Bill Thompson, a technology analyst for the BBC, has an intelligent and entertaining column, and I recently stumbled across this entry. It is a wonderfully concise daydream about what it would mean for Microsoft to adopt, consume, digest, and release a dominant version of Linux. A billion dollars here, a billion there, a few thousand developers redirected in the right spots and viola - instant success. It will remain a daydream, but a good read none-the-less.
To me, the thing Mr. Thompson's column washes over at the very beginning is how we view Linux, and how it most benefits customers. He wrote:
"The problem is GNU/Linux, a beast they [Microsoft] cannot destroy and cannot seem to tame, a beast that is encroaching on their markets by offering an alternative to the closed development and licensed software model..."
Linux represents strong product competition - and that is a good thing. Competitors on both sides of any match intend to win - thus they are called competitors rather than cooperators. A football match with both sides trying to help each other would not be much of a game to watch. In the operating systems game, the folks at Apple or SUN are more interested in bolstering their own operating systems rather than push the community OS. The Microsoft development teams are focused on improving our products. Even within the Linux community, the SuSE engineering team is working hard to create competitive differentiation from the Red Hat offerings. Linux has energized the competitive elements of the operating space and that is what benefits the customer most. More choice, more price competition, more product improvements, better applications ecosystems...the list goes on. Customers do not want vanilla solutions from the industry. They want to have manufacturers slugging it out and driving each other to produce better products. In fact, many enterprise customers have procurement policies in place that prevent single-source solutions for exactly this reason. Even if it means paying more for some subset of technology. The benefits to them from greater choice over the long-run far out weigh many short-term concerns.
Mr. Thompson goes on to suggest that Microsoft would:
"...allocate a billion dollars worth of programmers to shine and polish it [Microsoft Linux or Micrix] for a year, improving its compatibility with Windows Server technologies, donating parts of the Windows and Office code bases under the GPL and turning it into the world's best operating system."
Aside from the fact that we have no plans to do this, I think it may be a bit aggressive of a schedule. IBM has put more than $1B into Linux, redirected huge numbers of engineers, donated technologies and patents...and it is 4 years later. Linux is still not as mature as other UNIX versions out there never mind Windows or other commercial operating systems. I certainly believe that with continued investment and time Linux will continue to improve, particularly has customers use it for commercial deployments and drive requirements back upon the commercial vendors.
But, back to some old topics I have hit in the past, the more Linux matures and becomes commercial - the more closed it will become to customers. The more challenges the commercial community will face in dealing with standards and the more sustained engineering elements (such as security response and backwards compatibility) will drive increased dependence on a small set of vendors.
From my perspective, going down the path of overwhelming the Linux development community with our resources seems like a low-percentage bet. We have an excellent operating system already and continue to improve upon it. Perfection it is not - but it is a strong competitor in the market. That is what most commercial producers of software are looking to have.