On Fri, Dec 05, 2003 at 12:49:02PM -0800, rone <rone at ennui.org> wrote: > Matthew Hunter writes: > Actually, tradition is useful in precisely those situations where > we do not understand its genesis or the context in which it > originated. That is, tradition is the force that argues for > continuing to do things the way that *is known to work* (at > least, for evolutionary values of "work") rather than changing > something that we think we understand as being safe to change... > but are not necessarily correct. > A scientific people would prefer to understand why things are done a > certain way, not necessarily to change it, but to see if there is room > for improvement. Preference, sir, is irrelevent. I do not say understanding is an undesirable act, merely that the benefit of tradition lies in the successful function without requirement of comprehension. > You'll never see me argue with results, and i do lots of things by > tradition at work (one example is the classic "sync && sync && > reboot"), but i'd rather understand what's going on behind it. There was a bug in one of the early linux kernels that failed to sync disks properly on shutdown. If you called "sync" manually first, it would work properly -- unless something else wrote new data after the sync. By calling it twice in a row, you write all pending data, then write all data written while you were syncing the first set, then reboot. There's still a window, but it's a much smaller window than no sync and a somewhat smaller window than a single sync. > You should try to understand the root causes of a particular > tradition before advocating it be tossed out, but that > understanding is not required in order for the tradition to be > useful; in fact, the tradition is most useful in the absence of > understanding. > You're almost implying, it seems, that once you understand the process > behind a tradition, you destroy the tradition, much in the way that > explaining a joke destroys the joke. I am offended, sir. I do nothing halfway or almost. ;) -- Matthew Hunter (matthew at infodancer.org) Public Key: http://matthew.infodancer.org/public_key.txt Homepage: http://matthew.infodancer.org/index.jsp Politics: http://www.triggerfinger.org/index.jsp