Subject: Re: rocket science
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: Mon, 28 Feb 2005 14:43:55 +0900

>>>>> "Tom" == Tom Lord <lord@emf.net> writes:

    Tom> If you ask me `which is more likely to cripple all of
    Tom> civilization: a bug in rocket software or a bug in desktop
    Tom> software?' -- I'm pretty sure that the bug in desktop
    Tom> software presents the greater risk.  I'll add that I think
    Tom> that that particular risk is, already, non-negligable."

I have to disagree.  First, the fact that it's on a desktop means that
there are thousands or millions of human beings standing between it
and doing real damage.  The same social systems that cushion
civilization from buggy human beings will cushion it from buggy
desktop software.  Because in the end, the human pushes the lever and
we already know better than to trust the human.

And we already know where to look for civilization-crippling bugs: not
in space nor on the desktop, but in the software that controls the
electric power grid, or the telephone system, or railroads.  That's
not a matter of analysis---that's a matter of history.

    Tom> The hackers are good at cooperating among themselves on
    Tom> day-to-day issues across the boundaries between companies;

Heh.  Watching the reactions of several development organizations to
recent CAN notices, I think even this is wishful thinking.  I suspect
that what you're seeing is that obviously good technology gets adopted
rapidly across organizational boundaries.  It's a no-brainer if you've
got competent hackers, in the free software context---you just take
the good stuff and ignore the rest.  But let's assume it works.

At the other extreme, I notice that you carefully separate day-to-day
issues, which are the province of cooperative hackers, from
year-on-year project planning, which the very same hackers have
somehow managed to bungle so badly that the suits have taken firm
control.

I suspect that in the middle, the month-to-month design-level issues,
the cooperative hackers are just muddling through.

To me it's not surprising that the "vision" thing gets taken over by
entrepreneurs with substantially more emphasis on marketing than on
technology, and in many cases with substantially more talent for
marketing than for hacking.  That's as it should be.

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.