Well, I could easily question the overseer's judgement in locking the lever-pulling dwarves out of the lever room, but that's as it may be. Either way, it seems fairly trivial to make 1 or 2 spare keys to certain fortress-critical doors, and either entrust them to super-secure dwarves, or guard them with some super-secure militia. If that's your thing.
I would question the judgement of an overseer who doesn't.
If you expect to lock a dwarf in a lever room, then you basically ensure they will become unhappy as they fail to attend to their social needs, which will result in tantrums, which will result in wild lever-pulling. What could possibly go wrong?!
If you expect to leave a lever room unlocked, you open up the possibility of random tantruming dwarves running in and performing wild lever-pulling. What could possibly go wrong?!
If you expect, for that matter to have just
one dwarf who is assigned the duty of lever pulling, you're putting the fate of your fortress in the hands of a single dwarf that at any moment could decide it's time for a nap or a prayer. What could possibly go wrong?!
The only sane way to manage your levers is to have multiple dwarves ready for lever-pulling duties, and draft a few that are in handy locations and ready for jobs when the need strikes, usually locking them
in as well as locking everything else
out at the same time. I don't want my dwarves to have keys to
leave the lever room until I say the coast is clear, either. (Although, granted, I tend to accomplish that by drawbridging the lever room shut as an absolute line of defense.)
That requires letting some as-yet-undetermined dwarf from virtually anyone in the pool of dwarves who are likely to be near my fortress core have keys to get in, but not out until I tell them so... which is really a problem if dwarves have keys and the capacity to use them when not explicitly instructed to do so by me.
Oh HELL no. A lock that any dwarf can open is hardly better than no lock at all. IMO, every dwarf would carry the key to his/her own bedroom (possibly workshop/stockpile, too), and every militia captain would have a key to the Main Gate. Because if all your captains are dead at the same time, your fort is dead anyway.
This makes a ton of massive assumptions.
I don't know how you play, but I have emergency isolation drawbridges in key parts of my fortress, with emergency food and drink reserves as well as picks and axes to restart my fortress from scratch. My whole army can die, and I'm still in the game.
Also, I don't care about locking bedrooms, and neither do dwarves. That's trying to create an overcomplicated solution to a problem that doesn't exist.
True. I don't see personal possessions being implemented until dwarves get pockets.
Which is a potentially massive problem if you have hundreds of doors, including multi-layered doors, all of which need to be locked. Right now, especially, you need doors at least two layers thick just to stop people from dodging through locked doors.
If you set up doored hallways as speedbumps for FBs, then you need a few dozen keys just to get a patrol through their commute to work in the morning. A single dwarf could need hundreds of keys to get through their standard paths in my fortress.
Don't expect to be able to implement security on a setup that was specifically designed to prevent basic privacy.
You nonetheless want to break a basic functionality of doors as they currently exist without even bothering to consider the existing functionality valid, much less even attempt to propose a solution to maintain it.
Realism is not a benefit? Granted, as an overseer I like the ability to instantly lock/unlock any door in the fort, but that doesn't mean I feel it's an ability that I should have. How about . . . locked doors would slow down thieves, while barred doors would slow down building destroyers. Smart vampires would lock victims in their own bedrooms, greatly delaying their discovery. Locks would be pretty much the only way to deter petty theft.
OK, then, let's talk game-breaking consequences of what you're suggesting.
You know how
assigning doors to be pet-impermeable causes pets to just ram into those doors, continuously pathfinding into them? That's because an unlocked door (whether set for pets to pass or not) is counted in the connectivity map for pathfinding as an open space. Locked doors are considered impassible space. There isn't a separate connectivity map for pets, because it takes far too long to generate multiple sets of connectivity maps, which are honestly slow enough to generate that fluids breaking them is already a major FPS killer.
What happens when you try to set up a map where every dwarf has keys to different doors, and the capacity to go out looking for keys valid for them to find? A godawful clusterfuck of either pathfinding failures or the demand for
every single dwarf to have their own
individual connectivity map, which would literally cause lag on the order of several seconds so the game can regenerate connectivity maps every single time a new door is locked, a drawbridge is raised or lowered, or water crosses the transition from being 4/7 to 3/7 or back again, forcing a redraw.
So yes, there's a minor benefit of realism, but I still can't see how that could possibly outweigh the utterly massive amount of coding work it would take to try to mitigate the performance loss or fix the inevitable AI failures to use keys properly, or the insane amounts of micromangement it takes to manage tens of thousands of individual keys assigned to a hundred dwarves.