Kerim Aydin on Wed, 21 Nov 2007 18:39:13 +0100 (CET)


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: [s-d] [s-b] Registration, consultation, proposal


On Wed, 21 Nov 2007, comex wrote:
> Therefore, any non-null set of the requirements consists of *a*
> Membership Test.  The AFO can "send, and receive a reply to, an email
> to another entity" so it passes the Membership Test containing of just
> that requirement.  It is therefore capable of passing *a* Membership
> Test, but not (I think) *any* Membership Test.

Counterarguments:

1.  It can't send messages "on its own."  Its agents can, but there is no
proof that this is action on its own outside of its agents (it may be a 
legal fiction in some rules systems).  In terms of being an AI (e.g. an
autoreply), there's no proof that any setup wouldn't be wholly dependent
on its agents.

2.  It can't pass proof of uniqueness, as every message would be sent by an 
agent and therefore no single message can be said to be unique from a given 
agent.  A collection of messages from its set of agents might be unique if
the set is unique, but passing "a test" is a single action, not the collection.

3.  Even in Agora, it can't refer to itself as "I".  Messages are sent
"on behalf of" the AFO, etc., so even where its uniqueness is accepted (Agora)
it's awkward.

4.  It's definitely a Borg.

So I don't think it can pass any membership test.  It certainly can't pass
an arbitrary one.

-Goethe



_______________________________________________
spoon-discuss mailing list
spoon-discuss@xxxxxxxxx
http://lists.ellipsis.cx/mailman/listinfo/spoon-discuss