To: Robert Haas <robertmhaas@gmail.com>
cc: Kevin Grittner <Kevin.Grittner@wicourts.gov>, cousinflo@free.fr,
	pgsql-docs@postgresql.org
Subject: Re: Date/Time Types : internals 
In-reply-to: <11541.1335465978@sss.pgh.pa.us> 
References: <4F8F0F71.20500@free.fr>
 <4F8ED21402000025000470FA@gw.wicourts.gov>
	<9773.1334780084@sss.pgh.pa.us>
	<CA+TgmoZ11Ac_eyStw2bEh-Hk7KYqujAAES=WU_hDQvQHCH=JUQ@mail.gmail.com>
	<11541.1335465978@sss.pgh.pa.us>
Comments: In-reply-to Tom Lane <tgl@sss.pgh.pa.us>
	message dated "Thu, 26 Apr 2012 14:46:18 -0400"
Date: Thu, 26 Apr 2012 17:15:19 -0400
Message-ID: <14469.1335474919@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

I wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> So which calendar are we using, Julian or Gregorian?

> Gregorian.  I fail to see anyplace that suggests we use the Julian
> calendar.  We do understand the Julian date system, which is something
> entirely different with a confusingly similar name.  (Julian dates are
> just a count of days since a reference point in 4713 BC.)

On further reflection I wonder whether we shouldn't just get rid of
"8.5.6. Internals", which is at best rather inappropriately placed.
A person reading this portion of the docs probably couldn't care less
whether we use Julian-date calculations internally.  It might be
appropriate to mention that in appendix B.4, where there's an actual
explanation of what Julian dates are; but not here.

What I think we should have here is an externally-facing specification,
that is state that we follow the Gregorian calendar rules even for years
before Pope Gregory's edict, and then reference B.4 for more info.

			regards, tom lane