View from the Alpha Geek
Q&A: O’Reilly programmer and technology maven Rael Dornfest looks beyond Web services and social networking.
Rael Dornfest speaks quickly but calmly when he discusses trends. The words pour out with a hint of a foreign accent; he’s glad to release what his high-bandwidth brain has already absorbed and processed. Dornfest codes software, edits books, and organizes conferences for O’Reilly & Associates, the publisher of popular instructional manuals for programmers. The job gives him a close up perspective on computing’s grass roots movements.
Dornfest was program chair for the O’Reilly Emerging Technology Conference, or E-Tech, held this week in San Diego. Dornfest relied on a “kitchen cabinet of alpha geeks” to fill out three days of speakers and workshops that proceeded under the theme: “See the Future. Shape the Future.” Before the conference started, Dornfest told freelance writer Brad Stenger that he sees a future filled with social software, wireless devices, hardware hacks, and new interfaces, services, and business models.
TR: When you’re sizing up trends for O’Reilly, who do you pay attention to-and how do you keep track of them?
DORNFEST: A few years ago, the New York Times ran an article about a games company that wanted to find out the coolest kids to test out their stuff. So the company went around asking all these kids, “who’s the coolest kid you know?” And they’d tell them, “it’s Johnny over there.” Eventually the company came to the person who most kids pointed to, who would readily admit to the role-and that would be their tester. Finding these alpha pups, these alpha geeks, is a lot of what we do at O’Reilly.
TR: Can you connect that to what went on in putting together the Emerging Technologies Conference?
DORNFEST: We know who these smart people are and what they’re doing. We say these are the voices that we heard. We try to grow it every year. We’re bringing some stuff along from last year. We take a look at what’s new. We see what’s grown and what hasn’t grown, and what’s changed.
TR: How has the conference audience changed over the years?
DORNFEST: When it started five years ago, there were a lot of peer-to-peer hackers. The focus was Web services and there were a lot people talking about things like pipelining, coordination, and orchestration. That talk disappeared very quickly. The maelstrom kind of melted and became this grassroots thing that is happening now.
TR: What’s an example of that grassroots phenomenon?
DORNFEST: Web services are showing up all over the place. John Udell did a cool hack, called LibraryLookup. Here’s how it works. You go to any Amazon.com page and look for a book-and then click on a bookmark that takes you to your public library then look it up for you. Udell went to the library and saw that it was a place where people were sharing files freely. It wasn’t illegal and they were all copyrighted works. You could borrow them. When you were finished you could take them back. And it’s called the library. Stunning.
Little projects like that are what I call syndicated e-commerce-putting out pieces that you couldn’t do on your own, or that you wouldn’t want to do on your own. Put your project into the hands of people who otherwise wouldn’t help out. There’s not much programming. You can use Google for search and for ads. You can use Amazon for product shipment and sales and verification and availability, eBay for auctions, and Salesforce.com for contact management. In the past, you’d have to make all those tools on your own.
TR: Are there other cases of technologies that had been promoted before that are now becoming really useful?
DORNFEST: I think that RSS fits that description. RSS-which stands for RDF site summary, or really simple syndication, or rich site summary, depending on who you ask-is an XML metadata format designed to automate sharing of Web content between sites. I was author, along with a couple other folks, of the RSS specification.
TR: What makes RSS so important?
DORNFEST: It’s one of the most widely deployed Web services around. It’s easily decentralized. Blogging, because of RSS, has gone from a popularity contest-how many times has my site been pinged?-to discovery of information. The browse metaphor for search has gone by the wayside. I think that RSS and syndication and blogs broadened it out. There was a lot of hype about this a couple of years ago but it’s only in the last year that things really started bubbling. I use weblogs to find information more than I use Google. I follow trails. Google will find you obvious answers. Google will find you what you’re looking for. Blogs, with RSS, will find you what you never expected to find.
TR: What you just described blurs the line separating Web services from social software.
DORNFEST: Absolutely. It’s a very fuzzy line between “here’s everything you know about a product,” and “here’s everything you know about a person or a community.”
TR: Another thrust at E-Tech is mobile devices. Is there a common thread between social software and what’s going on in wireless?
DORNFEST: Moving from desktop to laptop and unwired laptop changed the place you park yourself but didn’t really change all that much about how you network. Think about it: I’m doing e-mail here versus doing e-mail there. I’m using the same tools. I’ve just picked up the wire and broken it. What I’m seeing that is fundamentally changing things, though, is the cellular stuff. When you go around the United States, you talk about mobility and you’re mostly talking about laptops and WiFi. The rest of the world is talking mostly about cellular.
TR: Having Meetup.com on your cell phone means that you’re never wanting for something to do.
DORNFEST: Absolutely. Or you’ve got conversations in your pocket, through IRC [Internet Relay Chat]. I can take my IRC friends with me, and not just talk to them when I’m sitting in the caf. I can be wandering around a store and say to myself, What was that thing I saw on the Internet? And I can ask my IRC friends right then and talk to them about the thing in the store. It’s a way of bringing the online life out into the world. There are things to figure out as you bring more of the online life off-line: How to know who to trust? How to know who it’s important for you to meet up with? How do you know where they are? Social software is going to play a large part in mediating some of those interactions.
TR: How well does today’s social software do that?
DORNFEST: I think there’s a long way to go, but I’m excited about it. The social software we play with everyday allows you to introduce me to a friend because you think we might have something in common. The interaction goes like this, basically: “Someone says you are their friend: Is that true? Yes or no?” The pressure is on! Do I say yes or no? So-and-so says you would take a bullet for them, or that you don’t care about them at all. It feels that extreme. If I had known that it was that easy to make friends I would have written Orkut. It’s a very primitive attempt at recreating the subtlety of real world interaction but it’s good that we’re trying these things.
TR: It seems as if for social software and mobile devices to advance, there will also have to be developments in business models, new services, and new interfaces. It’s as if they all have to move forward together.
DORNFEST: Yeah, that’s the supporting cast. Look, for example, at this SPOT watch [a wristwatch introduced by Microsoft in January]. Microsoft made a fundamental design decision that it would be read-only. You wouldn’t be able to send messages. You wouldn’t be able to respond. That changes fundamentally how you interact with it.
TR: Isn’t that rather limiting?
DORNFEST: Actually, that makes the device far more useful, in that I don’t have to do anything. It’s a notification service. And that’s what I wanted for my wrist. I wasn’t going to write on my wrist. So the interface has to change for the purposes of the device. And the service has to change.
Another example is cell phones. Look at WAP [Wireless Access Protocol], an open standard for handling Internet content on mobile phones. This was the first attempt to enable you to access the Web from small-screen devices. And it worked-sort of. Then phone makers managed to make the devices slightly bigger and put in a regular Web browser, experimenting with interfaces. Now, I love Sony and Ericsson because they’re managing to keep their phones simple. By contrast, using a Nokia phone feels very much like Windows 3.1: there are icons all over the place and you’re navigating without necessarily knowing what’s going on. The phone makers are experimenting. They have to because the data being thrown at these beasts has completely changed over the last few years.
TR: If the communities of innovators are geographically distributed, there’s no one place to look for the next big thing.
DORNFEST: It’s the “small pieces, loosely joined” idea, to use Dave Weinberger’s phrase. Yes, Microsoft can zoom in, but other people can still do pretty well. You can whip up an Orkut pretty quickly. But you should go and see what kind of stuff Microsoft is interested in with social software, and learn from that.
Innovation is all over the place. That’s what gets me most excited, seeing this stuff we’ve been talking about show up all over the place. It’s coming from all corners. It’s coming from the citizen engineers. It’s coming from the research labs. It’s coming from the logical progression of things. And it’s coming from those brilliant people who are either burnt out, or are out of jobs, or who are rebuilding after the bubble burst. There’s no one place to look for it.