This sounds like the right approach to me. I wouldn't expect "arbitrary"
changes to be proposed. Only, like you indicated, changes based on interop
testing.
Harry Lewis - IBM Printing Systems
harryl at us.ibm.com
"Hastings, Tom N" <hastings at cp10.es.xerox.com> on 01/08/99 04:45:20 PM
To: Ron Bergman <rbergma at dpc.com>, Harry Lewis/Boulder/IBM
cc: jmp <jmp at pwg.org>
Subject: RE: JMP> Ballot on Xerox Job MIB comments
We (Xerox) could go along with removing the mirror table as well, as long
as
we can get some agreement in the JMP WG that the internal PWG version 2.0
spec is agreed to for the two new groups:
the OPTIONAL mirror table
the MANDATORY new system group
Then implementer's who want or need to can build to the 2.0 spec for the
interoperability test next summer with some assurance that these groups
won't be changed arbitrarily, but only possibility from the results of the
interoperability testing. The best way to ascertain that we have agreement
on these two new groups would be to do a Last Call on them, but keep the
resulting spec as a PWG Working Draft. I.e., not forward the version 2.0
spec to the IETF until after the interoperability test (and any updates
that
are found to be needed).
How does that sound?
Thanks,
Tom
>-----Original Message-----
>From: Ron Bergman [mailto:rbergma at dpc.com]
>Sent: Friday, January 08, 1999 07:37
>To: harryl at us.ibm.com>Cc: Hastings, Tom N; jmp
>Subject: RE: JMP> Ballot on Xerox Job MIB comments
>>>Since the Mirror table is optional, I do not have a real
>problem with it
>either added or removed. But Harry does have a good point and
>it probably
>best that it not be included in version 1.0. This would also
>be a better
>position for us with the IETF.
>>Anyone else have comments?
>> Ron Bergman
> Dataproducts Corp.
>>>On Fri, 8 Jan 1999 harryl at us.ibm.com wrote:
>>>>>>> Tom. This is a much better proposal. I only have one concern.
>>>> The System Group was supposed to enable applications to
>determine whether
>> or not the (optional) mirror table was supported. Perhaps
>1.0 should just
>> be the first PWG standard Job MIB (as agreed initially) and the (new)
>> internal version (basis for next external) should have the
>mirror table and
>> Systems Group.
>>>> Harry Lewis - IBM Printing Systems
>>harryl at us.ibm.com>>>>>>>> "Hastings, Tom N" <hastings at cp10.es.xerox.com> on 01/07/99
>07:35:31 PM
>>>> To: jmp <jmp at pwg.org>
>> cc: Harry Lewis/Boulder/IBM, Ron Bergman <rbergma at dpc.com>
>> Subject: RE: JMP> Ballot on Xerox Job MIB comments
>>>>>>>>>>>> Harry,
>>>> I think that we (Xerox) can agree to forwarding the JMP MIB
>to the IETF now
>> with the System Group being entirely removed as you and Ron
>have suggested.
>>>> We can also agree that we should make the second EXTERNALLY published
>> version of the MIB AFTER an interoperability test and any
>updates needed
>> from the interpretability test, as you proposed. Currently, the
>> interoperability test is anticipated to be some time next summer.
>>>> However, we need to be able to build something with the
>System Group in it
>> NOW, rather than waiting until after the interoperability test (next
>> summer). So could we also produce NOW an INTERNAL PWG
>working draft with
>> version 2.0 that does have the MANDATORY System Group in it?
> This internal
>> PWG working draft would NOT be forwarded to the IETF for
>publication as an
>> Information RFC until after the interoperability test and
>any updates that
>> the interoperability tests show were added.
>>>> Thus the interoperability test would be for both version 1.0 products
>> without the System Group and for version 2.0 products with
>the MANDATORY
>> System Group. It is straightforward for any testing client
>software to
>> tell
>> the difference between the two by just querying the System Group and
>> getting
>> back no such object or not.
>>>> So for process as a result of the PWG Last Call, after Maui
>I would forward
>> the current JMP MIB document with the OPTIONAL Mirror Table
>but without the
>> System Group to the IETF as an Internet-Draft. Ron prefers
>that we call it
>> 1.0 (with a January 1999 date), rather than 1.3 so that we
>don't confuse
>> the
>> IETF. They have only seen the 1.0 spec from February 1998.
>If it looks
>> ok,
>> Ron will then request the RFC Editor to publish it as an
>Informational RFC.
>>>> Secondly, next month I would take the current 1.3 version with the
>> MANDATORY
>> System Group and rename it to be version 2.0. Then
>implementers that want
>> to can implement it for the Interoperability Tests.
>However, it would NOT
>> be forwarded to the IETF until after the interoperability test.
>> Implementers at the interoperability test could bring 1.0 or
>2.0 conformant
>> products.
>>>> If new attributes are proposed and agreed to, I can add them
>to both the
>> 1.0
>> and 2.0 specs until we make version 2.0 an EXTERNAL spec as
>published by
>> the
>> IETF and an Informational RFC. After that we only need to
>keep maintaining
>> the 2.0 spec.
>>>> How does that sound?
>>>> Thanks,
>> Tom
>>>> >-----Original Message-----
>> >From: harryl at us.ibm.com [mailto:harryl at us.ibm.com]
>> >Sent: Thursday, January 07, 1999 10:56
>> >To: jmp at pwg.org>> >Subject: RE: JMP> Ballot on Xerox Job MIB comments
>> >
>> >
>> >
>> >
>> >A few days ago, I stated my lack of support in Ron's ballot
>request...
>> >
>> >>I am requesting a formal ballot regarding the recent request
>> >>from Xerox to expand the object jmSystemAttributeSupport to
>> >three objects.
>> >
>> >I should elaborate that I do not object to the particular
>> >jmSystemAttributeSupport changes proposed by Xerox but to the
>> >entire notion
>> >of submitting an updated version of the Job MIB to the IETF if
>> >that version
>> >has new mandatory objects which were not part of the agreed
>> >PWG standard,
>> >closed months ago, and which is the basis of all prototypes and
>> >implementations thus far.
>> >
>> >The existing PWG Job MIB standard (v1) has been intended for
>> >submission to
>> >the IETF for consideration as an RFC. (It has been debated
>whether this
>> >would
>> >be standards track with the Printer MIB, Informational,
>> >Experimental etc...
>> >but, nonetheless, the Job MIB would be submitted).
>> >
>> >Recently, there have been proposals, resulting from
>> >prototypes, which have
>> >been discussed and agreed to be improvements. I agreed to the
>> >addition of
>> >an optional "mirror table" as part of the IETF submission
>> >because it was
>> >optional. Later there was the notion of a "system table" to
>> >differentiate
>> >Job MIB versions. This would be a new mandatory table.
>> >Finally, there are
>> >proposed modifications in the details of the objects in this table.
>> >
>> >While I welcome and have been very willing to discuss spec
>> >changes to the
>> >standard, I feel the PWG process recognizes a v1 PWG Job MIB
>> >standard which
>> >is in PWG maintenance mode. I believe this should be
>> >considered the most
>> >stable version and the one that is published to the IETF. A
>> >SECOND updated
>> >PWG Job MIB standard is entirely feasible but I don't believe
>> >it is valid
>> >to publish new mandatary objects in the first EXTERNAL
>version of the
>> >standard.
>> >
>> >I highly recommend at least one coordinated interoperability
>> >test prior to
>> >declaring the second major version of the Job MIB standard.
>> >
>> >Harry Lewis - IBM Printing Systems
>> >harryl at us.ibm.com>> >
>> >
>>>>>>>>>