>While I don't expect us to ever switch to a decent measurement of
>time*, I do wonder how much longer we'll use time zones.
Uh, gotta disagree here. Both a universal time and timezones that make
reference to a time relative to daylight are useful. RFC822 date fields
give a nice compromise:
Date: Tue, 1 Jul 1997 15:33:49 -0700
They show UTC time (Universal Time Coordinated, formerly GMT), but also
show the local offset. Not only can you see precisely when the message
was sent, you can also quickly figure out whether the recipient is
likely to be awake when you're responding!
Your URL was wrong. You meant to do
http://www.cs.wustl.edu/~seth/chrons/
This is brilliant, BTW. Of course, on a scale of impracticality it
really can't compare to the 28 hour day proposal that you link to.
Very good point on the lack of a 10^-4 prefix and 10^-5. (In case your
curious, prefixes are available at
<http://www.processassociates.com/process/convert/si_prefx.htm>.)
- dan