From: Jon Dreyer (jdreyer@ici.net)
Date: 03/19/97-02:22:40 PM Z
Message-ID: <33304B10.7F0A@ici.net> Date: Wed, 19 Mar 1997 15:22:40 -0500 From: Jon Dreyer <jdreyer@ici.net> Subject: Re: 1/1/1970 Lance Kibblewhite wrote: > > Faster for those systems that currently use two 32 bit values (or one for > seconds only). But not for those (such as PC's and VMS) which already use > a 64-bit value, albeit with a different unit and base. Also we should remember that protocols tend to last longer than computers. This usually implies that, when as protocol designers we are faced with a choice between an elegant solution and a hack, it should take a fairly compelling argument to make us pick the hack. I can just imagine ten years from now, when my computer has a 256-bit picosecond clock, wondering what posessed us to put the low-order seconds in one word and the high-order seconds and nanoseconds in the other!
This archive was generated by hypermail 2.1.2 : 03/04/05-01:45:33 AM Z CST