I'm both absolutely enthralled by it and kind of sad about it at the same time. It's obviously a phenomenal technological accomplishment. It would be even more amazing if Watson was actually "hearing" the question and responding to the sound and inflection, rather than being fed a string of text. However, something about the fact that we're pitting man vs. computer, and that the computer is absolutely mopping the floor with the humans, makes me feel a little unsettled.
The statistician in me is really enjoying watching this all play out - there's so much neat stuff to look at that I can't even take it all in. I've been DVRing the episodes and watching some parts twice - once to watch the swirling avatar and the colors and patterns light up as the machine churns through the data, and once to watch the three "top answers" that the machine spits out and the confidence levels associated with them. I want to know so much more about what it's doing and how it's working and PBS had better be planning an incredible documentary about it.
The sociologist in me, on the other hand, is wondering what these new technological developments mean for the future labor market. This is something that my non-sociologist husband has been talking about for the last few years, too. With all of these new technologies becoming widely (and cheaply!) available, we're going to eventually be able to replace huge swaths of the labor market with "smart" machines. Where will these workers go? Also, why not give it a feminine name and presence? (I know it's trite, but the gender sociologist background insists that I point it out)
And of course, the sci-fi geek in me is thinking about Skynet and AI and HAL and WarGames and Asimov's Laws of Robotics. It's almost enough to make you want to just stay in bed with the covers over your head.
If you're interested in more information about how Watson thinks, this article was pretty descriptive without using a lot of techie words. This one, too.
I *swear* we watched a PBS documentary about Watson last week!
ReplyDeletejust watched the third (final I think) game today. When it makes a mistake, it bombs out really well. The second day with Toronto being a US city was funny (answer was Chicago).
ReplyDeleteAs Katie pointed out above PBS did do a NOVA show on it already but maybe they will do a follow-up. IBM has a great site on the whole project.
thanks.
I have been watching as well. It is interesting and scary at the same time. The nerd in my has been wondering about the algorithms. It's like the categories don't mean enough hence the Toronto thing. I agree, it would be better if it had to understand what Alex was saying, not just being fed the text.
ReplyDeleteI have been watching as well. It is interesting and scary at the same time. The nerd in my has been wondering about the algorithms. It's like the categories don't mean enough hence the Toronto thing. I agree, it would be better if it had to understand what Alex was saying, not just being fed the text.
ReplyDelete