Futurist and technology pundit Esther Dyson focuses on emerging technologies, companies and business models. A founding member of ICANN, today her company, EDventure, invests in startups.

It's Important to Question Hype. I'm mostly a skeptic about emerging technologies. I started as a fact checker at Forbes and learned to be specific and skeptical.

Blogs and Social Networking Are Overhyped. I write a blog. I know from experience that they're not the ultimate destiny of mankind. And I don't think it's important for every executive or tech expert to have one. A blog is a useful outlet for someone who has things to say and wants to publish them. In terms of the qualities a blog should possess, authenticity is pretty high on my list. A ghost-written blog is not the real thing. I do have one difficulty with blogging: fact checking. Sometimes the process slows down the blog. Or I get too busy and don't post because I can't check all the facts. I can't see posting stuff that could be inaccurate (as opposed to opinionated).

I Don't Try to Predict the Future. I just try to understand the present. But since you asked, one place I was really wrong was about privacy, or more specifically, people's reaction to the issue. I thought the public would be much more concerned and careful. They are concerned about it but are still careless. They're both willfully ignorant and paranoid! But I am having my own problems trying to figure out how to install an upgrade from Symantec. You could argue that the vendors still don't make it easy enough.

Established Companies Can Learn a Lot from Startups. Startups can teach mature companies how to have a sense of humor and to be responsive and flexible. And to focus more on customers than on internal politics.

ICANN Taught Me a Lot. There's not much need for a lot of central policy direction on the Net. ICANN, the nonprofit entity that oversees the Internet's systems and protocols, did the wrong things in that regard. It designed a market with too much regulation and, ironically, too little enforcement. Its rules are so rigid and prices are set so that registries and registrars can't generally compete on anything other than sleazy marketing practices. There needs to be more disclosure and real penalties for breaking the rules but overall fewer rules. What did I learn from that experience? Fight harder for transparency. Avoid centralized power. And don't trust people to do the right thing; design systems that help them to do so.