That's the only qualm I have as well. I want to use correct language, it's just that if I ever used the term "Mebibyte", I'm afraid everyone would think I either had a terrible lisp or I was just generally a bit slow. It sounds ridiculous.

We need some better names here.

You're missing the point of the metric system. The entire idea is that I don't -have- to remember that there are 12 inches in a foot, that there are 14 pounds in a stone, that there are 4 quarts in a gallon, and all other such examples. Essentially any measurement that is not to do with time or computers will have a common divisor of 10.

It's less strain on your memory and easier to calculate. Yes, the mathematician in me realises that from an abstract point of view, the two systems are isomorphic. That doesn't mean that I want to go around using units with more complicated ratios where I don't have to.

As any good engineer will tell you, don't make a system any more complex than it needs to be.

Ah, the beauty of having your own computer that no-one else will use.

Of course, I'd say the number one porn protector is to install some variant of Linux on there. That way, the only people who will be able to navigate around it will also be able to use Linux, so you know you already have the firepower of "well you look at it too" in your back pocket.

To the post correcting me on composite, thanks. Stupid mistake.

Perhaps it's the extreme tiredness, but I still see nothing in elliptic curve cryptography which makes use of the fact that pi must not contains any patterns in its decimal digits. Unless you were talking about the stupid mistake re prime/composite that someone kindly corrected for me.

Cryptography has nothing to do with a prime "not being a prime". It's to do with quick factorization of primes.

Besides, I don't see why pi having any sort of repeating pattern would disrupt any theorems. I honestly can't think of any theorem that requires such a thing. Irrational and transcendental yes, but no repeating decimal pattern?

Maybe you can enlighten me to such a theorem.

All the time. Mainly because I'm a terrible coder, but it's ok, I'm a mathematician...

...but this reminds me of this, in a way.

http://bash.org/?117002 [bash.org]

In this case it's probably more akin to -

Given: ZFC
Prove: The continuum hypothesis

Given the legal systems of every country on Earth, I'm pretty sure you can find at least two contradictory laws.

That's because the model they use is pretty much the stock standard SRI (Succeptible-Recovered-Infected) infection model. They even say so themselves (while showing that their model contains an extra non-linear term, included via Mass-Action - another pretty common modelling "law").

Still, a fun application of stability analysis.

You can't seriously hope that the industry as a collective moves to $2.99 games. Sure, there's a market for those games, but don't kid yourself, there's a massive market out there for proper games (that sounds rather conceited, but so be it).

Perhaps it's the Australian in me that recognises I'm basically always going to get screwed on price when it comes to games, but I think there have been plenty of reasonably priced games released. I'll freely admit I've become something of a Steam fanboy, they have some fantastic deals on there. Without sounding too much like a shill, I don't know how anyone can complain about something like the Orange Box for $35 (that's about what it costs in Australia). I've bought more games in the past year than ever before - and paid far less for them than what I ever have in the past.

Slashdot seems to have yet another case of rose tinted glasses. Hell, I remember buying Street Fighter 2 Turbo on my Mega drive back in...whenever it was. That cost me $120. A lot of the other mega drive games of the day were $80+ (SNES owners can leave their sniggering for another day). I'd never pay that for a game these days, and thankfully, I don't have to.