Sunday, May 10, 2009

Clickstream map of science

An interesting example from academic research how our web surfing habits derived from clicksteram might facilitate another understanding or put it more fashionable another "new knowledge". I wonder if Wolfram Alpha could manage something on this scale, given we will feed it with needed data.
From the "paper" itself :
we investigated a methodological issue: can valid, high resolution maps of science be derived from clickstream data and can clickstream data be leveraged to yield meaningful insights in the structure and dynamics of scholarly behavior? To do this we first aggregated log datasets from a variety of scholarly web portals, created and analyzed a clickstream model of journal relationships from the aggregate log dataset, and finally visualized these journal relationships in a first-ever map of science derived from scholarly log data.


What I learned quickly, though I knew it before is a startling disconnection between what we call "brain research" and "brain studies"



Though you might think that we should be discussing one freaking entity, obviously there is Cartesian dichotomy, even if it is not named as was originally stated "Mind-Body Dualism".

Another curious fact - one of the first sciences - Geography is almost non existent, represented just by one dot :)

Friday, May 08, 2009

RepRap and democratization of manufacturing.

RepRap is short for Replicating Rapid-prototyper - a project to create an open-source self-copying 3D printer.

RepRap version 1.0 "Darwin" can reproduce 60% of its parts. The machine is manufactured from plastic and metallic rods. Next version RepRap 2.0 will be called "Mendel" it is author's incentive to commemorate famous biologists in the names of replicative machine designs. After all, they say it is mimicking biological evolution in the ability to reproduce himself. On my part, I would be very carefull to make long standing comparisons between RepRap and working evolution in action, since in biological evolution we have evolution of designs and here it is materialistic reproduction of human invented improvements, without any hints of competition and struggle for resources between different blueprints.



Nonetheless the project is alive and kicking by enthusiastic opensourcenicks. It is also loosely connected with another hardware open source project Arduino.

3D printing is nothing new, in fact, aside from RepRap there is ongoing push for introducing such devices right on your desktop.
For approx. 5000$ you can purchase Desktop Factory 125ci 3D Printer from "Desktop Factory"



In size comparable to early laser printers, uses an inexpensive halogen light source and drum printing technology to build robust parts layer by layer from composite plastic powder.
The cost of the build material is expected to be about $1 per cubic inch.

Those two examples are using different technologies, different input materials and more importantly different philosophies about how and who is going to use 3D printers. Certainly open source 3D production would've make a tremendous blow upon existing capitalistic system, but to make usefull things you need to be educated and to have IQ well above average. The real evolutionary struggle is between two worldviews : one seeing a lot of consumers and a handfull of creators, alternative claims - everybody can be a creator.

Monday, May 04, 2009

Cyberdyne Japan makes HAL

What is a common denominator between two science fiction movies: "2001: A Space Odyssey" and "Terminator" ? The answer gives Tsukuba University engineering professor Yoshiyuki Sankaican whose passion for Science Fiction gave birth to Japanese startup Cyberdyne and their product - HAL.

HAL (Hybrid Assistive Limb) is not an almighty AI computer but a cybernetic bodysuit when worn on human body your muscle strength can be facilitated as much as tenfold. Cyberdyne is said to start mass producing those HAL bionic exoskeletons from 4200$.



From the company's site

"Robot Suit HAL" is a cyborg-type robot that can expand and improve physical capability.
When a person attempts to move, nerve signals are sent from the brain to the muscles via motoneuron, moving the musculoskeletal system as a consequence. At this moment, very weak biosignals can be detected on the surface of the skin. "HAL" catches these signals through a sensor attached on the skin of the wearer. Based on the signals obtained, the power unit is controlled to move the joint unitedly with the wearer's muscle movement, enabling to support the wearer's daily activities. This is what we call a 'voluntary control system' that provides movement interpreting the wearer's intention from the biosignals in advance of the actual movement. Not only a 'voluntary control system' "HAL" has, but also a 'robotic autonomous control system' that provides human-like movement based on a robotic system which integrally work together with the 'autonomous control system'. "HAL" is the world's first cyborg-type robot controlled by this unique Hybrid System.


Meanwhile distribution is limited to Japan only.

I wonder if successful how long time it will before sombody takes them in court for infringement of "intellectual property" of trademarked names ?

Sadly, they can't rename it to "Android" too, Google is already strugling with this name. Frankly, I'm waiting for Karel Chapek's descendants to claim their right for the "robot" name. Apart from being enthusiastic about Cyberdyne HAL suits I also hope more people will read original Chapek's novels like R.U.R.
many of his works had deep philosophical roots. He cared about truth and justice, and these values he gained by his readings of philosophers, especially the then-modern philosopher Henri Bergson.

Thursday, April 30, 2009

Wolfram Alpha new blog

I'm eagerly waiting to see what kind of surprise is brewing under the guise of "computational knowledge engine" called Wolfram Alpha announced by Steven Wolfram in March and settled to debut in May. Steven Wolfram was widely credited for his Mathematica software package and equally criticized for the book "A New Kind of Science” ,abbreviated NKS , explained his new creation :

I had two crucial ingredients: Mathematica and NKS. With Mathematica, I had a symbolic language to represent anything—as well as the algorithmic power to do any kind of computation. And with NKS, I had a paradigm for understanding how all sorts of complexity could arise from simple rules.

But what about all the actual knowledge that we as humans have accumulated?

A lot of it is now on the web—in billions of pages of text. And with search engines, we can very efficiently search for specific terms and phrases in that text.
But we can’t compute from that. And in effect, we can only answer questions that have been literally asked before. We can look things up, but we can’t figure anything new out


While many dubbed it like a potential Google killer I don't expect from Wolfram|Alpha to be used in mass on the Google scale. Essentially it is not a general searching tool, but is meant to become a tool for "truth discovery". But the truth is that most people are not looking for computable "truth discovery", they are sufficed with finding the facts and then using their own brains to accomodate that data to build their own knowledge base.
I won't be as cynical as Ted Dziuba who wrote :
"That sounds an awful lot like the marriage of some Python scripts with a few hundred bucks spent hiring third world workers through Amazon Mechanical Turk.".


I do believe that people behind Wolfram research are one of the brightest you can find, but still they are numbered and their expertise is limited. Currently we are experiencing shift towards folksonomy and open APIs, semantic distributed web e.t.c. which is as i can understand not the chosen way for Wolfram Alpha project. Still, some believe that their work will be incorporated in the future with other projects like European Large Knowledge Collider

Yesterday after initial March anouncement Wolfram team started new blog covering Wolfram Alpha. From my understanding it is far from being search engine, at present they are not going to crawl intenet pages. Rather it is a closed proprietary expert system with some fancy computational algorithms ala John Conway's Game of Life capable to show results on-the-fly in digestable graphical manner. When Alpha will have web interface I'll take a closer look.

N.B.
Speaking of search engines and our latest global fever i tried Cuil (almost forgotten another Google-killer) to search for "swine flu" and i was kinda surprised. On the first page i found link pointing to one very good article describing 1976 swine flu case. And another link about recombination of birds and pigs flu virus genes, which dates from 2007, not exactly the last news as you might wrongly infer from the current media frenzy. If you had already encyclopedic info which you can find in Wikipedia those additional pieces are good hits and they are not easily found with general purpose search engines. What this is good for explaining knowledge ? When you have initial base you need the tools for finding more deep knowledge. And this is exactly what is lacking both in Wolfram Alpha project and even upcoming semantic web with all their ontologies. We don't have formal understanding what is "deep knowledge" which could also be adjusted according to the users needs.

Monday, April 27, 2009

Twittering from brain to PC.

Earlier this week my attention was caught by another marketing spin for twitter. Doctoral student Adam Wilson from University of Wisconsin-Madison has successfully tested a "brain twitter interface". The story made all the way up to the CNN headlines. According to CNN article twitter is simpler than e-mail.

"If I am locked in and I want to e-mail someone, the format is all wrong. You have to be able to select recipients and group them, copy, paste, send. ... We don't think about that much as normal people, but it can become unmanageable.
"Twitter takes care of all those things. They just have to get [the message] to a location where people can come and find it,
to Twitter publishing interface, allowing him to compose a message merely by thinking and publish it to the arguably too-popular microblogging service. "





Why it is simpler i can't say, if you look on the video demonstration of this experiment. After you've managed "to shoot with your brain" right letters from the flickering screen you shoot "twitter rectangle" and out of the blue appears twitter window. But in the same way it could be e-mail window, or anything else for that matter, another microblogging tool, "full blogging" interface, you name it. Overall all this experiment looked somehow fishy: I don't know how old was the monitor, but it was old. It is strange that you are using CRT monitor from web zero generation to demonstrate how web 2.0 tool will enhance your user interaction experience with PC.

If earlier i was ridiculing twitter, now I'm starting to hate it. May be for the wrong reasons, but i really don't like all this excessive marketing hype, propelled by the sites like Techcrunch: "Worldwide visitors to Twitter.com increased 95 percent in the month of March from 9.8 million to 19.1 million" and now more and more by the mainstream media like CNN.
Even academics from Universitiess are willing to ride on the wave to promote their research.

As for the technology itself, I'd like to recall another example of brain to PC interface from Emotiv Systems, which to me seems much more revolutionizing. Unlike this twittering hype which is based on letter to letter input, Emotiv technology enables recognizing mental states right from the brain waves skipping alone human language representation.