Simon Philpps, Chief Tech Evangelist, Sun Microsystems "Industry Perpective" Impressionistic transcript by Cory Doctorow doctorow@craphound.com July 9, 2003 Supernova Conference Washington DC -- Is social software just "Computer Supported Collaborative Work"? Why is it causing excitement now? I've built mainframe-based collab systems: threaded discussion, etc. A decade ago, I saw Wikis. Why is there fresh interest in this? I've stolen others' phrases to explain this: Paul Saffo: "Any tech has got to fail twice before it can succeed" [Ed: very Joseph Campbell-y] IRC hasn't failed yet. At IBM, we had ToolsRun. At CSCW, we had Wikis. There's nothing new under the Sun or at any other vendor. But the audience has changed. It took 10 years for "The network is the computer" to go from a slogan to a philosophy and 10 more to become a reality. "Decentralization" is too tech-centric. We are becoming "massively connected." This is the third coming of twice-failed tech into a massive connected society. -- Re: Metcalfe's Law: How many used BBSes in the 80s? [Lots]. IBM came up with SNA -- the architecture that would rule the world. What if the Itnernet had been built on BBSes or SNA: Prodigy. A central administrator would have to give permission to add your browser to the master table, and every site would use a different protocol. "Massively Connected" is all about shared, open, royalty-free loosely coupled standards -- not "everything joined to everything else," but "everything joined to everything else WITH THE RIGHT RULES." -- Every wilderness is paved, and every pioneer needs to cope with this. If you're on the leading edge, you need to move fast. -- Bill Joy: "Innovation happens elsewhere." Not all the smart people can work for you at the same time. Simple math says that most of them work somewhere else. The obvious way to develop software is to get other people to innovate for you. So OSS becomes obvious: it's the beta for software development in the massively connected society. It's not about "free" or "licenses" -- it's about Benkler's "Community-based peer-production." All the smart people come together and hack the tools they need. The licenses just facilitate that. What matters isn't the coolth of the licenses or the freeness of the code: it's the facilitation of a community that comes together and defines and builds the software. Sun's "Madhatter" project: we're going to try to build a business out of a community development model. -- "The most important consequences are the unintended ones." 1. How standards are created 2. How they impact society Hard to predict in a Massively Connected society. Web services is stalled because there's genuine disagreement about how the protocols should work. But blogs are a wildly successfull web service, with loose consensus and running code. If OSS is a beta-test for software-development in a massively connected world, then Echo is an alpha-test for standards-development in a massively connected world. Sam Ruby's wiki puts all the smart people in the same room at the same time. Standards used to be experts in smoke filled rooms specifying how plugs work. -- Digital ID: "You have no privacy get over it." The bad thing happened a long time ago, with your loyalty cards, credit cards, SSN, sat photos, etc. They already exist. DigID can redeem this mess. RFID's unintended consequences aren't being thought of at the moment. Blogs' unintended consequences aren't being thought of at the moment.