Very interesting! Thank you for sharing this with the group.
The only thing I noticed on a quick first skim is your Proposition 2.3, where you write “It is also remarkable that the lower bound established by Ashlock and Tillotson has never been improved.”
The lower bound
has been improved, I believe, to n! + (n-1)! + (n-2)! + (n-3). Unfortunately the proof of this is hard to cite, since it appears on a wiki page credited to “Anonymous”, in the section “The Lower Bound” of
http://mathsci.wikia.com/wiki/The_Haruhi_Problem. It was added on September 17, 2011 by user “Renaldo Moon”.
The proof is presented in a rather informal style, and it took a bit of work for me to convince myself that the measure N2 can be defined rigorously in a way that lets the proof go through as described. But I believe it can: I’ll post about that separately.
It’s exciting to see more people becoming interested in superpermutations.
Best wishes,
Robin