From: zingard@mcmaster.ca   
      
   Hi all,   
      
   On page 190 of "Introduction to Automata Theory, Languages and   
   Computation", 2E, Hopcroft et al., there is an observation about   
   derivability in CFG's. It states that if x1x2...xk =*>w, then we can   
   break w into w1w2... such that xi =*> wi. This makes sense, but I'm   
   wondering if anyone can further explain Hopcroft's proof hint that he   
   gives below this observation? Specifically, I can't figure out the   
   precise inductive hypothesis that is required, and why it allows us to   
   conclude the theorem.   
      
   Thanks for any ideas,   
   Dan   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)