From owner-freebsd-questions Sat Sep 25 15:44: 3 1999
Delivered-To: freebsd-questions@freebsd.org
Received: from scientia.demon.co.uk (scientia.demon.co.uk [212.228.14.13])
by hub.freebsd.org (Postfix) with ESMTP id B346C15377
for <freebsd-questions@freebsd.org>; Sat, 25 Sep 1999 15:43:44 -0700 (PDT)
(envelope-from ben@scientia.demon.co.uk)
Received: from lithium.scientia.demon.co.uk ([192.168.0.3] ident=exim)
by scientia.demon.co.uk with esmtp (Exim 3.032 #1)
id 11UzfC-000Brx-00; Sat, 25 Sep 1999 22:47:26 +0100
Received: (from ben) by lithium.scientia.demon.co.uk (Exim 3.032 #1)
id 11UzfC-0000PD-00; Sat, 25 Sep 1999 22:47:26 +0100
Date: Sat, 25 Sep 1999 22:47:26 +0100
From: Ben Smithurst <ben@scientia.demon.co.uk>
To: Arash Farahmand <afarah@mictlan.sfsu.edu>
Cc: Greg Lehey <grog@lemis.com>,
"Jon O." <netcmd@networkcommand.com>, freebsd-questions@FreeBSD.ORG
Subject: Re: Time and history
Message-ID: <19990925224726.B1470@lithium.scientia.demon.co.uk>
References: <19990925173053.F54407@freebie.lemis.com> <Pine.SOL.3.95.990925081211.9390B-100000@xolotl.sfsu.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.95.6i
In-Reply-To: <Pine.SOL.3.95.990925081211.9390B-100000@xolotl.sfsu.edu>
Sender: owner-freebsd-questions@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG
Arash Farahmand wrote:
> Unix time counter will reset sometime in the year 2037 (again, please
> correct me if this is wrong). Although this is relatively far in the
> future,
People in 1963 were probably saying that about the year 2000, and look
at the mess people have got into now.
> is there any plan to modify the time counter on Unix machines to cope
> with the next time bug? ;-)
With any luck, we'll all be using 64-bit machines by then, and this
won't be an issue, since time_t can be made a 64-bit value. I make 2^63
seconds to be about 288 billion years, which is almost certainly way
beyond the lifetime of our Sun and/or Earth. I can't see Unix's time
mechanism changing from secs since 00:00:00 1970-01-01, it would confuse
too many people.
time_t and friends could probably be made a 64-bit value even on 32-bit
machines, but I don't know what this would do to performance. (How do
32-bit machines to 64-bit math anyway? I guess I should find out.)
Incidentally, someone the other day mentioned in a newsgroup that when
the seconds-since-1970 value hits 10 digits (this isn't very far away,
only a couple of years I think), this might confuse some software
which has expected 9 digit values (which has been the case since about
1973). I wouldn't be at all surprised if such software existed, it's no
more or less stupid than storing two digit years.
--
Ben Smithurst | PGP: 0x99392F7D
ben@scientia.demon.co.uk | key available from keyservers and
| ben+pgp@scientia.demon.co.uk
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message