SAN FRANCISCO-When companies such as Adobe, AOL, Eye-Fi, JPMorgan Chase, Sun Telelogic and even Google want a certain something in their Web applications, they go to SitePen, a Web application development shop that does a little support, training and consulting as well.
SitePen also features Joe Walker, creator of Direct Web Remoting, also known as DWR-a Java-based library for AJAX development-and Kevin Dangoor, the creator of TurboGears, a Python-based Web application framework. And that's just a short list of the Web 2.0 name brands employed at the company.
I met with Russell here and asked him how a small, distributed operation could attract and maintain so many hot commodities. He credited the open-source model.
"It's not like there's some secret to this," Russell said. "People want to work on things that matter to them, and insofar as SitePen is deeply intertwined with a number of pretty successful open-source projects, we wind up as people-not as the company, but as individuals-being incredibly identifiable inside those communities."
Moreover, "it's kind of a weird thing about open source, but a lot of those people who have enough time to work on open source have enough time to work on open-source stuff because their jobs suck," Russell said jokingly. "So it kind of turns into this interesting scenario where a lot of people who otherwise might not be well-connected are very well-connected into these open-source communities and want to work on something more interesting.
"So in a lot of cases, it's just that people don't want to move, for one. And we all want to work with people we know we like working with. And we volunteer to work with these people in our spare time, so why wouldn't we want to work with them day to day? It helps that we pay competitively and have interesting work."
When I asked to meet Russell at the SitePen offices, completely prepared to drive to Palo Alto, Calif., he made no bones about not having an office. "We're truly distributed," he said.
Like many technology entities, SitePen is "headquartered" in Palo Alto but has no single brick-and-mortar office it calls home. So Russell and I met at a Gordon Biersch Brewery Restaurant at the corner of Harrison Street and The Embarcadero in San Francisco, smack dab on the route the Olympic torch relay was supposed to run (coincidentally right about the time of our lunch meeting)-except San Francisco authorities got smart and rerouted the relay away from throngs of protesters lying in wait for the torch to pass by. The protesters still marched and protested, though many gave up and some even wound up in Gordon Biersch to have a beer and watch the actual torch relay on TV.
Anyway, listening to Russell's smart, clear-headed thinking on OSS (open-source software), I couldn't help but view the SitePen brand of OSS as a passing of the torch-what OSS is really all about.
"We put a lot back into the open-source projects we work on, such that it's almost the easiest thing to do that if you're a super-valuable contributor to one of these open-source projects we'd love to have you if you play well with others," he said. "It comes down to basic, good community practice-give more than you take, create a lot more value than you personally want to siphon off and make it easy for people."
Among the open-source projects SitePen leads or contributes to are Dojo, DWR and CometD. The popularity of Dojo attracted investment and support from organizations such as IBM, Sun Microsystems, BEA Systems and the Mozilla Foundation.
Russell said it's absolutely right that software tool vendors develop products for and sell into the trend of distributed, collaborative development.
"But let me preface everything I've said about our model by saying that I think it's clearly preferable to have people in an office working together," he said. "Insofar as you can do it and still get the right people on the bus, then you should. But when you, for whatever reason, can't ... we feel that it's significantly more important to have the right people. And that's where we step in. We're able to be so much more effective than teams we wind up interfacing with, or replacing in a lot of cases, just because of who is involved. Not anything more sophisticated or important than that. It's just who we've got on the project."