Tech Talk iconTech Talk

Thirty startups in high tech show-and-tell at Launch Silicon Valley 08


Thirty tech startups, selected from a field of 266, took the stage at Launch Silicon Valley Tuesday afternoon. Held at Microsoftâ''s Mountain View conference center, the event is the flagship conference of the Silicon Valley Association of Startup Entrepreneurs.

The stars of the afternoon, by popular vote, were Dayak, Loadstar Sensors, Triggit, Sensear, uTest, and Dial2Do. Dayak makes an online tool that matches employers looking to fill jobs with recruiters able to find candidates; employers post for free, setting the fee that theyâ''re willing to pay recruiters for a successful hire. Loadstar Sensors builds compact capacitive load sensors for automotive, aerospace, medical device, and industrial applications, and is currently developing a 2mm by 2.5mm pressure sensor. Triggit is an ad management technology to help web publishers embed advertising into their sites. Sensear is speech-enhancing technologyâ''more on that in a bit. uTest is a web-based platform for software testing that lets testing be distributed to people around the world. And Dial2Do is a phone service that uses voice recognition to let users send emails, text message, and create reminders.

My picks were a little different. (I admit that I didnâ''t see the presentations of all the

selected winners; presentations went on simultaneously in two separate venues, I zipped back and forth but couldnâ''t catch them all.) I completely agreed that Sensear stood out from the crowd. This sp1.jpg

technology out of Australia uses microphone arrays and digital signal processing techniques to sort out speech from noise in very noisy environments, the kind of places in which ear protection is required, or at least a very good idea. It doesnâ''t completely eliminate background noise, so you can tell if youâ''re about to be hit by a car, for example, but cuts it down, and then amplifies the speech. Besides industrial applications, Justin Miller, CEO sees the earplugs (about $300 to $550, depending on whether or not they talk to Bluetooth cell phones) used in the military, by bartenders, and, eventually, by all of us baby boomers who went to too many rock concerts.

Another type of earpiece caught my attention as well. Silicon Valley company Kleer demonstrated a proprietary wireless Kleerbuds.jpgtechnology for transmitting audio, to wireless earbuds, headphones, or remote speakers. Given how tangled I tend to get in my iPod headphone cord when exercising, this definitely looks like a winner, and the company has already licensed the technology to consumer products manufacturers.

I also was impressed by Ion Applications, a Florida company making handheld ion detection devices that can be used for explosives 600_SUBJECT_SCREENING-1.jpgscreening in airports, drug detection, even air quality monitoring in semiconductor clean rooms; in its current configuration, it resembles a handheld hair dryer. Alexander Lowe, vice president of sales and marketing, said that devices with comparable sensitivity today are hugeâ''think of the large portals that screen for explosives at airports. He envisions airport screeners walking up and down the lines of passengers scanning for explosive residue, a method that would enable every passenger to be screened without slowing down the boarding process.

Theory behind Nanotechnology Trade Shows and a Nanotechnology Industry

Nanotechnology over the last seven years since the inception of the NNI in 2001 has had more than its share of conferences, seminars and other assorted meetings. These are usually informational or provide some kind of networking that could lead to some kind of transactions.

But one incarnation of event that there has not been a lot of is expos or tradeshows. Nonetheless, they do exist and the two largest events of this type of are the nano tech event held each year in Japan and the NSTI Nanotech event held each year in the US. With the latter US event, a large component of the meeting is that of the presentations, or the conference element.

Nonetheless there is some expo element, and I became intrigued to see what was being exhibited at a nanotech trade show seven years on from the launching of the NNI when I saw this blog entry from Information Week. I did a little research and discovered by my unofficial count that exhibitors for this yearâ''s NSTI event that concluded on June 5th reached about 221. Okay itâ''s not the nearly 3,000 exhibitors at CES last January, but no one was expecting it to be.

However, it is not exactly heartening to see what makes up a large portion of the list of exhibitors: regional economic development groups, university and national labs, a number of microscopy tool companies, and a few companies focusing on nanomaterials. There are no Intels, IBMs or AMDs, or large players in any of the market sectors nanotech is supposed to be impacting: pharmaceuticals, automotive, and the like.

This is not a knock against the organizers, who have annually put on the largest nanotech event in the US for years now, but is rather an observation that strengthens the argument that nanotechnology is not an industry but an enabling technology.

Think of nanotechnology, or specifically a nanotechnology trade show, in the terms of a trade show for gold during the great Gold Rush. Who would be your exhibitors? Certainly, your pick and shovel suppliers (read, microscopy tool companies), real estate brokers for the areas around the best gold sites (read, your regional economic development boards), etc.

On the other hand, if you had a tradeshow for one of the market sectors gold would impact, say jewelry, you could easily imagine a large number of jewelers and all the elements in the supply chain from gold to the final jewelry product wanting to exhibit.

Itâ''s the same for nanotech. There are huge trade show and exhibitions for electronics, pharmaceuticals, automotive, etc. Whatâ''s the difference? These are industries, nanotech is a technology.

New York State Gets Behind Oxyfuel Carbon Capture

In a somewhat startling development, New Yorkâ''s governor David Paterson announced on June 10 that the state will support construction of an experimental â''oxyfiredâ'' electric generation plant, in which coal will be burned in an atmosphere of almost pure oxygen, so that nitrogen emissions are eliminated and carbon capture simplified. Swedenâ''s Vattenfall and Franceâ''s Alstom are completing a similar demonstration plant in eastern Germany, as described in the â''winners & losersâ'' January issue of Spectrum, and Babcock & Wilcox has had a serious oxyfuel R&D program in the United States. But oxyfuel has not been the mainstream approach to carbon capture on this side of the Atlantic, and New Yorkâ''s Jamestown plantâ''if builtâ''may be the worldâ''s first quasi-commercial demonstration of the technology.

Earlier this year, the U.S. energy department cancelled support for Future Gen, a public-private IGCC plant, in which coal is gasified, yielding a post-combustion mixture of carbon dioxide and hydrogen. That had been the overwhelmingly dominant approach to carbon capture in the United States. With FutureGen and IGCC facing increasingly uncertain prospects, the door appears to opening for alternative paths to carbon capture such as oxyfuelâ''though Spectrum bloggers have had radically different takes on the implications.

The Jamestown project, located in western New Yorkâ''s Chautauqua County, emerged from an industrial alliance of Praxair and Foster Wheeler with Dresser-Rand, E&E, Battelle, SUNY Buffalo, and AES. Praxair, a leading supplier of industrial gases, would provide both the oxygen supply and carbon capture technologies, while Foster Wheeler would build the fluidized-bed stream generators. New York State will invest up to $6 million to support development of the project, which could end up costing hundreds of millions of dollars to complete.

The governorâ''s rather daring decision to put the state behind the project is all the more striking because Paterson only recently assumed the top job in the wake of a sex scandal that toppled his predecessor. His announcement this week drew praise from unions such as brotherhood of electrical workers and the boilermakers but criticism from environmental and public interest groups such as NYPIRG, Sierra Club, and the American Lung Association. They declared it premature, given that environmental reviews are not yet complete.

Whether or not the plant is ultimately built and succeeds, the decision to pursue it puts New York into a league with countries like Sweden that are moving aggressively to test carbon capture and storage technology. Already, as the governorâ''s office reminded its constituents this week, New York is the most important member of the northeast Regional Greenhouse Gas Initiative, which has been developing this countryâ''s first cap-and-trade system for carbon, and it has adopted a renewables portfolio standard that seeks to make 25 percent of the stateâ''s electricity green by 2013. Its â''15 by 15â'' initiative aims to cut energy usage, by 2015, 15 percent below business-as-usual projections.

Requiem for a technology savior?

Americans have a long history of offering technological solutions to problems in Africa that may or may not exist -- or certainly don't exist in the form that the Americans imagine. From the start, Nicholas Negroponte, the computer visionary and professor at the Massachusetts Institute of Technology, was handicapped by a glaring lack of knowledge about life in African cities and villages. Worse, he behaved as if his lack of knowledge was a blessing, since he could create a beauitful engineering solution unencumbered with messy "inputs" from the African ground.

Negroponte's One Laptop Per Child project, however much a mis-match with African needs, has proved to be a tremendous stimulus for new thinking about low-cost computers. Negroponte's XO Laptop, with its innovative design and valuable hand-powered supplementary energy source, raised the bar for what is possible at low-cost in portable computing. New offerings -- from Intel's Classmate PC to the tiny Asus Eee computer -- demonstrate that Negroponte, despite his sincerity, has no monopoly on creative thinking about appropriate computer technology for African kids.

In the end, Negroponte may be best viewed as a successful promoter of a groudbreaking idea rather than as a technological innovator. His quest to assist African children in meeting their computing needs seems already to be overtaken by various other groups who are offering either more appealing alternatives or simply conventional laptops that satisfy the requirements of better-off Africans for quality -- and at prices -- between $300 to $500 -- that are low-enough.

i-LIMB prosthetic hand grabs UK's top engineering award

A day in the life of the i-LIMB

The i-LIMB hand, an advanced prosthesis developed by UK-based Touch Bionics, has won the UK's prestigious MacRobert award. Since 1969, that award has rewarded the year's best engineering project with a gold medal and a cash prize of £50,000 (US $98,630.50). The committee announced the 2008 winner at last night's Annual Awards Dinner.

The i-LIMB has five individually powered articulating digits, and it has been commercially available since mid-2007. So far, more than 200 people have been fitted for the hand, including U.S. soldiers who lost limbs in the Iraq war.

"The hand has two main unique features," said Touch Bionics CEO Stuart Mead in a BBC News article:

"The first is that we put a motor into each finger, which means that each finger is independently driven and can articulate.

"The second is that the thumb is rotatable through 90 degrees, in the same way as our thumbs are.

"The hand is the first prosthetic hand that replicates both the form and the function of the human hand."

On May 15, TouchBionics made its U.S. presence a permanent reality by acquiring

LIVINGSKIN, a New York-based company that makes the lifelike prosthetic skin on the i-LIMB hand. The company's realistic looking "skin" resembles human skin by simulating the three dermal layers of natural human skin.

I don't know what's happening with DEKA's Luke Arm, but it looks like Dean Kamen has some competition.

Nano Blogosphere Comments on Spectrum's Singularity Report

The Foresight Instituteâ''s Nanodot blog has commented on Richard Jonesâ'' â''Rupturing the Nanotech Raptureâ'' and has rightfully pointed out the somewhat problematic assertion contained in the contents page leading to the article that states â''Tiny robots that can fix all our bodily flaws sound lovely, but they violate the laws of physics.â''

Nanodot illustrates how other statements of Jones seemingly contradict this position. While I blog for Spectrum, I am not involved in their editorial team, so I can only venture a guess that the subheading was not written by Jones. So, this is not so much self-contradiction, but rather Jones simply doesnâ''t hold this rather polemic position.

Surely Jones expresses many of the problems in physics for diamondoid mechanosynthesis such as van der Waal force and Brownian motion, but he presents these as rather significant challenges, among others.

On the other end of the nano blogosphere comes TNTLogâ''s response to Nanodotâ''s comments, which wonders about the Foresightâ''s own contradiction in now arguing that â''the diamondoid mechanosynthesis approach is only in the very early stages of computer simulationâ'' while ten years ago it was presented rather as an advanced technology.

However, despite all the debate, it seems that discussion of mechanosynthesis has become a little less contentious over the last few years at least since the rather polarizing debates between Richard Smalley and Eric Drexler back in 2003.

Maybe science, which is inherently skeptical, will continue to be the guiding principle as opposed to dogma. Perhaps one of the more encouraging examples of this has been last yearâ''s release of a Technology Roadmap for Productive Nanosystems, which was accompanied at its release by Drexler himself conceding that even if self-replicating assemblers may be feasible, they may not be the best method for achieving APM, and that further refinements of the vision are to be expected.

New Champion Crowned in World of Supercomputers

Say hello to the new greatest computer on the planet--weighing in at a svelte 500 000 pounds.

Working with computer scientists at the Los Alamos National Laboratory (LANL), engineers from IBM Corp. have created the most powerful computing machine to ever crunch a number. Dubbed Roadrunner by its LANL backers in New Mexico, after the official state bird, it clocks in at a blistering 1.026 quadrillion calculations per second on a standardized test. It now assumes the mantle of world champion over its predecessor, IBM's BlueGene/L at sister research institute Lawrence Livermore National Laboratory in California.

According to a statement from IBM, Roadrunner is the world's first hybrid supercomputer, which means it runs thousands of the firm's Cell Broadband Engine units in conjunction with thousands of x86 microprocessors. The US $133 million project at LANL was designed to build a revolutionary computing system that could run at 1 petaflop (or 1000 trillion floating point operations per second). IBM notes that this is twice as fast as the BlueGene/L, which itself is nearly three times faster than the leading contenders on the current TOP 500 list of worldwide supercomputers.

The tasks ahead for this calculating marvel will consist mainly of ensuring the safety and reliability of the U.S. nuclear weapons stockpile for the Department of Energy, as well as facilitating research into astronomy, energy, human genome science, and climate change.

In its announcement, IBM made the following points about its groundbreaking system:

  • Roadrunner connects 6948 dual-core AMD Opteron chips with 12 960 Cell engines on IBM blade servers. It has 80 terabytes of memory and is housed in 288 refrigerator-size IBM BladeCenter racks occupying 6000 square feet. Its 10 000 network connections require 57 miles of fiber-optic cable.

  • The machine was built, tested, and benchmarked by IBM in Poughkeepsie, N.Y. Software development was led by IBM engineers in Austin, Tex., and by researchers in Yorktown Heights, N.Y. It uses the open-source Linux operating system from Red Hat Inc., of Raleigh, N.C.

  • Compared to most traditional supercomputer designs, Roadrunnerâ''s hybrid format sips power (3.9 megawatts) and delivers world-leading efficiency--376 million calculations per watt.

IBM also noted that it is developing new software to make Cell-powered hybrid computing broadly accessible. With corporate and academic partners, IBM is developing an open-source ecosystem that will bring hybrid supercomputing to financial services, energy exploration, and medical imaging industries among others.

"You do these record-setting things because you know that, in the end, we will push on to the next generation, and the one who is there first will be the leader," Nicholas M. Donofrio, an executive vice president at IBM told The New York Times, in an article published today.

Still housed at its design center in Poughkeepsie, IBM plans to ship the 250-ton Roadrunner system to LANL's supercomputing center, loaded onto 21 tractor-trailer trucks, later this summer.

[Editor's Note: Please see our early article on the Cell micrprocessor, Winner: Multimedia Monster, from our Winners and Losers special issue of January 2006.]

IEA Energy Prognosis Confirms Stern Climate Review

The International Energy Agency, the Paris organization that reports to the advanced industrial countries, has just issued the latest of its biennial Energy Technology Perspectives, saying the worldâ''s present course is not sustainable. If current trends continues, weâ''ll be consuming 70 percent more oil and generating 130 percent greater carbon dioxide emissions in 2050 than we are now. To get emissions back to present-day levels or lower by 2050 will require a necessary but achievable revolution in global energy technology, IEA director Nobuo Tanaka said last week.

The IEA report evaluates three scenarios and broadly confirms the controversial conclusions reached by Britainâ''s Stern Review last year. Getting CO2 emissions back to present levels by 2050 would require that a price of $50 per ton be put on carbon emissions, and reducing them by 50 percent would imply a price of $200/t. Achieving the 50 percent reduction target would require a total of $45 trillion in new energy investments, equivalent to roughly 1.1 percent of average annual GDP during the next four decades.

The Stern Review found that to keep carbon dioxide levels in the atmosphere no more than twice as high as they were before the industrial revolution began, emissions levels would have to be about 25 percent lower in 2050 than they are now. Attaining that goalâ''which Stern would pay for itself in terms of damages avoided and risks avertedâ''would cost about 1 percent of global GDP per year.

To achieve a 50 percent emissions cut by 2050, the IEA says the world each year would have to outfit 35 coal and 20 natural gas generating plants to capture carbon, at a cost of $1.5 billion each. It would have to build 32 new nuclear power plants and 17,500 wind turbines each year. The whole transportation sector would have to become more efficient by a factor of eight.

To achieve the 25 percent reduction needed to stabilize carbon levels at twice pre-industrial, the Stern Review said that the power sector would have to become 60-75 percent carbon-free. Exactly as the IEA now says, Stern said, â''Deep cuts in the transport sector are likely to be more difficult in the shorter term, but will ultimately be needed.â''

Out of Africa: Safaricom's large profits co-evolve with African broadband gap

For anyone who still thinks that poverty alone holds back the growth of telecommunications in sub-Saharan Africa isn't reading the financial results from Safaricom, a Kenyan mobile-phone operator that is controlled by Vodaphone, the giant British mobile conglomerate. As recently as 2000, Safaricom was a struggling phone company in East Africa, with just 20,000 customers. Eight years later, the Nairobi-based Safaricom now has 10 million customers. More impressive, Safaricom earned a whopping $223 million in its financial year ending in March, 2008.

Communications networks are key to technological and economic development. Safaricom, as a new article from The Economist amply documents, is one of Africa's pacesetters. The Economist wants readers to believe that Michael Joseph, the company's chief, has found a way of profiting from African poor, providing pay-as-you-go service to plain folks who can only afford to dish out money for mobile calling in small increments.

The Economist quotes Joseph, who spent his youth in South Africa, as saying that the increased availability of mobile-phone lines -- a new one costs as little as one dollar in Kenya -- "has been hugely more effecient than aid."

That's open to debate. Safaricom's success underscores the rise of a new upper-class in Africa's fastest-growing countries, of which Kenya surely counts as one. Mobile telephone providers are essentially skimming the cream from the top of Africa's economic "tigers." Safaricom is only the most successul in doing so.

To be sure, the company has distinguished itself from rivals by innovating a key service that does assist the poor. The service, called M-PESA, allows customers to transfer electronic money to one another through text messages. The Economist says M-PESA moves about $1.5 million a day across Kenya. That's still only a tiny fraction of what Western Union and Moneygram and mainstream banks move electronically, but it is indeed an impressive achievement.

As The Economist concludes: "Mobile banking could be the next stage of mobile-driven economic transformation."

Privately-owned companies such as Safaricom will surely take leadership roles in this next phase. But they cannot do it alone. Africa remains hampered by significant gaps in public infrastructure. In the latest issue of Science magazine, Calestous Juma, a professor at Harvard University, argues that African governments must do more to bring "affordable connectivity to Africa in general."

Juma insists that international aid donors do have a role to play in this transformation -- to help insure that "bandwidth must be sold at a fair price to all buyers" rather to the monopolies emerging in Africa's telecom markets.

Good point. Which brings us back to Safaricom's enormous profits. Whether they are extracted from the bulging wallets of East Africa's new wealthy or more "democratically" removed from the working poor, Safaricom's impressive earnings fundamentally flow from very high prices.

In Africa, per unit costs for mobile phone calls are among the highest in the world -- and Kenyans pay some of the highest rates in Africa. In the long run, the combination of technological change and competition should drive down those prices. But not any time soon.

Federal Trade Commission Investigating Intel Again

It looks like another chapter is being opened in the ongoing series of inquiries into the status of Intel Corp. as a monopoly. According to an article in today's New York Times, the U.S. Federal Trade Commission (FTC) has begun a new investigation of the world's leading manufacturer of microprocessors on grounds of possible anticompetitive business practices.

Officials at the FTC told the Times that the agency has issued subpoenas to Intel, as well as its competitors, to seek evidence of any actions by the Santa Clara, Calif., chipmaker that might violate U.S. antitrust regulations. Today's news focused chiefly on Intel's pricing policies, which its critics have repeatedly called unfair in recent times. Those critics, mainly from Intel's rivals such as Advanced Micro Devices (or AMD), charge that the chip giant uses its position in the marketplace and superior resources to sell sophisticated microprocessors to computer vendors below the cost of their manufacturing, an illegal practice known as dumping.

While Intel is by far the leading maker of these chips (a market valued at US $225 billion), its status as a near-monopoly is not a violation of federal antitrust statutes in the United States, where practical monopolies are legal as long as they do not conduct what are known as predatory business practices against their smaller competitors.

The new investigation appears to be a re-opening of a similar inquiry from a decade ago by the FTC. That case escalated into a full-blown antitrust proceeding against Intel by the federal government. It was only resolved when newly appointed litigators for the nascent Bush administration decided to settle the matter out of court in 2001.

The Times report today notes that the re-opened case coincides with a change of leadership at the FTC, where outgoing Chairman Deborah P. Majoras (an architect of the Justice Departmentâ''s antitrust settlement with Microsoft in 2001 coincidentally) has been replaced by incoming Chairman William E. Kovacic, who has acted with the approval of the other members of the trade commission.

The Times reports that the investigation is still in its early stages and will undoubtedly require months to come to a preliminary conclusion as to whether more intensive measures are necessary. This will almost certainly leave the prosecution of a potential case against Intel up to the next presidential administration to consider, beginning in 2009.

Meanwhile, similar investigations against Intel concerning the same or similar charges are proceeding apace in European and Asian nations.

As of today, no information has been posted on the matter at the Web sites of either Intel or the FTC.


Tech Talk

IEEE Spectrum’s general technology blog, featuring news, analysis, and opinions about engineering, consumer electronics, and technology and society, from the editorial staff and freelance contributors.

Newsletter Sign Up

Sign up for the Tech Alert newsletter and receive ground-breaking technology and science news from IEEE Spectrum every Thursday.

Load More