Off the Beat: Bruce Byfield's Blog

Nonsense prevails, modesty fails,
Grace and virtue turn into stupidity.
While the calendar fades almost all barricades
To a pale compromise.
- Elvis Costello, "All This Useless Beauty"

Hearing the news that 75% of contributions to the Linux kernel are by paid developers, my first reaction is to recall Alec Guinness in The Horse's Mouth. In this classic from the 1950s, Guinness plays an eccentric and disreputable artist who cons his way into becoming a caretaker of a luxury flat so that he can paint a mural on its wall. When, after countless mishaps, he looks at his finished work, he mutters, "It's not the vision I had in mind," and walks away.

Then I wonder if I'm being too critical -- if the figure isn't a reason for cautious rejoicing rather than despair.

I discovered Linux and the rest of free software during the Dot-Com era of the late 1990s. That was business' first introduction to free software, and it was more radical than a modern newbie could imagine. Under the name of open source, free software was regarded as the victory of the amateur over the professional, of the at-home tinkerer over the paid professional. It was the epitome of the Internet (which was also in the process of being discovered then), and, like the Internet, it was supposed to revolutionize business.

Unsurprisingly, when The Clue Train Manifesto announced that business henceforth would be conversations, several of the first people involved in project were old FOSS hands like Doc Searls. Free software was going to change the way that people did business, and transform or replace corporations with a more humane way of operating.

From this decade-old perspective, the figures for kernel contributions seem reason to despair. Instead of replacing the professionals, the glorious amateurs have been overwhelmed by them, until now they contribute only 18% of kernel code. Instead of transforming business, free software has become simply another strategy, to be used when it offers benefits and ignored when it proves a handicap. It's all enough to make me think that the traditional distrust of business in the free and open source community is justified.

But then I start thinking of the implications, and I'm less sure that the situation is so negative as I've painted. The same news item mentions that the major contributions to the kernel come from Red Hat, Intel, IBM, Novell, and Oracle, who together contribute 29% of the code.

Today, that information is mentioned almost in passing. But, ten years ago, who would have imagined that these corporations would have realized that, in some circumstances, cooperating with each other is worthwhile? Viewed in this light, it only seems natural that Red Hat, the corporation most dedicated to open source ideals, should be the largest corporate contributor.

Sure, the corporations still compete fiercely in other matters. Yet the fact that they can cooperate under any circumstances does seem to indicate that some watered-down version of the old ideals has influenced them. If a complete transformation hasn't happened, a light makeover seems to have.

Very likely, too, this makeover is most confined to the IT Departments. No doubt in the upper echelons, among the senior management and corporate officers, the truisms of the business schools still prevail. Still, these are tech corporations, so they are undoubtedly places where the developers have greater influence than at most companies. Under these circumstances, if developers are engaged in the free software community, then some of their daily ideals are going to influence the rest of the corporations.

Don't forget, either, that this news came from one project. Companies that don't contribute to the kernel may contribute to other projects. Similarly, those that don't contribute at all may still use free software -- and, the last time I was out working as a freelance consultant, it was getting increasingly harder to find any large company that didn't. The few exceptions were freakish, and inexplicably gung-ho about Microsoft. But I suspect that results would be in many parts of the computer world would be comparable to those found in the kernel.

I'm not saying that free software has accomplished all we once hoped it would do, by any means. Yet, to a surprising degree, in the last decade, free software and its ethos has challenged traditional business wisdom, even if it has not supplanted it. Even if they are getting paid, thousands of contributors must be living by an ethos not much different than they would hold if they are volunteers. They may need to watch what they communicate in order to avoid violating their NDAs, but you can tell from their mailing lists and IRC channels that they are living with what, in the corporate world, is a surprising degree of freedom. And it's all happened so quietly that only when looking back can we realize that it has happened at all.

Of course, the result has none of the bright promise of revolution that the community once expected. Yet, even if the results aren't going to have us dancing in the streets, they are still reason for a bit of satisfaction -- even if, like Alec Guinness', they aren't exactly the vision we had in mind.

Comments