I remember a
very knowledgeable acquaintance of many years ago whose
business card read "Software Craftsperson".
Even in computing contexts,
many people would say that methodical and
consistent development work, i.e. the kind of thing someone familiar
with Knuth and who is a trained software engineer, is better than
something 'hacked together' by one guy at 4 in the morning.
The former is more likely to be good. The latter is more likely to be
brilliant. (Of course, in a world in far greater need of a lot of good
software than a few pieces of brilliant software....)
Also, the best hackers tend to also be good software engineers, though
it can take a decade or so if they have to learn things the hard way.
This is a very debatable topic, of course. But the
myth of the
'greatest software' coming out of hacking sessions is generally
rather dubious.
Well, there's some truth lurking in it; the greatest software usually
grows from a small seed, and that small seed is often the result of a
week of all-nighters or some such, and almost invariably the product of
a single mind.
Of course, that small seed is usually very much a hack job, the sort of
thing TNHD describes in the "Berkeley Quality Software" entry. (That's
the distinction between good and brilliant that I drew above.)
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse(a)rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B