OK,
If it takes a $1,000,000,000.00 USD to break Tor and the NSA is only spending
$250,000,000.00 USD a year how does anybody figure that the NSA has actually
broken Tor (at this point)? Maybe I'm just missing something although I
suspect all those systems probably cost $250,000,000.00 USD just to
run/aquire the hardware/power/etc. Was there additional money put out for the
custom IBM chips? Is there actual evidence that these chips exist?
Or is that what they are presuming the new huge NSA data center in Utah is
for? That was estimated at about $1,000,000,000.00 US.
In either case it sounds like the NSA does not yet possess the ability to
break 90% of Tor. It is something that is to be shortly. Which means there is
time (sort of) to fix the problem.
What people are probably missing is that Tor should probably be capable of
keeping secret it's users identifies for about 20 years. Short of this and it
may be worth cracking for nefarious purposes (persecution, etc).