Way OT: Adam Smith and Software Markets

["Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net>]

>   Unfortunately, as this points out in yet another way, the classic Adam
> Smith free market works very poorly for software.

There is a simple experiment that can show that the market for software, for
the most part, works fairly well. Do you have an old PC in your attic? You
know, an original IBM PC or compatible (floppy drive or cassette based
storage), or an XP running DOS 2.x with a 20 Megabyte hard drive? (Note: I
said Meg, not Gig!) If so, some weekend when you have the time, take it down
from the attic and set it up again. Recall the days before USB when all
hardware had to be installed in a motherboard slot, where it was up to you
to resolve IRQ conflicts by fiddling with DIP switches and jumpers, and it
was up to you to edit CONFIG.SYS to install a driver. Try out you old word
processor, the one that only worked with fixed pitch fonts, and could only
render a smooth right edge by periodically inserting extra blanks into your
text. And don't forget your first spreadsheet program, the one that is also
unaware of alternate fonts, not to mention graphics, embedded objects, or
connectivity to just about any other application. Remember when user
interface standards were no-existent, sort of like Moore's famous quote,
"Standards are wonderful, everyone should have one"?

Oddly enough, a lot of people who go through this exercise have found it
pleasing and nostalgic. They fondly remember the various clever tricks they
devised to get around the software limitations of that era. But let's face
it, the only reason anyone has any fun with this stuff is the fact that
these hassles are in the past. After this little experiment, all but the
most diehard archivist would return the old PC to the attic.

In my time, I have constantly seen market pressure fuel many hardware and
software improvements. The companies that failed to make improvements in
their products have often suffered tremendous falls. (Remember WordStar, the
MicroPro word processor that once held over half the word processing
market?) I am not saying that we in the software industry could not do
better, and I wholehearted agree that software development as it is done
today needs much better quality assurance techniques. But one cannot deny
how far we have come. The proof of that is just one attic away.