Day two of humans versus the computer on Jeopardy did not go well, at least not for the humans. Watson wiped the floor with the competition. And watching it was 30 minutes of well, not fun. Blake Eskin over at The New Yorker says it pretty well:
"When its data-processing algorithms are on target, it is indomitable, and even its runner-up guesses and slipups made you think about how it works. But after one round of questions, its novelty was beginning to wear off. Our television critic, Nancy Franklin, observed on Twitter, 'On the fun/tense scale, things on Jeopardy are leaning toward too tense, not enough fun. Watson just hits the button faster than people can.'...Of course, Watson’s servers have been optimized for correct answers, not for pleasure, ours or its own....If Watson does not eclipse the best human players by Wednesday, one assumes it is only a matter of months before it would blow them away. Watson has many potential applications in medicine, law, science, and other fields, but its prospects for a career in television are no better than Brad Rutter’s. Who would want to watch a computer win seventy-four nights in a row?"
Simply put, it just hasn't been enjoyable watching these last couple of days, maybe because playing Jeopardy with a computer is like playing Scrabble against a dictionary (yeah, I know that's not a perfect analogy...). Take the humanity out of Jeopardy contestants (their individuality, their strengths and weaknesses) and it's just not the same game.
No comments:
Post a Comment