I like TED Talks, and have been watching them on and off for the past few years. Well, I like some TED Talks, mostly the ones dealing with science / tech and social issues. There are a bunch of other talks about art and music that range from fluffy -- performers talking about their works in such broad terms that they manage not to say much at all (Bobby McFerin's talk where he gets the audience to sing the pentatonic scale by providing the first few notes of it and then hopping across the stage is an exception although it's not actually a TED Talk, just posted on the TED site and from another conference), to new-agey woo-woo talks where the speakers trying to explain an experience with methods that are decidedly unscientific.

I just finished reading The New Inquiry's Against TED, linked to from Hacker News with the headline The TED Talks: Silicon Valley Corporatism?. The article argues that TED Talks are nothing more than 16 minute Apple product release-style advertisements for whatever the latest new technology is, pushed forward by corporate interests.

I agree with that assessment of the presentation style. Many of the presentations are flashy and tend to focus on only a few positive aspects of their topic. By this I mean that they have excellent visual aids and distill complex topics down into a few interesting, easily digestible key points. Yes the presenters leave out a lot of the subtlety of their topics and yes, they all have an obvious agenda to promote whatever it is they're talking about, but let's not kid ourselves about what TED is, and what it has been since the talks they posted online started to become popular: it's infotainment, or as I like to call it, infoporn.

TED Talks get me off, intellectually speaking of course. Sixteen minutes and bam -- I feel like I'm an expert on anesthesia machines designed for the developing world. Lunch time at work and I'm feeling bored and uninspired? I'll just flip to that TED Talk with the awesome data visualization. Feeling like I need an in-depth discussion of how a complicated piece of technology effects the world? TED's not the place, so I'll head over to watch a Google Tech Talk or maybe even some MIT Open Course Ware.

The point is that TED is entertainment, not academia. And as far as entertainment goes, I'm sure you'll learn a lot more watching it than watching an episode of The Office, or even some basic cable infoporn like The History Channel.

Mark Zuckerberg and Paul Graham get ready to speak at Facebook Developer Garage - Startup Edition.

Today I visited the Facebook headquarters for a mini developer conference for Y Combinator companies. The turnout was huge for what I'd expect for this kind of event -- probably around 150 YC folks. It began with talks by Mark Zuckerberg, Brett Taylor (founder of FriendFeed, which was acquired by Facebook about a year ago, and now Facebook's CTO), and Paul Graham.

Zuckerberg talked a bit about the history of Facebook, and how our innate interest in humans makes social apps much more compelling than non-social apps. He said Facebook sees itself as primarily managing peoples' social graphs and allowing third-party services to build off it to create new services that Facebook wouldn't have thought of.

He also made the observation that games seem to be the first indicator of a good platform. Games were among the first applications to become popular when personal computers first became available, as well as on the iPhone and on Facebook when they started allowing thrid-party apps.

Brett discussed their public APIs, and also talked about how they allow some sites (such as Yelp) to access APIs not yet available to the public. He said they would be making new APIs available to Y Combinator companies before they released the to everyone, and offer all the support they're able to for companies who want to use them. One option he proposed is for companies to come work at the Facebook office for a few days next to the API team so they could get help and give input during development.

Paul Graham's discussed the advantages startups have over established companies -- mainly that there are lots them that are all willing to take bigger risks than companies that have to worry about things like maintaining revenue. Most startups will fail, but the few that succeed will have produced something their bigger competitors couldn't have.

After the talks, the three of them had a Q&A session. The most interesting response was from Zuckerberg to a question about privacy. He gave a history of Facebook's privacy settings, saying that when Facebook first began, it was a more private way for people to share photos and notes on the web than had previously existed since it would only be accessible to users' friends. As peoples' friends lists began to expand from only close friends to include acquaintances, privacy became an issue -- if people had content they didn't want everyone to see, some of the people they wanted to keep it private from were likely on their friends list. This prompted the creation of privacy controls that allowed users to specify which of their friends were allowed to see what content. As the types of content available on Facebook increased with the addition of apps and more features, so did the number of privacy options. Eventually there were so many that managing them all was impractical. He said their current (and future) strategy is creating groups of options that can be set together to make them more manageable.

The Q&A was followed by a barbecue on their lawn. At one point I found myself chatting in a group with Zuckerberg. Someone asked him if he was planning to see the Facebook movie. He said he wasn't sure -- he was there when Facebook was created and it wasn't all that exciting; mostly a lot of coding and Red Bull.

He also said that he makes it a personal policy not to read press about Facebook -- something like, "Sometimes the press will be saying great things about you while you know you're making mistakes, and other times they'll be saying bad things when you aren't really doing anything differently. It just goes in cycles. We have to trust our internal compass for what we should be doing, and make sure to always listen to our users."

On the way out I got a glimpse of their workspace. They have an open floor plan, with rows of desks facing each other -- no cubicles or offices. While I waiting for the shuttle home, I chatted with a Facebook programmer who talked about how much he likes the open floor plan. He said it made it really easy to talk to people since he could just walk over to their desk. (It sounds like Facebook's avoided the problem Steve Blank talks about in The Curse of a New Building. He was a new hire, and had just completed their six-week bootcamp, which gets engineers up to speed on the code base and company practices. You can read more about Bootcamp here.

Facebook's front door with the letters H A C K.