Minutes of the Feb-20-1997 conference call on Stardust Interoperability
test results:
Present: Bob Pentecost
Lloyd Young/Matt
Ron Bergman
Stuart Rawley
Chuck Adams
Harry Lewis
Continued Review of MIB WALK
Picked up where we left off at prtInputDefalutCharSet.
General question about removing objects that do not appear to have been
considered useful by the majority of vendors. Caution was expressed. If
the Printer MIB developers thought they were needed, we may find ourselves
just putting them right back in later. Software applications might have
been written which desire to use these objects but are not represented
in this discussion. In general, we decided not to deprecate objects without
a good deal of further discussion.
(Editorial) If someone wants to propose deprecation of specific objects,
the above comment should not stifle the proposal. Any proposal as such
should indicate reason (i.e. not implemented by anyone...) and status
of deprecated object (in general, the OID becomes a place holder for
maintaining compatibility with RFC1759 - the resulting Group is not
"compressed" as a result).
prtInterpreterFeedAddressability - Wide variance here. One vendor
(mis)interpreted this to mean "logical page size" since it does
apply to the Interpreter and references Marker Units. Bill Wagner
to clarify.
(Editorial) Can anyone give an example (applying the "Jay" test - who's
gonna use it) how and why this object is to be used? Recall, this is
not the CURRENT Interpreter setting, but the MAX.
prtInterpreterLangFamilyLevel - Wide variability. Predicted by Ray
Lutz. Should we look into Ray's MIME registration for a benchmark?
Unresolved. Implies volunteers.
prtLocalizationLanguage - ISO 639 list these abbreviations in lower
case. Several vendors responded with Upper case. The printer MIB
seems to reference these in Upper case. Does anyone (i.e. mgt app)
care? If so, we need clarification. If there was a volunteer, I
didn't write the name down.
Marker Supplies Group - Some vendors implemented partial optional
groups, just picking the objects of interest and not responding
to other values. This seems to be justified if those implementations
do not claim compliance with the optional group (however, the
process for claiming compliance is undefined). Operationally,
however, if the response for any object in a Var Bind is NV, then
all responses will be set to NV. Another approach (recommended)
when you need specific values from an optional group is to respond
with zero length strings or NULL rather then NV. That way, the
values you wanted will still be returned.
prtMarkerSuppliesSupplyUnit - One vendor used Impressions. This is
OK since this has been added to the set of Supply Unit enums
referenced by Marker Supplies in the latest MIB.
prtMediaPathDescr - Wide variability here. Leave it.
Thus, we completed the review of the MIB Walk (note we did not discuss
any MIB-II objects).
Brief discussion on subUnitStatus. At a high level, it appears much
of the discrepancy is between Avail/Idle and Avail/Active. Is this
so bad if some printers really consider themselves Active vs. Idle
when they are indicating available? Agreed we need more (e-mail and
conf call) discussion on subUnitStatus.
Chuck Adams has provided his proposal regarding the top 25 alerts. This
(perhaps along with subUnitStatus) will be the topic of the next call.
Next call was moved from 3/4 (as established while at Stardust) to
3/5 11-1pm Pacific. Please discuss Status and Alerts on the wire. This
call will be to reach FINAL consensus on the alert table.
Harry Lewis - IBM Printing Systems