A somewhat radical solution to the Traveller computer problem is to treat
all reasonably high tech level computers as AIs (artificial intelligences)
and use them as NPC's.
This has a number of advantages:
1) It is a (perhaps the only) reasonable extrapolation of current computer
science.
2) It often makes for a smoother and more natural interaction between the
Referee and players. The Ref can simply speak as the AI when giving
players info from data bases or ship's instruments rather than describing
what the players are seeing on a display in the usual narrative style.
3) Interactions with AIs often provide an additional "futuristic" element
or even challenge in the game. AIs should, of course, have interesting
and somewhat quirky (by human standards) "personalities" in some
plausable way (Think of HAL in 2001 but try to be original.)
There are some problems also, naturally:
1) If you are not careful the AIs will be so powerful as to overwhelm the
players or leave them with nothing to do except issue orders to the AI.
The best solution seems to be to handicap the machines in some ways; I
used (when I was a Ref) Asimov's famous 3 laws plus the stipulation that
AIs had to be shut down while preparing for, during, and following jump
due to an unfortunate interaction between the jump field and the
unspecified hardware technology of AIs. They can also be made relatively
fragile during combat. I found that these conditions left plenty of
need for the usual player skills.
2) The "computer" skill must be modified in some way. I usually let the
player choose between hardware skill (ability to fix AIs, can be very
important) and software skill (not programming which is all done by the
AIs themselves but gives favorable DMs in dealing with AIs, sort of a
machine equivalent of carisma).
Ed Turner