Slashdot Mirror
Underfunded NSA Suffers Brownouts
An anonymous reader writes "Almost ten years after the an internal report, and a year after a Baltimore sun story warned that the electrical system at the fort Meade NSA HQ couldn't keep up with the growing electricity demand ... the problem has got worse. The 'NSA has had to resort to partial, rolling brownouts at its computer farms and scheduled power outages and some offices are experiencing significant power disruptions'. NSA director Alexander testified to congress about this problem. It is suggested he wanted to add more than $800 million to the 07 budget. A recent public powerpoint presentation suggested 70% of of all intelligence spending goes to contractors. It also included a graph, without numbers, of this spending. It suggests that US intelligence spending is around $60 billion. An internal survey that showed NSA employees have problems trusting each other."
I wonder how much of these problems are really due to lack of funding and how much are just tactics to yank an even bigger chunk of money from the guys in Washington. After all, the problems that they describe should only exist if the person in charge purposedly screwed up the budget.
Slashdot, fix your code or at least hire someone who is competent at it to do it for you.
This doesn't sound like underfunding at all. It sounds like highly misappropriated funds going to prioritized sub-groups with an inherent motivation to see the other subgroups suffering and failing for the sake of their own relative gain. This is completely in keeping with the current administration's modus operandi of finding subgroups in organizations (lobbyists, regulators, etc.), that will play ball, and finding a way to eliminate or functionally undercut everyone else, then blame those who were undercut for the resulting general failure.
Ryan Fenton
As a graduate mathematician, I feel ashamed to see so many of my fellow students either going off to the City or to GCHQ (the UK's NSA); while it's true that the cryptanalysis work done by the latter is one of the few non-academic jobs requiring considerable "pure" mathematical skill, that's really not what the huge amount of money spent on infrastructure is for.
Because the War on Terror/Evil Of The Day really isn't about challenging mathematical genius terrorists to ever more complicated ciphers - yes, GCHQ created RSA a few years before R, S and A, but occasionally beating open academia and having a lot more horsepower isn't ever going to put you beyond the mathematical principles you're faced with (*). Massive horsepower is for statistical analysis of insanely large quantities of data. This might occasionally find you something saucy, but it's mostly going to allow you to profile, and profiling reduces risk - past trends are a useful indicator of future performance, whether you're analysing a financial market or the behaviour of groups of humans.
None of this will help if some random guy decides, tomorrow, to commit some nefarious deed involving an IED - something I'd say 90% of graduate scientists either have the knowledge to do, or could read up on overnight. Which goes to show that the reason everyone's not blowing everyone else up is not because there are any technological measures in place to stop them, but because by and large, for whatever reason, people don't want to.
(Oh, and the NSA/GCHQ do have some obvious legitimate uses - such as decrypting messages between known ne'er-do-wells. If that's all they did, I might even like them.)
Oh, and before people forget, the problem of whether the NSA is allowed to spy on Americans is easily solved in principle by GCHQ and NSA doing the dirty work for their friends across the pond; in practice, an extra-judicial agency couldn't care less anyway: he who is not accused (for there is no-one allowed to witness the crime), is not judged.
(*) This is why I love my discipline. Men can only discover mathematics, never beat it!