What happens when leaders quit?

Off the Beat: Bruce Byfield's Blog

Jun 24, 2015 GMT
Bruce Byfield

Recently, Swapnil Bhartiya published an article that quoted Linus Torvalds speculating on what might happen if he quit leading kernel development. To my surprise, I have seen the article condemned here and there as being in poor taste. Yet the more I think, the more I think that many free software projects need to start similar discussions.

We live in a death-denying culture, and the sub-culture of developers is not known for its maturity. All the same, as free software advocates and contributors are aging. On social media, you may have noticed gray hairs that weren't there a decade ago, and people who were once cradling infants now congratulating those same infants on graduation from university.

Or to make the matter concrete: of those who founded free and open source software, Richard Stallman was 62 this year. Eric Raymond, who provided the philosophical underpinnings for open source, is 58. John "maddog" Hall is 64, Bruce Perens somewhere in his mid-fifties. At 44, Linus Torvalds is relatively a boy, but has teenage children if he needs any reminders of the passing of time. These are ages and circumstances in which the illusion of youth is difficult to maintain and  concerns about what happens after you are gone start to sprout.

Ten years ago, Eben Moglen reached similar circumstances in his mid-forties, and founded the Software Freedom Law Center to provide for his own legal successors. Admittedly, some of free software's first generation of leaders are already inactive, but, for others, the time is rapidly approaching when not thinking the unthinkable will be irresponsible. Even if most of today's leaders stop short of establishing their own foundations the way that Moglen did, the question of who the next leaders will be, and how leadership will be transferred will only become increasingly important.

What happens next
Part of the reason why these concerns are worth facing is that changes of leadership are not that common among the larger free software projects. In many cases, the same leader has been in place for over two decades, and a change would be as startling as the announcement of a new host for a leading late night talk show.

A notable exception is the kernel itself, where Torvalds has moved increasingly into a managerial role, and encouraged the rise of lieutenants to take on the daily responsibilities that were once his. Under these conditions, Torvalds is probably right in suggesting that his departure would disrupt kernel development hardly at all -- an observation that, contrary to some of the recent complaints about his belligerent manner -- demonstrates considerable leadership skills.

By contrast, who can imagine the Free Software Foundation without Richard Stallman? Probably, it is impossible to replace a symbolic figurehead, but, at least in the past, Stallman has more to do with the FSF's daily operations than many assume. With Stallman gone, who would make the final decisions? Would a president be appointed, or would there be some sort of realignment of the working relationships between the board of directors and the executive director? Some sort of internal culture change seems to be inevitable, and perhaps the FSF would run more smoothly if it settled such questions now rather than letting the answers evolve on their own.

Another question that retiring leaders need to face is how to ensure that free software is more than a single generation's concern. The FSF has tried to encourage people under 25 to become involved in its programs, but you only need to attend a typical conference to notice that the average age of free software developers and advocates is solidly middle-aged, and not that much younger than its leaders. If free software is going to have a next generation, then projects need to discuss what policies and practices could help guarantee it.

Some leaders might want to ensure a continuance of their own policies -- a desire that is almost certain to remain largely unfulfilled. Yet without focusing on egos, project members may want to discuss how their successful practices can continue without them. Projects as explicitly governed as Debian can likely ignore this concern, because their organization will tend to preserve the project culture. Less organized projects, however, may need to add the concern to their discussion lists.

The final responsibility
Few people care to think of the world going on without them. Yet, for leaders, thinking about that circumstance is a responsibility -- and one better thought about in advance than in the confusion of necessity.

Maybe you need to be a certain age to understand, but thinking such thoughts is no more morbid than making out a will: You hope it won't be needed any time soon, but you don't want to leave behind chaos. Moreover, it is a discussion that only becomes increasingly urgent the longer it is ignored.

comments powered by Disqus
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Subscribe to our ADMIN Newsletters

Support Our Work

Linux Magazine content is made possible with support from readers like you. Please consider contributing when you’ve found an article to be beneficial.

Learn More

News