Re: nfs4time

New Message Reply About this list Date view Thread view Subject view Author view Attachment view

From: Dickon Hood (dickon-ml@fluff.org)
Date: 07/23/98-01:35:30 PM Z


Date: Thu, 23 Jul 1998 19:35:30 +0100
From: Dickon Hood <dickon-ml@fluff.org>
Subject: Re: nfs4time
Message-ID: <a35e486a48%dickon-ml@splurge.fluff.org>

In message <199807231809.LAA01351@emf.emf.net> you wrote:

: I think timestamps should be passed as ascii strings, representing absolute
: calendar/clock time in some easy-to-parse format:

: 	1969-12-31-23:59:59.999999831

: Typical clients can convert that, discarding precision.  Best possible
: clients can use all the bits.  Typical servers can generate that from
: a reduced-precision approximation.

: It never expires, refers to no epoch, and is arbitrarily precise (server's
: determination).

Erm... yuk.  The epoch that refers to is 0001-01-01, anyway, isn't it?  NFS
is a binary protocol; why make something arbitrarily ASCII?  Or are you
advocating ASCII permissions, ASCII filesystem IDs, etc., etc., as well?

BTW, anyone else think that signed time from 2000-01-01 is a fun idea? :-)

-- 
Dickon Hood

I've now managed to find my .signature file, normal service
will be resumed when Connex South Central get their act together.
We apologise for the inconvenience in the mean time.


New Message Reply About this list Date view Thread view Subject view Author view Attachment view

This archive was generated by hypermail 2.1.2 : 03/04/05-01:46:02 AM Z CST