Co-DMs are a thing that should probably be done more, especially in large groups. Have one person handling story, the other handling rules, and they split the monsters in fights and such.
When I was running for a 12-player table in a 2-hr time slot, I got a co-DM, sometimes un-planned, just 'hey, run the monsters for me,' sometimes by design, and we'd even split the table and run in parallel if it seemed warranted. (A similar tactic at encounters was to split a table if it got too many players to run officially one of the players would just step up, read that one chapter of the module and run it, then return to playing when there were enough DMs available.) I've also had or been a co-DM at convention games.
The closest I've come to using that technique since 5e was a 4e/5e hybrid session, a treasure-hunting scenario that crossed universes...

It was fun, but...
[quo9te]That's actually kindof unfortunate. I find that the game runs more smoothly and enjoyably for everybody(including the DM) when the players are ultra dedicated and knowledgeable about the game.[/quote]
...for the most part, I don't see the 'everyone knows the rules super well and collaborates to help with some DM responsibilities' or the closely-related 'co-DM' techniques as a great idea in 5e. 5e depends on DM judgement at every turn, and taking the curtain off that judgement by discussing it with a co-DM in front of everyone or building a consensus undermines - trust isn't quite the right word - a sort of mystique or confidence that's of benefit when so much of what's happening all rests on the DM. The more the DM can create an illusion that there's a rule and/or roll and/or pre-written scenario factoring into resolution, the easier it is for players to accept & participate in the story he's telling. 'Seeing the wires' blows it.
In 3.x dedicated & knowledgeable was almost required, and could certainly work, as long as the players are pretty nearly exactly as dedicated & knowledgeable, or those that are higher on the system mastery scale exercise restraint and or tackle the challenge of building to & playing disfavored concepts that heavy optimization & skilled play merely brings up to the par or the rest of the group.
In 5e, so much of the resolution and balance of the system rests on DM rulings that, while it's certainly possible to have a campaign built around dedicated players of high-but-comparable system mastery (Celtavian's campaigns sound exactly like that from what he's recounted), it limits the DM's latitude in delivering the best possible experiences (and we get complaints the game is 'too easy' for instance).
1. 1e and earlier didn't care nearly as much about encounter balance.
And class balanced was already eff'd on purpose, as 'imbalance over levels equal balance.' So, if a party did rest as much as possible at low level, and, consequently, the magic-user used his 1 spell in more encounters and shone more brightly, it was really just bringing him up from virtually worthless to merely under-performing. Likwise, through the 'sweet spot's, approximate balance/greater playability, shorter days might just mean harder combats and casters clearly being more important. Once out of the sweet spot, casters were going to dominate, regardless.
2. Resting in 1e didn't give back nearly as much as it does now. Spell recharging was about it, really - you didn't get back much at all by way of h.p., for example - and so this often meant a party might be resting for more than just one night. An overnight rest to reload the healer, then a day of only curing and resting, then another overnight to reload the healer was and sill is fairly common IME. Which leads to...
Which meant more resting rather than less, and those very early levels with only 1-3 spells make more sense in retrospect...
3. The expected pace of play is different. In 1e it was no big deal if a party went into a dungeon, did what they could, then pulled back and rested for a week before trying again - bursts of in-game action interspersed with long periods of in-game quasi-downtime. But now the expectation has gone more towards more action more of the time, coupled with less downtime - which puts resting rules front and centre along with driving the game towards both giving more resources back on a rest and making resting easier via the short rest.
It's been a good week for me, confirmation bias-wise. First someone else finally remembers the 'caller' concept, now someone else finally remembers old-school 1e pacing the way I always have.