Also, if the 4's were the 'be all, end all', why did the 2.8 trinity get released first? We only have the 2nd lens of the f4 trinity and the 3rd might make it this year if lucky.
Because they show what the market demands, right? Canon makes what will sell, even if that means making sales of what I am arguing are the wrong products for the wrong reasons.
I agree to most of what you say here but I wouldn't call the 24-105/4LIS a "kit lens," implying a lens good enough to let a first-time buyer get at least a taste of photography before they save up for a real lens. Unlike the EF MkI, it's sharp enough for most usage. Unlike the EF MkII, it's small enough for most usage. Unlike the f/2.8's, it has wide enough range for most usage. Unlike non-L glass it's sturdy enough for pro use. Unlike smaller-aperture glass, its 25mm aperture (at 100mm f/4) gets enough bokeh to make your subject pop, when that's what you need. Unlike non-IS lenses, it can work well in very dark environments. (Subject motion, yes, not ideal, but in terms of camera shake and grain, it's gold.)
It's not a jack of all trades, master of none. It's instead a master of a huge core of what most photographers spend much of their time shooting, and what many photographers spend all of their time shooting. It's a perfect example of what I'm talking about. I've had f/2.8 trinity since it was 20-35, 28-70, 70-200 and I've traded up with every new model on EF. I haven't even LOOKED at the f/4 EFs... but likewise I have also not even looked at the f/2.8 RFs.
You know, of the people arguing that I'm wrong and that f/2.8 continues to have a role, not a single one is posting a photo taken at f/2.8 and explaining how it couldn't have been taken at f/4, how the resulting increased DOF or increased noise, or some failure of AF, simply makes the shot not work at f/4+. (In contrast, half the photos shot wide-open with fast primes are inarguably different than what could be achieved with even an f/2.8 zoom. So while the argument for 50/1.2, 35/1.0 and 135/1.4 is still there, I'm happy to double down that there just is no longer any argument for f/2.8 trinity.