Re: [Gnu-arch-users] Re: s/GMT/UTC/

[Phil Frost]

The issue as i see it is not the accuracy of the value of the dates, but
which of "GMT" or "UTC" is correct. In previous posts I have asserted
that any clock set to a modern time source such as NTP, WWV, or other
civil time service is set to UTC, not GMT. The difference between UTC
and GMT will never be more than 0.9s, but I believe there is still some
value in labeling the time with the correct timescale simply for the
sake of being correct.

On Thu, Sep 09, 2004 at 02:14:22PM -0400, Adrian Irving-Beer wrote:
> Dates are just rough ideas of when changes were implemented. Two years
> later, I can look back and say 'that project lasted from September
> 2003 to April 2004'; leap seconds aren't exactly my top priority at
> that point.
> 
> The only time it would really matter would be with CVS -- and only
> because CVS didn't have atomic commits or automatic full-tree
> versions.  As an informational tool, I think the date is pretty
> ubiquitous across RCSes.  They are, after all, supposed to track
> changes over time. :)