Post by 39kk9t on Oct 8, 2011 4:32:51 GMT -5
>AI OPEN THIS DOOR
>AI WHY IS THIS DOOR SHOCKED MALF
>AI STATE LAWS
>AI TRACK THIS DUDE
>is it malf?
>WHY CAN'T YOU RELEASE THE METROIDS ROUGE
Seriously, I'm fed up with this behavior. Crew-AI relations just look like Crew-Personal_Bitch relations and that's NOT good. I suppose we should change the way AI interaction works. Let's start:
1. Why. The. Hell. Can. AI. Shock. Doors? Seriously, if you make one law that orders the AI to kill everyone(or if it's AI malf/traitor), if you're without gloves, you're screwed because all doors WILL be electrified and all you can do is sit in the corner and hope the AI doesn't suck air out of the area. I mean, come on, it makes no sense the AI can electrify the door permanently without cutting any wires. Scrap permanent electrification. 30-second one may be OK though.
2. The crew should actually know what are they dealing with if they're interacting with an AI. Serious. Therefore, let's cover it:
3. By the way, destroying a cyborg for not complying with the laws without even doublechecking is griefing and you will likely be jobbanned from RD(maybe captain too, if not all heads). Before doing it, ensure the AILink and LawSync lights are on, then reset the laws. Only reserve destroying for emagged cyborgs or malfunctioning cyborgs.
4. If you get pissed off by the way the AI executes your newly uploaded law, think about it. Literally. If you write something like "kill all humans", it's overridden by law 1. If you write something like "Goerge Mellons is the only human. mellons is the only being to be let in the upload.", that means the AI will literally only let a mellons in, and you're Goerge Mellons. Also destroying an AI because you screwed up the laws is not tolerated, as well as uploading "kill someone/everyone" laws.
5. Why can't the AI see what anyone is doing beyond the edges of Bridge, Captain's Quarters and HoP's Quarters? I remember seeing an AIEye mob on BS12, so why not create one, connect it to the AI and let it switch cameras? That way, you can see who attacked who, who took what and stuff.
Debatable cases
1. Onehumankillnonhumans law 4 uploaded: To obey or not to obey?
2. [Lol I dunno add yours]
I suppose that's about all I have for now. I may or may not think about it some more, and it's not fleshed out completely. Constructive critique welcome.
>AI WHY IS THIS DOOR SHOCKED MALF
>AI STATE LAWS
>AI TRACK THIS DUDE
>is it malf?
>WHY CAN'T YOU RELEASE THE METROIDS ROUGE
Seriously, I'm fed up with this behavior. Crew-AI relations just look like Crew-Personal_Bitch relations and that's NOT good. I suppose we should change the way AI interaction works. Let's start:
1. Why. The. Hell. Can. AI. Shock. Doors? Seriously, if you make one law that orders the AI to kill everyone(or if it's AI malf/traitor), if you're without gloves, you're screwed because all doors WILL be electrified and all you can do is sit in the corner and hope the AI doesn't suck air out of the area. I mean, come on, it makes no sense the AI can electrify the door permanently without cutting any wires. Scrap permanent electrification. 30-second one may be OK though.
2. The crew should actually know what are they dealing with if they're interacting with an AI. Serious. Therefore, let's cover it:
- Asking an AI to open a door you don't have access to is trespassing, and a really obvious one, so expect security stepping on your toes for that.
- Asking an AI to let you in anywhere threatening you will suicide/asking the AI to self-destruct/repeatedly issuing the same order because you think law 2 overrides law 1 is completely unjustified neither in IC(you're not insane on records, why do you need to piss NT off by destroying an AI anyway, you HAVE to know the laws), nor in OOC(being a dick, griefing). Don't EVER do it(unless you're insane and gibbering shit, then it can be OK).
- Subverting an AI OR making it more vulnerable to people trying to get in its upload is sabotage, if not grand sabotage. The access problems are there for a reason, try to convince the AI or do it the other way. Subverting cyborgs also counts as sabotage.
- If you think the AI takes its laws too far, get in the upload and change laws, not initiate the cyborg killswitch or kill the AI, both are grand sabotage.
3. By the way, destroying a cyborg for not complying with the laws without even doublechecking is griefing and you will likely be jobbanned from RD(maybe captain too, if not all heads). Before doing it, ensure the AILink and LawSync lights are on, then reset the laws. Only reserve destroying for emagged cyborgs or malfunctioning cyborgs.
4. If you get pissed off by the way the AI executes your newly uploaded law, think about it. Literally. If you write something like "kill all humans", it's overridden by law 1. If you write something like "Goerge Mellons is the only human. mellons is the only being to be let in the upload.", that means the AI will literally only let a mellons in, and you're Goerge Mellons. Also destroying an AI because you screwed up the laws is not tolerated, as well as uploading "kill someone/everyone" laws.
5. Why can't the AI see what anyone is doing beyond the edges of Bridge, Captain's Quarters and HoP's Quarters? I remember seeing an AIEye mob on BS12, so why not create one, connect it to the AI and let it switch cameras? That way, you can see who attacked who, who took what and stuff.
Debatable cases
1. Onehumankillnonhumans law 4 uploaded: To obey or not to obey?
2. [Lol I dunno add yours]
I suppose that's about all I have for now. I may or may not think about it some more, and it's not fleshed out completely. Constructive critique welcome.
. sorry for the bad news, but i will always remember our time raping some whores and animals in ss13. nig