Re: [PATCH] Cleanup of GUC units code - Mailing list pgsql-hackers

From Alvaro Herrera
Subject Re: [PATCH] Cleanup of GUC units code
Date
Msg-id 20080904131956.GA5786@alvh.no-ip.org
Whole thread Raw
In response to Re: [PATCH] Cleanup of GUC units code  (Hannu Krosing <hannu@krosing.net>)
List pgsql-hackers
Hannu Krosing escribió:
> On Wed, 2008-09-03 at 20:01 -0400, Alvaro Herrera wrote:

> > Yes there is --- it's the SI.
> > 
> > http://en.wikipedia.org/wiki/SI#SI_writing_style
> > 
> > I don't know about it being "evil" and punishment, but it's wrong.
> 
> SI defines decimal-based prefixes, where k = kilo = 1000, so our current
> conf use is also wrong.

Actually, this has been a moving target.  For a certain length of time,
some standards did accept that k meant 1024 "in computing context"; see

http://en.wikipedia.org/wiki/Binary_prefix

So we're not _absolutely_ wrong here; at least not until KiB are more
widely accepted and kB more widely refused to mean 1024 bytes.  The
relevant standard has been published just this year by ISO.

http://en.wikipedia.org/wiki/ISO/IEC_80000#Binary_prefixes

So this is new territory, whereas case-sensitivity of prefixes and unit
abbreviations has existed for decades.

-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support


pgsql-hackers by date:

Previous
From: "Tobias Anstett"
Date:
Subject: xml2 vs XMLFunctions
Next
From: Andrew Sullivan
Date:
Subject: Re: [PATCH] Cleanup of GUC units code