Kind of puzzled about timestamps

[Ben Scott]

On Sat, Mar 6, 2021 at 9:17 PM Curt Howland <Howland at priss.com> wrote:
>> Say you find a file that has a stored time of 2007 MAR 31 17:00 UTC.
>
> With GMT as the standard time stamp, one can at least know relative
> times of files, even if one does not know such real-world details.

There are a few problems with that theory:

(1) Not everything is the last modified date of an inode.  For
example, a file storing a time, like I gave as an example.  (Think
meetings in a calendar, messages in a mailbox, etc.)

(2) Clocks don't exist in a vacuum.  If the computer is deriving the
time from NTP or some other thing using UTC, then yes, you're good.
If the computer is deriving time from what the human typed in, you've
got trouble.  These days that's more the exception than the rule, but
not in 2007.

(3) Computers don't exist in a vacuum.  The whole point of a computer
is to automate real-world details.  True, if we didn't have humans, we
wouldn't have to worry about humans changing the time zone, or wanting
to read the time out of the computer.  But if we didn't have humans we
also wouldn't need to worry about the time being wrong either, because
the humans are the point of all this.

If we could get all the humans to agree to follow UTC everywhere and
just abolish time zones, things would be better, yah.  But we can't
even get the US on the metric system, so I hold little hope for that.

-- Ben