>    
   > Although you claim to be being scientific, you nevertheless seem to be    
   >    
   > using a private definition of "information", so that we're all talking    
   >    
   > past each other here.  What exactly do you mean by "information"?   
   >    
   >    
   I never claimed to be scientific, just that the Shannon people falsely do. Nor   
   do I know what "information" is, but just that they don't either. There is   
   nothing private about my questions, except the answers. As I mentioned,   
   Szilard was trying to find a    
   new dimension of physics, and if you look at what physics uses as their   
   fundamental units, mass, energy, space, and possibly entropy - then there is a   
   refinement from mass to energy to space to entropy. So "information" was going   
   to be a new dimension of    
   physics.    
   After doing a bit of AI, and looking at the scientific method, I naturally ask   
   these questions. But they may not relate to what you are doing in engineering,   
   and I am interested in your methodology and thought processes regarding   
   particularly measurement.   
    Let's put the IT issue aside (for now), as it does relate to AI, until and if   
   it reappears. (1)   
   =============   
      
   OK, Here is a thing of investigation on my part. What controls the rise time   
   and decay time of an led? It's an accepted rule of measurement that rise time   
   time is directly related to Band Width. (at least I hear it with   
   oscilloscopes.) So, I have to    
   review your paper on paper on circuitry to increase the BW of the photodiode.   
   It's an amazing paper, not least for the fact that this approach was   
   overlooked in a fairly mature field. I have to re-read it and ask questions.   
   So my question is: For the led,   
    is there any circuit approach to increase their BW? Or is this a wrong track   
   of thinking? (ie works for a PD, but a LED another matter entirely.) And a   
   corollary question: Once you have a fast blinking light source and fast data   
   collection. (Say ADC at    
   1 MSPS), is there any SNR-type advantage there?    
      
   I ask these things out of scientific curiosity only, whatever that is.   
      
   JB   
      
   1. I will probably post more on this, as I have been thinking about more - and   
   how to extend Shannon a bit. Shannon tried to develop various other   
   applications of his theory, in the AI area, such as a chess program, but no   
   luck.   
      
   =======================   
   "New ideas are like foreign proteins to a body: the mind rejects them" -   
   Medawar, biologist   
      
   "People ask me when I'm describing electromagnetic radiation and wave my arms   
   in the air, whether I know what EM 'IS.' I have no idea." - Feynman   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)