I played during launch, and the system was honestly imbalanced to the point where I ignored most of the experience gain mechanics for items unless there was a specific item I was targeting.
I'd get upgrades quicker than I would actually find an item interesting, and often times it was faster to just re-farm an item until it procced up in quality, or leave an item in the slot and forget about it until it stopped generating messages. Or leave my PC on overnight while mobs die.
My experience may be slightly twisted by the fact that
the entire item leveling system was broken due to a code bug on launch for several months.
As others pointed out, because you could reliably AFK farm, the interesting parts of the game (ie; grouping with other players) were generally automated in private instances alone. Then, it became a system of farming lower bosses by effectively one shotting them or autoattacking them until they die once per day.
This leads to the game being devoid of social interactions outside of maybe early attempts where no one has powerful gear, or some later times when someone catches you up.
As you've pointed out - buying BIS gear was a problem for how the game was structured and wasn't the only way in which that structure was broken. Being able to group with significantly more powerful characters that could get you the same gear was also a way to skip the entire design.
The only gameplay that was left after this is initial choice in character builds via their pre-defined AAs and skills/spells loadouts, and with trying to get your already-powerful character more powerful via items. The 'current best build' would generally change with the attempt to balance everything.
The best incremental upgrades came from grinding AA, and items went from 0 to 100 on power creep.
I would say that the biggest flaws in balance came from the ability to automate the best increase in power. There were a ton of folks cheating ingame by automating AA XP gains over the server's lifespan.
Forcing yourself to engage with that loop in solo or self-found just meant that you, yourself, alone, did all of the AFK killing / boring automation parts - and no one was allowed to help you. And at the end of that, you could tell the world, "Hey! I did this myself!"
Effectively, the core design was a skinner box with an EverQuest skin.
And they weren't the first server to come up with that concept.
I would say the strongest part of the gameplay loop for THJ was the ability to interact with content alone based on nostalgia. It allowed players to interact in a familiar world and that nostalgia itself pulled what was otherwise 'potentially good design done slightly wrong' ahead.
I would not expect a game with THJ's exact balance to be sustainable long term, because you would eventually run dry of content. It's the same problem EQ Classless ran into, and why future iterations tried to correct that design flaw, but ultimately it boils down to needing a treadmill for the rats to run on.
Also, I don't think a THJ-like game would draw initial attention if it didn't have EverQuest attached to it - or if it didn't have EQ's years of content available without addressing its core balance and content design issues first.
To be clear, this is one of the most fatal flaws of MMOs in general, and isn't just limited to this design, but that's a different discussion. They run out of content quickly and only those with years and years of content are generally interesting for more than a cursory glance. You can see this with Pantheon, for instance - it only had so much content to go around, and to even land on the ability to make content, you had to spend a ton of time in predevelopment.
And the current methodology in which games are made will not allow for that in AAA published games - they want to get their money ASAP, and to flesh out months of playable content while making a game at MMO scale requires rapid design and content iteration. Those tools for content creation are generally reinvented each time a project attempts something like this, or if they are already invented, need to be adapted to the design of the game, which results in the same level of time consumption.
Utilizing AI would ultimately result in a lower quality product, and players will be able to easily clock AI design and narrative if they see it. It would take the right content designers to be able to modify what it outputs and make it reliably interesting.
