If you actually drilled into Garriott's argument here -- instead of just recording 'bombshells' and briefly trying to establish some room for him to backpedal later -- his argument would dissipate into a lot of hand-waving. Because the particulars are just going to identify exceptions and display a lack of concrete evidence for his view, beyond slinging arrows at designs that 1. he personally doesn't like 2. are beyond any designers' ability to influence (e.g. things handed down, or shortcuts taken due budget/scheduling issues, etc)
So yeah. PR stunt.
He is right that most "game designers" suck, and that the core reason is because they were people who didn't cultivate a real skill, so went through the QA -> level designer -> designer path. A path which doesn't make any sense, and for which there is very little useful training or education available. There is so little analysis of what makes games fun and how to design fun games, and 99% of game designers never see or hear about the little that is available.
Until he actually defines what is and is not good design, he hasn't actually made a point that can even be evaluated.
The bit about the career path is somewhere between ad hominem and red herring. The path is irrelevant to the results. If bad designers are getting through, it'd be due a failure of the filter between level/encounter designer and lead designer. Not a product of whether the level/encounter designer came from QA or development or ops or whatever else.
There's also no way for us to evaluate how much a given 'bad design' was a function of 'no effective training' as opposed to 'time crunch', 'no resources', 'dictate from management', 'tech limitation', or even 'novel theory that just didn't pan out'. And absent that, how could we lay the blame at the feet of career path?
When I say "most people think most things suck, but they don't agree on what sucks" - that applies to industry people too. And nothing about their being industry people means they magically all agree on what, in particular, sucks nor in what the non-sucky alternative would be.
One last time: Garriott's statement is uselessly vague. If we ask people "do most designers suck" we will be lumping together the affirmative answers of those who think (e.g.) Halo's designers suck alongside those who think they do not.
So both groups agreeing to the over-broad phrasing "most designers suck" gives us only the illusion of consensus.
So as long as Garriott avoids defining "what sucks" and "what does not suck", he is not saying anything true or worthwhile.
I'd have thought not, but you seem impervious to it.
Your consensus includes people who disagree on what sucks. How in the world can you pretend their seeming agreement on a broad phrasing remotely matters?
8 out of 10 people might says shooters suck, but if 4 think health packs are the problem and 4 think auto-regen health is the problem and health packs are the solution, they do not agree. The 8 out of 10 number is useless in deciding not only which is the design that sucks and need to change, but in deciding how to change it to not suck.
Saying "most people think most design sucks" is at the level of abstraction of "most people think most human interaction sucks". "getting punched in the face" is a specific, defined human interaction. If you or Garriott wanted to specify an example design at that level of specificity, we could evaluate it.