You are what you watch. That’s what the “Kill Your TV” people used to say. But if TV is mindless, where does that leave YouTube?

Apart from search engines, YouTube is now the second most popular website in America, drawing the average visitor for a solid sixteen minutes of video surfing–a web eternity. The site hosts a long tail of clips on every item imaginable, but the top videos actually track the vices of television: sex, celebrities and sensationalism. And as the web morphs from endless text to an increasingly video-focused platform, YouTube is ground zero for some of the dumbest crap online. Yet web videos don’t have to be vapid, according to the entrepreneurs behind Big Think, YouTube for the Harvard set.

After working as producers for The Charlie Rose Show, Harvard grads Peter Hopkins and Victoria R.M. Brown saw an opening for thoughtful, short-form intellectual videos targeting online audiences. The idea was simple: take the brightest, most creative thinkers alive, plunk them down for a conversation straight to camera–reality-show style–elide the moderator and provide an intimate window into the “big ideas” of our time. The erudite site drew investments from heavy hitters like Peter Thiel, a PayPal co-founder and Facebook angel investor, and Larry Summers, the former Harvard president and treasury secretary.

Compared with the experts on serious television, let alone the pundit circuit, Big Think’s interview subjects have remarkable depth, diversity and credentials. There are famous professors and renowned writers, award-winning scientists and prominent theologians, political activists and tech futurists. In other words, the site is full of intellectuals with ideas that can make for compelling video–but without the sound bites and sizzle that dominate TV and YouTube. (There are also interviews with traditional newsmakers like senators, governors, former government officials and celebrities.)

The prolific author and conservative Judge Richard Posner, for example, offers a meandering but intriguing answer to the open-ended question “What’s your counsel?” After lamenting the cost of the Iraq War, he notes that only government can tackle existential problems like global warming and disaster prevention. “It’s actually kind of heresy, but I think the American people are undertaxed,” he says in a low-key confessional. It’s the kind of policy-driven argument that would rarely make a cable news debate, let alone a viral hit. “We ask a range of questions that are open-ended, forward-looking and nonpartisan,” explains Brown, who works out of one of the spare photo booths in Big Think’s Manhattan office. The start-up does not have enough desk space for its five employees.

Big Think strains to transcend traditional media framing, self-consciously shunning categories like “news” and “opinion” for more trippy headings. A “physical” section lists videos on architecture and music, while a “meta” category covers concepts like identity, wisdom, death and inspiration. It’s more nuanced than YouTube, but also more confusing. (Why is “justice” meta? Why is “media” physical?) Yet Big Think is not just striving to be a hipper PBS, blasting highbrow content at enlightened Millennials. The founders say they’re aiming for a meaningful, interactive dialogue–the kind of audience participation that makes good blogs lively, social networking sites sticky and YouTube profitable.

In an interview with TechCrunch, an influential Silicon Valley site that analyzes Internet firms, Hopkins, the 24-year-old Big Think co-founder, said his site is special because it empowers visitors “to contribute actively and in the same manner as our invited participants.” So while any media company can deliver “lots of high-quality content in one direction,” he explains, “Big Think is about using some high-quality content to begin an exchange of ideas in two directions.” A great goal, but it’s not happening yet.

After a three-month beta run, it is too early to judge the site’s interactive potential. So far, however, the vast majority of videos feature handpicked experts. Unlike on YouTube, user videos are quite rare. And most visitors choose to reply to featured videos through written comments, like on any blog. Big Think drew roughly 22,000 unique visitors in February, according to the web analysis firm Compete.com, with an average visit of six minutes. For comparison, Bloggingheads.tv, a wonky hub for lengthy video debates founded in 2005, drew 17,000 unique visitors in February.

Developing an interactive video dialogue with the public is a new challenge, since Internet debates are anonymous by default. Most people use pseudonyms to write blogs or comments. Video obviously requires more intense exposure; fewer people publicize their personal and political views when the cost of entry is sharing their faces, voices and identities. On YouTube, even users who post video “responses” don’t usually provide a linear dialogue, let alone intellectual discourse. (Many people post random replies to popular videos to boost their channel’s traffic.) There are exceptions. In the summer of 2007, for example, a random user’s YouTube query generated twenty-seven direct, logical video responses–and an impressive 6 million views. The question: “What is the best sex you have ever had?”

Highbrow video can also work, at least when its high-powered. The Davos Economic Forum launched a YouTube channel last fall, drawing more than 100 user videos about how to “make the world a better place in 2008.” The forum also uploaded ideas from stars like Bono and world leaders such as Hamid Karzai and Henry Kissinger. But who cares about your global vision when Bono is one click away? “There is a tension between the wisdom of the crowd and wisdom of experts,” explains John Palfrey, executive director of Harvard’s Berkman Center for Internet and Society. After looking over Big Think for The Nation, Palfrey said that by making expertise a priority, the site may have a harder time fostering a participatory culture.

Everyone knows video is taking over the web. The question is whether that augurs the triumph of a warmed-over broadcast model–the one-way communication of Hulu–or something more interactive, like YouTube’s grassroots carnival and the experimental Current TV, with breathing room for the long tail niches of specialized online communities and salons like Big Think.

“As TV and the Internet converge into something generically known as broadband, the distinctions between the two will soon become nugatory from a consumer point of view,” writes Michael Hirschorn, a former VH1 executive and prescient technology journalist. “But will this resulting hybrid be more like TV, plus interactivity; or more like the Internet, plus TV?” he asks, predicting that eventually one format will offer the desired medley of user-made, broadcast and web-exclusive video. That billion-dollar blend, Hirschorn contends, will be the “Web 3.0.” Hopefully, it will still be big enough for big ideas from the bottom up.