Ooh! Contact those Conficker guys to add that to their payload!

Moblin? What's their slogan, "It's a secret to everybody"?

A brief reading of a wikipedia link about the neocortical column provided from another poster above seems to suggest that a similar topology is present in the brain. From the section on columnar functional organization:

So if their goal were to only simulate the neocortex, it seems they would really only need dense inter-chip connections between those chips that would make up their version of a cortical column and sparser connections to its neighbors, rather than needing to connect each simulated neuron in each chip to every other. (IANA neurobiologist)

If you change it to 'any cheap way', it makes more sense. Still not completely true, but at least it's closer to reality...

Your first point is valid enough, and your second one does hold some water, but it seems that it has already been addressed by other books on programming in PHP. Further, if I fully understand what your second point is talking about, it would seem that solving it would require getting rid of some of the flexibility (and hence, utility) of the language.

Notably, points 3.2 and 3.3 in your post have already been addressed (not sure about 3.1 or 3.4... never ran into anyone who did 3.1 since it seems incredibly foolish, and haven't heard of any changes yet for 3.4).

3.2: 1 == '1more' is true for two reasons: you're using the weak equivalence operator (hence why a string and an int can be equivalent) and your string starts with valid numeric data which can be converted properly while the rest is ignored. See here for more examples of what would and wouldn't be converted.

3.3: Using the strict equivalency operators ('===' and '!==') in your conditional statements solves this. According to the link, this has been available since PHP 4 so I'm not sure why you're still going on about it...

Your fourth point doesn't necessarily make PHP any less secure than any other programming language that can use external libraries, unless the ones PHP uses are particularly poorly written. While it is true that most of the issues with buffer overflows and memory corruptions in C can be more easily avoided with other languages (e.g., Java), it doesn't mean they can't also be avoided in C with careful coding.

If those aren't already options in the install GUI, they should be. Having a package manager for which you can set preferences to look for only free driver updates, etc. or automatically install whatever drivers it thinks you need (regardless of whether it's proprietary or not) would be nice, and it might help to keep things usable for newbies while still being flexible enough for more advanced users. Granted, advanced users would just do a command line install, but whatever.

As a side note, I see the disparities mentioned in TFS as a good thing, as it indicates more people are switching to Linux.

Disclaimer: I don't usually install Linux boxen, I just SSH into them.

Commander Keen to the rescue!

Why not just pick up some plutonium from the corner drugstore?

If my calculations are correct, you're gonna see some serious shit.

I believe it was picked to help foster in laymen the vague notion that we're just making shit up to mess with them. In order to solve this issue, I propose we now start calling them kibibikibitibobitiboobytes.

It's that image burned into our retinas when the power to our monitors dies, right? Right?

For a second I read that as 'Hydrogen can be metal' and thought of this...

Haven't we already seen enough of the consequences of cross-contamination in the previous Phobos Grunt missions?

While I agree the reason we don't reprocess is probably political, it would seem that addition of reagents virtually eliminates the possibility of developing weapons-grade plutonium from reprocessed nuclear material.

There's only one way to be certain: repeat the experiment.

I would suggest having the blind man and an average-sighted man separately walk down a very well-lit corridor with randomly positioned (i.e., changing every iteration) obstactles 10 times, recording the amount of time each takes and the number of collisions. Reduce the amount of lighting by some increment and repeat the experiment. Continue reducing lighting until total darkness is achieved.

If the blind man is truly navigating by blindsight, both his course times and collision rates should roughly scale positively with those of the sighted man and inversely to the light levels. But then, that would be using the scientific method like the international team of neuroscientists in TFA (whom you are accusing of incompetence) did, so of course you wouldn't believe it...

Blindsight is 20/20.