CURRENT_MEETING_REPORT_


Reported by Glenn Trewitt/DEC

NPP Minutes

We reviewed the goals of the Working Group (problems to be solved) and
looked at how the problem could be partitioned into smaller subproblems.
Three subproblems stood out:


  1. Wire Protocol
     There is an immediate need among some vendors (notably those who
     make terminal servers) to have a standard that they can implement
     that provides the capability to create a TCP connection to one (of
     many) hardware ``byte-stream'' interfaces (either serial or
     parallel).  The path must be capable of being 8-bit clean.  It
     would be a good thing for the protocol to provide a mechanism for
     supporting ``rotary groups'' for groups of printers.
  2. Printer Access
     Communication to a printer that has some sort of direct network
     connection.  Presumably, the printer has its own IP address.  The
     source of the print job is unspecified -- it might just be
     someone's PC, or a full-blown print manager and spooler.  There are
     many requirements here:  authentication, accounting, capability
     negotiation (what page description languages are supported, paper
     sizes, special features, etc.), etc.
  3. Job Submission Communication from someone who has a document to be
     printed to a printing manager/spooler.  The current popular example
     of this is the lpr/lpd protocol, which most people seem to feel is
     inadaquate for more complex printing environments.


Decisions and Action Items

We quickly decided that problem (1), the wire protocol, deserved a
general solution in a broader context, since there are a number of other
applications that require it, such as data collection, modem pools, and
``milking-machine'' concentrators.  Russ Hobby agreed, and suggested
that we take that protocol project to the Telnet Working Group.  Bill
Westfield (cisco) agreed to do this.

In response to problem (2), Ajay Kachrani (DEC) proposed that we adopt
the ``Printer Access Protocol''.  This is the protocol used to
communicate with Digital's networked PostScript printers.  He handed out
a description of it, which I will make available via anonymous FTP.

Richard Hart (DEC) has proposed that we address problem (3) by adopting
the Palladium printing archictecture, developed at MIT's project Athena.

I suggested that we contact Adobe Systems, to see what input they could

                                   1






provide about printing architectures.  I have since talked to Carl
Orthlieb and Sherri Nichols at Adobe.  They will be providing four
documents about Adobe's model of printing architecture, and will
participate in our activities as well.

I will shortly be setting up a directory (accessible via anonymous FTP)
containing the Adobe documents, some relevant DEC Western Research Lab
technical notes, as well as the documents about the two proposals (PAP
and Palladium).

Other Activities

Leo McLaughlin and Robert Knight finished RFC 1179, documenting the
Berkeley Line Printer Daemon Protocol.

Attendees


Philip Budne             phil@shiva.com
Anthony Chung            anthony@hls.com
Richard Hart             hart@decvax.dec.com
Russell Hobby            rdhobby@ucdavis.edu
Ajay Kachrani            kachrani%regent.dec@decwrl.dec.com
Stuart Lynne             sl@wimsey.bc.ca
Robert Morgan            morgan@jessica.stanford.edu
Glenn Trewitt            trewitt@nsl.dec.com
Bill Westfield           billw@cisco.com



                                   2