Ed Murphy on Tue, 27 Jul 2010 12:32:28 -0700 (MST)


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

Re: [s-b] [Oracle] CFI 122


JamesB wrote:

>> Date: Sat, 24 Jul 2010 19:01:16 -0400
>> From: teucer@xxxxxxxxx
>> To: spoon-business@xxxxxxxxx
>> Subject: [s-b] Let's have fun with timing, shall we?
>>
>> A recent proposal attempted to put my name back to what it was prior
>> to the point where I stopped having one, but it didn't give me the
>> name I most recently intended to give myself. I'm therefore currently
>> named Rule 700. So to fix this tragic oversight, I change my name to
>> "Respected One."
>>
>> (Note: this is still permitted, and it still appears to have the same
>> effect as before. Not having a name is of course now on the LOGAS,
>> though you have three ndays to acquire one if you don't have one.)
>>
>> Then I change my name to Craig B. Daniel, and then at last I change it
>> to teucer, which is what I actually want it to be.
>>
>> I now submit a CFI: "I have exactly one name."
>>
>> Arguments: Rule 21 tells us that events occur on reaching the Public
>> Forum, and that events not caused by agents (including the loss of the
>> name "Respected One") occur when the rules indicate. Rule 49.B.4,
>> which defines the unique behavior of the name "Respected One",
>> indicates that the well-documented vanishing happens "whenever a
>> player who has the title 'Respected One' fails to meet the
>> aforementioned condition" - which I fail to meet.
>>
>> I see two obvious interpretations of this situation. First, the name
>> changes all apply, in order, on this message reaching the PF, after
>> which I would lose my name if it were to be Respected One. In this
>> case, the CFI is plainly TRUE - my name is teucer, and I have no other
>> names. I find this hard to believe, since it requires positing a gap
>> between name changes during which I failed to meet the specified
>> condition and must have lost my name. Second, the name changes occur
>> sequentially, and immediately after the first one but before the
>> second, I lost my name. My name is now blank, and I have three ndays
>> to correct this unfortunate situation on pain of ass-kicking. In this
>> situation, obviously the CFI is FALSE - I have *zero* names. But it
>> also seems incorrect, since it suggests that somehow two things that
>> reach the PF at precisely the same moment can not only be ordered but
>> can in fact have things happening in between them - in which case, the
>> obvious question the judge should be prepared to answer is precisely
>> *when* I could possibly have lost my name; it must be after the
>> "Respected One" name change reached the PF and yet before the "Craig
>> B. Daniel" change did so.
>>
>> Given that these are both clearly wrong, the judgement of UNDECIDED
>> seems indicated. But there is a third possibility, and the fact that I
>> see no clear reason why it is incorrect means that the falsity of the
>> other two is not paradoxical after all. The last interpretation is
>> that the name changes are simultaneous, and thus gave me not one but
>> three names. I lose one of these, obviously.
>>
>> This third case is confused, however, by the fact that the CFI is
>> happening in the same post as the name changes, reaching the PF at the
>> same time they do. The judgement is supposed to indicate the truth of
>> the statement at the time of the Call, rather than of the judgement.
>> Yet the call is problematically simultaneous with the name changes! If
>> it came before them, it would be trivially TRUE - my name at that time
>> was Rule 700. If it comes after them, it is trivially FALSE, because I
>> have two or three names (two if it somehow also comes after the loss
>> of "Respected One," which has the same criticism as the second
>> interpretation above, and false if it somehow comes before that part,
>> which has the same problem as interpretation number one - this
>> quasiparadox is why I phrased the CFI such that this doesn't have to
>> be decidable for the CFI to remain FALSE in this condition). But how
>> many names do I have during the process of changing my name? This is
>> not an easy question for a judge to answer, and I'm glad I'm
>> submitting this CFI rather than judging it.
>>
>> The third interpretation is problematic for another reason - game
>> custom presumes, and the rules certainly used to indicate (though I
>> can't find where they do so now, if in fact they do) that actions
>> posted in one message are resolved in order. But if this is no longer
>> supported by the text of the rules, then the phrasing of Rule 21 means
>> we must instead favor simultaneous resolution. While I can't at this
>> moment see any problem with this, other than the fact that it allows
>> me to have multiple names (not actually problematic, but it might
>> become a headache for the registrar if it catches on), I suspect
>> strange corner cases will be found, whether by accident or design.
>>
> 
> 
> This is CFI 122. I assign CFI 122 to Judge Murphy. 		 	   		  

As previously noted, Rule 47 fails to do anything (it only
refers to entities defined by current proposals, not rules).

Also as previously noted, Rule 2 prevents the self-assigned
name "Respected One" from being conflated with the rule-defined
title "Respected One".  Because of this, Rule 49 does not apply
(it only applies to the rule-defined title "Respected One").

The timing of the actions directly announced in teucer's message,
relative to each other, is uncontroversial (at the same time, but
sequential in the order they were announced; there were no attempts
to make the order ambiguous, e.g. via ASCII art).  The only perceived
controversy was the relative timing of Rule 49's effect, but as noted
above, Rule 49 has no effect in this case.

TRUE.

For completeness, if Rule 49 had applied to names instead of just
titles, then I would have judged as follows:

Rule 49 attempted to take effect after the event it was reacting to
(the first change), but before the next event (the second attempted
change).

However, Rule 2 ("All game entities must have uniquely identifying
names") took precedence over Rule 49.  If there was a rule-defined
procedure for automatically assigning a unique name to an entity
that would otherwise lack one, then that procedure would have prevented
Rules 2 and 49 from conflicting at all; but there isn't one, so Rule 2
prevented Rule 49 from taking effect.

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