Editor's Note: Is Ximian Practicing Appeasement or Practicality?
Ximian Turns Heads and Gets Real
Ximian, Inc. made an announcement today that will no doubt turn heads, especially among people who've been watching Miguel de Icaza over the years since he first kicked off the GNOME Project, ostensibly to wrest the future Linux desktop out of the hands of KDE, which was saddled with the controversy over its use of TrollTech's Qt toolkit.
In addition to the much-anticipated release of Evolution 1.0, which is possibly the best GUI mail client going on the Linux desktop, Ximian announced that it is developing a component for the software that will allow it to interoperate directly with Microsoft's Exchange. If Ximian lives to see the other end of the tunnel in the current economy, it will be taking its desktop strategy in a new direction: from simple mimicry of the things it admires most about Microsoft's own desktop products to a stance of collaboration or, if more angry rhetoricians among us feel like skirting Godwin's Law, appeasement.
This isn't the first time Microsoft has faced a challenge to its Exchange/Outlook combination. Bynari's TradeServer is designed to replace Exchange outright, attacking the server side of the equation and allowing Microsoft clients in an organization to remain in place. Ximian is going after the client, acknowledging that organizations will have investments in Microsoft server software they'll be unwilling to write off.
Without making comment on the health of Linux on the desktop, perhaps the most well-gnawed subject of 2001 and possibly the most bone-headed litmus test of one's "true believer" status concocted since it was fashionable to check suspected witches for buoyancy, this is a good announcement for a lot of organizations that have not decided to make the move to Linux for productivity apps but maintain a sizable cadre of Linux-using engineers, admins, and developers (think Google) who get their work done with Linux.
This will be a nettlesome issue to the element of Linux supporters who want nothing to do with Microsoft at all. This same element is hostile to attempts to work with Microsoft's .NET, thought Evolution itself was a bad idea to begin with if only because it looks an awful lot like Outlook (which implies tacit acknowledgment that there's something to like about a Microsoft product), and generally frames interactions between the world of Linux and the world of Microsoft as a zero-sum game.
We don't agree with these folks and believe they'll ultimately find themselves frustrated as the mainstream computing world discounts what is perceived to be a stance of perpetual self-marginalization and continues, for the foreseeable future, to find a use for Microsoft's offerings. We know of at least one major technology company that found itself rocked hard enough by Code Red and Nimda to muster up some momentum to consider replacement of thousands of back-office servers with Linux machines but refused to consider a parallel move to Linux desktops despite the part Windows software had in exacerbating the company's woes.
Since Evolution works well with e-mail and calendaring solutions built around completely open standards and Open Source/Free Software, an organization seeking to implement a 100% Free Software mail and calendar solution can still do so with Evolution as its primary client, or any other to be found in the Open Source/Free Software world. An organization unwilling to toss out the money, time, and expertise it has invested in Exchange will be able to more easily integrate Linux desktop machines into its operation, providing an opportunity to demonstrate Linux's resistance to the worms and viruses that continue to bring company networks to their knees. Outsiders to the Linux world will see the ability to communicate with Exchange servers as a feather in Linux's cap, and a sign of maturity.
A second issue raised by Ximian's announcement today will likely anger another element in the Linux community. Without providing much detail, Ximian announced that the Outlook-friendly Evolution plug-in will be proprietary and won't be available for free (as in beer): it will presumably be closed-source and will carry a price-tag of $69 per seat.
With this move, Ximian takes its place with a variety of other Linux and open source companies that have begun the move toward mixing their licensing models, trying to find the balance between contributing to the collection of Open Source/Free Software that comprises what most people think of as 'Linux' while also finding a way to pay the bills. Caldera has gone down this road with its Volution offerings and per-seat licensing for shrink-wrap versions of its distribution, VA Linux has announced similar plans for its SourceForge offerings, and SuSE has had a non-free piece of software at its heart for years, which has presumably helped it to pay the salaries of KDE, kernel, and XFree86 developers who sport SuSE addresses.
While we'd prefer a world where companies and individuals could solely develop GPL'd code and make a living at it, too many people in the software business who started out with exactly that intent have backed away for us to believe it's a model that will work for everybody, or that these companies have simply suffered a failure of imagination or nerve during hard times. Even companies that haven't succumbed to including proprietary software in their internally developed offerings have acknowledged that it has a place somewhere: consider Mandrake's Gamer's Edition.
Ximian, like the other companies mentioned, has given something to Linux in the form of outright contributions of useful, GPL'd code or polishing (as with its ongoing work with GNOME). We'd rather have a world where these talented developers continue to work on Free Software and produce the occasional proprietary offering than one where those same developers fade away into shops that produce nothing but closed-source software and follow the numbers in deciding which platforms for which they'll develop: something that, at the moment, people with dreams of a prominent Linux desktop can ill afford.