IPP Mail Archive: RE: IPP> OPS - Comments on Set2 and Set3 operations: IPP Printer

RE: IPP> OPS - Comments on Set2 and Set3 operations: IPP Printer

harryl@us.ibm.com
Tue, 7 Dec 1999 22:49:00 -0700

Actually, during our conversation, I suggested to Tom that we "skew" the
terminology from quiesce and shutdown to shutdown and poweroff. I agree,
quiesce is beyond the 8th grade vocabulary. Anyway, Ira's dissertation
actually shows the intent of quiesce is correct. Dormant is a good
visualization. So, if we don't like the word quiesce, may I suggest
shutdown in it's place with the understanding that shutdown is all but
powered off?

Not sure I buy the "when breaks security infrastructure" argument (I'll
need to understand that point better). It's really meant to be fairly
concrete and practical. What are the natural boundary's that make sense
for printing. Sheet, copy, job, queue or buffer of jobs... I accept that,
overall, it adds complexity, but to just say "orderly" doesn't provide
much predictability. My printers orderly might be next page... yours might
be end of (large) job. Operator might have really wished for copy
boundary.

Harry Lewis
IBM Printing Systems

"McDonald, Ira" <imcdonald@sharplabs.com>
Sent by: owner-ipp@pwg.org
12/07/99 07:07 PM

To: "'Manros, Carl-Uno B'" <cmanros@cp10.es.xerox.com>,
"Mcdonald, Ira"
<imcdonal@sdsp.mc.xerox.com>, "Hastings, Tom"
<hastings@cp10.es.xerox.com>, IETF-IPP <ipp@pwg.org>
cc: "Manros, Carl-Uno" <manros@cp10.es.xerox.com>
Subject: RE: IPP> OPS - Comments on Set2 and Set3
operations: IPP Printer versus D
evice object and operation semantics

Hi Carl-Uno,

The verb 'quiesce' is used in both IBM's SNA (Systems Network
Architecture) and ISO's OSI upper layer protocols. The adjective
'quiescent' is of course derived from the verb (where else would
it come from?). IBM's SNA distinquishes the symmetric, graceful
Quiesce protocol from the assymetric, immediate Shutdown protocol.

The use of illiterate, inadequate spell checkers rather than real
dictionaries is a common problem in our industry.

I just had my wife Nancy read out the definition of the verb
'quiesce' to me from my copy of Chambers Concise 20th Century
Dictionary (1985):

1) To become dormant.

And in the IBM Dictionary of Computing (1994), the verb 'quiesce'
is defined:

1) To end a process by allowing opoerations to complete normally.
2) In a VTAM [SNA] application program, for one node to stop
another node from sending synchronous-flow messages.

Nonetheless, using 'quiesce' is a bad idea in a standard (IETF IPP)
that must be clear to speakers of English as a second language.

The serious problem with the 'when' parameter of 'Pause-xxx' or
'Shutdown-xxx' is that it means access control lists (and filtering
by firewalls) cannot be at the granularity of operations. This
generally breaks most existing security infrastructure.

Rather than introducing the 'when' parameter, I would suggest that
instead we infix 'Immediately' and 'Orderly' (used in IETF's TCP
for the same distinction) into the IPP operation names:

Pause-[Immediately|Orderly]-[Printer|Device]

Shutdown-[Immediately|Orderly]-[Printer|Device]

Or perhaps 'Now' versus 'Soon' (short and sweet)?

Tom asked me what verb conveyed this distinction in existing
protocol suites. Unfortunately, I suggested the only one I'd
ever seen used that way...

Cheers,
- Ira McDonald (outside consultant at Sharp Labs America)
High North Inc

-----Original Message-----
From: Manros, Carl-Uno B [mailto:cmanros@cp10.es.xerox.com]
Sent: Tuesday, December 07, 1999 4:56 PM
To: Mcdonald, Ira; Hastings, Tom; IETF-IPP
Cc: Manros, Carl-Uno
Subject: FW: IPP> OPS - Comments on Set2 and Set3 operations: IPP
Printer versus D evice object and operation semantics

Ira,

The spell checker tells me there is no such word as "quiesce".

I must admit that I also have problems with "quiescent", it
is not exactly an everyday word, and I suspect that any non-native
English person would:

1) Have to look it up in a dictionary to understand the meaning

2) Be totally clueless about how to pronounce it

I am speaking from personal experience here, revealing my ignorance.

Not recommended for an international standard if you ask me...

Can't we find a synonym word somehow?

Carl-Uno

-----Original Message-----
From: McDonald, Ira [mailto:imcdonald@sharplabs.com]
Sent: Tuesday, December 07, 1999 4:17 PM
To: 'Hastings, Tom N'; ipp
Subject: RE: IPP> OPS - Comments on Set2 and Set3 operations: IPP
Printer versus D evice object and operation semantics

One minor nit:

'devices-quiescent-failed' should have been
'devices-quiesce-failed' (the verb, not the adjective)