Finding The Invisible In The Visible

A team of scientists at the Massachusetts Institute of Technology has developed a computer program that reveals colors and motions in video that are otherwise invisible to the naked eye.

Amazing AI Robot

Courtesy Of NOVA

You can view the full episode here: "PBS"

Arepo Simulation Of Galaxy Formation

This computer animation, created using new software called Arepo, simulates 9 billion years of cosmic history. 

Arepo can accurately follow the birth and evolution of thousands of galaxies over billions of years. 

Arepo generates the full variety of galaxies seen locally, including majestic spirals like the Milky Way and Andromeda.

Credit: CfA/UCSD/HITS/M. Vogelsberger (CfA) & V. Springel (HITS)

The Computer That Never Crashes

A revolutionary new computer based on the apparent chaos of nature can reprogram itself if it finds a fault

OUT of chaos, comes order. A computer that mimics the apparent randomness found in nature can instantly recover from crashes by repairing corrupted data.

Dubbed a "systemic" computer, the self-repairing machine now operating atUniversity College London (UCL) could keep mission-critical systems working. For instance, it could allow drones to reprogram themselves to cope with combat damage, or help create more realistic models of the human brain.

Everyday computers are ill suited to modelling natural processes such as how neurons work or how bees swarm. This is because they plod along sequentially, executing one instruction at a time. "Nature isn't like that," says UCL computer scientist Peter Bentley. "Its processes are distributed, decentralised and probabilistic. And they are fault tolerant, able to heal themselves. A computer should be able to do that."

Today's computers work steadily through a list of instructions: one is fetched from the memory and executed, then the result of the computation is stashed in memory. That is then repeated – all under the control of a sequential timer called a program counter. While the method is great for number-crunching, it doesn't lend itself to simultaneous operations. "Even when it feels like your computer is running all your software at the same time, it is just pretending to do that, flicking its attention very quickly between each program," Bentley says.

He and UCL's Christos Sakellariou have created a computer in which data is married up with instructions on what to do with it. For example, it links the temperature outside with what to do if it's too hot. It then divides the results up into pools of digital entities called "systems".

Each system has a memory containing context-sensitive data that means it can only interact with other, similar systems. Rather than using a program counter, the systems are executed at times chosen by a pseudorandom number generator, designed to mimic nature's randomness. The systems carry out their instructions simultaneously, with no one system taking precedence over the others, says Bentley. "The pool of systems interact in parallel, and randomly, and the result of a computation simply emerges from those interactions," he says.

It doesn't sound like it should work, but it does. Bentley will tell a conference on evolvable systems in Singapore in April that it works much faster than expected.

Crucially, the systemic computer contains multiple copies of its instructions distributed across its many systems, so if one system becomes corruptedthe computer can access another clean copy to repair its own code. And unlike conventional operating systems that crash when they can't access a bit of memory, the systemic computer carries on regardless because each individual system carries its own memory.

The pair are now working on teaching the computer to rewrite its own code in response to changes in its environment, through machine learning.

"It's interesting work," says Steve Furber at the University of Manchester, UK, who is developing a billion-neuron, brain-like computer called Spinnaker (see "Build yourself a brain"). Indeed, he could even help out the UCL team. "Spinnaker would be a good programmable platform for modelling much larger-scale systemic computing systems," he says.

This article appeared in print under the headline "Machine, heal thyself"

Build Yourself A Brain

The systemic computer takes its lead from nature (see main story), but so does Spinnaker, an ambitious project at the University of Manchester, UK, to build a one-billion-neuron computer from microchips. The idea is to create a supercomputer that works just like the human brain using the same ARM chips that power most smartphones. The team wants to do parallel simulation of large-scale neural networks using the equivalent of 1 per cent of the human brain's neuron count. They are well on their way: using chips that model 1000 neurons each, their system has created the equivalent of 750,000 neurons. "We're advancing slowly but steadily," says project leader Steve Furber.

Via: "New Scientist"

Google To Create An Artificial Super-Mind

Famed AI researcher and incorrigable singularity forecaster Ray Kurzweil recently shed some more light on what his new job at Google will entail. It seems that he does, indeed, plan to build a prodigious artificial intelligence, which he hopes will understand the world to a much more sophisticated degree than anything built before–or at least that will act as if it does.

Kurzweil’s AI will be designed to analyze the vast quantities of information Google collects and to then serve as a super-intelligent personal assistant. He suggests it could eavesdrop on your every phone conversation and email exchange and then provide interesting and important information before you ever knew you wanted it. It sounds like a scary-smart version of Google Now (see “Google’s Answer to Siri Thinks Ahead”).

Kurzweil says this of his project at Google, in a video posted by The Singularity Hub:

“There’s no more important project than understanding Intelligence and recreating it. I do envision a fundamental approach based on everything we understand about how the human brain [works]. And there are some things we don’t yet understand so I plan to go off and explore some of my own ideas about how certain things work.”

Kurzweil makes it sound like the effort will be based on the theory of the put forward in his new book, How to Create a Mind. In this work, based largely on observations about current trends in AI research, and his own work on speech and character recognition, Kurzweil suggests a fairly simple mechanism by which information is captured and accessed hierarchically throughout the neo-cortex, and posits that this phenomenon can explain the miracle of human conscious experience.

Kurzweil’s claims are certainly bold, and some have criticized them as hopelessly naïve. Indeed, it’s easy to dismiss any predictions he makes because of the outlandish ones he’s made in the past. But Kurzweil is nothing if not a brilliant inventor, and he indicates that at Google he’ll be rolling his sleeves up and doing real engineering. It’ll be fascinating to see how far this remarkable project takes both the inventor and the company.

Via: "Technology Review"
Image by Anatole Branch
(AKA: Morpheus)

The Year 2038 Problem

The year 2038 problem may cause some computer software to fail at some point near the year 2038. The problem affects all software and systems that both store system time as a signed 32-bit integer, and interpret this number as the number of seconds since 00:00:00 UTC on Thursday, 1 January 1970. The furthest time that can be represented this way is 03:14:07 UTC on Tuesday, 19 January 2038. Times beyond this moment will "wrap around" and be stored internally as a negative number, which these systems will interpret as a date in 1901 rather than 2038. This is caused by integer overflow. The counter "runs out" of usable digits, "increments" the sign bit instead, and reports a maximally negative number (continuing to count up, toward zero). This is likely to cause problems for users of these systems due to erroneous calculations.

Further, while most programs will only be affected in or very close to 2038, programs that work with future dates will begin to run into problems much sooner. For example, a program that works with dates 20 years in the future will have to be fixed no later than 2018.

Because most 32-bit Unix-like systems store and manipulate time in this format, it is usually called Unix time, and so the year 2038 problem is often referred to as the Unix Millennium Bug, or s2G.

In May 2006, reports surfaced of an early manifestation of the Y2038 problem in the AOLserver software. The software was designed with a kludge to handle a database request that should "never" time out. Rather than specifically handling this special case, the initial design simply specified an arbitrary time-out date in the future. The default configuration for the server specified that the request should time out after one billion seconds. One billion seconds (approximately thirty-two years) after 9:27.28 pm on 12 May 2006 is beyond the 2038 cutoff date. Thus, after this time, the time-out calculation overflowed and returned a date that was actually in the past, causing the software to crash. When the problem was discovered, AOL server managers had to edit the configuration file and set the time-out to a lower value.

Vulnerable Systems:

Embedded systems that use dates for either computation or diagnostic logging are most likely to be affected by the 2038 bug. Telecommunication systems and transportation systems from flight to automobiles increasingly use embedded systems. Automobiles, electric vehicles, and hybrid vehicles are increasingly using embedded systems to maximize efficiency and reduce pollution. Other automotive safety systems include anti-lock braking system (ABS), electronic stability control (ESC/ESP), traction control (TCS) and automatic four-wheel drive. New aircraft contain advanced avionics such as inertial guidance systems and GPS receivers that also have considerable safety requirements. Another major use of embedded systems is in communications devices, including cell phones and Internet appliances (routers, wireless access points, etc.) which rely on storing an accurate time and date and are increasingly based on UNIX-like operating systems. For example, the bug makes some Android devices crash and not restart when the time is changed to that date.

As of 2012, most embedded systems use 8-bit or 16-bit microprocessors, even as desktop systems are transitioning to 64-bit systems. Despite the modern 18–24-month generational update in computer systems technology, embedded systems are designed to last the lifetime of the machine in which they are a component. It is conceivable that some of these systems may still be in use in 2038. It may be impractical or, in some cases, impossible to upgrade the software running these systems, ultimately requiring replacement if 32-bit time_t limitations are to be corrected.

The use of 32-bit time_t has also been encoded into some file formats,^{[citation needed]} which means it can live on well beyond the life of the machines on which such file formats were originally supported.

MySQL database's inbuilt functions like UNIX_TIMESTAMP() will return 0 after 03:14:07 UTC on 19 January 2038. The limit reached will vary in time, based on the time zones as in the following table:

Auckland	19 January 2038, 16:14:07 NZDT (UTC+13:00)
Sydney	19 January 2038, 14:14:07 AEDT (UTC+11:00)
Tokyo	19 January 2038, 12:14:07 JST (UTC+09:00)
Beijing	19 January 2038, 11:14:07 CST (UTC+08:00)
Mumbai	19 January 2038, 08:44:07 IST (UTC+05:30)
Dubai	19 January 2038, 07:14:07 GST (UTC+04:00)
Nairobi	19 January 2038, 06:14:07 EAT (UTC+03:00)
Cairo	19 January 2038, 05:14:07 EET (UTC+02:00)
Paris	19 January 2038, 04:14:07 CET (UTC+01:00)
London	19 January 2038, 03:14:07 GMT (UTC±00:00)
Brasília	19 January 2038, 01:14:07 BRST (UTC−02:00)
Atlantic Time	18 January 2038, 23:14:07 AST (UTC−04:00)
Eastern Time	18 January 2038, 22:14:07 EST (UTC−05:00)
Central Time	18 January 2038, 21:14:07 CST (UTC−06:00)
Mountain Time	18 January 2038, 20:14:07 MST (UTC−07:00)
Pacific Time	18 January 2038, 19:14:07 PST (UTC−08:00)
Hawaiʻi	18 January 2038, 17:14:07 HST (UTC−10:00)

Data Structures With Time Problems:

Many data structures in use today have 32-bit time representations embedded into their structure. A full list of these data structures is virtually impossible to derive but there are well-known data structures that have the Unix time problem.

• file systems (many filesystems use only 32 bits to represent times in inode)
• binary file formats (that use 32-bit time fields)
• databases (that have 32-bit time fields)
• COBOL systems from the 1970s, 1980s and 1990s that have not been replaced by 2038-compliant systems
• embedded factory, refinery control and monitoring subsystems
• assorted medical devices
• assorted military devices

Each one of these places where data structures using 32-bit time are in place has its own risks related to failure of the product to perform as designed.

Solutions:

There is no universal solution for the Year 2038 problem. Any change to the definition of the time_t data type would result in code compatibility problems in any application in which date and time representations are dependent on the nature of the signed 32-bit time_t integer. For example, changing time_t to an unsigned 32-bit integer, which would extend the range to the year 2106, would adversely affect programs that store, retrieve, or manipulate dates prior to 1970, as such dates are represented by negative numbers.

Most operating systems designed to run on 64-bit hardware already use signed 64-bit time_t integers, effectively eliminating the Year 2038 problem in any software that has been developed to use the extended format. Using a signed 64-bit value introduces a new wraparound date that is over twenty times greater than the estimated age of the universe: approximately 292 billion years from now, at 15:30:08 on Sunday, 4 December 292,277,026,596. The ability to make computations on dates is limited by the fact that tm_year uses a signed 32 bit int value starting at 1900 for the year. This limits the year to a maximum of 2,147,485,547 (2,147,483,647 + 1900).^[7] While this solves the problem for executing programs, it does not, however, solve the problem of storing date values within binary data files, many of which employ rigid storage formats.

Starting with NetBSD version 6.0 (released in October 2012), the NetBSD operating system uses a 64-bit time_t for both 32-bit and 64-bit architectures. Applications that were compiled for an older NetBSD release with 32-bit time_t are supported via a binary compatibility layer, but such older applications will still suffer from the Year 2038 problem.

Alternative proposals have been made (some of which are in use), such as storing either milliseconds or microseconds since an epoch (typically either 1 January 1970 or 1 January 2000) in a signed 64-bit integer, providing a minimum of 300,000 years range.^[8][9] Other proposals for new time representations provide different precisions, ranges, and sizes (almost always wider than 32 bits), as well as solving other related problems, such as the handling of leap seconds. In particular, TAI64^[10] is an implementation of the Temps Atomique International standard, the current international real-time standard for defining a second and frame of reference.

Via: "Wikipedia"

Gadgets To Recognize You Through Touch

We’ve already seen the first wave of this technology in Disney’s amazingmulti-touch houseplants. Now, it’s being turned to humans. Whenever a user touches a display or object of any sort, their Capacitive Fingerprintingtechnique fires electrical frequencies through that user’s body. And by analyzing the exact impedance--the frequencies’ interaction with the particular biochemical makeup of your body--the system can identify you and whomever else is around you tapping at the screen, too.

In fact, the ID technology is so hypersensitive, it won’t even recognize your body’s state eight hours later. “It is not that it is too sensitive, but that the condition and electrical properties of the human body change with time and location,” Poupyrev tells Co.Design. “So, at this point we’re investigating its use for short-term interaction that happens in simple games or in location-based entertainment.”

Indeed, Capacitive Fingerprinting is almost most exciting when it’s connected to devices that aren’t normally as omniscient as smartphones and tablets, but the countless slew of absolutely perfect objects--the tables and the doorknobs--that our world has labeled “dumb.” Or as Pouprev puts it a bit more poetically: “A physical, inanimate object can ‘feel’ who is touching. So, it is kind of magical.”

Via: "Fasct Co. Design"

Soldiers To Increasingly Tap Military Apps

By S. William Matthews
Monday, June 4, 2012
Courtesy Of "The Daily"

Modern warfare? There’s an app for that.

Well, not from Apple, but there are a handful from the Pentagon — and more to come. They’re being downloaded on tablets and smartphones to help combat troops complete their missions.

Right now, in transports and fighters, pilots and navigators are on Air Force-issued iPads, checking their orders and reviewing their flight plans.

And this summer, the Army is putting Droids in the hands of grunts headed to Afghanistan so they can receive streaming surveillance video from drones, scan their surroundings for threats, call for fire support and request medical evacuations. Sailors and Marines are connected, too.

Though not a nuclear warhead or an M-16 rifle, the smartphone and the tablet are indeed weapons — helping the warfighter defeat the enemy.

The Air Force has led the electronics revolution and understandably so since it is considered the most technical of the services. The Army is close behind and the Navy and Marines are coming on board, too.

This spring, the Air Force bought 20,000 iPad 2s so air crews in Air Mobility Command and Special Operations Command no longer have to haul around 70 pounds of technical orders, flight manuals, charts and the like. These new “electronic flight bags,” which weigh about a pound apiece, are expected to speed up responses to in-flight emergencies and will most definitely eliminate stacks of paper that can quickly clutter cockpits, officials with the commands said.

Inter-service rivalry might make the Air Force want to claim credit for the Army’s entry into the world of gadgetry, but the Big Green Machine came into it on its own.

Soldiers have their own apps store, which the Army opened in March with prototypes of training manuals mostly.

But dozens of warfighting applications are in development: one tracks the blood pressure and temperature of a wounded troop still on the battlefield, another searches continuously for updated information on nearby enemy activity and yet another translates English into Pashto and Dari. 

The next step is to get all that information to every soldier by making smartphones, and someday tablets, standard equipment, just like boots and rain gear.

The first smartphones and tablets go to Afghanistan in a matter of weeks. Some will use whatever connectivity is available on the battlefield — satellites, cellphone antennas attached to aerostats, aircraft and towers, tiny cellular base stations. Others will boost the brain power of standard military radios for communication.

Michael McCarthy has spent two years testing smartphones and tablets in simulated combat on ranges at Fort Bliss, Texas, and he’s convinced the devices are ready to go to war.

“Imagine getting off a helicopter on a moonless night at 3 a.m. The first thing you have to do is to determine where you’re at,” said McCarthy, director of operations at the Army Brigade Modernization Command.

A smartphone’s global positioning capability can pinpoint its location to within a meter. The phone can then connect to Army databases to download “situational awareness” intelligence — where nearby friendly forces are, where the enemy is, what recent attacks and ambushes were in the vicinity.

A few doubters had worried that a commercial smartphone wouldn’t be rugged enough for combat, but the Army made it more durable by putting it an inexpensive silicon case and sticking a thin film glass protector on the screen, McCarthy, said.

“We don’t have to spend a couple thousand dollars to harden a $200 phone,” he said.

Still, there are limits. McCarthy learned the hard way that a smartphone won’t survive being run over by an armored vehicle, even in a protective case.

“That kind of thing is going to happen,” he said.

In a program called Nett Warrior, the Army uses smartphones to get more out of its Rifleman Radios.

A Droid smartphone is plugged into the radio, and the phone’s ability to run applications and process data transforms the radio into a “dismounted situation awareness and command and control tool,” said Maj. Gregory Soule, one of the officers overseeing the project.

The slick new radio and smartphone combination weighs about 3 pounds, significantly lighter than the clunky radio and computer combo, and costs substantially less. The computers they replace ran $15,000; smartphones cost a few hundred dollars.

As Soule aptly puts it: “This is designed for someone at the tactical edge.”

The Silicon Heart Of The CIA's Drone Program

Netezza v. IISI

By JOANNE MARINER
October 21, 2010
Courtesy Of "CounterPunch"

The Suffolk County courthouse in Boston, Massachusetts, is an unlikely place to learn about the CIA's drone program. Yet a contract dispute currently being litigated in Superior Court there offers a rare glimpse into the computer systems and software that are at the heart of the program.

The suit gives worrying indications that the CIA may have knowingly relied on untested and substandard software to operate its drones. It also raises important questions regarding potential civil and criminal liability for civilian casualties that could result from flawed/erroneous drone strikes.

The CIA is not a party to the Massachusetts case. But its unmanned aerial vehicle program, whose operations are very much at issue in the case, was responsible for at least 20 missile strikes that are believed to have killed more than 150 people last month in the tribal areas of 

northwest Pakistan.

Netezza's Claim

The lawsuit was initiated in November 2009 by Netezza Corporation, a publicly-held computer company that IBM recently announced it was acquiring for $1.7 billion. Netezza, whose speciality is data storage and analytics, filed suit against Intelligent Integration Systems, Inc. (IISI), a software firm that Netezza had worked with since 2006.

According to Netezza's complaint, IISI breached an August 2008 contract between the two companies by refusing to adapt its software to run on a data warehouse appliance, called the TwinFin, that Netezza had developed. The complaint also asserted related claims regarding intentional interference with business relations, breach of the implied convenant of good faith, and conversion, among others.

Page 11 of Netezza's complaint made an intriguing but unspecific reference to "a large government agency" that was one of Netezza's clients. Without going into great detail, the complaint stated that during protracted discussions with IISI before the suit was filed, Netezza has warned IISI that its delay in developing the requested software was endangering Netezza's relations with this agency. The complaint also alluded to the agency's "increasing pressure" on Netezza, suggesting that IISI was breaching its promise to provide the software at a moment when Netezza was in real need of it.

IISI's Counterclaim

IISI's counterclaim, filed in January, told a different story. Accusing Netezza of demanding additional software-development work at no charge -- work beyond that outlined in the contract -- the counterclaim also asserted that Netezza had made false and defamatory statements about IISI and stolen IISI's trade secrets.

The counterclaim provided important clues about the uses to which Netazza's government agency client meant to put the computer system. In the fall of 2009, IISI's counterclaim explained, Netezza had began asserting that "an agency of the United States Government needed immediately to purchase and deploy TwinFin to process geospatial data relating to vital military operations." The need for the system was so urgent, apparently, that "the agency was willing to accept an incomplete beta product designed to perform that function."

The counterclaim went on to state that IISI was "incredulous" regarding Netezza's claim that "a U.S. Government agency would seek or use an incomplete beta product for the military purposes in question, where lives could be at stake."

The IISI counterclaim also gave more detail about the software product at issue. Called Geospatial, the software permits users to process vast amounts of data, enabling "events (such as a tornado heading towards a population center ... ) to be matched with personal characteristics in a database (such as telephone numbers for houses in the path of the tornado ... ) and mapped and analyzed quickly and efficiently."

Summary Judgment Motion and Ruling

By April 2010, when IISI filed a motion for summary judgment in the case, it had identified the US government agency as the CIA. It also said that in the end, after IISI had refused to provide the requested software, Netezza had illegally reengineered it, creating a hacked but flawed version of the product to run on the TwinFin appliance.

The motion said that Netezza had delivered the hacked version of the product to the CIA in October 2009, and that the CIA had accepted it. After that, IISI alleged, Netezza commenced a full-scale project to develop a version of the software that would work on the TwinFin, relying on IISI trade secrets and proprietary material.

The Superior Court ruled in favor of IISI in August 2010, finding that IISI did not breach its contract with Netezza by refusing to perform the software-development work that Netezza had demanded. The court's opinion did not mention the CIA or the software's military uses.

Left pending, among other issues, was IISI's claim that Netezza had misappropriated its trade secrets. And left entirely unaddressed were questions about the CIA's negligence in accepting substandard software and possible legal liability for flawed targeting decisions that could result from its use.

Joanne Mariner is a human rights lawyer working in New York and Paris.