Originally Posted by dddd

I agree with what you just wrote pertaining to the mounting screw putting pressure on the (stationary) inner barrel.
And I also agree that the spring clutch exerts a fixed amount of drag (variable only via use of alternate lubricants).

The wording used in each of your linked write-ups, suggestive of adjustment of friction via the screw, is however mis-leading, but each in it's own very different way. So a reader is easily mislead. Good detective work (or luck?) coming up with those two.

The proof that the screw is in no way intended to be an adjustment screw is in that it features no anti-rotation washer to prevent any back-and-forth movement of the fixing screw in response to the rotation of an insufficiently-fixed inner barrel. All other friction levers have such anti-rotation washers (or plates) which either have A) flats to engage mating flats on the stud, or B) a cross-piece "leg" keying into a stationary inner plate fixed to the band clamp via a square-shape engagement.

Again, I myself found myself "adjusting" the tension on one of my bikes which came with such levers, but which proved to be very flawed in it's inability to prevent loosening in use. With conspicuously (tellingly) no provided or optional D-ring (or wingnut or knurled knob), a racer might see their race ended the very first time that the screw loosened even slightly from their fine-tuned, adjusted setting!
As well, no friction lever that I've ever seen used such a small-diameter friction surface as could be imagined to exist on the ends of the Retrofriction lever's spring-wrapped inner barrel, nor such steel-on-steel mating surfaces, to control their friction level.

I've used quite a few different front and rear derailers with Simplex Retrofriction levers by now, and have yet to have experienced any problem of slippage with any of them when the fixing screw was properly/fully tightened.
This doesn't mean that they will work quite optimally-as-intended with every single derailer out there, but I have yet to come across even one front or rear derailer where the connected Retrofriction levers had any friction level issue. Kind of like with the actuation ratio of vintage derailers, there was usually very little variation among any of the popular parallelogram derailers (outside the various Huret Allvit/Luxe/Schwinn designs).

Now you can use a partially-tightened fixing screw to reduce a Retrofriction lever's resistance to cable tension, but particularly on the left lever, the larger friction torque acting as the lever moves forward will quickly loosen the fixing screw, while the right lever fixing screw will more likely simply lose tension from wear, which you then have no way of adjusting from the saddle.

Huret, for their part, sought to address the lack of friction adjustment and actually designed a self-adjusting retrofriction-type lever which added return friction only in proportion to the cable tension.
They likely managed to skirt any patent protections on Simplex's design at the same time!

They did this by anchoring the cable head to the end of the spring rather than to the lever itself (each end of the spring getting un-wrapped by applied lever force/motion in one direction or the other). So adding cable tension proportionally increased the spring-wrap friction!

The genius of their design was that while the wrap clutch always sustained the cable tension, pushing the lever forward applied un-wrapping motion to the other end of the spring, making for a lightest-of-light, fingertip lever feel in both directions.
So we can't blame riders for trying to "adjust" their Simplex Retrofriction levers "lighter", they just needed Huret's (albeit less-sexy) version.

This would have also better accommodated the unusual variable actuation of Huret's own Allvit designs (which featured a regressive actuation ratio approaching the largest cog).
Oddly, the Huret version never seemed to make headway into the pro peloton, or on other than a few mid-level production bikes.