Re: what ever happened to npp?

Steve Smith <steve@next-s.lanl.gov> Wed, 02 December 1992 19:47 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa17871; 2 Dec 92 14:47 EST
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa17867; 2 Dec 92 14:47 EST
Received: from inet-gw-2.pa.dec.com by CNRI.Reston.VA.US id aa21821; 2 Dec 92 14:48 EST
Received: by inet-gw-2.pa.dec.com; id AA21077; Wed, 2 Dec 92 11:40:18 -0800
Received: by nsl.pa.dec.com; id AA21278; Wed, 2 Dec 92 11:20:12 -0800
Received: by nsl.pa.dec.com; id AA21274; Wed, 2 Dec 92 11:20:11 -0800
Received: by inet-gw-2.pa.dec.com; id AA20008; Wed, 2 Dec 92 11:20:09 -0800
Received: from next-s.lanl.gov by p.lanl.gov (5.65/1.14) id AA17333; Wed, 2 Dec 92 12:20:05 -0700
Received: by next-s.lanl.gov (NeXT-1.0 (From Sendmail 5.52)/5.17) id AA04512; Wed, 2 Dec 92 12:17:13 GMT-0700
Date: Wed, 02 Dec 1992 12:17:13 -0000
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Steve Smith <steve@next-s.lanl.gov>
Message-Id: <9212021917.AA04512@next-s.lanl.gov>
Received: by NeXT Mailer (1.62)
To: print-wg@pa.dec.com
Subject: Re: what ever happened to npp?

Brad Clements writes -

> The last I knew, the big open question was `what protocol are we
> supposed to be defining?'

My charter includes the entire range from submission to delivery of output
so I would like to encourage or participate in all of the above.

However, for the sake of getting things moving, I suggest either describing
in more detail the entire model, Brad's following sketch is a good place to
start -

>
>Consider three or four possibilities:
>
>  client ----> server/spooler  ---> server/spooler ---> printdevice --> printer
>
>Protocol     		 Is
>
>Client to Server/Spool	Communication between client (PC, process, etc) to spooler
>			or other server device
>
>Server/Spooler to same	Possible spoolers forwarding to another spooler
>
>Server/Spooler to printdev   Perhaps defining how a spooler/server should 

>			     communicate with a smart print device, ie:
>			     an HP Lan Card, Microplex box, etc
>
>Printdevice to printer		Something like the NPA spec, or NPA itself
>
>---
>
>There are four possible protocol challenges shown above (not counting
>Client to Printdevice, bypassing spooler). 

>
>Which protocol (or all) are we supposed to be working on?

And can we get some work started, at least some discussion?

Areas:
	Page Description Languages
		What "document types"/"file formats" are important, does it matter?
	Existing Protocols
		What are the existing protocols, how do they fit current
		environments, do we care about working with current solutions?
	Media Extensions 

		Printing is no longer black characters on while 8.5x11 paper.
		There are emerging lots of other media types of interest and
		options among the types.  Do we care?
	Accounting
		Accounting is not an issue for everyone all the time, however,
		without accounting, sharing between administrative units is
		impossible.  Does anyone else care?
	Authentication/Registration/Access
		How do we manage resource usage?  Similar issues to above.
		Do we care?
	Distribution
		Most people just want a way to print from their favorite
		platform to their favorite printing device but others care
		about providing and using "print services" which includes
		the requirement for returning the product to the customer.
		For very ad-hoc situations, most people want at least a
		banner page.  Does anybody else care?
	Prioritization
		Print devices are usually low-bandwidth with lots of 

		very sporadic use.  How to manage priority during 

		conditions of contention?
	Scheduling
		Most people think they want ASAP service.  Unfortunately,
		that is contradictory with low-bandwidth, low availability
		services (printers go down, run out of media, jam).  A
		"scheduled" service is important in providing consistency.
		"here's my job, can I have it in less than an hour?" 

		"How soon can I have this job completed?  With what probability?"
	Inquiry
		"here's my job"
		"how soon might I expect it back?"
		"how is that job of mine doing?"
		"is this resource available?"
		"how likely is it that this resource is going to continue to
		 be available?"
		 et cetera
		 

There are many more issues involved here not explicitely mentioned.  Is
anyone else interested at this level here?

Steve Smith
Los Alamos National Laboratory