plaxoed!

Terry lent me his copy of Hackers and Painters by Paul Graham and I finished reading it a few days ago. As I started reading, some of the topics were quite intriguing. Here are some of Paul’s main points in the first half of his book:

• Nerds aren’t popular in school because they’d rather be smart - Maybe, but I think the social dynamics and causal relationships in American high schools are a little more complicated than smart kids simply wanting to be smart more than they want to be popular. I wonder if there have been child psychology studies on this particular hypothesis.
• It’s natural and OK for the disparity between rich and poor gets larger - This may be natural and at first, I’ll admit, I thought his argument was well reasoned. After all, I am part of the population in the middle bracket, striving to get into the upper brackets. But then I talked to Terry, because this idea sounded a little strange to me (it didn’t jive with the ideals they discuss in economics about the Gini Index and etc.). Terry pointed out a few case studies that brought things back into perspective. Without a wealth gradient, we become almost a two-caste system; the poor get poorer and the rich get richer. Worse yet, the poor become the majority and the rich remain a small group that continues to gain more and more power. In history, this social inequity is a catalyst for instability, revolutions, and decreased overall quality of life (yes even for the rich people, look at South Africa).
• To succeed as a start-up, the most important factor is for you to be different than your competitors - It may be an important factor for success, but I think timing and luck play a bigger role than Paul recognizes. There have been plenty of failed start-ups with great ideas, great execution, solved hard problems, and differentiated themselves from competitors.
• To solve spam, all you need is a highly tuned Bayesian filter - Terry has the most to say about this particular piece. As Terry pointed out to me, the problem with this approach is that it just increases the amount of spam sent. At the end of the day, spammers will figure out how to create a spam message that passes through your Bayesian filter or you’ll start getting real e-mails filtered as spam, or worse yet, both.

As for the second half of the book, I definitely lost interest when Paul starts talking about what programming languages will look like in 100 years and why Lisp is the bomb. I wouldn’t venture to guess what the software development landscape would look like in 20 years let alone 100 years… 20 years ago, I don’t think people would have predicted the internet, web services, scripting languages like PHP, or programming inside browsers via AJAX. What’s the future of computing going to look like in 100 years? I wouldn’t want to make any bets

The one thing Paul has convinced me to do is to take a look at Lisp. I’ll add that to my to-do list; any one know a good place to start?