Re: Org mode timestamps on the Moon ;] (was: [FEATURE REQUEST] Timezone support in org-mode datestamps and org-agenda)

[Tom Gillespie <tgbugs@gmail.com>]

Oh boy. In short, I think we can only hope they come up with
LTC and we already have a syntactic space to say whether
our reference seconds are TAI/UTC/LTC/MTC/JTC etc. And
being the privileged squats that we are if the time system is
left out then it means UTC. The friendly thing to do would be
to always include the time system code in our timestamps,
but I don't think those have standard reference codes yet?

> would gain about 56 microseconds over 24 hours

And here I was naively hoping to avoid having to deal with
relativistic effects. That would seem to be a pretty serious
issue, but my assumption further down that comment is
hopefully what they will do. Specifically:

> let's just assume that clock synchonization will happen between earth and
> mars so that the unix epoch could be synchronized between the frames

The only generalized solution is to record the full location
(see intro to http://naggum.no/lugm-time.html which I'm surprised
hasn't been linked in this thread yet, but was linked on the same
topic back in 2011 it seems) and let people sort out how to translate
that local time from that reference frame in that particular gravity well.

I don't think we want to force users to start embedding their gps
coordinates, their current acceleration, etc. along with their
timestamp. But it turns out that the unix epoch is only standard
at 9.8m/s^2. So ... minutes should be ok right?

Joking aside. Clocks run faster on the moon, slower on the ISS,
so either clocks are synchronized back to earth seconds, or
they use some alternate standard reference time.

My expectation is that there will be some time keeping standard
that will ensure uniformity and that what we will want to capture
is whose seconds they are using to keep track of time, because
unix epoch on the moon has a different integer value than on
earth, and is likely off by multiple seconds.

I guess as long as the moon can participate in earth NTP it will
be ok? Otherwise I suspect that the NTP network standard for
the celestial body will be what we want to go with/record. I think
that will fit in the opening you have created for this in the syntax.

So instead of UTC it would be LTC (lunar coordinated time).
If you're getting logs from servers on the moon knowing that the
system clock was running on LTC and not UTC will be critical.

Some future engineer: "Leaving earth was a mistake."