I'd hoped the full text of the paper would make things clearer, but I hadn't read the paper itself yet because I've got so many browser windows open that I dared not try to fire up Acrobat Reader. Well, I just went ahead and tried the "full text" links at http://xxx.lanl.gov/abs/cond-mat/0303110 (http://xxx.lanl.gov/abs/cond-mat/0303110), which claims "14 pages, 6 figures", but all I get in either the PDF or the PS versions is just another copy of the abstract. Bleah.
Re: That prime-numbers-not-random one...