I suspect a major factor in whether or not the short adventuring day issue appears in a particular group is the personalities of the player or players who make the decisions in the group. Impatient decision makers tend to press ahead regardless of resource issues, and possibly accept high casualty rates as a standard part of play, not attaching much feeling to any one PC. Whereas cautious mastermind types like to plot and plan everything, and will exploit features of the system used for their own and hopefully the group's benefit.
I personally think there's a darwinian selection process towards mastermind type players for casters, as they are so vulnerable to sudden death when played in a more gung ho style. It's easy to run casters badly - just choose less effective spells, use them inappropriately or worse don't use them at all. Badly played casters of the "accidentally fireball their own group" type tend to have a very high casualty rate in my experience as the group themselves force the offending PC to retire or just gank them.
Groups with ineffective casters are less likely to see the short adventuring day problem. There's a vast gulf between naively used casters and those who ruthlessly exploit the potential of the most broken spells in most editions, especially at higher levels.
And for many editions of D&D the issue is the system rewards effective use of "nova and rest" tactics as written, as it generally reduces group risk and importantly, makes the casters players feel really powerful and dominant, and the non-casters less relevant. Caster players are incetivised to look for "nova and rest" opportunities as a consequence, and if they are the ones making the group decisions, since it really does reduce group risk and increase goal success probability, it's hard to argue against on a purely in-game basis.
Steps can be taken to make the tactic less effective, and they have been discussed in this thread. There are options that can't be used effectively in some game styles or editions. Mechanics matter, and options in one edition may not work in another.
The most notable examples of short adventuring day in my experience have been in sandbox style games at mid to high level where the players largely determine their own goals and schedules, especially when they avoid personal entanglements.
What won't work is asking players to act contrary to their own interest for no reward "for the good of the game". Some players don't find taking higher risks more fun than being more cautious.
And asking players to press ahead will backfire when there are casualties or a TPK from pressing ahead due to DM pressure when the group tried to rest. Players will do what's successful, what they are taught to do. If short adventuring days work they may stick to that. If taking risks results in lots of failure - casualties and/or failure in achieving goals, they will become risk adverse and work hard to avoid taking them.
