It is really concerning how terribly this is written. I'd expect such poor wording from the early parts of the edition, but at this point the writers should have already learned to avoid such pitfalls.
And yeah, if it allows you to cast a high level spell that you don't even need to know without using slot, it is obviously hella broken.
Odd. I would say the opposite: with a few exceptions (e.g. 3.5e, because they hardly playtested it at all), it's the earliest stuff which will be closest to the mark, because that's the stuff that actually got extensive playtesting. The standards become looser over time, due to a variety of factors like power creep, changing of the guard, re-evaluating past decisions, etc. I also find that, in general, editorial standards are at their highest with the core books, because those are so important to not screw up, whereas supplements are treated much more laxly because eh, it's
just a supplement, if it goes bad it's not the
whole game affected (even if, realistically, it is.) 4e had the much worse balance of Essentials (and the massive over-emphasis on Wizard subclasses...), 2e had Skills and Powers and the crap-tastic Complete Book of Elves, Pathfinder had the
very poorly-handled Gunslinger with its "misfires become more common as you gain levels due to making more attacks per round" problem, etc.
It's why we've gotten Twilight Cleric and Cartomancer and these other things several years after 5e's launch, not at the start. Why the Storm Sorcerer was initially much more powerful than other Sorcerer subclasses
for good reason, and they neutered it before publication, because they couldn't justify drifting that far from the PHB Sorcerer, even though it really did need the boost.