Here's a really nice overview of character sets and character encoding for Japanese [Japanese Text Encoding]. 11:36:35 AM    comment []  

Coming from a background in hard sciences and computer science, I have often been biased when I hear the theories and approaches of the psychological profession. I generally find their theories to be simplistic and have always found it doubtful that they were on the right track.

For the past 100 years, we are supposed to have been living in “the psychological century”. Psychoanalysts and psychotherapists did not exist before Freud, yet by the 1970s there were 200 therapies on offer, ranging from Synanon to Insight to Rolfing. At certain points in the 20th century there were more psychology graduates than any other kind. The psychological jargon that has entered the language stretches from paranoia to penis envy, and from id to identity crisis to inferiority complex. So why, amid all this new professional apparatus, are behavioural problems more widespread, depression more common and psychotherapeutic drugs taken in unprecedented quantities? Could it be that psychology is failing? I think so. In fact, I think it has failed big-time. Furthermore, it has failed not just in the sense that more people are ill or unhappy, it has failed technologically, philosophically and is already in an advanced stage of decomposition [Times Online].

I think the study of networks and simple systems are giving us some important insights into why artificial intelligence and psychology research has failed; the underlying systems may turn out to be simple, but the resulting behavior can be irreducible. In other words, the outcome of a system can not be shortcut. Why is this important? Well, if it turns out to be true, then the only way to predict the outcome of a system is to run the system and see what comes out. This defies the general thrust of existing sciences that are geared around shortcutting physical processes by defining systems that can predict the outcome of a given system. However, it is only possible to shortcut a system if that system is reducible.

Psychology tries to shortcut the processes of the human mind by defining rules of causality that attempt to express universal truths about human beings. If the human mind is a reducible system, then this approach has some chance of success. However, I think the subtle differences in the nature and nurture of a given individual will result in potentially significant differences in the behavior of an individual. My intuition tells me that the human mind has more in common with the irreducible systems that we see in chaos theory and Wolfram's automata than it does to the predictable (and reducible) systems of physics and the hard sciences.

The understanding that we are developing of systems is going to eventually come around to save some of these sciences. However, it isn't a silver bullet. It is a salvation only in the sense that we are going to understanding the failings of the past and make the best of the future. However, the new face of these studies will not look anything like the old. 10:55:02 AM    comment []