The implications of Wolfram|Alpha
It's a real-world test of the proposition that complex answers do not require a complex agency
Wolfram's Alpha is not a competitor to Google. I've been playing with it since it went live the other night, and its limitations are glaring and clear. It has trouble answering even the simplest and most intuitive query, which makes it seem like it's a spectacularly stupid system. But what's impressive is that it is able to answer any questions at all.
If I understand Stephen Wolfram's description of the system (and others') correctly, Alpha is an attempt to create a knowledge engine out of a very large library of fairly small algorithms. Its database is vast; but the code that operates on it is not necessarily complex. In other words, Alpha's not a monolithic "thought engine" but a collection of heterogenous mini-engines that Wolfram hopes will interact in unpredictable but creative ways. As Stephen Wolfram puts it in a recent blog entry on the subject:
There is an immensely complex web of systematizable knowledge out there in the world. And before NKS [Wolfram's book A New Kind of Science --K], I would have assumed that to handle something of this complexity would have required building a system that is somehow correspondingly complex—and in practice completely out of reach.
But from NKS we have learned that even highly complex things can have their origins in simple rules and simple programs.
This last statement is the important one--it speaks to what I've been saying for a while now, that the vision of a 'technological singularity' that comes as a result of increasing complexity of information processing systems, is mistaken. (It is, in fact, an example of the erroneous theory of Intelligent Design.) Creativity is not correlated to complexity; and as well as being a potentially useful tool, Alpha is an attempt to prove this very non-intuitive idea.
As Wolfram goes on to say in his blog post:
Today, Wolfram|Alpha uses existing models from science and other areas, then does computations based on these models.
But what if it could find new models? What if it could invent on the fly? Do science on the fly?
That is precisely what NKS suggests should be possible. Exploring the computational universe on request, and finding things out there that are useful for some particular specified purpose.
Stephen Wolfram expects Alpha to be more than a data regurgitator or formatter. He expects it to be creative. And, he expects this creativity to emerge, not from complexity, but from simplicity.
These are very interesting ideas. The next year of Alpha's growth should be interesting to watch.