The problem with the 'EA for testing mechanics' argument is that we can currently only test a tiny set of *low-level* mechanics. Characters play very differently as they level in 5E. I would like to play some higher level classes/content to see how the game systems/mechanics scale. Instead, we haven't (and may not) see *any* of that. That is concerning - given how messed up many system are at levels 1-4. The assumptions seems to be: if we can get it working with levels 1-4 (which have limited complexity) then it will just work for more mechanically complex higher level play. I don't think that follows at all.
I can understand your point, especially since Larian has shown having problems translating some core mechanics like reactions. But, it seems we will be given lvl 5 atleast. Sven wouldn't have hinted at that if it weren't in the plans. We just don't know when.
Overall I feel that Larian has taking on this project in the completely wrong order. When you have a source material so vast and and a rule system evolved through trial and error over 40 years, I feel the first thing they should have done would be to make sure they can translate the core mechanics well to the new media format. Resting, character creation, reactions, spell integration(single vs aoe, melee/ranged touch, saves and projectile pathing, action economy etc). Once that is put in place they can start worrying about adding the fluff that is their thing. It just seems like they try so hard to work around the rules, making us players upset, giving them extra work to fix it later.
