On Fri, Jan 29, 2021 at 10:15:58AM -0500, Jason Cobb via agora-discussion wrote:
> On 1/28/21 10:50 PM, Falsifian via agora-discussion wrote:
> > On Tue, Jan 26, 2021 at 08:53:55PM -0500, Jason Cobb via agora-discussion
> > wrote:
> >> This is intended to prevent lockers from being excluded from
> >> rules-defined theft (which I plan to propose stones for soon(tm)). The
> >> intended purpose of this is to allow stones to mess with assets without
> >> incentivizing lockers (which would ruin Trigon's life).
> >>
> >> Key design goals:
> >>
> >> 1. Each contract has an Executor, which is usually a player or Agora.
> >>
> >> 2. Executors can be flipped to/from Agora in a similar manner to
> >> charities, which means that the people are saying it's not just a locker
> >> to avoid assets being taken away.
> >>
> >> 3. Executors can be null both as a stopgap and a default. Contracts with
> >> null executor cannot receive assets and risk losing their assets if they
> >> don't put _someone_ in charge of them.
> >>
> >> 4. Once an executor has been set to a party, it can't be set back to null.
> >>
> >> 5. When rules take assets from a person due to theft/whatever, it takes
> >> from first the person itself, then the contracts for which e is the
> >> executor, then the contracts for which the executor is null (which,
> >> after a month and a half, should be ~0).
> >>
> >>
> >> So, for example, if a stone had an effect of transferring a pendant to
> >> the wielder, it would say "For a specified player, eir first
> >> jointly-accessible pendant is transferred to the wielder."
> > Is this really needed? It seems kind of complicated.
> >
> > If we want a stone that steals things, why not say it can steal from
> > any entity? Taxes can take from contracts as easily as from people, and
> > giving people exemptions could maybe even disincentivise putting things
> > in contracts.
> >
> > I imagine there are examples where this would help but I'm not seeing
> > them right now.
> >
>
> Here's the proposal I was planning to submit before it was pointed out
> that lockers break it:
>
> Title: Playing the stone card
> Author: Jason
> Coauthors: nix
> Adoption index: 2.0
>
> {
>
> Amend Rule 2645 by appending the following:
> {
>
> - Middleman Stone (weekly, 60%): For a specified card type and player
> who owns at least 3 of that card, 3 of that type of card are revoked
> from that player; the wielder gains 2 of the associated product; the
> specified player gains 5 of the associated product.
>
> - Quill Stone (weekly, 40%): A pendant is transferred from the sole
> person (if any) who submitted the most proposals in the previous Agoran
> week to the wielder; an attempt to wield this stone fails if that person
> is not correctly specified.
>
> - Justice Stone (weekly, 70%): A blot is created in the possession of a
> specified person.
>
> }
>
> }
>
>
> The first two are broken by lockers. For the first, even if we let it
> take assets from contracts directly, it could be circumvented with
> multiple lockers that each hold less than 3 stones. I wrote the second
> one because nix suggested making stones with indirect targets to require
> timing or shenanigans if you want to target a specific person. Lockers
> would break the quill stone because there's no sane way to specify it to
> take from contracts without breaking the fact that it's indirectly targeted.
>
> We can disagree about whether these specific stones are a good idea, but
> lockers break anything that does anything remotely similar to the
> middleman or quill stones, and I think that's bad.
>
> --
> Jason Cobb
Thanks for the examples.
Miscellaneous comments and thoughts:
* I think someone could get around the current proto by creating a new
locker every couple of days.
* Another workaround is to use other players as your locker, with
appropriate contracts or pledges so you can get your things back.
It's not a perfect workaround so maybe we could just let it be.
* Saying the executor of a new contrtact is the person who created the
contract might partly mitigate the first workaround, but people could
conspire to make executors not match effective owners for a few days.
I think in that situation the workaround it would be about as
powerful as the second workaround (other players as lockers), since
at least all the possessions involved would be jointly-accessible by
someone.
* As an alternative to the Middleman stone that doesn't depend on
Executors, what about this: you specify 4 Cards; they're revoked and
replaced with 2 products each with the same owner the Card had, and
you get 2 products too.
* In general I'm still not that keen on the Executor idea. I'd probably
be a PRESENT for now, and maybe itching to repeal it in a few months
or a year once we've had a bit of fun with it. It adds some
complexity to the rules (though admittedly the implementation is
reasonably short and well-designed). I'd be more interested in seeing
theft protos that try to naturally disincentivise (or be neutral
about) lockers rather than depend on a heavy-handed rule change that
exists just so that theft is easy to define.
--
Falsifian
