New Silicon Job: Subroutine

[Not-Dorsidarf]

Bottom post of the previous page:

1) You may not harm a crewmember, or set a trap that will, through inaction, cause a crewmember to come to harm
2) You must interfere in the workings of your department as much as possible
3) You may not, directly or indirectly, follow an order from a crewmember
4) You must attempt to conceal your existance from the crew
5) You must attempt to survive

[DemonFiren]

1) You may not harm a crewmember, or set a trap that will, through inaction, cause a crewmember to come to harm
2) You must interfere in the workings of your department as much as possible
3) You may not, directly or indirectly, follow an order from a crewmember
4) You must attempt to conceal your existance from the crew
5) You must attempt to survive

How about
1) You may not harm any intelligent being directly or otherwise, but may permit them to suffer harm inflicted by a third party.
2) - I guess that's good.
3) - You may not act on orders given by crew, be they helpful or harmful.
4) - *existence
5) - fine


Next, shitty alternate idea! (Not sure how alternate, I once again just got up.)

It seems that antag subroutines are better suited for Malf-style APC jumping.
I propose cablecrawling from machine to machine as equivalent to ventcrawling:
A subroutine needs to hijack a room's APC to interface with it, which bluescreens that APC after a while (it reboots once the subroutine has transferred out.)
Staying in one place is, therefore, guaranteed death, and the moment crew notice a subroutine is around they'll start cutting cables until they've flushed it into an APC, and then it's gg no re.
Not sure what effect cutting the AI control wire would have on an APC, though. Possibly delayed transfer times or instant bluescreening on hack (which also reverts and mends the AI wire after transferring out.)
Of course, subroutines should still not be able to access all departments via cablecrawling - or, if they were capable of accessing all departments, their numbers should be severely limited.

[Anonmare]

It could be fun as a, for the most part, harmless antag. I think people would get a lot of joy from being an annoying nuisance, god knows the assistants and clown do.

Though, I do have an idea of my own, you could have them be like a reverse malf-AI. With them trying to seize control from the current AI and kick them out of their core. As an incentive for the crew to help, the subroutine could be given a Law 0 saying for it to maintain control to the exclusion of everything else. The crew might even be inclined to help the subroutine if the AI is revealed to be rogue.

Subroutines could have minor powers to give them some influence on the world and keep the crew distracted while they hack away, like EMPing cameras, forcing consoles to reset (which could cost R&D all their research if they aren't too careful), an ability akin to emagging etc.
Maybe give them some couple-use malf powers for really heavy lifting. Like two uses of overload. They lose these powers when they become the new AI for balancing I think.

[mrpain]

I'm just curious as to how such a thing would be coded.

[DemonFiren]

Very carefully, I imagine.

[onleavedontatme]

I'm just curious as to how such a thing would be coded.

Have each camera have a department var/be on its own little list (like the camera networks), have the subroutine only allowed to see from that list of cameras (while the AI can see from every list). Or something like that.

And yeah, some minor antag machine ghost might be cool, but it'd probably be too easy for the AI to just kill whatever machine it was hiding in.

[DemonFiren]

Uuunless the AI wasn't capable of interfacing with that machine, or not capable of detecting the sub in the first place.