From: grahn+nntp@snipabacken.se   
      
   On Wed, 2 Sep 2009 20:17:27 +0000 (UTC), Rick Jones  wrote:   
   > Jorgen Grahn  wrote:   
   >> OK; I just thought TCP was such a protocol. Making read()s fail and   
   >> telling the peer on the other side that the socket RX buffer is full   
   >> seemed right to me -- given that I thought shutdown() on a TCP   
   >> socket kind of transformed it from two octet streams into just one.   
   >   
   > Indeed, TCP is supposed to support "simplex" operation where data can   
   > flow in only one direction.  I seem to recall some sort of ancient   
   > remote job entry mechanism using that feature, but I cannot recall its   
   > name, nor anything more recent which actually used the simplex   
   > feature.   
      
   And if I read between the lines, you believe that this shows that   
   using TCP in simplex mode is almost always a bad idea, right?   
      
   I realize that application-level ACKs cannot be replaced by   
   TCP-level/socket API ones. And I guess most people who think "I don't   
   need ACKs" realize sooner or later that they *do* need them. Is that   
   what you're thinking of?   
      
   /Jorgen   
      
   --   
     // Jorgen Grahn    O  o   .   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)