I don't think I've seen this come up before here, so here goes...
Thanks I suppose to Mr Ermita, it seems that
US Patent #4118611 Aug 30, 1977 by Richard Hunter Harris for the "Buckling spring torsional snap actuator" is pretty familiar around here.
It's immediately obvious (at least to us lot) that it refers to keyboard switches, though there is an
earlier Harris patent from '71, referenced in the '77 patent, which appears to be the initiation of using a spring like this, developing the idea from snap action switches. He also refers to a
'71 Holzer patent with a very interesting system, but I digress.
However, The Harris Patent of '77 refers to "Individual electrical contacts... shown mounted on a circuit board or other similar non-conductive substrate" and furthermore states that the contacts "constitute plates in a capacitive switching system".
This sounds distinctly like Model F technology.
There is a later patent by Edwin T. Coleman, III which references The Harris Patent, for a "
Rocking switch actuator for a low force membrane contact switch", filed in October '83, issued July '85, which describes "a pivoting rocking actuator" with a "membrane contact switch assembly". The membrane assembly "includes an upper layer... of an electrically insulating material, an intermediate layer... of an electrically insulating material, and a lower layer... of an electrically insulating material."
Coleman makes clear that the new "invention is an improvement of the aforesaid Harris patent in that the cost of the present invention is substantially less, approximately half, than the apparatus of the aforesaid Harris patent.". This corresponds to the view presented by DeFosse et al. in "
Development of a membrane switch-type full-travel tactile keyboard" in the IBM Research Journal in September '85, which opens with opens with the words "To reduce cost in the area of keyboards...".
In other words, I'm pretty sure that this Coleman patent is a better reference for the Model M than The Harris Patent.
Note that while Coleman refers to a "low force" switch, the key travel/key force plots are remarkably similar (click for the entries at Google Patents):