Why the temptation for dial users to crank back rwin/mtu?

In a previous episode Forrest W. Christian said...
:: > MTU - at least this makes a little bit of sense.. If they're doing
:: > HTTP/1.0 stuff with parallel connections then a smaller MTU is going
:: > to make that parallelization latency much more effective and perceived
:: > performance will go up some.. it doesn't impact full document
:: Just for my information, does the MTU setting affect <received> packets
:: in some way? My understanding was that a machine wouldn't send packets
:: over the MTU size, but could recieve anything up to whatever the TCP/IP
:: stack writer included in the stack. Guess I'll have to go dig out the

windows uses the mss option in it's SYN.. it is supposed to be set to
the max mtu of all the boxes interfaces - 40. (and dial users probably
only have 1 interface)...the other end has to use that as a max
starting point for pmtud.. thus lots of small packets headed towards
the windows box..