Re: [Nmh-workers] date math

[Ken Hornstein]

>It actually is hard, there's no one-one mapping between numeric offsets
>and timezone names.abbreviations - you alo need to know the origin of the
>message (of course, if you've converted to local time, and are just
>printing the local offset, then it is relatively easy).
>
>For example, if you see a message with -0500 as the numeric offset, is
>that US Eastern Standard time, or US Central Daylight time?   Or
>might it just be from Chile (which also uses -0400 and -0500, but
>given it in the other hemisphere, at different times of year).

In the EST/CDT division, existing routines already "know" if a
particular date falls under DST so there is no ambiguity there.  But
like you said, it's hard to know if the sender is in the "wrong"
hemisphere :-)

>There's a lot of work required to get this particularly useless piece of
>information - just stick with numeric (except possibly when you are 
>deliberately printing the local equivalent - which also makes it more
>clear to the viewer whether it is an original or a converted time they're
>seeing).

That's fair enough, although the original MH developers did not think it
was useless; the code to get this information existed back then.  But
if people are fine jettisoning the last vestiges of it, then I'm fine
as well.

>On the other question, of defaults, I quite like the exmh method,  for your
>message, for example, the Date: header is shown as ...
>
>       Date: Mon, 15 Dec 2014 12:56:46 -0500 (Tue 00:56 ICT)

That ... does not seem to be something that exmh does, actually, at
least not out of the box AFAICT.  There's no code in exmh I can find
which rewrites the date header for you.  I'm an exmh user, and your
message had the date header:

        Date: Tue, 16 Dec 2014 02:49:59 +0700

As far as I can tell, exmh just displays the date header directly
as it was in the message.  Are you sure your local MTA isn't doing that
for you?

--Ken