[blml] Positronic brain

[Eric Landau]

At 04:48 AM 10/9/06, Herman wrote:

>richard.hills at immi.gov.au wrote:
>
> > I agree that UI completely identical to AI is rare.
>
>The situation we disagree upon may well be very rare.
>But there are many situations that we (and all the TD's around the
>world) agree upon, but which you don't acknowledge as being of this
>same type. Just think of all the correct explanations (which ARE UI)
>if you don't believe me.

I think Herman is making the semantic error (for which I blame not 
Herman, but TFLB, which is far from precise in its use of the terms) of 
treating "unauthorized information" as a synonym for "extraneous 
information".  But L16A ("Extraneous Information from Partner") 1 
("When Such Information Is Given") tells us that it is not; it allows 
"the opponents" to "dispute the fact that unauthorized information 
might have been conveyed", which means that "extraneous information" 
may be "given" without "unauthorized information [having] been conveyed".

Once we recognize that it is possible to receive "extraneous" 
information from partner without any "unauthorized" information having 
been conveyed, correct explanations of the partnership's methods would 
seem to be an obvious example of this.

Other examples of EI from partner which do not convey UI might include, 
for example, his estimate of the partnership's current score, or his 
evaluation of the strength of the opponents.

It is critical to recognize, however, that the manner in which 
"authorized" EI is transmitted may convey UI by its means or manner of 
transmission, as opposed to its content.  "We're vul, pard" 
is  harmless EI when you're pulling the cards out of the board, but 
will certainly convey UI when partner is in a huddle over a high-level 
decision.  That doesn't make the vulnerability itself -- or "all the 
correct explanations" themselves -- UI.


Eric Landau                     ehaa at starpower.net
1107 Dale Drive                 (301) 608-0347
Silver Spring MD 20910-1607