The main issue is having to dedicate people to militarism, while making the fort actually work. I would have to have 2 training, so that would leave me with 5 to both produce food and to collect silk + process it. Also, if I set the pressure plates to activate even for friendlies, surely they would also trap any liasons... oh, and I am planning to set this up FAR quicker than 3-4 years anyway, hopefully in only about 1 and a bit... (and I think the idea of dropping immigrants into a room and then have them try to kill all of my invaders is cool). And another thing is that giving dwarves full battle gear takes time in it of itself...
It wouldn't take 3-4 years to be self-sufficient, legendary wrestling would be enough to be fairly fearsome, and then they could continue to train after you stop paying attention (switched to a weapon + continuing armor and shield training). I mean, you plan on enemies walking through your barracks, where they'd be sparring, right?
Full battlegear could easily be available sometime year 2, and is unnecessary for wrestling training. (Although you'd want to give them some armor and shield so they can start gaining skill - say Leather Armor and a Wood Shield?)
How fast do you expect the fortress to become self-sufficient anyway? I mean, you'll still be laying out rooms by the end of the 5th year, and you'll need a military larger than 2 dwarves to withstand sieges, even if they are multi-legendary.
And on that clock idea, I would assume that it would rely on the predictable slowness of water or something, so by a certain date, the water would have reached a pressure plate to activate both the magma and the drainage... is that correct?
Ok, so imagine something like the following:
%%-BB-PP-_
Where %% is your screwpump (pumping from the west),
BB is a drawbridge (attached on the west side),
P is a pressure plate, one set to 0-1 water, and the other to 6+ water
- is blank ground,
_ is a channel (there's also one under the eastern bridge tile)
%% is powered (and structure is otherwise encased in walls)
Without P and B down, water is at 7/7 for most of the length and empties naturally. (This avoids multi-triggering of a P with a depth trigger when the bridge opens)
When water is high, P should trigger to raise bridge, causing water to drain from both side and emptying the chamber (bridge blocks new flow)
When water is low, P lowers bridge.
The bridge has a delay.
Thus, a cycle of the system takes well defined time. You hook one of the Ps up to a counter system, adjust the length of the course so you can time a day on integer multiples of an operation cycle, and can thus have a system which knows which day it is during the year. Since caravans always arrive on the 15th of the 1st month of a season, and you can time their transit time, you know when to burninate, and can thus set it up that when its the right day, the system should flood with magma.
This is of course a complicated engineering project that will probably take years to build. I'm still working out the timing, although i have other fortresses i want to 'finish' before i dedicate a lot of time to it.
Yeah well the reason I am not doing it with military dwarves is the fact that I would have to devote a fair amount of time to making the weapons/armour and training them up, and I wouldn't be able to use them for anything else, with my 7 dwarf limit. That was my point. If you read my first post, I described the bridge system which is designed to dispose of enemies, liasons and immigrants alike. (The main reason I am keeping it to the starting 7 is that otherwise I would need to produce food for 200 dwarfs right from the start and there would probably be a high risk of tantrum spirals.
Also, with the noble mandate problem, if I have an engraved bedroom and dining room, along with hopefully decent food, will that negate the effect of unfulfilled mandates? Aside from this, my main worry is strange moods... would I have to wait until they had all had one before I could call my fort fully sustainable?