Too many lords, not enough stewards

I think this talk was pretty interesting, so I wrote a lengthy mail to Daniel personally. These are some of those thought and some other ones. It is too long so no-one will read it except me, so I can go back and feel smart and smug about myself.

I would say my condensed appraisal is that the problems are real, but is not in any way limited to the kernel community and in some cases deeply founded in ~10000 years of culture.

The reference to the book about domestic relationships focus on some aspects of the problem that are psychological and individual. We can all be better people. Some have more potential for improvement than others. Some are more fragile, some are more stoic, some likely have distorted emotional abilities on the autism-spectrum making them especially hard to deal with. Some people will react with hostility to anything resembling psychology simply because they are insecure people and do not like the idea about thinking critically about themselves and their own behaviors, also they tend to place the problem elsewhere than within themselves when put to the point. They will have to be reached by other methods than directly pointing out personality flaws, it's just not going to help, they will feel intimidated.

The problem with leadership is harder to get at and has sociological, anthropological, ethnological and political implications. Nobody really knows what good (community) leadership is, I haven't ever seen a scientific deductive reasoning idea of that, ever, all statements seem to be inductive, even to the point of being just a bin of anecdotes about people "many think were great leaders". Different cultures have very different ideas about what constitutes good leadership on top of that (google "power distance index" or "Hofstede's cultural dimensions theory" for a wakeup call).

The same goes for organization. What is a good open source organization principle? We know that some of the most *successful* ones have developers that are organized hierarchical and despotic, the despot may be benevolent or far from, they can still be successful. Despots or maintainers rule by governors called submaintainers.

Whether a successful leadership is a good leadership is a philosophical question. If we take the stance that it is desirable that leadership makes the developers feel good about their work environment, that is a direct utilitarian claim. (Most happiness to the most people.) What is the goal: to get the technology in place at any cost or to make the developers feel reasonably OK or even great about their work-life? History certainly do not lack examples of great technology being developed by people under threat even to their personal security and their families lives. Was that "good" then, in the philosophical sense of technological imperative?

What we have today is neither overly (tyrant) despotic or ideally friendly. There is room for improvement.

I think the main takeaway from Daniels talk and other talks he made is that when you get a group of developers and (sub)maintainers with very similar abilities and strong trust among them, it is possible to make the organization more decentralized. More anarchist, less autocratic if you like. And that works great for everyone involved.