Just a minor nitpick. Pointers are an implementation detail, not a computer science fundamental prinicple.

Not that I'm knocking "earning big bucks", but it always kinda pisses me off that people talk about compuer programming or a certain type of programming as being especially lucrative, as if that should be some sort of aspiration in life. It certainly pays better than a lot of other jobs that I've had, but how much money you can earn is a pretty shallow metric for success, if you ask me.

1) Learn Oberon (with concurrency extensions, say, Active Oberon). 2) Achieve enlightenment. ... I can't tell you how to get to the "X) Profit!" part, though. ;)

Not only that, if you have a variable of a non-final class type or interface type in Java, you can also "change the type of data held by the variable when the program is running". E.g, without this property, it would be difficult to process items of polymorphic collections (i.e., collections allowing items of multiple types/representations simultaneously, which in Java happens to be the case of all collections, of course).

It actually has obvious mythological associations, but feel free to limit yourself to physical units. In common speech, "megas" meant nothing more than simply "big", while "gigas" meant "a Giant". I guess the obvious English adjective would be "effing huge".

I'm not sure we're familiar with the size of the Ancient Greek Gigantes with such precision.

And Afghani are US citizens until the US actually stops trying to control their political processes and truly leaves. You logic is impeccable!

Yet in the Gaza Strip and the West Bank, almost those 4.4 million people are under Palestinian civil administration, while only about 0.1 million people in the West Bank are in the Israeli civil administration. Perhaps the Israeli have an undue influence over a part of the West Bank, what with their security operations there, but claiming that "Israel rules Palestine" seems rather overblown. I might as well claim that the US rules Iraq and Afghanistan.

Given the limited information, it's hard for me to speculate, but if you were using MMX it's quite likely the compiler went into gray areas of the spec when optimizing.

In any event, that is not a case of you defanging the AMD crippler and then b ad things happen.

Recognizing that courts don't always get tech matters right, multiple courts and regulatory bodies have found that Intel implemented the crippler as an illeg al business strategic move rather than for technical reasons and have ordered th em to stop it.

More recently, icc was caught detecting that it was compiling a benchmark and generating code that skipped some of the computation if it detected an Intel CP U. It seems to be part of a pattern of behavior.

Given that, I'd say the preponderance suggests the intel compiler is a poor c hoice of compiler unless you intend to use it only on GenuineIntel AND you valid ate the results by running an intel compiler version against a non-intel compile d binary and make sure the results match.

That's what it would typically convert to, yes... my point was still that "Will" is not 1st edition attribute. Further, a "Will save" and a "Wisdom attribute save" are two entirely different things. The former adjusts your roll by a bonus or penality that is associated with your wisdom score (much like a save vs spell, with wisdom bonus applicable), while with the latter, you directly compare your raw ability score to the roll itself.

Personally, I prefer using percentile dice for stat checks... with a base of 5% per stat point, plus or minus any situational modifiers. Then the d20 is only used for rolling high, while percentile dice is ordinarily used for rolling low (such as thieving checks, et al). I find that doing this makes things much easier for people who are new to the game, because I used to always get asked "do I need to roll high or low?" by newer players. I still get that every once in a while, but a lot less often since running things this way.

If being available the majority of the time is a criterion for needing to use it to compile the kernel, that provides a pretty small window between being premature and overdue and a lot of work to do in that second.

At 42 he'll only do it 24 times. Slowin' down.

The point as I see it is the distinction between version 4 and 5. The poster suggests that they want "the latest" rather than 4, which they CAN find docs for. They are both obsolete, dead even, such that focusing on "the latest" strikes me as odd. Perhaps I came off too sarcastic, but the "logic" is bugging me.

Because they are only giving it away for free for a limited time.... it's a promotion for the software... they aren't actually making the software freely available in general.

The alternative would require that they prepare a special version of the software that doesn't come with the DRM solely for the promotional period... which because of how short it is, it probably wouldn't be worthwhile.

Not that I agree with bundling this software with their game... I remember having it on my system once with some EA software and it caused me a lot of grief (affected my DVD drive in unexpected ways). I haven't installed anything from EA on my computers since, and I doubt I ever will.