PMP Mail Archive: PMP> Considering "The Best Spec" vs. "Time to Market"

PMP> Considering "The Best Spec" vs. "Time to Market"

JK Martin (jkm@underscore.com)
Fri, 18 Jul 1997 13:08:27 -0400 (EDT)

Ira,

It's very clear that you're quite frustrated with the PMP group's
rejection of the many localization and internationalization proposals
you and other Xerox folks have submitted recently.

All of us over the last four years have had our share of similar kinds
of frustrations, so you're in good company. However, this kind of
statement might be just a bit too extreme, IMHO:

> If none of you folks want to play in the international
> marketplace or none of you care if you violate the conformance
> rules stated in the Printer MIB, Xerox and Fuji Xerox and
> Rank Xerox do care, very much.

Please understand that the recent decisions have nothing to do with
the desire of the PMP group to address international markets or
conformance. Rather, we set a deadline for ourselves (quite a while
ago) to get the best possible spec completed with the time we had
available.

I am personally hoping we can start up the next version of the Printer
MIB, whereby L10N and I18N can serve as the primary motivations for
the new version.

We realize you have published considerable skepticism of the possibility
for another version of the Printer MIB, but please try to be positive
and help the effort along. Hey, maybe you can even chair the new
IETF working group!

I (and others) wish L10N/I18N could have been done in time for the
completion of the latest Printer MIB draft, but it couldn't.

...jay

----- Begin Included Message -----

Date: Thu, 17 Jul 1997 20:41:35 PDT
From: imcdonal@eso.mc.xerox.com (Ira Mcdonald x10962)
To: pmp@pwg.org
Subject: Re: PMP> ISSUE: OCTET STRING ...

Jay and others,

This is a last gasp, but...

NVT ASCII from IETF Telnet (the mandatory ONLY code set underlying
'DisplayString' with MANDATORY semantics for CR and LF - unlike
Dave Kellerman's inaccurate comment about LF as just a delimiter)

does NOT permit the use of EVEN UTF-7 (which uses the same code
points with a shift sequence to encode a DIFFERENT character
set, Unicode UCS-2 from ISO 10646)

If you folks leave 'prtChannelInformation' as 'DisplayString'
then Scott Isaacson may NOT put Unicode (no matter how it's
folded into UTF-7 or UTF-8) in for NetWare channel types -
Jay, Scott made clear on a telecon I attended months ago
that NetWare NDS uses Unicode and NOT ASCII names for
PServers and RPrinters - there's your example of 'someone
who uses another code set.

Conformance in SNMP MIBs means just that - literal, exact
conformance - not 'close enough' - NVT ASCII does NOT
permit the use of UTF-7 (same code points, but different
displayed graphical characters, per RFC 2130).

If none of you folks want to play in the international
marketplace or none of you care if you violate the conformance
rules stated in the Printer MIB, Xerox and Fuji Xerox and
Rank Xerox do care, very much.

Cheers,
- Ira McDonald (outside consultant at Xerox)
High North Inc
PO Box 221
Grand Marais, MI 49839
906-494-2434

----- End Included Message -----