Former facebook acquiree here.
The metaverse is what happens when you let your leadership/product team convince you that the key to speed up what you want to deliver is to throw people at the problem, and not put any constraints on deliverables.
The original plan for oculus is to establish a VR eco system that would have transitioned into AR glasses, allowing facebook to have a platform of its own.
VR was/is a bit niche, because it required lots of expensive hardware, and there were limited games/uses.
first logical step: remove the need for a high end PC, make the thing cheap.
That drops one barrier to adoption: expense.
The next one is, great I have this $400 device that does VR, but what can I actually _do_ on it? That means you need content and features. This is where it all turned to shit. Zuck looked at steam, and itunes and said: "make it so", and they started tapping up devs to make small games, and AAA to make big ones.
But, its expensive to port games, and it takes time, why not buy studios that are making great games and get them to make more? so they bought a bunch of indie studios. Those studios had to fight to keep their devs, because facebook normally fires/rehires, forcing everyone to re-interview for their job. Games devs aren't really hired because they don't pass the technicals (Don't know why, given that games devs need to be good or the FPS drops like shit.)
with all that upheaval, those games studios don't really produce extra games to sell.
All the while a small team had been making a roblox clone. It was slow and a bit buggy, and you could make shitty games. During lockdown we all had a play. Needed a new generation of hardware to work properly, because it was a unity game with a bunch of hacks to allow custom maps and rules.
Never mind, we are doing E N T E R P R I S E now. enter work rooms. Again a small initiative, which basically asked, can we make better VC if we are in VR? The answer is yes, yes you can, but selling it is hard. There were a lot of hard problems to solve, like needed to detect keyboards, how do you present your screen if you can see your computer? how can you do computer passthrough or virtual monitors in VR?
Zuck saw this and jizzed his pants, so made it a priority. This meant the small team (probably less than 40) swelled to like 4000. Most of the people who moved were not games devs, or had ever worked in graphics/3d. This meant that loads of silly lessons had to be learnt in prod. Nothing was stable, everything was high friction, and no, there was no public API to allow you third parties to integrated into the app.
For the longest time it took >5 minutes to join a VR meeting.
Basically Zuck loves features, and cant understand that user experience is way way more important than features. He throws engineers at the problem which means that instead of solving product issues, they endup solving people issues.