In the “interesting new application” department, I ran across an announcement about a new application developed by a research group at Northwestern University that helps manage your laptop’s power management subsystem. This is a very interesting development that leverages existing hardware in a new and different way. It’s rather like the earthquake-detection network based on the accelerometer built into many modern laptops.

The objective is to “shut off the display immediately when the user leaves the computer rather than using slow and error-prone mouse/keyboard activity timeouts.” We all know and, generally, hate these timeouts. They’re what cause your display to go blank at exactly the wrong time — usually while you’re on the phone, taking notes, or (worst of all) watching a movie. But these screen-blank systems are supposed to help save power while running in battery mode, so it’s not a good idea to set them to a long delay like 30 minutes.

The new application makes use of the laptop’s built-in microphone and speakers, and basically acts like sonar. The researchers “developed software that, when the user is not using the mouse or keyboard, plays a tone at a high frequency and records the tone’s echo. The computer then processes the tone and filters out everything except the frequency and looks for variance.” If the application “hears” the same tone for 10 seconds or more, it can be pretty sure no one’s in front of the machine. User movement, even slight hand or head movements, should produce variations in the tone that will prevent the software from turning off the display.

Even cooler, it “can detect movements from up to about eight feet away.” So you don’t have to be right in front of the system for the software to work properly.

That’s pretty clever.

If you’re interested, you can help test it out. The group is “currently looking for users to download the software and test it to see just how much power it saves.” Download the application from their website, set it up on your laptop, and let it do its thing. It’s supposed to record the amount of power it’s saved over time, transmitting it (with no personally identifying data) back to the researchers who developed it. Is this worthwhile? Try it out and see!

The keyboard and mouse are almost certainly the most recognized I/O (Input/Output) devices ever made for computers. Type on the keyboard, move the mouse (or “rodent” as we used to call them). What could be easier? This model is certainly a lot more intuitive than punch cards, entering instructions via individual switches, or old-style paper terminals. But they’re both fairly old technologies. The keyboard, of course, is derived from (even older) typewriters and teletypes. The mouse has been infesting desktops since at least the 1960s, and tablet-style pointing devices may be even older.

Manufacturers and researchers have been looking for better, even more intuitive I/O devices since the first keyboard rolled off the assembly line. The whole problem involves what’s known in psychological and design circles as “mapping,” or the way humans interpret and use tools. For instance, look at a standard round doorknob. Its very appearance suggests one should grab and turn it. Its function is mapped into our brains, so it’s fairly intuitive. Likewise with a mouse. Move it up and down, or left and right. The pointer on screen mimics that motion, so the movement is natural. Good brain mapping translates to good design.

Given all this, one of the holy grails of computing design is the touchscreen monitor. If it works properly, it eliminates the need for the mouse. It also provides direct mapping. Touch the screen in the right place and an event occurs.

This is probably why HP has been developing what are arguably the best touchscreens on the market. A new one, the model 600-1055, has just shown up on the market. It looks pretty good overall, and has full integration with the soon-to-be-released Windows 7 OS.

Probably the only real reason touchscreens haven’t taken over the market is cost. They’re far more expensive than regular monitors, and have had reliability problems in the past. At $1600 (street) the new HP continues the cost trend, but reviews indicate it’s way better than earlier models. This may be just the boost the touch paradigm needs. But I’m not sure I’d give up my desktop mouse, since I don’t like the idea of reaching for the screen. It’d be less of an annoyance on a laptop!

One of the big advantages of the Internet as a whole is that, to some degree, it puts the sum total of human knowledge at anyone’s fingertips. Rather than sifting through masses of paper and dusty books, you can run a search query from the right archive and probably will find what you want in short order.

I’m reminded of the Harvard professor who used to hand the same assignment to his freshman students every year. He knew which references they’d require, where they’d to look, and how long it should take to complete the work. The exercise was designed to teach them how to do research, use the libraries, and deal with source material.

In 1995 or so, he came to us and demanded that the university cut off student access to this “Internet” thing. Why? Because he’d given the assignment to his students again, and they’d finished it in a day. He was astounded, and asked how they’d done it. Guess what the answer was?

This said, there are dangers to having everything available only in digital form, like the big Google Books project. What if the scanned books include errors? What if certain “inconvenient” works (maybe controversial or little-known resources) aren’t made available? What happens if the main Google data center gets hit by an asteroid? Or, what if Google simply loses interest in the project sometime in the future? They’ve done it before, and could easily do so again.

Personally, I like books. Real, paper books. I hate reading long documents online, and usually print them out (the “paperless society” is a huge myth, incidentally). I like digging through huge libraries, since I never know what I’ll find next to the book I was looking at originally. Sure, this experience can be duplicated to some degree in a digital format. And yes, electronic documents make it easier for multiple people to read certain material simultaneously. But for me it’s just not the same.

Central data storage, whether of books, government documents, or medical records, is a good thing. But it shouldn’t be the only option. Decentralization means less disruption in cases of a major data center outage or other disaster. And keeping all ones’ eggs, or books, in a single basket is just asking for trouble.

Nearly everyone has been affected in some way by the global economic downturn. Unemployment has risen to 9+% while markets have stagnated and housing prices have plummeted in many areas. As I’ve noted before, high tech suffered a pretty significant hit as well: layoffs, pay cuts, benefits cuts, and other bad news has hit the industry hard in the last year.

Surprisingly though, it appears the industry wasn’t hit as hard as many others were. The loss of high tech jobs, by comparison with auto workers, retail, and other industries, was relatively minor. This might not be very comforting if you’re a laid-off techie now looking for work, but things could probably be a whole lot worse.

According to some recent statistics, job losses declined in the June-July period and tech actually gained somewhat. The analysis suggests that “despite an uptick in jobs in June, the high-tech industry overall employs fewer people than in 2008. The industry as of June 2009 employed some 5.81 million workers, a decline of 224,100 jobs (3.7%) since June 2008.”

The per-sector numbers mirror industry trends, with more jobs lost in networking (45,000), engineering/tech services (21,500) and communications (13,000) than in the more lucrative services & consulting areas (10,500). None of these numbers are surprising. And if you’re in the industry, you probably know that “customer-facing” positions — consulting, professional services, and even customer support — are where the jobs are.

The good news is that the economy seems to be on the mend…somewhat. There are no guarantees, but we may start seeing a better job market starting later this year. Many companies held off on upgrades, purchases, and new projects during the deepest portion of the downturn. They should start buying again soon, as the next round of major upgrades becomes unavoidable.

If you’re out of work and looking, brush up your skills. If you’ve managed to retain your job so far, keep learning new technologies and don’t let yours become “stale.” Engineers have a shelf life, just like any other product. What does your sell-by date look like?

Several months ago I posted an article about a university project called “Vanish,” which allowed documents to become gradually unreadable using distributed encryption key elements scattered across a large network. The researchers who created this little scheme figured (as do all code makers, I suspect) that their system was unbreakable, since the key would eventually degrade as elements were lost over time. This would render “stale” messages unreadable, even if they still existed somewhere.

These guys, it appears, should have taken a lesson from the “unsinkable” Titanic.

Today it was announced that some competing researchers at the University of Washington have figured out how to defeat Vanish. And their method was so obvious that one wonders why the original inventors didn’t think of it. They simply caused a single computer on a Vanish network to masquerade as more than one node. This caused the computer to accumulate more than one component of the encryption key, which made it much easier to reassemble.

As the researchers said, “rogue machine would simply need to capture and store anything that looked like a Vanish key fragment. The researchers said that this was simple, as the Vanish fragments are identifiable because of their size. Later it would be possible to reconstruct a Vanish message by simply consulting the Unvanish archive.”

Of course, the original team was quick to fire back. They said they’d already modified their initial approach, thus making it much harder for a single machine to be used to accumulate multiple key fragments. “The newly discovered weaknesses with our initial research prototype are not an invalidation of Vanish,” said Tadayoshi Kohno, a University of Washington computer scientist.

All I can say is this: no system is invulnerable to penetration. Any encryption can be defeated, and in some cases that’s not a bad thing. As some companies are now finding, encrypted documents for which they have no decryption key can cause legal issues (including contempt of court charges) if they need to be produced during a legal discovery hearing. This is an emerging area of technology, and business is still catching up.

I’ve never been one for big, flashy office spaces. And despite being a long-term geek, I don’t have an office filled to the brim with tech. I suffice quite nicely with a single monitor, three office machines (laptop, Linux development server, Windows general-use box) and a general home network. I don’t have the time, or the energy, to devote to designing a huge office filled with toys. Plus, I’d have to dust around all that hardware. That’s just not worth the effort.

That said, I must admit to a major case of geek envy when I saw Stefan Didak’s new office. Wow. Just wow. Heck, I’d take his old setup off his hands, just to get all those great monitors. This place looks like something you’d find at NASA or in a relatively recent Star Trek movie. And it’s all functional, which is even better. He actually uses all this equipment. It manages his various websites, provides a development environment for his distributed programming team, and (shocking!) is almost never used for gaming.

As he notes on his site, he uses all the monitors pretty much simultaneously. Unlike some of us, he hates minimizing windows or having one application obstructed by another, so he buys enough screen real estate to make sure that never happens. I can understand how he feels, and have sometimes considered adding a second monitor to my current environment. Screen real estate is a major consideration, especially when you’re writing, testing, and debugging code in certain environments.

This type of setup would also be useful for network monitoring offices (and I’ve seen several that look very similar). Such offices often need to display statistical and availability information for multiple networks, systems, and other resources.

It should be emphasized that all of the systems Mr. Didak has in his office aren’t used as workstations. He has hundreds of gigabytes of storage space, probably set up in a RAID array for safety and redundancy purposes, that are used for large code development offices. Effectively, he has a small server room in his house.

Do you have a room like this, or (scarier still) one that’s even more elaborate? Post a URL of yours in the Comments section, and we can all have a geekfest.

Anyone who hasn’t been living under a (technical) rock knows that Intel has announced its latest series of processors, code named “Lynnfield.” I always find technical-project code names interesting, but have no idea where some teams come up with theirs. This one promises to be interesting.

As evidenced by PC World’s recently published tests, the new release appears to be pretty typical in terms of features…but is at least somewhat faster than the current crop of CPUs, as one would expect. Over the last few years, processors have tended toward “multi core” architectures, in which one chip holds several actual CPU layouts. Even the fastest single CPU can only perform one instruction at a time, so the multi core design allows them to work like dual, or even quad processor systems.

In general terms, old-style programs were basically linear. They expected to request execution of an instruction, then waited for the CPU to deliver the result. Newer programs designed for multi-core or multi-CPU systems (especially things like games, transaction processing software, search engines, and some webservers) can detect the presence of more than one processor. They’ll then fire off multiple independent processes that each perform a certain task; or they’ll use “threads” to do the same thing. A thread is a task that runs under a single process (they’re also referred to as “lightweight processes”), executing a task and terminating when it’s done.

Thread-aware programs can fire off multiple threads simultaneously, increasing overall speed by having each processor perform a separate task. But since they don’t start and stop separate processes to do so, lots of time and memory is saved.

The new Intel processors expand on the multi-core offerings of previous releases. You probably won’t see them in cheaper (low end) systems, but they’ll start appearing in mid- and upper-range systems soon. Hey, just in time for the holidays! What a coincidence.

It was interesting to see that one of the United Kingdom’s first computers, the Harwell, is being taken off display for a full restoration project requiring a full year’s work. Even better, they’ve apparently scared up enough spare parts and expertise to render it again fully functional, which means future students of computing (as well as the general public) will be able to see it work.

The Harwell was built in 1949. Remember, that’s long before the arrival of this tiny little thing called the transistor. Hence the whole system is based on relays and vacuum tubes (yes, really!). Each of the 900 tubes in the system could hold 1 digit worth of information. Now, remember that each tube is probably at least 1×3″ in size, and probably larger than that. The system filled most of a room, and printed its output on paper tape.

The original system was used until 1973, when it was mothballed and put on display. It’s exciting that the whole unit was actually preserved, unlike other early systems like the Harvard/IBM Mark I. Only one section of that landmark system was actually preserved. It’s understandable in a way, since even the preserved section is 40′ or so long and 6′ high, but it would be nice to have the whole system available for teaching and demonstration purposes.

These early systems had no “operating system” in the modern sense. They were solely used for computational purposes, speeding up calculations and taking the burden off humans who ordinarily had to perform them by hand. Early systems were used mainly for defense purposes during World War II, and spread from there into civilian businesses.

Pioneers in early computing would be astounded to see even the oldest, clunkiest PC. The idea that the average homeowner or student could have more computing power on their desk (and largely unoccupied) than existed on the whole planet in the 1940s would absolutely blow their minds. Sixty years after these first systems appeared we have a worldwide network providing instantaneous communications, not to mention enough online storage to hold whole libraries worth of data. Not bad.

Where will we be in another sixty years?

Disk drives are cheap. Anyone can now have several Terabytes of storage in a home PC for a few hundred dollars, whereas the first “home computer” Winchester-style drive I recall seeing was a 5 Megabyte (yes, really) model that retailed for around $5,000. That was in the late 1970s. You could run your whole OS on it while storing hundreds of files. Amazing!

Given that Terabytes are commonplace, Petabyte-level storage is what the big boys — ISPs, large corporations, and so forth — now manage. You can buy Petabyte-sized storage arrays from EMC, Sun, and other big companies for hundreds of thousands (if not millions) of dollars…or you can do what these guys did and build one yourself. And it costs less than $8,000!

As this company found, a lot of the hype behind big iron (that’s “large computing systems” to non-techies) is in the support, marketing, and so-called “features” they provide. I’ve worked in high tech for several decades, and I know this is true. For instance, I write patient management systems for researchers and hospitals, and can deliver a basic system for under $100,000 even if I have to develop it from scratch. By contrast, commercially available systems go for millions of dollars, not to mention ongoing support costs.

Big companies want you to believe you need all the support, warranties, and other services they provide. If you didn’t believe in “the name” and “the reputation,” you’d go elsewhere. And yes, if you’re in business and want a pre-built system for which someone else is completely liable, by all means go pay for one. Many companies are better off outsourcing bits of work that lie outside their “core competencies” — finance companies, banks, airlines, and others may want to put their computing and other needs in the hands of professionals who do those things for a living. But they’ll pay out the nose for those systems and services.

So it’s not far-fetched to believe that you can build the same storage array Amazon charges $2M for using $8,000 worth of commonly available parts. This “backblaze” array uses standard 1.5TB drives, SATA controllers, and other off-the-shelf components. The only custom item they have is the cage and backplane assembly, which they bought from a fabricator for under $800 per unit.

Whether you’re aware of it or not, most of your PC’s capacity is most likely totally untapped. Unless you render complex graphics (a task more often handled by the video card these days) or run heavy duty computational tasks, your PC’s CPU is very likely at least 50% idle…possibly even 75%. And when you’re not typing or running applications, it’s closer to 99%.

Not only that, you’re probably using only a small percentage of the memory and other resources available on your machine. So why not contribute them to a good cause?

I’ve mentioned it before, but numerous “distributed computation” projects would love to use all those spare cycles for good causes. If you’re skeptical, check out a list of top projects you can donate cycles to by downloading a client such as BOINC (a distributed processing application from the University of Berkeley). Depending on what type of project you find worthwhile, you can help climate change research, earthquake prediction, cancer prevention, or even LHC (Large Hadron Collider) physics/chemistry modeling.

If you get a copy of the legitimate BOINC client (as always be careful where you obtain any application) you can register and connect to these, or a dozen other projects. You won’t notice the processor time and memory use as long as your system and the application are configured correctly. The client is aware of other activity on your PC, and only does work when the system is relatively idle.

I started using the BOINC client years ago, when it was originated by the SETI@Home project. I and a number of friends helped search for signals from outer space, which was a fun way to use spare cycles. It was even better, since I had a lab filled with mostly-idle systems and racked up a huge number of processing hours. Today’s projects are at least as interesting, and even more relevant.

Go donate some spare processing time. It’s a lot better use of your PC’s idle time. And who knows what your home PC, or even one at work, might discover.