No-one is suggesting either of these things needs to be the case, though, are they? Not a single person on this board has ever suggested either AFAIK.
So I think hyperbole like that is unhelpful to the discussion re: balance.
Yeah, when I hear "balance" I think of the ability of the GM to translate a desired campaign narrative into a design idea for a scene, and then... the scene plays out as desired. Balance is when the GM wants to plan an encounter to accomplish something and -- with minimal information about the composition and abilities of the party -- is able to create encounters that all players can participate in and enjoy that also achieves the narrative goals of the wider game.
Critically, the GM should never have to stop and
completely redesign an encounter just because one of the PCs is out sick that week. It should not take a huge amount of system mastery to achieve this, nor should the discrepancy between a PC from a player with low system mastery and a PC from a player with high system mastery really be
that different.
The bottom line is that PCs of the same level should always feel like
peers. 5e is moderately better than 3e was at higher levels, and the "sweet spot" where 5e largely works is levels 3-8 or 4-10. That is as good as or better than 3e or AD&D. They just don't seem interested in extending that.
5E actually isn't too bad re: combat balance. Only 4E did better. At least up to the low teen levels, full casters for example aren't "outright better" than non-casters in actual combat, unless you're doing like, 2-3 encounters/day specifically and they know it.
The issue is that full casters generally do well in all three pillars - some do amazing - whereas the non-casters typically do less well outside combat. At best they might do well in one of the other pillars, but that's far from assured, especially given their lack of "fiat" abilities. 1D&D's rules/class changes, so far, make absolutely zero improvement to this. In fact they arguably make the problem worse by making all casters preparation casters, casting from new, even bigger lists, which increases the strength of casters outside combat.
That's really the entire balance issue with 5E - play a full caster and you get to participate in the whole game. Play other classes and you get to participate in some of the game. It didn't have to be that way - it's purely down to the legacy/sacred cow decision to give casters incredibly broad utility in their spell lists.
Part of it is that nearly all social skills are siloed into CHA. This means if you play a character class that happens to have Cha as the prime requisite -- which is essentially always a full caster or equivalent -- then by
sheer coincidence you also happen to be better at the social tier than any other class while also being just as good at combat.
I really wish someone could explain to me why Persuasion and Deception
need to both be CHA skills, or why they should be separate at all. Why are Dex and Cha so overrepresented? It's really ironic that Str, Int, and Wis aren't the top attributes given how the game started out!
What's particularly sad is the easiest, cheapest "fix" would have been to make all casters spontaneous rather than prep, forcing people to make real decisions about what they could cast, and to chop the spell lists down a bit, but they went the precise opposite way.
There are so many potential designs that could be real improvements on the casting system.
Nope, I think OGL and the OneDND have shattered the illusion of D&D for me. There's so many glaring flaws, and there's no real impetus to address any of them.