Algebraic surfaces

I like this list of mathematical formulas for describing shapes (or, in fancy terms, 'algebraic surfaces'). You can try these out with the 'Grapher' program that Apple ships with its system (or any other 3D graphing program).

Three related announcements?

I am starting to notice a lot of development on large-scale, well-funded attempts to make computers answer questions posed in ordinary language. Three in particular are of note:

i) IBM's 'DeepQA' Project: "IBM has unveiled the details of its plans to build a computing system that can understand complex questions and answer with enough precision and speed to compete on America's favorite quiz show, Jeopardy!."

ii.) Stephen Wolfram's Wolfram|Alpha: "Although it’s tempting to think of Wolfram|Alpha as a place to look up facts, that’s only part of the story. The thing that truly sets Wolfram|Alpha apart is that it is able to do sophisticated computations for you, both pure computations involving numbers or formulas you enter, and computations applied automatically to data called up from its repositories." (From the Wolfram|Alpha Blog)

iii.) The Large Knowledge Collider Project: "The aim of the EU FP 7 Large-Scale Integrating Project LarKC is to develop the Large Knowledge Collider (LarKC, for short, pronounced “lark”), a platform for massive distributed incomplete reasoning that will remove the scalability barriers of currently existing reasoning systems for the Semantic Web."

Will we soon see a leap forward in computer intelligence? Stay tuned! These are exciting times!

[Image from The Large Knowledge Collider Project]

Complexification

I have long been a fan of www.complexification.net, a website devoted to artistic computer programming. It uses a free computer language (specialized for visual programming) called Processing that I have always wanted to spend some time learning. There are many more examples here.

Microchip turns 50 today!

The microchip that drives all our computers was invented 50 years ago today. This is a picture of the very first one, constructed by a guy named Jack Kilby working at Texas Instruments. He also invented the hand-held calculator.
Via: Gizmodo, which has more details if you want'em.

Cost per gigFlOPS: REALLY cheap now!

I love this table (adapted from Wikipedia) on how the cost per gigaFLOPS has changed in the last 50 years or so:

Hardware costs:

* 1961: about US$1,100,000,000,000 ($1.1 trillion) per GFLOPS
* 1997: about US$30,000 per GFLOPS
* 2000, April: $1,000 per GFLOPS
* 2000, May: $640 per GFLOPS
* 2003, August: $82 per GFLOPS
* 2006, February: about $1 per GFLOPS
* 2007, March: about $0.42 per GFLOPS
* 2007, October: about $0.20 per GFLOPS (in a Sony PS3 console).

Imagine: we pay twenty cents today for something that cost over a trillion dollars around the time I was born! Now that's a bargain!

Petaflop barrier soon to go down?

Computer speeds are often measured in 'FlOPS' = floating point operations per second, essentially a measure of how many calculations the machine can do in a second (skipping over some technical details). A megaFlOPS is about one million floating-point operations per second. A gigaFlOPS is one thousand times faster (actually, 1024 times faster = 2 to the 10th power), or (roughly) one billion floating-point operations per second. Most desktop PCs can do a few (< 50) gigaFlOPs now. A teraFlOPS is one thousand (1024) times faster: roughly one trillion floating-point operations per second. Only a few machines in the world can do a teraFlOPS. Now IBM says their RoadRunner computer (pictured above) will soon bust through the petaFlOPS barrier, which is again 1024 times faster, or 1024 trillion FlOPS. You could play some cool video games on that baby.

Via: dvice.com