percent report not having enough computers to meet demand, and 45 percent say their Internet connections are too slow.

Libraries help tens of millions of job seekers do their seeking; in fact, 72 percent of libraries report helping out with online applications.

A 2010 infographic by the non-profit Online Computer Library Center had a few other nice stats:

• More libraries (12 000) provided free Wi-Fi than Starbucks (11 000)
• Libraries lent out almost as many movies (2.1 million) as Netflix (2.1 million)
• Libraries were more popular (1.4 billion visits) than movie theatres (1.3 billion tickets)

It's common practice among those who travel internationally to unlock a smartphone so that it will work with SIM cards from multiple service providers. That's the best way to get around massive roaming fees that would otherwise be assessed by your home provider. But come tomorrow, in the United States at least, this will be illegal. Here's why.

Every few years, the U.S. Librarian of Congress signs off on exceptions to the Digital Millenium Copyright Act, a far reaching bill passed in 1998 that, among other things, makes it illegal for consumers to circumvent any barriers that manufacturers put up around their copyrighted material. Recommended exemptions find their way to the Librarian via the U.S. Copyright Office, but he makes the final decision.

In 2006, he approved an exemption that allowed consumers to unlock their phones. In 2010, he renewed the exemption. Last year, he let it expire but provided a 90 day grace-period for the benefit of people who still wanted to unlock legacy phones. After today, unlocking a smartphone without the maker's approval will be a criminal offense, even if the contract on the device has expired.

Some 5 to 10 percent of photovoltaic cells shatter at some point during fabrication, thanks to microscopic cracks in the silicon wafers on which they’re built. The fabrication processes—oxidation, annealing, purification, diffusion, etching, and layering—account for about half of the solar cell’s total installed cost, so this breakage wastes significant money, time, and material.

Researchers at the National Renewal Energy Laboratory (NREL) have built a device that identifies the friable wafers before they go into fabrication. The Silicon Photovoltaic Wafer Screening System (SPWSS), developed by Bhushan Sopori’s NREL team, uses intense light from an optical cavity furnace to heat narrow bands on the wafer up to 500ºC. This produces strong thermal gradients that splinter the already-cracked wafers while leaving solid wafers intact.

The SPWSS is built around the technology Sopori’s group used to develop the innovative optical cavity furnace, originally designed to support high-temperature photovoltaic-cell fabrication operations. The heart of the SPWSS is a chamber roughly 40 cm on a side lined with a highly insulating, reflective ceramic and lit by high-intensity lamps (the patent mentions three to five computer-controlled 1600-Watt tungsten-halogen lamps). The ceramics retain the radiation almost perfectly, capturing the light and diffusing it into a highly uniform thermal glow. By cutting a 15-millimeter slit in the bottom of the chamber, the group produced a very uniform light source for localized heating. Then they added a wire-mesh conveyor (think of the toaster at a breakfast buffet), computer-controlled to produce the spot-heating duration appropriate to the material’s thickness and other characteristics.

The conveyor can test as many as 20 wafers per minute—fast enough to keep up with other production processes and cheap enough to cost just “some fraction of a penny per wafer,” according to Sopori.

Images: Dennis Schroeder/National Renewable Energy Laboratory

Editor's Note: An earlier version of this post was removed because of inaccuracy in unemployment figures.

A few recent signs show that job prospects for tech professionals are good this year.

Recent Bureau of Labor Statistics numbers show the unemployment rates for techies are much lower than the national average. The U.S. unemployment rate at the end of 2012 was 7.8 percent, more than twice the 3.3 percent for tech professionals.

As Ted Samson points out in this InfoWorld article, BLS numbers “should be consumed with a dash of salt.”

“First off, the bureau has a way of lumping arguably nontechnical jobs into its computer-oriented categories. Second, the BLS doesn't factor in people who are unemployed because they have given up on finding jobs.”

That being said, here are unemployment rates for specific professions: 1.5 percent for database administrators; 1.9 percent for network architects; 2.9 percent for software developers; 3.3 percent for computer systems analysts; 4.3 percent for network and systems administrators; and 4.6 percent for programmers.

Other reports seem to confirm the positive job prospects for tech employees. Tech salaries are expected to see a big average increase, according to a 2013 salary guide released by staffing and consulting firm Robert Half International. A December 2012 report released by the Engine Advocacy and the Bay Area Council Economic Institute stated that, between 2002 and 2011, job growth in Science, Technology, Engineering and Mathematics (STEM) fields has outpaced all other occupations by a ratio of 27 to one. And this demand is expected to continue through 2020. What’s more, high-tech and STEM employees are paid between 17 and 27 percent more than workers in other fields.

Finally, for what it’s worth, of the 15 best jobs for 2013 that CBS News’ Market Watch picked out, about half were in tech fields, including software developer, computer systems analyst, information security analyst, mechanical engineer, and computer programmer.

PHOTO: Biappi, Flickr