Twitter founder Evan Williams announced a new platform for the popular microblogging service during a keynote interview today at South By Southwest Interactive (SXSWi). The new platform, called @Anywhere, allows websites to integrate Twitter more deeply.

The platform is launching with a number of media partners, including The New York Times, Digg, The Huffington Post, and Bing. Williams said it will allow users to, for example, discover and follow a writer's Twitter account simply by hovering over a byline in a news article, or to follow a band after finding its name in the text of an article.

The idea of the platform, Williams said, is to reduce the friction that prevents more engagement. If the platform is adopted, he said, it "should result in more followers for a site," as well as more community interaction. He referred to the platform as an "@" platform rather than an "app" platform.

Williams was thin on details, despite some prodding during the keynote interview. In some respects, this seemed to be because the platform is designed to be flexible and he's not yet sure how sites will use it. I did, however, wish that he would use more than 140 characters to describe how it was intended to be used.

Usually, when engineers try to make technology better, they add new features. After all, the more options a user has, the more she can do with a gadget, piece of software, or Web application. This approach is championed by Microsoft, which crams an increasing number of functions into each iteration of its operating system and Office applications.

But sometimes the most successful products have fewer features. At the Web 2.0 Summit yesterday, Evan Williams, one of the founders of Blogger and Twitter, talked about the value of keeping a product bare. Blogger came from a more complex project-management tool called Pyra. Pyra failed, but Blogger helped spur the blogging revolution. Twitter is even simpler than Blogger. It's a microblogging tool that forces people to condense their thoughts into 140 characters, without titles or any special formatting. (For a profile of Williams, see "What Is He Doing?" For a review of microblogging services, see "Trivial Pursuits".)

At the summit, Williams said that one of the "things we've learned about running Twitter ... is that aggressively applying constraints could have unexpected benefits for users and yourself."

To be sure, Williams doesn't consider himself to be anti-feature. "Features are critical, especially as a product evolves," he said. "But I'm also a fan of asking a different question, especially in the beginning: what can we take away to create something new?"

He gave examples of a number of successful products that followed this path. Fotolog is a photo-sharing site similar to Flickr that, due to technical limitations in the beginning, only allowed users to post one picture a day. Williams posited that one reason for Fotolog's popularity is that the pictures were higher quality, and users commented more often. He mentioned Hot or Not, a dating site that simply posts people's pictures and provides a yes-or-no button instead of a lengthy profile. When Facebook was first launched, he said, it resembled MySpace but was limited to a select group of people--the Harvard student body.

What other products could benefit from some constraints?