[OT] Grand Unified Theory of Computer Suckage (fnord)
Fred
puissante at lrc.puissante.com
Wed Dec 28 08:42:00 EST 2005
On Tuesday 27 December 2005 20:42, Jon maddog Hall wrote:
> dragonhawk at gmail.com said:
> > Sometimes it's like having your entrails ripped out with a rusty
> > pitchfork and ground into the mud by a herd of stampeding llamas with
> > mad cow disease.
>
> Let's see. Let's take about several billions lines of code, written by
> several thousand people (and that is a conservative estimate), most of who
> never communicated with each other in any way, and most of whom never read
> the documentation (if it existed) all the way through, (or even started
> reading it). And they definitely, definitely never read the "erata"
> sheets or "manditory update" information.
... (wonderful description of reality gracefully skipped)...
> Finally, even when the systems don't work, and continuously fail, we do
> not throw them out. If other systems fail, we throw them out. But with
> computers we keep trying to use them, refusing to look at other answers.
>
> And you wonder why things don't work? I am continuously amazed they work
> so well.
Well put. It is my educated opinion that software development is becoming
more and more like biological ecosystems. Use whatever works by way of
Natural Selection, no matter how ugly it is. Elegance be dammed. A model of
co-evolving systems would probably be surprisingly appropriate!!!!! And like
real evolutionary biological systems, once you start down a certain road,
there is no looking back. Once a system, no matter the ugliness, is
actually working and in place, there is little incentive or hope to improve
it. It's used as is, with more kludges piled on top later.
Does that mean that the computer world, taken in as a whole, is a living
system? Um, yep.
And you should pray that this big morass we hackers and software "engineers"
created does not ever become self-aware! ;-)
-Fred
More information about the gnhlug-discuss
mailing list