Author: Rolf Tueschen
Date: 05:29:47 09/27/02
Go up one level in this thread
On September 27, 2002 at 07:41:07, Uri Blass wrote: >On September 27, 2002 at 06:57:57, Rolf Tueschen wrote: > >>Who is "you"? - The candidate was innocent. So how could he have a clue about >>the exact relations? That was my point. > >The question is not what the candidate should do based on his knowledge but what >is best for him based on your knowledge. Uri, thanks for this post. Unfortunately I can't omit to add a few "corrections". This is a very fundamental question. Of course you should answer the question with your best knowledge, - but, you must consider the position of the candidate. And of what he could do on the base of his informations. You can't give your advice on your information. We have here another dilemma. BTW scientists do it in their normal daily routine. Take a medical expert, he 'knows' what is the best for a 'patient', but in his practice he must also consider the patient in his very unique personality, and it could well be that the doctor must conclude that it might be better to advise something 'sub-optimal', but that would exactly be the optimal for this patient. If you read my contributions here and on my webpage at http://hometown.aol.de/rolftueschen/monty.html (all in English), you'll see that I was always talking about the psychological situation for the candidate. Because the question of Mr. Whitaker was not theoretical but very much to the point. If such and such, could the candidate (the concrete, not a number in science) make an advantageous choice. Ok, then Marilyn should have differentiated between the optimal anwser and the realistic answer for the innocent candidate, just the way the medical doctor above. I thought it could be a good idea to give the topic here to the computer chess people, because we suffer under the lack of 'differentiations'. With Sune I had a short conflict before he realised that we might have the situation of two possible discussions. Well, that was exactly the point. It's important to notice that the whole problem is not based on the bad English of s foreigners. The problem happens in every language. also if it's your native one. The reason is that out of science you suffer under the ambiguity of terms. But also in discussions between scientists most problems result of misunderstandings caused by different definitions. Then you can perfectly discuss for years without making any progress. Let me put it into a bon-mot. People seem to miss that they insult themselves if they offend the other for "his" mistakes. > >If your knowledge is that the host's strategy is to always open >a door then it means that it is better for him to switch. But it wasn't the stratedy. At least not declared as. > >Understanding that the host's strategy is always to open a wrong >door from the question does not seem to be wrong to me. I agree. That was indirectly mentioned with the host knows where the car is. > >I did not say that marilyn was right because data is missing but I said that she >was right because it is not clear from the question that >data is missing and it is possible to understand it in >the way that sune does it. And I say that she digged not deep enough to the question of what the candidate could have done at best on the base of _his_ informations. > >If someone ask you how much is 5+7*8 you can say 5+56=61 >another solver can say that it is (5+7)*8=96 > >Who is right? > >I do not think that you can say that the first was wrong because >it is normal to assume that multiplying come first and only >later comes adding. > >This case is similiar. Not at all. You confuse a mathematical law and the candidate's knowledge of the host's strategies. ;) > >Misunderstanding of the intention >of the question is possible. Yes. And in such acase who's to blame? The scientist, the expert or the female IQ genius? I think the answer is clear. All three but not the candidate. > >I suspect that it is even possible to say that people >who say that data is missing are >wrong in understanding the question because of lack of knowledge >in english but I am not sure. >I cannot say that they are wrong in solving the question. Also in that case a thorough analysis should mention the problem. BTW that is a very basic neccessity in science. You must document all what is relevant. > >People who solve the question by saying that the probability is 1/2 are clearly >wrong because even if you do not know the strategy of the host you >cannot say that the probability is 1/2 and it even can be a bad idea >to switch your choice if the strategy of the host is to open >a door only when you choose the right door. Here we ave two variations, not a single problem. Of course 1/2 is the solution for the candidate. But in the second variation it's still unclear if the candidate knew this bad possibility. In that case, if he only guessed that the host could have such a strategy, he couldn't be sure and therefore it still remained at 1/2. 1/2 is a gambling, but at least with even chances. :) Rolf Tueschen > >Uri
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.