>  Why make that assumption?

Well, this is a slippery slope.  We're already assuming the spy has a lot of 
information, and the ability to get more, and that I'm
generating random numbers frequently enough to make it worth his while to 
continue spying (he's already been spying for a couple
months now, in order to get his 100 samples of 1000 numbers each).  We also 
have to assume he knows how I'm using these numbers (or,
more specifically, when I'm using a number in an application that interests 
him, e.g. cryptography as opposed to playing with a J
verb).

If we assume he also knows that I'm using GB_FLIP, then why not assume he also 
has access to my initial seed and be done with it?

>  Let us first tackle the easy case where we know
>  that it was generated by GB_FLIP.

But sure, in the interest of learning how this stuff works, let's assume he can 
sample as you described earlier, and also knows I'm
using the GB_FLIP algorithm.  How would he go about predicting future generated 
numbers?

One easy way I can come up with is dispense with the sampling, and just wait 
until I've generated <:2^55 numbers (which he knows is
GB_FLIP's period).  Then he'd be able to predict the entire future stream.  Of 
course, he'd have to wait many months or years before
that happened, depending on my appetite for random numbers.

-Dan



----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to