Worse is better

One of the most saddening aspects of computers and software is that the best alternative does not always win. In fact the best alternative is usually the one that fails in spectacular ways while mediocre to crap programs become very popular. Here are some examples in case you are not familiar with them.

Amiga

A wonderful, efficiently engineered computer system. Had an OS light years ahead of anything else. Was ruined by owner company Commodore, who reportedly could not have sold ice water in the desert. Begat the term Amiga persecution complex.

OS/2

A great operating system from IBM and Microsoft. Was stable, nice to use and so on. Was destined to run some ATM machines and not much else

BeOS

An OS designed from the ground up to be responsive and efficient in handling multimedia. Had some truly revolutionary concepts, such as a data base in the file system. Never made much of a splash.

X windowing system

Horribly over-engineered mess from the late 70s. Was just efficient and extensible enough that no-one bothered writing a better product from scratch.

I could go on, but I'd rather not. The moral of the story seems to be the following: If you want to be successful in computer business, never ever make the best product. If you are the best, you will bite the dust very soon. Maybe this is due to marketroids who buy software: they like products that have a clear future rather than a solid present. If your program is crappy, your road map is clear: remove bugs X, add features Y. If, on the other hand, your product is the best, the future is less clear. For some reason which avoids us brain-enabled people, the former one is more appealing to some lame-os.

But wait

The above reasoning has been common knowledge for years. But recently I started thinking. If we assume that the conjencture holds, and follow the logical chain to its end, what conclusions do we reach?

Recently Microsoft has been making a lot of noise about their next version of Windows called Longhorn. Windows has never been very good, which may be why it is so popular. But Longhorn seems to have some rather kewl, but not original, stuff like monitor resolution independent graphics WinFS etc. Therefore there can be only two possible outcomes.

1. Longhorn appears on time and has all the killer new features working. In this case Longhorn would (arguably) be the best OS on the market. "The curse of the best player" immediately strikes and MS becomes a niche player in five years or so.
2. Longhorn is delayed even more, misses most of the greatest new features, but offsets that with a ton of new remote root exploits. MS's sales jump through the roof, Apple goes bankrupt and Linux gets torpedoed by software patents.

If I were a gambling man I would, given MS's track record, put my money on choice number two. Sometimes being a realist sucks big time.

Back to front page.