[blml] Positronic brain

Herman De Wael herman at hdw.be
Sun Oct 8 10:16:11 CEST 2006


Hello Steve,

by trying to agree with me in different words, you are actually 
advocating the exact opposite of what I'm trying to say. Let's analyse:

Steve Willner wrote:

> From: Herman De Wael <herman at hdw.be>
> 
>>I believe that a more accepted interpretation is that once there is AI 
>>of the same content as UI, the UI restrictions fall away.
> 
> 
> I'd put it somewhat differently, though the result will often be the
> same as Herman's.
> 
> 1) The AI determines the LA's.
> 

Good starting point. Without the AI, there are probably more LA's than 
with it. If there remains only one LA, then we have no problem, so 
let's assume that there remain 2 LA's: A and B.

> 2) UI that is the same as AI will seldom if ever suggest one LA over
> another.  However, if in some rare case the UI does suggest one over
> another (perhaps in Tim's probableistic sense), the usual rules apply.
> 

So let's assume that the UI suggests A over B. Then the AI also 
suggests A over B. So with the AI, the player will choose A. Yet since 
this is also the suggested LA from UI, you will rule against him. I 
won't. Let's put it this way: You give the auction, without UI, to a 
number of players. 3 choose alternative A, 1 chooses B. That means 
that B is a LA. You now add the UI, and all four tell you that it does 
not alter their choice, even if they wanted to cheat. You would rule 
against this pair, I wouldn't.


-- 
Herman DE WAEL
Antwerpen Belgium
http://www.hdw.be



-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.407 / Virus Database: 268.13.1/466 - Release Date: 7/10/2006




More information about the blml mailing list