ZDNet columnist Stephan Somogyi wrote this lengthy and insightful commentary about the recent attack by Microsoft on open source software . . .

Anyone with even a passing interest in open source will have heard about the latest Microsoft take on open source, as interpreted by Craig Mundie. It's an interesting read if only to see what thoughts are bouncing around in the collective noggins up in Redmond.

I must confess, however, that it's not entirely clear to me what Microsoft aims to get out of such pronouncements. After all, Jim Allchin's jingoistic chest-thumping wasn't considered a successful argument in the court of public opinion. Consequently, Mundie's treatise comes across primarily like a finer tuned beta of a Microsoft policy statement, but far short of any kind of coherent strategy.

Shared vs. Open

I won't deconstruct the speech piecemeal, especially since others, such as Alan Cox, have already done a perfectly good job. However, I cannot resist making a few general observations of my own.

As Cox points out, Mundie's discussion of business models is a straw man, since the demise of numerous dot coms cannot by any sane measure be blamed on open source. Indeed, not only was Cygnus Solutions a successful endeavor, but companies like Cobalt Networks used open source to their clear advantage.

When I add Apple and IBM to the other open source success stories, I have to wonder why Microsoft is expending prodigious amounts of energy to escape and evade open source instead of looking for ways to make it profitable. After its late arrival to the Internet, one would've thought that as a company, Microsoft had realized that being hidebound is a competitive disadvantage.

Indeed, I'm surprised that Mundie wasn't leerier about raising the issue of competitiveness. By taking this kind of pot-shot at open source in general, Microsoft leaves itself open to criticisms that it's afraid of competition.

It's well understood that Linux is eating into Windows' market share on servers thoroughly indeed, and with the advent of Samba 2.2 and its ability to act as a (limited) domain controller, this trend looks not only likely to continue, but like it's only now starting in earnest.

Microsoft's Shared Source effort is a simplistic attempt at applying Embrace and Extend to open source. It seems that Microsoft apparently lacks the creativity to find a way to profit from open source; oddly enough, it doesn't seem likely to fiscally profit from Shared Source, either.

Apple gives back

One of the many compelling features of open source software is that companies and individuals not only have the ability to inspect source code as with the Shared Source model, but they can actually make changes, either customizations useful in their limited context, or fixes and enhancements suitable for worldwide use. It's the latter feature, and not the former, that has made open source the force that it is.

Apple was probably the first large computer company to base a its product strategy on open source technology. Apple's engineers and lawyers spent a lot of time and effort on figuring out how to make open source work for the company. While the Apple Public Source License was initially greeted with outright resistance by the open source community, a few iterations later it is accepted as a reasonable license for a company such as Apple, which has the responsibilities of a real company, which differ considerably from those of a typical not-for-profit open source project.

Apple provably gives back to the open source community, too. The Darwin Streaming Server is a provably useful piece of software. Apple's changes to GCC are also being submitted upstream to the FSF as the requisite paperwork is completed. While there are plenty of substantive, legitimate criticisms that be leveled at Apple's Darwin and OS X efforts, being a black hole is not one of them.

Big Blue, too

Despite IBM's current fervor for all things Linux, it's worth noting that Big Blue's first foray into open source was into the land of Apache, whose license is BSD-style, and not the GPL.

This is significant, because I harbor a suspicion that it's far easier for an established commercial entity to get comfortable with open source via a BSD-style license. If such an initial foray is successful, as was the case with IBM's participation in the Apache project and the subsequent release of IBM's WebSphere products based on Apache, then it only makes sense to look at other ways of integrating open source into one's business, which is what IBM is currently doing with Linux, despite Linux being GPL'ed software.

Good business sense

The bi-directional nature of open source is what makes it genuinely useful to businesses and non-businesses alike, and this is what Microsoft doesn't quite want to admit. From a customer's perspective, what's the use of being able to look at source if one can't do anything useful with it?

Given the choice of reviewing source code and feeding back to Microsoft in the hope that the feedback is acted upon, or making necessary changes immediately and getting on with progress, the latter strikes me as more efficient, not to mention more cost-effective. Even looking at it from a business perspective, open source offers more than Shared Source, and it does so with less investment in both time and money.

---

---

 
This article was originally published on LinuxDevices.com and has been donated to the open source community by QuinStreet Inc. Please visit LinuxToday.com for up-to-date news and articles about Linux and open source.