Good stuff, TeeDee! With the collective thinking of this group we have iteratively converged to a pretty sweet suspension setup, with subtle differences that should be expected when optimality criteria differ slightly.
Just to ramble a bit, if I may, about how my thinking evolved -- I originally got started with toggles based on the double descender ring suspension. I'd visualized a toggle as a short segment of an infinite radius ring, and gravitated to it because of its potential for quick (dis)connect. Even tried using a toggle with a carabiner in place of the second descender ring; while that worked,the strapping was still heavier and more prone to slipping or misalignment than I liked. I then tried the Garda hitch with the AmSteel line but that still left me with alignment and slippage concerns, and Grizz's misgivings about the line getting squeezed gave me some pause. Concomitantly, HC4U's SLS approach looked pretty good to me, but I really wanted a removable ridge line. Then your marlinspike approach caught my eye and something about it intuitively felt rignt. The only other thing I wanted was to be able to swap out 'modules' quickly and easily, as well as free up the ridge line. Finally came the discussion following ZA206's UCR post and harmonic convergence ensued.
As a further aside, one of the projects I'd been involved with in my waning days in uniform was a so-called "Modular Building Block
" program for communications systems. It was a pretty cool approach - almost like a set of Legos for deployable comms. Mac's 'Molly Mac Pack' is another one of those things that feels 'right' to me, and I look forward to seeing how that goes. That mindset has been ingrained in me for a few decades now. I think I even still have the Byte magazine issue on Smalltalk somewhere in my files.
Anyway, all told I'm pretty happy with how this has evolved. If this were math I'd call it an 'elegant solution'.
This has been a fun thing to think about and play with.