Off the Wall

Early this week, developer Sarah Sharp complained about lack of politeness on the Linux kernel mailing list, singling out Linus Torvalds as a prime example of rudeness. Having not persuaded many people, she took her complaints public. Soon dozen of bloggers were condemning Torvalds, and hundreds of commenters were either joining in the condemnation or defending him. But, looking at all the responses, I wonder what everyone imagined they were doing. Feeling good about themselves, maybe?

Don’t get me wrong – as an ex-university instructor, who used to give out criticism to students, I know that rudeness is a poor way to motivate people to improve. They’re more likely to resent the criticism than to change their ways. True, Torvalds is more interested in issues than people, and seems to accept people answering his rudeness with their own, but I still suspect that the Linux kernel is a successful project despite his manner, not because of it.

However, the difference between me and the people publicly expressing outrage is that, as fond as I am of expressing my opinions (that is, after all, what I get paid for), I don’t make the mistake of believing that my opinion matters. It certainly isn’t going to improve the tone of the kernel list.

Condemning Torvalds in public may satisfy the need for self-expression. It may publicly align you with the forces of Progress and Good. However, one thing it will never do is to improve civility within the kernel project. Even if thousands of people express their outrage, it won’t do anything.

In every way imaginable, Torvalds is immune to such criticism. If nothing else, I doubt he reads it. Furthermore, he rationalizes by assuming a dichotomy between effective bluntness and politeness that, for all its falseness, gives him no reason to consider reforming himself (although he might welcome a good argument as amusing).

But how anyone imagines him vulnerable to public criticism is beyond me. He has never been before. The kernel is his project, and few developers are going to work on a fork in preference to the cachet of working with Linux.

Even pressuring him through the Linux Foundation, his nominal employer, has no chance of success. Sponsoring Torvalds legitimizes the Linux Foundation, which means that it needs him far more than he needs it. In the unlikely event that the Linux Foundation did try to chastise him, as a mufti-millionaire he can walk away any time he pleases.

In the end, being outraged by Torvalds’ rudeness is no more than elaboration on the idea that you are helping world hunger or some other cause by clicking Like on a Facebook link. It’s easy to do, and leaves you pleased with yourself, but you have confused an expression of concern with effective action.

The Internet allows considerable freedom of speech, and the free software community often has a democratic appearance. However, neither of these facts means that all of us have an equal say in every situation. In this case, the only people whose opinion matters are those who have any chance of making a change – kernel contributors.

Instead of mouthing off in public, if you really want change on the kernel list, lobby kernel contributors, especially Torvalds’ lieutenants. If you are a kernel contributor, come out in support of a code of conduct, and try to enlist other contributors. None of these actions have any guarantee of success, but they have a far greater chance of encouraging change than saying, “Me, too!” in public.

Put so bluntly, the choice of tactics sounds obvious. But the fact that so many seem to think that condemning politeness is enough to cause change suggests that, on some levels, for most people it is not at all obvious. In the end, campaigning for change is slow and unglamorous, and lacks the immediate satisfaction of posturing as rebel via a token gesture, yet it has two unarguable advantages – it is unhypocritical, and it just might work.