Author: Mogens Larsen
Date: 17:06:36 05/05/00
Go up one level in this thread
On May 05, 2000 at 19:38:24, Walter Koroljow wrote: >I have noticed two problems with this thread: very high standards for >experimental results and very loose standards for accusations. The experimental standards goes hand in hand with the conclusions IMO. Out of curiosity. What accusations are you talking about? >With limited resources it is impossible to achieve mathematical certainty in an >empirical study. Just like in our everyday experiences. However, I think we >can (and do!) absorb the information from both sources and become more >knowledgeable. I would say that we can all learn from even a poor study if the >study conditions are explicit. In this case the study was not poor and the >conditions were explicit. You're unable to conclude whether the study is poor or not and neither am I with the information available. The study in question was explicit. However, it is not adviseable to extend the conclusions beyond the emphirical data, which is one of my complaints. But you're right, emphirical data is always interesting, the problem is interpretation. >I would suggest that the standards for a useful accusation should be much much >higher than those for a useful report. If we're talking science that would be a wrong approach. It's the purpose of the study to present the question that needs analysis, and how that question is to be answered within the obvious uncertainty of any emphirical approach. Furthermore, it's also necessary to present the uncertainties, and the importance of them, involved as the study progresses. In other words, planning is essential. I don't like the word accusation as it implies intention on behalf of the person conducting the study. Suggestions and questions are more reasonable terms. Sincerely, Mogens
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.