Maybe my claim of "majority" is a bit of an exaggeration, but clearly, to a huge chunk of the player base, the only version of the Ranger they have experienced is the Half-Casting Ranger in 5e.
That doesn't mean they think it's right.
That's a key thing. You've played RPGs. Did you immediately think "THAT'S CORRECT!" about every class you read in every RPG? I never did. Jeez it took me YEARS to come to terms with how Clerics worked in D&D, and I'd never played an RPG before that, and only been playing videogames for two years (mostly pretty simple ones).
And I've introduced people to RPGs with 5E (not as many as 4E but w/e), and they are often mildly vexed by the fact that Rangers use spells, because they don't expect it, and it doesn't make sense to them. They have a pre-existing, pre-D&D notion of what a Ranger is. D&D "getting wrong" doesn't change that.
I agree that the ranger is a poorly defined mess. However, I don't think that spellcasting is the reason for that.
It demonstrably is the reason for that. It's the reason Rangers don't really have abilities or any kind of consistent identity, and it doesn't even match their own lore, which barely mentions magic.
I don't think that the majority of players would want a non-Half-Casting Ranger.
Uh-huh, and your reasoning for this is deeply flawed as I've illustrated.
I think the issue is that some players do like Rangers with magic - but even of them, the 5E Ranger is sometimes seen as excessively magical/magic-heavy. And a lot of what the Ranger can/can't do is because it has magic.
I think if we had a Ranger that only had fairly subtle magic, and wasn't doing stuff like being utterly reliant on magic for combat prowess (as the 1D&D Ranger 100% is), then there'd be less of an issue.
Also, according to all surveys WotC's done, the Ranger needs a pet. It's a huge part of what people expect from a Ranger. And yet it doesn't have one.