Re: [art] [saag] Date formats: RFC3339 vs. RFC 5322

[John C Klensin <[e-mail]>]

--On Tuesday, April 13, 2021 19:00 +0000 "Eliot Lear (elear)"
<elear=[e-mail]> wrote:

> The question is whether you need something that is easy to
> parse or something that is human readable and can be
> localized.  It SEEMs that this draft is intended to be human
> readable, and so 5322 doesn't seem out of bounds.

I suggest that even for reading by humans in 2021 --as distinct
from 1982 (RFC 822) or 1977 (RFC 733, which used day-month-year
ordering)-- the 5322 dates are not easy to understand and use...
at least unless one is an English speaker on this side of the
pond.  It was quite wise at the time to spell out the month
name, thereby eliminating the ambiguity associated with, e.g.,
5/10/1977, but still bad news for someone who might think the
fourth month in the Gregorian calendar is, e.g., апреля,
أبريل , or 四月.

So I would argue that, for new protocols or data structures in
this increasingly global/ international Internet, and even for
elements visible to humans, sticking as close to ISO 8601 as
possible (with minimal profiling) is the Right Thing to Do.
Much too late now to change the 822/5322 format, turning
supplemental protocols for email into a gray area, but, for new
work, ISO 8601 formats are not just easier to parse but easier
to understand globally and in an unambiguous way.

Just my opinion, of course.


    john