|About InfoWorld : Advertise : Subscribe : Contact Us : Awards : Events : Store|
Follow the money
BLOGTROLLING CONTINUES to reward. Hopefully, one day RSS (Really Slick Syndication) tools will automate information gathering, but for now we have to rely on trusted sources for the daily scoop.
I say "too often" because those drivers, in and of themselves, can fade with time and wear. Even the other common motivator -- peer recognition -- can dissipate without a coherent foundation of resources. To quote the philosopher Frank Zappa and his Mothers of Invention, "We're only in it for the money."
Actually, InfoWorld lead analyst Jon Udell and Google steered me to Peter Drayton's Weblog. Drayton's reworking of Groove's edge services into the Groove Web Services (GWS) architecture is a marvel of economical system design. Now Drayton is cashing in with what he calls his dream job -- program manager in Microsoft's Common Language Runtime (CLR) team. Apparently, he didn't think he had the credentials for the job back in July when Microsoft's Jim Miller posted a job description in the .dotnet-sscli mailing list. "Apart from making me wish I had a Ph.D. in computer science and 'high visibility in the external computer science research community', " Drayton wistfully wrote, "this is an interesting indicator of how Microsoft is thinking about transferring technology from research (in Rotor) into production (in the CLR)." Rotor is the shared-source implementation of elements of the .Net Framework and the CLR.
But Miller's job description actually read "or" instead of "and," as in "a Ph.D. in computer science or ... a minimum of eight years industry experience at the cutting edge of software technology, interacting with the research community." Drayton's success in transferring Groove's technology from a locked box to an open treasure chest certainly qualifies as the cutting edge.
Blogtrolling uncovered Mitch Kapor's open-source, peer-to-peer PIM (personal information manager) project, though it has been an open secret in the community for at least a year. But Kapor, Lotus founder and Lotus 1-2-3 designer, launched a Weblog to go public with his Open Source Applications Foundation (OSAF) and its Chandler project.
Chandler is named after the great detective novelist, but it might as well have been called Colombo in honor of its style, as in "Oh, I just had one more feature ..." The feature list runs the gamut from the usual e-mail and calendaring suspects to the science fiction at the edge of the network -- remote querying and peer-to-peer browsing of others' data, publish/subscribe auto-updating , integrated IM, and presence.
Kapor has invested $5 million in this open-source project, to build on top of a carefully chosen suite of technologies: the cross-platform wxWindows layer, Python's interpreted object-oriented programming language, the ZODB (Zope Object Database) persistent store, Jabber's XML IM platform, and Mozilla's cross-platform browser project.
And Kapor has other significant investors, most notably Mac UI pioneer Andy Hertzfeld, who's volunteered his time to build the first prototype. Initially, Kapor's target was not Microsoft Exchange but market leader Outlook. "While feature-rich, [Outlook] is very complex, which renders most of its functionality moot," Kapor says on his Weblog (http://blogs.osafoundation.org/mitch).
"Its information-sharing features require use of Microsoft Exchange, a server-based product, which is both expensive and complex to administer," Kapor writes, suggesting that Exchange is overkill for small-to-medium organizations, "which we think creates on opportunity we intend to pursue (as well as, of course, serving individual users)."
Five million dollars doesn't go far these days, so enterprise customers will have to wait until commercial developers build on top of Chandler's code base. Kapor says he worries "about migration paths from existing products, synchronization with PDAs, and a whole host of details beyond core functionality ... required to make a truly first-rate product."
Doesn't an open-source Outlook clone such as Ximian's Evolution already fill the bill? "[It's] an approach which benefits users by being familiar, but doesn't take design risks which could have big payoffs for users in power and simplicity," Kapor writes. "We're building the product using up-to-date architectural components (peer-to-peer networking, integrated instant messaging, an RDF-compatible semantic database), and are not saddled with legacy code."
Ironically, OSAF's nonprofit model both intersects and competes with Groove, a company Kapor invested in both privately and as partner in a venture capitalist firm. He talks of Chandler plans for "perhaps half a dozen killer features elsewhere unavailable," one of which would leverage "the general and powerful information-sharing technology (built on top of Jabber) we are embedding."
It's shared vs. open source, investment vs. intellectual capital -- two models for financing innovation. "If I were Microsoft I'd be worried about open source in general, not about losing Outlook/Exchange market share," Kapor blogs. "With or without OSAF, I believe all the applications in Office will be commoditized with equivalent free versions."
And then there's the Redmond approach: "A successful candidate will have high visibility in the external computer science research community, interpersonal skills to deal with both that community and the Microsoft product development community, and a deep passion around the transfer of experimental technology into product."
Place your bets.
Steve Gillmor is director of the InfoWorld Test Center. Contact him at firstname.lastname@example.org.
Click here for all of Steve Gillmor's past columns.
SPONSORED WHITE PAPERS