Hello Kees,   
      
   On Friday January 11 2013 19:05, you wrote to me:   
      
    MvdV>> If there is a problem in Fidonet that nobody bothered to whine   
    MvdV>> about in 20 years, then it can't be all that big a problem.   
      
    KE>  It is not when reading messages, I can guess when you use an accented   
    KE> e, and I see some rubbish. I still understand what you want to tell   
    KE> me. If I get a full page of rubbish, that stems from apparently a user   
    KE>  writing, in Cyrrilic, I don't bother, because I cannot read the   
    KE> message in either notation. So there appears to be problem.   
      
   Hmmm..  it seems to me a  "no" dropped out of that last sentence....   
      
    KE>  But now you want to appaly the same to texts, that should be read by   
    KE>  in this case dumb machine,   
      
   That "dumb machine" has nothing more to do than just pass the text "as is".   
   Nodelist processing software does not interpret names of people or places, it   
   just moves them around without doing any interpretation on them at all.   
      
    KE> without the capacity of the IBM Watson system. You can be quite shure   
    KE> that there will be plenty of problems reported.   
      
   Sorry, I think you are seeing problems that are not there.   
      
    KE>  How will I handle such addresses with my Fido 11T system and my ASR   
    KE> 33 terminal?   
      
   You can't and you never could. When Fidonet was born the ASR 33 and other   
   Upper case only terminals over 7 bit channels had already been left behind.   
      
   Fidonet is fully 8 bit channel.   
      
   You can't program in C with an ASR 33 either. C is case sensitive, you need   
   upper and lower case. Plus that the ASR 33 misses the curly brackets and the   
   '| '. Indespensible, you need them for the "or" function.   
      
   Backwards compatibility is fine, but there are limits..   
      
   Cheers, Michiel   
      
   --- GoldED+/W32-MINGW 1.1.5-b20110320   
    * Origin: http://www.vlist.eu (2:280/5555)