Re: DIS: (Proto) Raybots
ais523 via agora-discussion [2023-05-19 04:51]: > Here's an idea I had as a way to a) shake things up in a way that's > likely to lead to lots of interesting CFJs for the next few months (I > came up with it after reading the CFJ archives for cases that looked > interesting), and b) let us experiment with mechanisms for awarding > Radiance that don't need a whole proposal cycle to go through. I like this A LOT. It seems the right way to implement what I was going for with Golems. Actually, I would like to suggest the following: * Rebrand Raybots into Golems (aesthetic personal preference) * Make Golems into another kind of entity. The Agoran consent mechanic seems a nice way to throttle golem creation, and the promise-granting is a terrific way to make golems act-on-behalfable. As to the “other kind of entity”, I think it'd be better to define these golems as just something else other than people. We just grant them precisely the habilities we need them to have (acting on behalf, for example). But I'll elaborate. I notice a quirk in this proposal. R2152 “Mother, May I?” defines POSSIBLE as such: > CAN, POSSIBLE: Attempts to perform the described action are successful. R478 “Fora” read: > Where the rules define an action that a person CAN perform "by > announcement", that person performs that action by, in a single public > message, specifying the action and setting forth intent to perform that > action by sending that message, doing both clearly and unambiguously. And R2466 “Acting on Behalf” (needed for cashing in promises) reads: > When a rule allows one person (the agent) to act on behalf of another > (the principal) to perform an action, that agent CAN perform the action > if it is POSSIBLE for the principal to do so, taking into account any > prerequisites for the action. So, a person cashing-in a raybot's promise CAN do so if it is POSSIBLE for the raybot to do so. R478 possibly overwrites R2152's definition of CAN, which gives two possibilities: 1. It does overwrite. Then, that a person CAN perform something by annoucement means that they perform it by sending a message, etc, etc. But then, I'd argue it is not POSSIBLE for a Raybot to do it, as it can't (if it even exists) send a message. 2. It doesn't overwrite. Then, it is just explaining what “by announcement” means when under a CAN. In that case, it *is* POSSIBLE for a Raybot to do that action because the rules specify, by using CAN, that sending a message to perform certain actions *is* successful, when done. And Raybots can NEVER send messages. So “attemps to perform” such action must be successful: they don't exist, so can't possibly fail. This might be interesting. But anyway, I think better to just define Raybots/Golems as a new kind of entity and just duplicate the few capabilities they need. What do you think? -- juan
Re: DIS: (Proto) Raybots
On Sun, 2023-05-21 at 13:13 -0700, Kerim Aydin via agora-discussion wrote: > On Sun, May 21, 2023 at 1:01 PM ais523 via agora-discussion > wrote: > > We've had artificial persons in the past, and they ceased to exist with > > no real issues. That predated Promises, which probably need a fix to > > cease to exist when their creator does. > > When you say "no real issues", IIRC we had to think really carefully > about things like rights, and repeal parts of R101 guaranteeing rights > to persons before we "destroyed" artificial persons, then put it back. > Someone might make a slight case that "Persons" are the core of the > game in the current R101 (though that's really a stretch) and > interestingly, reading the rule on Banning, we could probably Ban > artificial persons - a "runaway bot" should something like that ever > exist seems to meet some of the banning criteria of "the person's > actions have been deemed harmful to Agora". > > While I haven't been following the Raybots proposal, I think the main > issues were learned from zombies - there's lots of things (like having > your bot support your own tabled action announcement) that need to be > strictly off-limits for bots in general - the zombie nerf list from a > couple years back is worth a cross reference if that hasn't been done > already. Well, Raybots are safer than zombies in that everything they can do is known in advance while the Agoran consent action to create them is pending – it would probably even be safe to allow them to have full voting strength, because people should in theory object to abusive Raybots being created in the first place, but Agorans have been known to vote for bad ideas on occasion so it's probably worth the safeguard of preventing them voting. I agree that it's probably possible to ban a runaway Raybot, something that seems useful rather than broken (although just exiling it would have the same effect). It's worth noting that although zombies got abused a lot, that sort of abuse seems to have been mostly unique to "artificial person with a single owner" systems. Agora has in the past tried things as simple as the Partnerships mechanic: the more recent version of it was "any group of two or more non-artificial persons can create an artificial person, if they come to some agreement between themselves as to how to control it, and can only create one at a time", and that (amazingly) wasn't abused to any major extent despite there being few limits on what they could do. There definitely *were* abuses in the partnerships era (e.g. the AFO was used for several scams, and P1-P100 – immortalized in the Writ of Fage report – were definitely part of a scam, although one that IIRC didn't work as intended). But that was a minor part of it, and there were partnerships that became a major part of the game for a substantial length of time without any major drama stemming from the fact they were persons. The abuses also tended to generally be interesting rather than boring (e.g. partnerships designed to create interesting scenarios for CFJs but that ultimately didn't do anything, or that time I achieved a "blot everyone" victory condition by using the equivalent of today's PM's Dive power on a partnership to blot numerous players at the same time). The other thing about zombies is that they weren't artificial players, but rather former players being controlled by the rules – many of the zombie protections were to protect *the zombie*, rather than to protect the game from its owner. That's the primary reason why it was dangerous to let them self-blot, deregister, or agree to contracts. I think the only zombie protections that were designed to prevent them being overpowered were the protection against table/support/oppose (which was in my proto), and the protection against recursive zombie use (which probably isn't necessary here because Raybots don't have owners). -- ais523
Re: DIS: (Proto) Raybots
On Sun, May 21, 2023 at 1:01 PM ais523 via agora-discussion wrote: > > On Sat, 2023-05-20 at 23:43 -0400, Janet Cobb via agora-discussion wrote: > > Actually, in general persons ceasing to exist is likely to cause > > problems, and the current ruleset is careful to avoid it (R869/51's "is > > or ever was"; you remain an Agoran person after you die). > > > > I'm not sure there's a good solution here. Having disabled Raybots just > > sit around doing nothing isn't ideal. Auditing the whole ruleset for > > issues caused by this is probably good to do anyway but error-prone (and > > future proposals are reasonably likely to introduce new problems). > > We've had artificial persons in the past, and they ceased to exist with > no real issues. That predated Promises, which probably need a fix to > cease to exist when their creator does. When you say "no real issues", IIRC we had to think really carefully about things like rights, and repeal parts of R101 guaranteeing rights to persons before we "destroyed" artificial persons, then put it back. Someone might make a slight case that "Persons" are the core of the game in the current R101 (though that's really a stretch) and interestingly, reading the rule on Banning, we could probably Ban artificial persons - a "runaway bot" should something like that ever exist seems to meet some of the banning criteria of "the person's actions have been deemed harmful to Agora". While I haven't been following the Raybots proposal, I think the main issues were learned from zombies - there's lots of things (like having your bot support your own tabled action announcement) that need to be strictly off-limits for bots in general - the zombie nerf list from a couple years back is worth a cross reference if that hasn't been done already. -G.
Re: DIS: (Proto) Raybots
On Sat, 2023-05-20 at 23:43 -0400, Janet Cobb via agora-discussion wrote: > Actually, in general persons ceasing to exist is likely to cause > problems, and the current ruleset is careful to avoid it (R869/51's "is > or ever was"; you remain an Agoran person after you die). > > I'm not sure there's a good solution here. Having disabled Raybots just > sit around doing nothing isn't ideal. Auditing the whole ruleset for > issues caused by this is probably good to do anyway but error-prone (and > future proposals are reasonably likely to introduce new problems). We've had artificial persons in the past, and they ceased to exist with no real issues. That predated Promises, which probably need a fix to cease to exist when their creator does. The main potential issue I could think of is "what happens to a CFJ if its judge ceases to exist", but it turns out that there's a specific allowance for that in rule 991 (the nonexistent person remains assigned as the judge). Likewise, rule 649 allows non-persons to bear patent titles (oddly, it even allows non-persons to be *awarded* patent titles). According to the FLR annotations, we were fixing bugs with loss-of-personhood as recently as 2020, so it's historically been considered desirable to have rules that make sense in that context. I checked every use of "person" in the rules to find uses that might cause issues: * Rule 1742 - what happens to a contract if a party ceases to exist? * Rule 2659 - stamps - already addressed in my proto * Rule 2644 - lockout on Stone win condition ends early if the winner ceases to exit - probably not going to matter in practice * Rule 2464 - tournaments have no Gamemaster if their creator is no longer a person, but work just fine in that state * Rule 869 - playerhood - already addressed in my proto * Rule 1023 - definition of "round" - may need fixing, although the definition is used only to fix the First Speaker rule, which wouldn't be affected * Rule 1728 - if an officer tabls an intent as an official action of their office, then ceases to be a person, the new officer can't then resolve the intent if it's an action without objection: potentially buggy, but unlikely to be a major issue in practice * Rule 2530 - potentially weird if a proposal's coauthor ceases to be a person, we might want to reinforce that (although I don't think anything is actually breakable there) * Rule 2493 - regulations - the definitions here break if the promulgator of a regulation ceases to be a person, although I don't think that causes any actual rules to break as a consequence * Rule 2127 - conditional votes - an attempt to endorse a voter breaks if the voter ceases to be a person (even though the non-person's vote is still valid) * Rule 2210 - self-ratification - not broken, only persons can CoE but the CoE remains valid even if the CoEer ceases to be a person * Rule 2478 - investigation of infractions - potentially broken, could most simply be fixed by allowing Favoritism towards non-players * Rule 991 - CFJs - judgehood works fine, but recusal is broken Probably the best approach here is to ensure that loss-of-personhood works: it's something that used to happen at Agora all the time (there have been many eras where a couple of conspiring players could create and destroy an legal-fiction person pretty much at will, and the rules used to use the terminology "first-class person" and "second-class person" so that the legal fnctions could be easily identified). I suspect that regardless of how Raybots goes, it's worth a big fix proposal to make sure that loss of personhood is something that the rule can handle. This does make me think that something like the Raybots proposal is worthwhile, though: the best way to ensure that the ruleset can handle loss of personhood is to make sure it's something that regularly gets tested, thus incentivising us to fix the bugs in it. (The proposal's inspiration came from the direction of "legal-fiction persons would be interesting to have again and we haven't had them for a while, can we find a way of doing them that's significantly different from what we've had before?".) -- ais523
Re: DIS: (Proto) Raybots
Maybe we can keep "persons" as is, and make a new definition that pretty
much encompasses everything persons can do? For example, adding something
like "If a person CAN do something, an Agent CAN do so as well, other rules
notwithstanding", and just refer to Agents for the Raybots.
On Sun, May 21, 2023 at 5:44 AM Janet Cobb via agora-discussion <
agora-discussion@agoranomic.org> wrote:
> On 5/20/23 23:30, Janet Cobb wrote:
> >> In rule 2618, amend
> >> {{{
> >> A consenting player CAN, by announcement, grant a specified entity a
> >> promise, specifying its text and becoming its creator.
> >> }}}
> >> to
> >> {{{
> >> A Raybot or a consenting player CAN, by announcement, grant a specified
> >> entity a promise, specifying its text and becoming its creator.
> >> }}}
> >> [It's an interesting philosophical question as to whether Raybots can
> >> consent to things, so avoid the issue by making it possible for Raybots
> >> to create promises by announcement even if they don't consent to them.
> >> For what it's worth, rule 2519(3) means that the Raybot probably is
> >> consenting, but it's better to make it clear.]
> > What happens to such promises when the Raybot ceases to exist?
> >
> >
>
> Actually, in general persons ceasing to exist is likely to cause
> problems, and the current ruleset is careful to avoid it (R869/51's "is
> or ever was"; you remain an Agoran person after you die).
>
> I'm not sure there's a good solution here. Having disabled Raybots just
> sit around doing nothing isn't ideal. Auditing the whole ruleset for
> issues caused by this is probably good to do anyway but error-prone (and
> future proposals are reasonably likely to introduce new problems).
>
> --
> Janet Cobb
>
> Assessor, Rulekeepor, Stonemason
>
Re: DIS: (Proto) Raybots
On 5/20/23 23:30, Janet Cobb wrote:
>> In rule 2618, amend
>> {{{
>> A consenting player CAN, by announcement, grant a specified entity a
>> promise, specifying its text and becoming its creator.
>> }}}
>> to
>> {{{
>> A Raybot or a consenting player CAN, by announcement, grant a specified
>> entity a promise, specifying its text and becoming its creator.
>> }}}
>> [It's an interesting philosophical question as to whether Raybots can
>> consent to things, so avoid the issue by making it possible for Raybots
>> to create promises by announcement even if they don't consent to them.
>> For what it's worth, rule 2519(3) means that the Raybot probably is
>> consenting, but it's better to make it clear.]
> What happens to such promises when the Raybot ceases to exist?
>
>
Actually, in general persons ceasing to exist is likely to cause
problems, and the current ruleset is careful to avoid it (R869/51's "is
or ever was"; you remain an Agoran person after you die).
I'm not sure there's a good solution here. Having disabled Raybots just
sit around doing nothing isn't ideal. Auditing the whole ruleset for
issues caused by this is probably good to do anyway but error-prone (and
future proposals are reasonably likely to introduce new problems).
--
Janet Cobb
Assessor, Rulekeepor, Stonemason
Re: DIS: (Proto) Raybots
On 5/18/23 23:51, ais523 via agora-discussion wrote:
> Create a new power-3 rule, "Raybots":
> {{{
> A Raybot is a type of entity that has been created using the process
> described in this rule. Raybots CANNOT be created except as specified
> by this rule, and entities that came to exist by any other means are
> not Raybots.
>
> Raybots are persons. Raybots are created with their Citizenship switch
> set to Registered and their Radiance switch set to 40. Raybots agree to
> abide by the Rules.
May want to say "Immediately after a Raybot is created, eir Citizenship
switch is set to...", just to avoid the fencepost issue?
> Motivation is an untracked Raybot switch whose possible values are
> texts, and whose default value is "I deregister."
I'm not sure this exactly matters, but this promise wouldn't be
resolvable since deregistration on behalf is prohibited.
> A player CAN create a Raybot with a specified Motivation with 2 Agoran
> Consent, unless a Raybot with an identical Motivation was created
> within the previous 14 days, and SHOULD specify a name for the Raybot
> when doing so.
>
> If, for any given Raybot, at least one of the following conditions is
> continuously true for at least 10 seconds, that Raybot ceases to exist:
> * e is not a player, and/or
> * e is not the creator of any currently existing Promises, and/or
> * eir Radiance is 0.
>
> When a Raybot is created, it grants the Library a promise, becoming the
> creator of that promise, and whose text is that Raybot's Motivation.
>
> Raybots CANNOT support or object to tabled actions. The voting strength
> of a Raybot on an Agoran Decision is 0.
May want to add a RttCN clause to the support/objection prohibition.
Also, this should probably add a large fixed decrease to voting strength
to ensure it really stays 0 (which we may also want to do for festivals,
come to think of it).
Here's the list of zombie prohibitions as of its repeal:
> The master of a zombie CAN act on behalf of em, except a master
> CANNOT act on behalf of a zombie to:
> - initiate, support, object to, or perform a dependent action;
> - act on behalf of that zombie's zombies;
> - bid in a zombie auction;
> - enter a contract, pledge, or other type of agreement;
> - initiate a Call for Judgement;
> - create blots;
> - deregister.
> Players SHALL NOT cause Raybots to perform ILLEGAL actions.
This turns any infraction into a class-2 infraction if it can be done
through a Raybot.
> In rule 2618, amend
> {{{
> A consenting player CAN, by announcement, grant a specified entity a
> promise, specifying its text and becoming its creator.
> }}}
> to
> {{{
> A Raybot or a consenting player CAN, by announcement, grant a specified
> entity a promise, specifying its text and becoming its creator.
> }}}
> [It's an interesting philosophical question as to whether Raybots can
> consent to things, so avoid the issue by making it possible for Raybots
> to create promises by announcement even if they don't consent to them.
> For what it's worth, rule 2519(3) means that the Raybot probably is
> consenting, but it's better to make it clear.]
What happens to such promises when the Raybot ceases to exist?
> Create a new power-1.5 rule, "Raybot Transfer":
> {{{
> A Raybot CAN spend a specified amount of radiance to grant that much
> radiance to a specified player.
>
> A player CAN spend a specified amount of radiance to grant that much
> radiance to a specified Raybot.
> }}}
"non-Raybot"? Also, this arguably makes a Raybot directly transferring
to another Raybot ambiguous, as there are two methods that could be used
(even if they do the same thing).
> In rule 2659, amend
> {{{
> For each person there is a corresponding type of stamp.
> }}}
> to
> {{{
> For each non-Raybot person there is a corresponding type of stamp.
> }}}
What happens if a stamp would be created of a Raybot's type (e.g. if it
had the Wealth dream)? Might be cleaner to continuously destroy stamps
of Raybot type instead.
> {{{
> Any player CAN win by paying N Stamps, where N is the current number of
> active players and each specified Stamp is of a different type.
> }}}
> to
> {{{
> Any player CAN win by paying N Stamps, where N is the current number of
> active non-Raybot players and each specified Stamp is of a different
> type.
> }}}
> [Prevent Raybots from being counting towards Stamp victories, as they
> would badly unbalance them if created in number.]
>
Just for defense-in-depth, "each specified Stamp is of a different
non-Raybot type"?
> I'm interested in feedback about both the general idea, and the wording
> of the proposal to implement it. I am encouraged that, despite being an
> apparently major mechanic, it doesn't add much text to the rules,
> because it's mostly building on what's there at the moment.
>
I'm not sure that I want this gameplay, though I agree that it's
interesting. (This is "I'm legitimately undecided", not "I weakly think
I don't
Re: DIS: (Proto) Raybots
I dig this idea; the idea of being able to set up limited-purpose
Radiance automata is very fun, and I feel like it will almost certainly
give rise to some absolutely wild stuff. I do also like that Raybots can
create other Raybots (though the identical motivation proviso means that
you'd need to get creative to be able to have that work on any sort of
scale). That could get used to give a Raybot effectively infinite
radiance (I think, with the way that promises are structured, that you
could have a Raybot create a second Raybot that promises to give all its
radiance to the first bot, then cash that promise), but again, having
the Raybot's full capabilities be public from the jump should prevent
that from being too much of a problem.
On 5/18/2023 11:51 PM, ais523 via agora-discussion wrote:
Here's an idea I had as a way to a) shake things up in a way that's
likely to lead to lots of interesting CFJs for the next few months (I
came up with it after reading the CFJ archives for cases that looked
interesting), and b) let us experiment with mechanisms for awarding
Radiance that don't need a whole proposal cycle to go through.
The basic idea is to reintroduce the idea of artificial / legal-fiction
persons, but this time, instead of treading back over the old ground of
"let's let players create new persons that they have control over more
or less at will", the new persons are created with 2 Agoran Consent and
are effectively "powered by promises", so everyone knows what the new
persons will and won't do, and any abusive or unfair design can be
objected to. (Using Promises rather than having things happen
platonically makes things easier to track, as the Raybots won't do
anything unless someone cashes the promises.)
In addition to being powered by promises, they serve as a source of
Radiance, being created with some and being able to transfer it to
other players. So the basic economic idea is that if you have a good
Radiance award condition in mind, you can try it out without needing to
go through a whole proposal cycle, and it disappears naturally after
paying out a certain amount of Radiance so there isn't too much cost to
experimentation. In addition to the economic side of things, I'm hoping
there'll be a lot of gameplay simply stemming from trying to create
weird situations, e.g. can we get a Raybot to play the game as a semi-
autonomous player (with the only human action being to cash its
promises when they become cashable)? Could we get one to win? Could we
(and should we) get one to do the duties of an office?
In rule 869, amend
{{{
Rules to the contrary notwithstanding, no other entities are persons.
}}}
to
{{{
No other entity can be a person, unless explicitly defined to be so by
a rule with power at least 3.
}}}
[Makes it possible to create legal-fiction players again.]
Create a new power-3 rule, "Raybots":
{{{
A Raybot is a type of entity that has been created using the process
described in this rule. Raybots CANNOT be created except as specified
by this rule, and entities that came to exist by any other means are
not Raybots.
Raybots are persons. Raybots are created with their Citizenship switch
set to Registered and their Radiance switch set to 40. Raybots agree to
abide by the Rules.
Motivation is an untracked Raybot switch whose possible values are
texts, and whose default value is "I deregister."
A player CAN create a Raybot with a specified Motivation with 2 Agoran
Consent, unless a Raybot with an identical Motivation was created
within the previous 14 days, and SHOULD specify a name for the Raybot
when doing so.
If, for any given Raybot, at least one of the following conditions is
continuously true for at least 10 seconds, that Raybot ceases to exist:
* e is not a player, and/or
* e is not the creator of any currently existing Promises, and/or
* eir Radiance is 0.
When a Raybot is created, it grants the Library a promise, becoming the
creator of that promise, and whose text is that Raybot's Motivation.
Raybots CANNOT support or object to tabled actions. The voting strength
of a Raybot on an Agoran Decision is 0.
Players SHALL NOT cause Raybots to perform ILLEGAL actions.
}}}
[The basic mechanic: Raybots are created with 2 Agoran Consent, and act
only as a consequence of players cashing their promises. The idea is
that the Motivation – the initial promise – will specify everything
that the Raybot can do, probably by creating more promises. The
Motivation is untracked because it has no effect beyond the Raybot's
initial creation.
Being players, Raybots are (under this version of the proposal) tracked
by the Registrar. It doesn't seem like that should be enough additional
work to require a new officer?
Raybots are made unable to support/object/meaningfully vote as a
precaution, in order to prevent them being used to flood our consensus
mechanisms if someone finds a way to mass-produce them.
The starting value of 40 Radiance is a guess.]
In rule 2618, amend
{{{
A
Re: DIS: (Proto) Raybots
I think the general idea could be interesting. It's similar to juan's golem
idea, which I was also fond of.
I'm likely not the best to help with wording but, "Raybots agree to abide
by the Rules." seems weird to me. How can it 'agree' to anything?
On Fri, May 19, 2023 at 5:52 AM ais523 via agora-discussion <
agora-discussion@agoranomic.org> wrote:
> Here's an idea I had as a way to a) shake things up in a way that's
> likely to lead to lots of interesting CFJs for the next few months (I
> came up with it after reading the CFJ archives for cases that looked
> interesting), and b) let us experiment with mechanisms for awarding
> Radiance that don't need a whole proposal cycle to go through.
>
> The basic idea is to reintroduce the idea of artificial / legal-fiction
> persons, but this time, instead of treading back over the old ground of
> "let's let players create new persons that they have control over more
> or less at will", the new persons are created with 2 Agoran Consent and
> are effectively "powered by promises", so everyone knows what the new
> persons will and won't do, and any abusive or unfair design can be
> objected to. (Using Promises rather than having things happen
> platonically makes things easier to track, as the Raybots won't do
> anything unless someone cashes the promises.)
>
> In addition to being powered by promises, they serve as a source of
> Radiance, being created with some and being able to transfer it to
> other players. So the basic economic idea is that if you have a good
> Radiance award condition in mind, you can try it out without needing to
> go through a whole proposal cycle, and it disappears naturally after
> paying out a certain amount of Radiance so there isn't too much cost to
> experimentation. In addition to the economic side of things, I'm hoping
> there'll be a lot of gameplay simply stemming from trying to create
> weird situations, e.g. can we get a Raybot to play the game as a semi-
> autonomous player (with the only human action being to cash its
> promises when they become cashable)? Could we get one to win? Could we
> (and should we) get one to do the duties of an office?
>
>
> In rule 869, amend
> {{{
> Rules to the contrary notwithstanding, no other entities are persons.
> }}}
> to
> {{{
> No other entity can be a person, unless explicitly defined to be so by
> a rule with power at least 3.
> }}}
> [Makes it possible to create legal-fiction players again.]
>
> Create a new power-3 rule, "Raybots":
> {{{
> A Raybot is a type of entity that has been created using the process
> described in this rule. Raybots CANNOT be created except as specified
> by this rule, and entities that came to exist by any other means are
> not Raybots.
>
> Raybots are persons. Raybots are created with their Citizenship switch
> set to Registered and their Radiance switch set to 40. Raybots agree to
> abide by the Rules.
>
> Motivation is an untracked Raybot switch whose possible values are
> texts, and whose default value is "I deregister."
>
> A player CAN create a Raybot with a specified Motivation with 2 Agoran
> Consent, unless a Raybot with an identical Motivation was created
> within the previous 14 days, and SHOULD specify a name for the Raybot
> when doing so.
>
> If, for any given Raybot, at least one of the following conditions is
> continuously true for at least 10 seconds, that Raybot ceases to exist:
> * e is not a player, and/or
> * e is not the creator of any currently existing Promises, and/or
> * eir Radiance is 0.
>
> When a Raybot is created, it grants the Library a promise, becoming the
> creator of that promise, and whose text is that Raybot's Motivation.
>
> Raybots CANNOT support or object to tabled actions. The voting strength
> of a Raybot on an Agoran Decision is 0.
>
> Players SHALL NOT cause Raybots to perform ILLEGAL actions.
> }}}
> [The basic mechanic: Raybots are created with 2 Agoran Consent, and act
> only as a consequence of players cashing their promises. The idea is
> that the Motivation – the initial promise – will specify everything
> that the Raybot can do, probably by creating more promises. The
> Motivation is untracked because it has no effect beyond the Raybot's
> initial creation.
>
> Being players, Raybots are (under this version of the proposal) tracked
> by the Registrar. It doesn't seem like that should be enough additional
> work to require a new officer?
>
> Raybots are made unable to support/object/meaningfully vote as a
> precaution, in order to prevent them being used to flood our consensus
> mechanisms if someone finds a way to mass-produce them.
>
> The starting value of 40 Radiance is a guess.]
>
> In rule 2618, amend
> {{{
> A consenting player CAN, by announcement, grant a specified entity a
> promise, specifying its text and becoming its creator.
> }}}
> to
> {{{
> A Raybot or a consenting player CAN, by announcement, grant a specified
> entity a promise, specifying its text and becoming its creator.
> }}}
>
