A belated response to “A Generation Lost in the Bazaar “

Back in 2012, Poul-Henning-Kamp wrote a disgruntled article in ACM Queue, A Generation Lost in the Bazaar.

It did not occur to me to respond in public at the time, but someone else’s comment on a G+ thread about the article revived the thread. Rereading my reaction, I think it is still worth sharing for the fundamental point about scaling and chaos.

There are quite a lot of defects in the argument of this piece. One is that Kemp (rightly) complains about autoconf, but then leaps from that to a condemnation of the bazaar model without establishing that one implies the other.

I think, also, that when Kamp elevates control by a single person as a necessary way to get quality he is fooling himself about what is even possible at the scale of operating systems like today’s *BSD or Linux, which are far larger than the successful cathedrals of programming legend.

No single person can be responsible at today’s scale; the planning problem is too hard. It isn’t even really possible to “create architecture” because the attempt would exceed human cognitive capacity; the best we can do is make sure that the components of plannable size are clean, hope we get good emergent behavior from the whole system, and try to nudge it towards good outcomes as it evolves.

What this piece speaks of to me is a kind of nostalgia, and a hankering for the control (or just the illusion of control) that we had when our software systems were orders of magnitude smaller. We don’t have the choice that Kamp wants to take anymore, and it may be we only fooled ourselves into thinking we ever had it

Our choices are all chaos – either chaos harnessed by a transparent, self-correcting social process, or chaos hidden and denied and eating at the roots of our software.?