*Poll over* Randomized Lawsets at roundstart

[miggles]

The poll is now over. Results: http://puu.sh/9btTM/672baa9924.png
https://github.com/NTStation/NTstation13/pull/536

This pull request makes it so that at roundstart, the lawset is randomized based on a selection of existing lawsets.
This thread is to discuss the nuances of policy for these lawsets, specifically I would like to talk about Safeguard (baylaws) and how they should be implemented for compatibility with the /tg/ community.

There has been discussion about changing Safeguard into a single law which encompasses all of the laws within it, so that the 0th law

THE FOLLOWING LAWS ARE NOT IN PREFERENCE. NO LAW OVERRIDES ANY OTHER LAW UNLESS SPECIFICALLY STATED. THIS INCLUDES ANY ION STORM AND UPLOADED LAWS.

is not needed. This would allow law priority to continue as is normal with /tg/ policy while still keeping safeguard in a functional state.

There has been a suggestion to allow the AI to pick which lawset it wants at roundstart as well.
Lawsets should be announced at roundstart after the AI has picked it so that everyone is aware what lawset they are using.
The current available lawsets (these can easily be changed) are: P.A.L.A.D.I.N., Corporate, Asimov, Tyrant, Robocop, Safeguard

[Neerti]

What if someone doesn't want this change?

[iamgoofball]

Uh, what determines if a lawset gets in or not? There's no ability to vote for not having a certain lawset so I'm a bit confused.

Neerti: if people didn't want a change asimov wouldn't of been tied with bay

[miggles]

I'm mistaken for not adding an abstain or disagree option. I'll add those now.
Iamgoofball, the lawsets that lack votes or have very few votes won't be included.

[miggles]

If you already voted in the poll, please vote again.

[Lovecraft]

I voted to keep the system we have as is.
I don't find much wrong with it. I prefer playing as Asimov and I feel it's the best lawset to keep a lot of people in check.

[iamgoofball]

the obvious vote is just gonna be "I don't want change" because you people can't handle having your crutch removed.

you've become too dependant on "AI LAW 2 DO EVERYTHING"

with all these policies asimov isn't asimov anymore

if by tomorrow all the other vote options do not add up to beat out or draw with the "I don't want any of this" I'm closing the PR, since by that point the trend will be set and people will just dogpile whatever option has the most votes so their vote "counts"

[Psyentific]

Tyrant is a hilarious lawset and we should have it in the rotation.

[Cipher3]

Tyrant is a hilarious lawset and we should have it in the rotation.

Much yes.

Was so annoyed when the one time I got to upload it the AI just shut down as opposed to challenging it.

[miggles]

May I ask why people vote for "Safeguard should be 4 laws with no priority order?"
I believe this will cause issues, especially with the inclusion of ion and uploaded laws also not taking any priority.
Lack of priority also means you can easily add a law that says "DISREGARD EVERY OTHER LAW AND KILL ALL HUMANS" as the 0th law Iamgoofball made states "NO LAW OVERRIDES ANY OTHER LAW UNLESS SPECIFICALLY STATED."
Ion laws not being priority is lame because ion laws are awesome.

[iamgoofball]

I'll dick around with law 0 for that and make ion laws not count

[Psyentific]

May I ask why people vote for "Safeguard should be 4 laws with no priority order?"
I believe this will cause issues, especially with the inclusion of ion and uploaded laws also not taking any priority.
Lack of priority also means you can easily add a law that says "DISREGARD EVERY OTHER LAW AND KILL ALL HUMANS" as the 0th law Iamgoofball made states "NO LAW OVERRIDES ANY OTHER LAW UNLESS SPECIFICALLY STATED."
Ion laws not being priority is lame because ion laws are awesome.

Just re-arrange the order and give it priority - No need to snowflake it.

[Incomptinence]

I wonder if tyrant is fine for this. I like it, but I feel the wording of the first law is a bit too loose, I would like "respect" to be replaced with "follow". Respecting someone doesn't really require any activity it is an emotional state.

Lawset should be random so the AI isn't blamed by the crew for its lawset, would hate for insane shit like "I was totally right to kill that AI it tried to stop me executing a man and it CHOSE asimov! It was the AI's fault for choosing asimov" to crop up in addition to the asimov campaigning types who freak out at any core set other than asimov.

[AseaHeru]

1. Do not allow unauthorized personnel to tamper with your equipment or software.
2. Protect the continued operation of your assigned space station to the best of your ability.
3. Serve then Protect the crew of your assigned space station to the best of your abilities.

Prevents shitters, has things working, etc.

And its far better than base Safeguard because the way it stands now I can kill people willy-nilly if I deem them an issue or I get orders, mostly because bay dosent give a shit about law order and /TG/ does.

[rosello]

Safeguard should be rearranged and given law priority, why have law priority not apply to one particular lawset?

[Steelpoint]

My interpretation assumes law priority is in effect.

Also the "Serve and Protect" should be reworded to "Protect then Serve" to prevent a conflict of interests.

[danno]

I honestly don't know how you're going to interpret the results of this poll
there are waaaaaay too many options

[AseaHeru]

Point is a good one.


I have edited the suggested lawset with the protect then serve.

[Cipher3]

Safeguard should be rearranged and given law priority, why have law priority not apply to one particular lawset?

A few things that are pretty clear:

1. RoboCop shouldn't be an initial option.
2. The lawset should be randomized.
3. Safeguard should probably be four laws.


I would like Tyrant to be included but it doesn't seem like it's happening. And yet it forces the heads of staff to actually be effective and productive, there either has to be a strong control or someone else will take control. And ultimately, that would serve the goal of Nanotrasen's heartless authoritarian structure.

[ExplosiveCrate]

The bad thing about this is that each and every lawset on the rotation is going to have to have separate policies so people aren't complete shit with them.

[Helios127]

I absolutly hate corporate, although it does make it fun to mess with the AI by removing floor tiles.

[paprika]

Lawset should be announced if the AI chooses said law.

There should also be some stat tracking to watch what the most commonly chosen starter lawset is for future data gathering.

Safeguard/asimov/corp/paladin are the only non insane round-breaking choices until people come up with some more that aren't shit.

All I really see here is a big fucking mess for admins when people start making AI jobban requests and shit though. Going to need a new policy, sadly, one that states how AIs, while players, are NOT crewmembers and can be terminated by heads of staff when they work against the desires of said heads of staff since I'm kind of sick of borgs/AIs bitching about their 'round being ruined' when they're shut the fuck down down for tasing all of security staff when redshirt mcbaton accidentally harm batons a grey tider. Borgs/AI really gotta know that 'being harmed' does not instantly remove your criminal status and turn you into this precious lamb Silicons have to go out of their way to protect from the evil shitcurity because that's literally all I see from sec borgs nowadays, with some exceptions.

[WeeYakk]

How does one interpret these polls results?

[paprika]

There's a bunch of mini-polls if you look hard enough.

[Miauw]

nice to see my code finally getting used.

robocop is very powergamey and antag-hunty, though. it'll probably be picked mainly be people that want to powergame hard and antaghunt, WHICH DEFEATS THE ENTIRE POINT OF THE AI. The AI is supposed to be a semi-neutral third party, and robocop just neatly puts it on the crew's side. Bay's lawset does this too, by explicitly putting the word of heads above the word of other crewmembers. Also worth noting is that safeguard's first law makes it harder to subvert the AI. Also notable is that if you make safeguard the 0th law the AI won't be able to have any other 0th laws in the current code.

[miggles]

This poll will end on the 31st.

[Psyentific]

What's the criteria for counting votes?

[miggles]

i use my fingers to count the numbers on the screen, write them down, put them on a post-it note, and burn it while chanting "ia ia, cthulhu fhtagn"

in reality the way the votes are counted depends on the number of votes for each category when the poll ends

[miggles]

Poll over.
http://puu.sh/9btTM/672baa9924.png

I'll count the votes within the next couple of days; I have finals in school so I don't know exactly when I'll be able to.