To expand on [MENTION=18736]reckful[/MENTION], people like
Wilde have come up with quantitative approach for generating something like
functions from preference strength (of course this only works for dominant and auxiliary). And "dichotomy" (really a preference pair) really indicates a continuous—not dichotomous—strength of preference, with people tending toward middling prefernences.
So, while this tends to rule out tertiary and inferior functions as meaningless (despite many efforts in support for finding them), it does open the door to other preference pairs (and triads) as being meaningful. It rescues the validity of the Keirsey temperaments (with non-functional pairs like "NF" and "NT") and others preference combinations as well (like "IJ" and "IP").
All that being said, I personally like Thomson... partially because she has good "function" (actually preference pair or triad) descriptions and breaks away from traditional type dynamics. I would rank her significantly less silly than Beebe (who I think is smoking crack beyond the dominant and auxiliary
function preference pairs), since she has some non-traditional-type-dynamics insights to offer and some keen observations.
I wouldn't take her "lasagna" approach too seriously (although lasagna is delicious), but I think for people with middling T/F and/or S/N preference, her model is closer to accurate that the traditional four functions models (far more so than Beebe's).
