Hello mark!

31 May 03 12:05, mark lewis wrote to Bob Jones:

 BJ>>> I believe there was something else at the time that also
 BJ>>> drove the 2 second decision -- that was out of other fido
 BJ>>> based sosftware implementations.

 WM>> Two seconds is the limit of resolution in the FAT filesystem.
 WM>> That's ultimately where it comes from.

 ml> and that only because the date/time stamp was taken from the original,
 ml> one message per file, format (ie: *.msg)... the stamp in the message
 ml> header can/could be different than the one on the disk but that the
 ml> same structures were (apparently?) used, lead us down this road...

I'm sure you'll find both even and odd secs represented by both the binary
and ASCII datestamps if you randomly LIST a PKT or two.  Scott was
deliberately rounding to produce the dupecheck CRC in 1.01 and stopped in
1.10.  File system and local messagebase formats concerns were not the
driver.

.\\ike

--- GoldED 2.50+