practical annoyances from multi-level security?

[Simon Spero]

Does anybody have any pet peeves and gripes about using 
systems with a compartmentalised security model- things that took ages to 
do even though they didn't really impact security, features that just got 
in the way or were too complicated to really use, etc?

I'm currently working on a problem that seems to have some analogies to 
the compartmentalised problem, running code with various levels of trust 
were deliberate read-up and accidental write-down could be a breach. 
I've  personally never used a system of this sort, but I've heard from 
friends in the RAF doing COM/SEC (more COM than SEC) that MLS can be a 
real pain at times. [for the truly paranoid, the RAF operates a network 
of communication satellites called Skynet :-)]

Simon
p.s. 
  still no luck on the Hackers soundtrack - I did recognise a few prodigy 
tracks though

(defun modexpt (x y n)  "computes (x^y) mod n"
  (cond ((= y 0) 1) 
	((= y 1) (mod x n))
	((evenp y) (mod (expt (modexpt x (/ y 2) n) 2) n))
	(t (mod (* x (modexpt x (1- y) n)) n))))