Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Dangers in CC - SSDF: Terminology, Statistics

Author: Rolf Tueschen

Date: 04:13:24 02/21/03

Go up one level in this thread


On February 21, 2003 at 06:49:24, Jonas Cohonas wrote:

>On February 21, 2003 at 06:15:29, Rolf Tueschen wrote:
>
>>On February 21, 2003 at 06:02:26, Jonas Cohonas wrote:
>>
>>>You do know that the SSDF does not claim to be 100% accurate right??
>>>
>>>Jonas
>>
>>I know what they wrote, and simply read the advertisement now for SHREDDER, that
>>a new program had climbed number one place. Exactly that is not only inaccurate
>>but simply plain false! And not by a slim margin but with a big number! Note
>>that they had a difference of 8 points and tht with >30 points incertitude.
>
>If you fail to take their conciderations into account, your reasoning is
>contaminted by your own interpetation/s of the SSDF and therefore not valid in
>the context you present here.

Perhaps I fail a lot in my life but I know one thing, in stats I am not
contaminated by ignorance which seems to be the case with other persons here
involved. You start the usual pratronizing. Without success. You are surley not
the one who is entitled to tell other people to take it or leave it. You are
basically telling me that this is a tabooed topic. But as you see, I have a
different opinion. As to your defense line you are more than false. You simply
don't get the point that SSDF can't dave their results by implicite
confirmations of inaccuracy. To present a number one is a qualified decision and
it's simply not possible on the base of the given data. Now call me stupid, do
what you want, but you can't change facts. I don't need a spin doctor for my own
views. I have studied stats and such basic stuff is so trivial that all the
attempts to maipulate the debate here are breathtaking. One can only assume one
thing. The Lobby for SSDF must have big commercial interests. oherwise it would
not be explainable why NONSENSE is sold here as wise decisions. NONSENSE remains
NONSENSE, no matter how much it's worth for you.

Rolf Tueschen



>
>>Ok, I answered you here but all others please have a bit of understanding that I
>>can't reply on all the usual claque.
>
>Knowing that they do not claim to be 100% accurate is an important factor as it
>nullifies your critique...
>
>Hey, I also do like SSDF! And it tears my
>>heart in two when I see what they are doing and how they "defend" the
>>impossible.
>
>It is just a list, where they present the numbers produced by different chess
>programs, not a soap opera. Take it or leave it there is flaws in anything if
>you look hard enough.
>
> But please I am not the official who now must defend scientifical
>>trivialities. So, unless I see a really heavy weight argument, I want to abstain
>>from debates.
>
>But the problem is, you seem to overlook that they do not claim that the SSDF is
>a 100% accurate scientific conclusion.
>
>Jonas



This page took 0.01 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.