Reinventing the intranet

In an interview long ago, Marc Andreessen told me about the moment he knew Netscape’s business plan would succeed. That plan, as you may recall, was modeled on Gillette’s: give away razors (browsers and mail/news clients) and sell blades (enterprise servers). For Andreessen, the magical moment came when, shortly after the word “intranet” was coined, he heard it echoing all around him in a restaurant.

Free IT resource Open Source Business Conference (OSBC) May 22-23, 2007 Sponsored by OSBC Free IT resource Virtualization Insights from Top Experts - Learn how virtualization gets real! Sponsored by Dell

As it turned out, Netscape didn’t sell many of the blades. It’s been a while since I’ve heard people in restaurants chattering about intranets. And Andreessen’s new venture, Ning, seems to have nothing to do with the enterprise. It’s a Web 2.0 service for building social Web applications.

Ning may not be the next Netscape, but ad hoc social software will likely flourish in the enterprise. Is this the next-generation intranet? If so, we should sort out what we got wrong on the first try, and what we’ll get right this time around.

The key challenge is scale. And I don’t mean scaling up, as we usually want our enterprise applications to be capable of doing. I mean scaling down -- because no matter how large your company may be, it’s a drop in the ocean of the public Web.

Even on the public Web, useful network effects aren’t a sure thing. On an internal network that reaches hundreds, thousands, or even tens of thousands of people, the odds are much slimmer. There never was a great solution to this problem, but there are some commonsense ways to improve the odds.

Consider search. Can you instantly find everything -- at company, departmental, or workgroup scope -- that you’re allowed to see? I thought not. Posting information to the public Web is far more likely to guarantee successful search than posting to the intranet.

It ought to be the other way around. Inside the enterprise, teams, tasks, products, and services define metadata vocabularies that the Internet search giants would kill for. Exploiting those vocabularies to deliver search results that are better than what’s available on the open Web is low-hanging fruit. As we roll out SOAs that route well-formed messages through a fabric of intermediaries, it’ll get even easier. What Verizon’s CIO calls “Googling the enterprise” should be the rule, not the exception.

Shared bookmarking, coupled with tagging, is another piece of low-hanging fruit. Sprinkling Web 2.0 pixie dust won’t solve every problem, but the benefits of public services such as del.icio.us and Furl can be realized within the enterprise, too. That’s true because they benefit the individual first, and then, as a useful side effect, the community.

Given the opportunity, people will want to bookmark and tag the resources they publish internally. It’s the easiest way to create, manage, and share dynamic lists of such resources. This system pays for itself in improved personal productivity alone. Everything else is gravy, and there’s plenty of that.

Saved bookmarks chart the current and historical levels of interest in what their URLs represent, and they identify groups that share those interests. (Note that behind the firewall, bookmarks can refer to public resources as well as private ones inside the enterprise.) Tags identify sets of related resources and groups related to those sets. They also extend the metadata vocabularies that can be used to improve search. What’s not to like?