Hi,

a point worth noting about HTTP pipelining is that it prevents the "TCP
slow start" from happening whenever you wait for data on an idle
connection.

Goswin, you could have a look at libwww, but I do not recommend it; while
pipelining works, it has a number of other issues. Instead, either just
live with persistent, non-pipelined connections or use libcurl with 2 or 3
parallel connections to the same server. :-/

Cheers,

  Richard

-- 
  __   _
  |_) /|  Richard Atterer     |  GnuPG key:
  | \/¯|  http://atterer.net  |  0x888354F7
  ¯ '` ¯