Tech Talk

p>The maverick British billionaire whose team built the first private spacecraft to reach sub-orbital flight has his sights set on commercializing the high frontier in the not so distant future. Yesterday, in New York City, Sir Richard Branson showed off a design of the customized cabin for the first of the new SpaceShipTwo (SS2) vehicles—to be christened the Enterprise.

Branson said the Virgin Galactic spacecraft, to be built by Scaled Composites, of Mojave, Calif., will be completed in the second half of 2007, begin testing in early 2008, and start commercial operations in 2009. Speaking at Wired Magazine's NextFest conference, the British entrepreneur said: "Construction of SS2 is underway at Burt Rutan's factory ... and this is the first opportunity for the public to get a sneak preview of the sheer scale of what is under construction there."

The cabin mockup was designed by leading British product designer Seymour Powell. It accommodates a crew of eight—two pilots and six passengers—who will have to fly to sub-orbital space experiencing a 3-G force and then recline during the period the vehicle is under zero G. During this time, expected to only last a few minutes, passengers will be allowed to unbuckle and float freely before they return to their seats for the descent to the surface. The price tag for the ride of a lifetime will initially be US $250 000, according to Branson.

The music and travel tycoon has planned for a fleet of five SS2s in the coming years. Each will be hoisted into launch position, mid-air, by carrier craft known as WhiteKnightTwos. The work being done on the new project extends the successful effort culminated two years ago by launches of SpaceShipOne, which soared to an altitude of 100 kilometers to claim the $10 million Ansari X Prize for the first privately owned vehicle to fly to outer space.

"Our vision is to successfully build the world's first environmentally benign space launch system and prove once and for all the commercial viability of a safe space launch system that we believe will eventually be capable of taking payload and science into space as well as people," Branson said yesterday, promising to make the SpaceShipTwo program as ecologically friendly as possible (last week, he pledged $3 billion of his fotune to support environmental causes). "If you're going to build a spaceship, you've got to build a green spaceship."

Once more, Senior Editor Tekla S. Perry reports from the DEMOfall show in San Diego on the latest gadgets looking to prove themselves to consumers. Here are her thoughts on Day Two. (For her Day One musings, click here.)

Tekla S. Perry

I've seen 67 product demonstrations at a distance and gone one-on-one with about two-thirds of these new products. Some make absolutely no sense to me.

For example, for about 40 cents a pop, I can add cartoon speech bubbles to photos I take with my cellphone and then send the altered images to my friends. I don't have a camera phone; but even if I did, I doubt I'd spend time and money "cranking" the images (which is what the company, photocrank, is calling this editing). Or I can download music for free to my cellphone or MP3 player, with Lirix, as long as I'm willing to listen to advertisements along with my music. I'm not. And then there's the wireless rabbit that will read the New York Times to me when I wake up in the morning and send me secret messages by tilting its ears. The Nabaztag from Violet has an appealing shape, a cross between a Pokemon and a lava lamp; but at US $150, I think not. The developers of those products, however, can console themselves with the fact that, as a 40-something mom, I'm not in their target demographic; they're mostly looking for 18 to 25 year olds, maybe 30 at the outside.

Most of the other products seemed reasonable but, for the most part, blur together in my mind. However, in spite of Demo overload, a few products that I saw on Day Two did manage to stand out. They fill a real need, are well designed and, while I can't be sure without testing them outside of the Demo environment, seem to work.

Living in Silicon Valley, I can't take a walk down the street without seeing people that appear to have Bluetooth earpieces permanently implanted on the sides of their heads. Could be, though, that they're going to want a device transplant after seeing the Mvox Duo. At $199, this unit looks like an ordinary Bluetooth earpiece, but it converts to a speakerphone, with impressive clarity and volume. It uses an omni-directional mike for noise canceling and a directional mike for picking up conversation. The thing I liked most is that Mvox got the little things right. A magnetic switch in the lapel-clip turns on the speakerphone; pull the earpiece out of the clip and the speakerphone switches off; this protects your ear from an accidental up-close overdose of sound. The lapel-clip fits nicely on a seatbelt; a feature I'll need in California very soon, because driving while holding a cellphone will soon be against the law. And a twist of the lapel-clip turns it into a microphone stand for conference table use. Very clever.

As the person in my household who is in charge of tech support, Retrevo is definitely for me. I rarely bother with specialized search engines, but this free one looks for specific solutions to problems, sorting past chit-chat and product reviews. And it looks inside manuals, not just on Web sites and blogs. Given that I tend to lose manuals, I can use this feature several times a year. The Retrevo demo involved figuring out how to reset a Slingbox that had stopped slinging, and easily found the procedure. But I had a trickier test for it. About a month ago the keyless remote for my Mazda MPV died; changing the battery didn't work, it just wasn't talking to the car anymore. I spent over an hour scouring the Internet for a solution beyond "take the car to an authorized service center." Eventually, I found the answer (an arcane combination of key turns and door slams that resembles the chicken dance) buried deep in a blog. When I proposed this search to the executives at Retrevo, they prepared me for failure. A car remote isn't really a consumer electronics product, they pointed out, definitely on the fringe at best, but they would give it a try. Retrevo found that deeply buried answer in about five seconds. We were all amazed.

The idea of having one phone number you can use for all your phones forever isn't a new idea. Call-forwarding is an early implementation; sometimes useful, but not simple or powerful enough to bother with regularly. GrandCentral gives you that single phone number and a Web interface. You can manage incoming callers by phone number, sending them selectively to your different phone numbers (work, cell, or home) directly into voice mail, or into the purgatory that is a spam folder. The voicemail service alone seems worth the price of admission—Grand Central's $14.99 a month is cheaper than the monthly fee for my AT&T voice-mailbox. And I like that you can use voicemail to record conversations; I can't do that with my existing voicemail.

I'll try Grand Central, and you can to, for free, for the next 60 days. You can also try Retrevo for free as well—can you come up with a better challenge than I did? Contact me at tDOTperryATieeeDOTorg to tell me about your experiences.

Surgeons are generally perceived as cool, precise operators who focus on their work with the concentration of a machine. One day soon, though, critical-care patients in places where immediate surgical response is not possible may find themselves under the scalpels of wireless robotic systems, which could be even more cool and precise than the skilled doctors at their controls. In our cover story this month "Doc at a Distance", authors Jacob Rosen and Blake Hannaford explain how remote surgery technologies are maturing and where they could be headed in the future.

They note that surgeons have remotely commanded robotic systems before, even with real patients under the knife—situations that took place in "well-equipped hospitals and relied on dedicated, wired communications channels." Their research, on the other hand, looks at how we might be able to cut the cords and perform true electronic telemedicine operations on severely wounded patients in chaotic conditions, such as battlefields. They ask, essentially, how we can break free from the wires.

Rosen and Hannaford (IEEE Fellow) are the co-directors of the University of Washington's BioRobotics Laboratory. They recently deployed a distance-medicine system to the hills of California for field-testing, under the aegis of the U.S. Army, and ran it through surgical procedures on, thankfully, anatomical dummies. The first conclusion they drew is that "surgical robots need plenty of improvement." They go on to say, however, that future deployments "will use imaging technologies such as ultrasound, MRI, and CT scans as their 'eyes', and they will break free from centuries of surgical convention, entering the body through existing openings and moving inside the patient as they make their way to the surgery area." They surmise:

As the technology matures, surgical robots promise to improve a wide range of procedures in terms of patient recovery time, cost, and safety. Medicine, however, is, as it should be, a conservative field, following Hippocrates' mantra: "I will keep [patients] from harm and injustice." In the next several decades, surgical robots, like many technologies introduced in medicine, will prove their value and become mainstream tools—tools always guided by a physician's judgment and dedication to the delivery of the best health care.

The BioRobotics Laboratory team in Washington participates in another military-funded initiative, the US $12 million Trauma Pod program, begun last year by the Defense Advanced Research Projects Agency and managed by SRI International, of Menlo Park, Calif. It seeks to develop an unmanned, mobile operating room that is equipped with a host of automated surgical systems and could be quickly dispatched anywhere in a war zone, according to the authors. You can learn more about this effort at the Army's Telemedicine and Advanced Technology Research Center.

Robotic medicine is coming—and it's coming faster than you think to a hot-spot near you.

Showing that sometimes you need to maintain your sense of humor in today's world of advanced technology, author Brian R. Santo this month examines the tech-speak we use routinely to convey complex concepts to others—and finds it to often be LOL funny when taken out of context (or sometimes even in context). In "Acronym Addiction", Santo compiles a list of acronyms and initialisms that should leave you shaking your head in wonder that we even manage to communicate effectively with each other at times.

To verify his thesis that the alphabet soup of contemporary technical jargon has spilled over our collective plate, Santo spoke with a number of new IEEE Fellows. Sandra Johnson, chief technology officer for IBM's global small and medium businesses, told Santo that she recalls being at a presentation that was so chock full of esoteric acronyms that she "leaned over to the people next to [her] and asked if they knew what the presenter was talking about, and they didn't." Her question got all the way around the room, but no one was familiar with all of the acronyms being used. "It was amusing," she said. "This guy was going to town, and no one knew what he was talking about."

In that spirit, here are some of our favorites from Santo's feature:

• ABT: Advanced BiCMOS technology. Building BiCMOS chips, which combine bipolar transistors and field-effect transistors, started out as a fairly complicated process; apparently it's become even more so.

• MHEMT: metamorphic high-electron-mobility transistor. An MHEMT is a variation of a high-electron-mobility transistor (HEMT), which is a type of really fast switch. MHEMTs are found in adaptive cruise-control radar in cars you probably cannot afford.

• PECL: positive emitter-coupled logic. A way of constructing logic circuits so that they operate faster. PECL is just ECL operated between positive voltage and ground. Another ECL variation is LVPECL (which we dearly wish were pronounced "love peckle").

• SED: surface-conduction electron-emitter display. That should be SCEED, right? We thought so. Normally we disdain cheating, but on the other hand we admire the chutzpah required to simply discard 40 percent of your word count to get to a marketable acronym.

• STRIFE: stress plus life (testing). A portmanteau posing as an acronym—there's no reason for this word to be in all capital letters other than the perversity of whoever minted it. We're fond of it anyway, because it actually means what it says.

• WAF: wife acceptance factor, wife approval factor. A product feature or modification sufficiently appealing to women that they will permit their husbands to buy the product.

Our absolute favorite, though, has got to be TLA: three-letter acronym. As Santo, a senior editor at CED (Communications Engineering and Design) and a former editor at EET (Electronic Engineering Times), notes: 'Two letters are rarely enough. And four letters or more give people the inexplicable urge to try to pronounce them, even if they shouldn't. Thus, the electronics industry's penchant for acronyms is so powerful it has its own acronym.'

With 26 letters in the Roman alphabet, there are a possible 17 576 TLAs in English, and sometimes it seem that every one of them is currently in use. Maybe we should offer a contest at Spectrum Online (SOL) to see who can come up with the most TLAs not in use—and see how long it takes clever technologists to fill them. That would be BAD—broken as designed. Or just plain bad, seriously. Let us know what you think.

>One of the most famous people alive has had to live for nearly 40 years with a lingering cloud over his head—all because of something he supposedly did not say. Recently, though, a technologist in Australia employing software that enhances speech for persons with disabilities stumbled upon a word that had been missing from the historical record for decades. When Neil Armstrong first stepped off the Apollo 11 lunar landing vehicle on 21 July 1969, he spoke the words that he had formulated only after landing on the moon and considering his role in an immensely human-intensive project: "That's one small step for a man, one giant leap for mankind."

The problem with these words, of course, is that they were not what was heard back at NASA Mission Control in Houston and subsequently transmitted around the world. Those words were: "That's one small step for man, one giant leap for mankind." Strictly speaking, if you parse the latter statement, you get the rather banal construction: "One small step for mankind, one giant leap for mankind." Which is, essentially, nonsense. Armstrong believed he had pronounced the sentence correctly; but when he returned to Earth a few days later, he was apprised that his famous statement had been garbled, tainting what should have been one of the towering moments in human achievement. And language mavens and critics of technology have never ceased since to make the remark a subject of mockery.

Then along comes Peter Shann Ford, CEO of Control Bionics, in Sydney, Australia, who while working on software that allows disabled people to communicate through computers using their nerve impulses decided earlier this year to run a sophisticated program on an archival sound file from NASA. Ford, an admitted "space junkie" (and someone who had covered the space program for CNN and NBC during the Eighties), found the missing "a" using a graphical analysis tool. In a brief account on his firm's Web site, "That's one small a...", he relates the story of his discovery and offers several links to relevant information—including his paper "Electronic Evidence and Physiological Reasoning Identifying the Elusive Vowel 'a' in Neil Armstrong's Statement on First Stepping onto the Lunar Surface" on the science of it.

Ford soon contacted Armstrong with his findings, and two weeks ago he presented his analysis to the first man on the moon at a meeting at the U.S. National Air and Space Museum. According to Ford, Armstrong had a vivid recollection of the moment and of the equipment used to transmit his first words from the lunar surface. After reviewing Ford's presentation, the world-famous astronaut offered a statement that said simply: "I have reviewed the data and Peter Ford's analysis of it, and I find the technology interesting and useful... I also find his conclusion persuasive. Persuasive is the appropriate word."

NASA has all along maintained that Armstrong spoke the famous words correctly. In its World Book entry for Neil Armstrong, the space agency states clearly that the first words spoken by a human on the moon, amidst the static, were: "That's one small step for a man, one giant leap for mankind."

Taken in context, and now proved by modern computer science, those words resonate through time and sound triumphant for all of us. A very humble man once spoke them on behalf of all of humanity. It's only fitting now that we finally give him credit for speaking them properly.

Earlier this month, the United States and Canada issued a final report by the U.S.-Canada Power System Outage Task Force on the causes of the August 2003 North American blackout and on recommendations for the prevention of future such events. The task force makes numerous recommendations to both governments to minimize the likelihood of future blackouts, reduce the scope of those that do occur, and improve the security of the North American power grid.

The 14 August 2003 blackout was the largest power outage in North American history, plunging some 50 million citizens of Ontario, Ohio, Michigan, Pennsylvania, New York, Vermont, Massachusetts, Connecticut, and New Jersey into darkness. The report notes that, while the origin of the blackout may have started in a power system in Ohio, the ultimate impact of the source failure was compounded by "long-standing institutional failures and weaknesses that need to be understood and corrected in order to maintain reliability."

The 221-page report states at the very beginning that: '[T]he blackout could have been prevented and that immediate actions must be taken in both the United States and Canada to ensure that our electric system is more reliable. First and foremost, compliance with reliability rules must be made mandatory with substantial penalties for non-compliance.' To this end, the task force emphasizes that 'significant accomplishments' in the last three years have been achieved.

An accompanying announcement from U.S. Department of Energy Secretary Samuel W. Bodman and Minister of Natural Resources for Canada Gary Lunn states: 'Mandatory reliability standards are being implemented in the United States and in jurisdictions across Canada. The North American Electric Reliability Council is submitting 118 new standards to the U.S. Federal Energy Regulatory Commission (FERC) and appropriate authorities in Canada for review and approval. Additional standards are also being developed.'

Bodman writes: "I appreciate the hard work and diligence that went into this important report. It demonstrates that while improvements are being made to enhance grid reliability, we still have a very complex system that is subject to possible mechanical and human failures. We must remain vigilant."

Lunn adds: "The Task Force has been an outstanding example of close cooperation between the governments of Canada and United States, and we have established a Bilateral Electric Reliability Oversight Group for collaboration between authorities in both countries on issues of common concern."

Additional documentation on the response to the blackout is available at the Department of Energy's August 2003 Blackout Web site.

p>

Today, Associate Editor Samuel K. Moore offers a mea culpa to a small tech firm that has confounded his doubts by waging a patent fight against some of the microprocessor giants—and winning.

Samuel K. Moore

Back in 2003, when I first read a press release from Patriot Scientific, of San Diego, I must admit I laughed out loud. This troubled, profitless speck of a company said it was suing five of the biggest consumer electronics makers in Japan for infringing patents it held that described technology used in just about every microprocessor made since the early 1990s. That's a lot of processors. To top off this brazen claim, the company started filing suits on Christmas Eve. I didn't think Patriot had a chance.

Perhaps I should not have been so dismissive. After years of legal hardball, the company has made good on its claims, to the tune of US $35.9 million in revenue for the fiscal year ending 31 May. Patriot's annual report, filed on 13 October, gives the first comprehensive glimpse of how well the company has convinced major electronics makers to knuckle under. In the first nine months of 2006, the company's jointly owned licensing company, Phoenix Digital Solutions, hooked nine big fish: Casio, Fujitsu, HP, Seiko Epson, Sony, Nikon, PENTAX, Olympus, and Kenwood. AMD and Intel gave in last year. From 1 June to 3 October Phoenix brought in $32.7 million from some of these deals; Patriot should see a little less than half of that. As you'd expect, Patriot doesn't disclose how much each one paid. In conversation last year, a lawyer involved in the licensing indicated that Patriot wanted to attract customers by cutting the first few licensees a cheap deal. I don't know if any of the cheap licenses are left, but clearly the strategy has worked.

The issue concerns the so-called Moore Microprocessor Patents (MMP), named for Charles Moore, the inventor of the Forth programming language. A good example of an affected technology involves clocks, the drumbeaters that keep all a circuit's soldiers marching in step. In the old days, microprocessors ran off the same clock signal as the rest of the computer, and that was no problem at speeds below around 120 MHz. The trouble is that the computer's clock signal has to follow many centimeters of copper wiring all over the computer, leading to delays. Charles Moore and Russell Fish dreamed up the solution of letting the processor clock run as fast as you want while synchronizing it to the computer's much slower clock, enabling speeds in the gigahertz range. And that's the way things have worked for the past decade or so.

How a seeming pipsqueak like Patriot wound up with the keys to the microprocessor kingdom is a complicated tale, but here goes.

Fish and Moore came up with their technology while developing the SH-Boom microprocessor in the late 1980s. Sometime before the patents issued, Fish transferred his interest in SH-Boom to a family trust, which sold it to a company called Nanotronics, which sold it to Patriot in 1995. Patriot apparently didn't realize the value of the patents until the early 2000s, a time of stock taking for many holders of intellectual property. (Remember when British Telecom briefly thought it owned hypertext linking?)

Patriot saw it had been leaving money on the table and embarked on its Christmas quest for cash, but it quickly got stuck (unless all owners of a patent sue, a court case will get nowhere), and Moore wouldn't go along with Patriot's plan. His Silicon Valley licensing firm, Technology Properties, Ltd. (TPL), began a legal battle for full ownership, and Patriot's dreams of riches seemed to recede.

But the big technology companies apparently feared infringing on the patents and rushed to cut a deal no matter what the legal niceties. At one point, AMD took a $1.7-million license from Patriot, and Intel took a license from Moore. (AMD put both oars in, paying additional cash for Patriot's original product, the processor that succeeded Moore and Fish's SH-boom processor.)

The promise of vast sums of money can make even bitter rivals kiss and make up; and in 2005, TPL, Moore, and Patriot set aside their suits and formed Phoenix Digital, a 50-50 joint venture for licensing the patents.

Patriot and its shareholders are enjoying their first profitable year ever, and Moore has moved on to found a fabless multicore processor firm, Intelasys. But the tale of the patents may not be over. Russell Fish and his family trust sued Patriot seeking a piece of the action. The Fish family and Patriot have been in mediation since 11 September. Stay tuned.

Luck plays a role in any technology project, but critics rarely give it due credit, preferring instead to blame failures on design. This tendency was on display last week, when pundits called North Korea's nuclear test a failure because it had the unusually low yield of about half a kiloton. Even the comedians got in their licks.

"The blast was so small that many scientists are saying it was a dud," said late-night television host Conan O'Brian. "Apparently, the nuclear bomb didn't work well because it was made in Korea."

No strategist ever went right by bad-mouthing an adversary, and so we welcome the corrective supplied by Richard L. Garwin, a distinguished physicist who helped design the first hydrogen bomb and has advised the U.S. government ever since. In a Q&A with Spectrum's William Sweet, Garwin agrees that the North Koreans must have wanted a bigger bang and thus may indeed have miscalculated. But he maintains that a chance occurrence could have caused a perfectly fine bomb to fizzle.

The chance occurrence, called predetonation, happens when stray neutrons trigger a tiny chain reaction before enough plutonium can be crammed into a small enough volume for a long enough time. Energy released in that reaction blows the plutonium apart before much of it has reacted.

Garwin points out that the possibility troubled the sleep of the original bomb-makers in Los Alamos, N.M., during World War II. "Oppenheimer said there was a 2 or 3 percent chance of a fizzle," he says, "and that there might be a substantial reduction in yield—but that it could never go lower than 5 or 10 percent of the expected yield." The North Koreans may thus have expected 5 to 10 kilotons. [The post originally said megatons, a rather large error, and we thank Dinesh Bansal for pointing it out to us in his comment, below.]

If so, then the designers may not need to make changes; they can just try again, and hope for better luck. "I'd expect another test within a few months that's likely to be a successful 4- or 5-kiloton device," Garwin says.

If current assessments of the North Korean plutonium stockpile are correct, such a "destructive test" would still leave enough material for another five bombs. That's one for each neighboring country and Uncle Sam besides.

>With nationwide elections in the United States now only two weeks away, many are increasingly focusing renewed attention on the hardware and software that will enable about 80 percent of Americans to cast their votes for political candidates. A federal law known as the Help America Vote Act of 2002 (HAVA) requires election officials throughout the country to replace paper-based voting machines with controversial electronic equipment. It has led to quite a few reports that the new machines have flaws that could lead to vote tampering. Moreover, a subsequent provision of the law mandates that the states must build databases of registered voters and use these to check on the eligibility of individuals to cast ballots. In this month's news analysis "The Next Voting Debacle?", Senior Associate Editor Steven Cherry reviews the use of the databases that will guard the nation's polling booths on 7 November—and his findings are cause for concern.

Cherry writes that HAVA gives states wide latitude in reacting to database mismatches. If a voter's registration information and the data in other government databases differ, the act does not require that a state keep registrants off the rolls; but it also doesn't forbid the states to do so. State officials have therefore set up their own rules, which have resulted in mass purges of registrants in California, Iowa, South Dakota, Texas, and Washington. If you're a U.S. citizen interested in the technology used in elections, this item is must reading.

Cherry goes further into the topic on Spectrum Radio. In "Electronic Voting: Computerized Voter Rolls Pose Problems, Too", currently our top story (available as a podcast), he speaks with Wendy Weiser, deputy director of the Democracy Program at the Brennan Center for Justice at the New York University School of Law, in New York City. She notes that, among other suspect practices, some states are using the databases as a barrier to voting, for the most trivial of mismatches.

Of course, database problems are not the most controversial issue with electronic voting. That remains the province of the new polling machines. Recently, a group studying e-voting at Princeton University demonstrated how a malicious hacker could access a supposedly secure machine and, within a minute or so, download software that would cause it to tabulate false results—and even infect other units. (We blogged about this topic last month in "Worst Machine Ever?".) In this streaming video presentation by Edward Felten, director of the Center for Information Technology Policy, you can see for yourself just how easy this form of election fraud can be committed.

Expect even more scrutiny of the role of computers in e-voting after 2 November when HBO airs a documentary called "Hacking Democracy", in which the producers note that, "[C]oncerns over the integrity of electronic voting are growing by the day. And if the voting process is not secure, neither is America's democracy." (They point out, for example, that in the 2000 presidential race an e-voting machine recorded negative 16 022 votes for Al Gore in Volusia County, Fla.)

With the future direction of the United States riding on its national elections, Americans should be doing all they can to require officials to tighten the security of the new machines and systems that enable them to exercise their franchise and count their votes accurately as intended. It's the cornerstone of democracy. If it fails, democracy will fail with it.