|Just another Perl shrine|
Re: Favorite Weaponsby mstone (Deacon)
|on May 31, 2002 at 02:40 UTC||Need Help??|
Erm, let's look at that statement in its original context..
Musashi was a swordsman -- a warrior. His job was to kill other people before they killed him. And like any decent military strategist, he knew he'd have a much better chance of doing that if his opponents couldn't predict his moves well enough to put a sword right where he was about to duck.
(And speaking of ducks, we fossils who grew up with Daffy Duck cartoons will remember Daffy's immortal performance as Robin Hood: "Ho! Haha! Guard! Turn! Parry! Dodge! Spin! Thrust!", and what happened when Porky caught him at "Spin!" the second time around)
But since when is programming a mutually-antagonistic field?
Sure, there are some good, direct corollaries to Musashi's admonition in areas like cryptography: don't use a Linear-Congruential Random Number Generator for hard encryption, because the numbers it generates have a tendency to clump when graphed in multiple dimensions.. in other words, the LCRNG has a 'favorite' set of lattice points from which it likes to choose numbers. That makes it more predictable, and thus weak in an antagonistic setting.
Most software development isn't antagonistic, though. We tend to cooperate with each other, to the benefit of all. The whole point of the open source movement is to lay things out so everyone can see what we're doing, and the strong growth of open source development suggests that the idea has legs.
In a cooperative environment, predictability can be a Good Thing. We call it 'standardization', and it keeps us from having to solve every problem from the ground up every time. Heck, even in cryptography, the experts prefer to use a known algorithm over a new one, because they've seen the old one stand up to more types of attack.
Now, I realize that getting too perky about standards can lead to stagnation, and I'm a big fan of reinventing the wheel just to be sure that you can. I appreciate the value of diversity, and dislike seeing people work themselves into a religious fervor over the glory of a monoculture.
That ain't what Musashi was talking about, though.. he was talking about how to fillet others before they kebab you.
So let me return an aphorism for an aphorism:
Be sure of your objective before planning your strategy.