Back to article
The StartX Files: Between the Sheets With abs
IF Crew = Dave THEN DoorPosition = Close
December 7, 2001
I managed to get a hold of the television remote the other evening after I distracted my wife by gently tossing one of the cats' catnip mice at my wife's head and ordering the cats to go get the toy mouse. In one sense, my gambit worked, as my wife's reach for the remote was diverted into a desperate flinging of hands to get the mouse off of her before the cats could charge across the room and attack, thus leaving me with the all-powerful remote in my possession. In my house, possession is no mere nine-tenths of the law--it is the law.
Of course, ultimately my gambit failed, because my wife was later able to get back at me in ways that I would rather not mention.
All part of the magic of the Proffitt home.
Anyway, to the victor go the spoils, and I began to peruse the channels to see what was on. Being a guy, I naturally avoided such oddities as the WE channel, Oxygen, or Lifetime. Being a geek, I concentrated on such channels as Discovery, Sci-Fi, PBS...hello! What was this? I had stumbled on "2001: Hal's Legacy," a program on Arthur C. Clarke's vision of 21st Century computers (like Hal) and the reality of current efforts for artificial intelligence. Intrigued, I settled in for a hour-long journey into serious geek territory. Fuming, my wife rolled her eyes and settled in to plot her revenge.
I won't bore you with unnecessary details, but the upshot of the program was a look at the various aspects of AI manifestation that Hall seemed to comfortably manage: speech, vision, common sense, and emotive response were just some aspects that were examined. Various scientists are taking different approaches to each of these obstacles to producing artificial intelligence, in projects that are divided piecemeal throughout the artificial intelligence community. IBM is hammering on problem solving with Big Blue, for example, but all that Big Blue can do is play chess (albeit very well). It could not even play checkers if you asked it to.
There are projects on teaching object recognition, speech recognition, and there's even an open source project on teaching a computer common sense. I'll believe this last when I see it, since I know several human beings that seem to blithely go through life with a complete lack of it.
Strides are being taken, but we are nowhere near the computer technology Clarke envisioned with Hal. Certainly there have been few (if any) efforts to try to combine all of these different aspects into one cohesive system.
To some, this seeming failure of not living up to the timeline of one man's dream could be discouraging. I myself have often joked about the fact that it's the 21 Century and there's still no flying cars yet.
But beyond this layer of disappointment, maybe it's a good thing we have yet to figure out how to create intelligence with technology yet. It's a nice reminder that although we have figured out how to split an atom, we still have not figured out how we, our inner selves, really work. Perhaps when we have a better handle on that, we will be able to bring the technology up to speed.
Because I am reminded every day that it's not the technology that's the important thing. A funny thing for a technology writer to say, I know. Here in the Linux community, there are countless examples of people using their technical knowledge to show what's good about the human race. And sometimes people put the technology before the people around them and say and do things they might otherwise regret--all in the name of that same technology.
This week's shocking statement: Linux is not the be all end all. I know even Linus and the rest of the keepers of the kernel don't buy into this. Ultimately, Linux is just another tool. Just like UNIX. Just like Windows. Something I think we all could use a reminder of as we conduct ourselves in this community and all the rest of the IT world.
Great things will not be done by Linux. Great things will be done with Linux, driven by the people who know how to use it and can build more tools to work with it.