Ira,
OK. It sounds like the best path.
Thanks,
Bill Wagner
-----Original Message-----
From: McDonald, Ira [mailto:imcdonald@sharplabs.com]
Sent: Thursday, June 03, 2004 2:39 PM
To: 'Harry Lewis'; Wagner,William
Cc: McDonald, Ira; Zehler, Peter; wbmm@pwg.org
Subject: RE: WBMM> Re: Need quick decisions on schema changes
Hi,
I'm delighted to define Resource object in WIMS spec - it's already been defined
in IPP 4 years ago - we've reviewed my Resource schema (and requested only
slight changes) - I can write the WIMS appendix.
I'm also delighted to define Subscription object in WIMS spec - PSI doesn't need
it, but WIMS does - WIMS already has to define Alert object (see the schema),
so this is consistent - I can write the WIMS appendices.
The difficulty with the Resource events (and any other non-Printer events) is that
for some months (while it's approved and adopted and finally posted) the Events
schema MUST remain unchanged. A way forward is to let PSI also formalize
the Events schema (Printer-only) and publish it. Meanwhile, continue to use
an expanded version of Events schema in WIMS directory for WIMS development.
I dislike removing the WIMS Management and Admin actions from either the
Schedule schema or the WIMS spec. They're well-specified and work fine now.
In order to have even Counters for non-print services in an XML schema, WIMS
is going to have to define at least System object (and probably Service) and WIMS
can't wait for some hypothetical future MFP WG to define them.
My two cents,
- Ira
Ira McDonald (Musician / Software Architect)
Blue Roof Music / High North Inc
PO Box 221 Grand Marais, MI 49839
phone: +1-906-494-2434
email: imcdonald@sharplabs.com
-----Original Message-----
From: Harry Lewis [mailto:harryl@us.ibm.com]
Sent: Thursday, June 03, 2004 2:48 AM
To: Wagner,William
Cc: McDonald, Ira; Zehler, Peter; wbmm@pwg.org
Subject: RE: WBMM> Re: Need quick decisions on schema changes
I agree with Bill's points and I believe they show us a way forward. I share Bill's concern that we need to avoid getting tangled in our own shorts (that is the way Bill put it... isn't it ;-). I believe what still needs to be sorted out is how much (or little) skeleton imaging system model do we need (or can we get by with) to support the (appropriately) limited initial WIMS scope yet not block or disrupt a future SM expansion embracing the MFP, entirely.
----------------------------------------------
Harry Lewis
Chairman - IEEE-ISTO Printer Working Group
http://www.pwg.org
IBM Printing Systems
http://www.ibm.com/printers
303-924-5337
----------------------------------------------
"Wagner,William" <WWagner@NetSilicon.com>
Sent by: owner-wbmm@pwg.org
06/03/2004 12:07 AM
To
"McDonald, Ira" <imcdonald@sharplabs.com>, "Zehler, Peter" <PZehler@crt.xerox.com>
cc
<wbmm@pwg.org>
Subject
RE: WBMM> Re: Need quick decisions on schema changes
I agree with Harry and Pete on most things, and I will admit to not following how your argument below applies to their objections (nor, indeed, many of your other undoubtedly logical relationships)
Although having everything follow a grand scheme is desirable, the effect appears to introduce so much interdependency that everything requires that everything else be done, thereby defeating the ability to complete anything. Since PSI deals only with printers but needs alerts done, why can it define just printer alerts (and perhaps general alerts)? Alerts for WIMS would be the combination of printer alerts, scanner alerts, fax alerts, resource alerts etc, as these become defined. Dropping the register for alerts capability because the only alerts we can currently support are printer related seems undesirable.
With respect to "Does WIMS WG accept the need to model Resource?", it is unclear to me what other group would do this. Although I would not put this on high priority, I would prefer not to loose Resource. Again, since we may not at this time fully understand all the aspects of Resource, can we not work with a skeleton model, filling it out later?
I understand the notion that the current semantic model is only for printers, and that elements belonging to a non-printer part of an MFP need to be in some other part of the semantic model. As such, forcing them into WIMS at this time may upset the future logical model if it cannot be remove from WIMS in favor of a reference to a more general semantic model at a later time.
Indeed, the compromise forcing these cuts was in response to the understanding that we needed your skeleton imaging device model filled out before we could use it, and that it was unclear when (and if) that would be done. But it seems that we need some interim way of handling references, realizing that they may change in detail and completeness.
Anyway, it would seem that, insofar as the schema are informational adjuncts to the spec, the changes you identify must also be reflected in the spec. I would, however, like to wait until the smoke clears before making those changes.
Thanks for your continuing efforts.
Bill Wagner
-----Original Message-----
From: McDonald, Ira [mailto:imcdonald@sharplabs.com]
Sent: Wednesday, June 02, 2004 4:20 PM
To: 'Zehler, Peter'; McDonald, Ira
Cc: 'wbmm@pwg.org'; Wagner,William
Subject: RE: WBMM> Re: Need quick decisions on schema changes
Hi,
For PSI to advance, the Events spec (psievents10) and the Events
schema MUST be formally approved for addition to the next version
of the Semantic Model.
Because there is no such thing as a Resource or a _Subscription_
defined in any standards-track IETF or PWG approved spec,
there CANNOT be any Resource operations or events in WIMS.
And more cogently, there CANNOT be any Subscription object
defined in the Alert schema (so there is no source for the element
bindings of most Alert types). So RegisterForAlerts doesn't work.
This just gets worse...
----
The PWG Semantic Model element PrinterOperationsSupported
ONLY contains the base IPP/1.1 operations. There is no obvious
way that it could easily be extended to cover all PSI operations,
WIMS operations, etc.
And arguably, the PWG SM/1.0 Printer object should NOT be
the source of any WIMS connection. The abstract System or
Service object in front of Printer should start WIMS connections.
Near-term the WIMS Agent should NOT be conflated with Printer.
----
Cheers,
- Ira
Ira McDonald (Musician / Software Architect)
Blue Roof Music / High North Inc
PO Box 221 Grand Marais, MI 49839
phone: +1-906-494-2434
email: imcdonald@sharplabs.com
-----Original Message-----
From: Zehler, Peter [mailto:PZehler@crt.xerox.com]
Sent: Wednesday, June 02, 2004 2:36 PM
To: McDonald, Ira
Cc: 'wbmm@pwg.org'; 'Wagner,William'
Subject: RE: WBMM> Re: Need quick decisions on schema changes
Ira,
Harry covered my views. My inserts are only on 7 & 8.
Peter Zehler
XEROX
Xerox Innovation Group
Email: PZehler@crt.xerox.com
Voice: (585) 265-8755
FAX: (585) 422-7961
US Mail: Peter Zehler
Xerox Corp.
800 Phillips Rd.
M/S 128-25E
Webster NY, 14580-9701
-----Original Message-----
From: Harry Lewis [mailto:harryl@us.ibm.com]
Sent: Wednesday, June 02, 2004 1:57 PM
To: McDonald, Ira
Cc: 'wbmm@pwg.org'; 'Wagner,William'
Subject: WBMM> Re: Need quick decisions on schema changes
Inserted...
----------------------------------------------
Harry Lewis
Chairman - IEEE-ISTO Printer Working Group
http://www.pwg.org
IBM Printing Systems
http://www.ibm.com/printers
303-924-5337
----------------------------------------------
"McDonald, Ira" <imcdonald@sharplabs.com>
06/02/2004 11:47 AM
To
"'wbmm@pwg.org'" <wbmm@pwg.org>
cc
"'Wagner,William'" <WWagner@NetSilicon.com>, Harry Lewis/Boulder/IBM@IBMUS
Subject
Need quick decisions on schema changes
Hi,
PLEASE answer quickly with your opinions on edits below,
so I can begin the edits needed in all of the WIMS schema
after last week's PWG Vancouver meetings.
Last week, we reduced the scope of the PWG Std Events spec
and Events schema to Printer-only (Printer, Job, Document,
and Subunit). Fine, but...
(1) Alerts schema - Should I delete 'AlertResource'?
- it depended on the now _deleted_ ResourceXxx events
in the Events schema and Resource object in the
(abandoned) Imaging System Model draft
HL - Yes, for now... but we need to put these back in later
(2) Alerts schema - Should I change 'NotifySourceState'
to delete 'Testing' and 'Down' from 'hrDeviceStatus'
in Host Resources MIB (RFC 2790)?
- this change will make support of coherent Printer
state harder to harmonize with HR MIB
- I think that it's a bug that Printer state in IPP/1.1
requires state reasons to report Down or Testing
HL - No
(3) Alerts schema - Should I rename 'NotifySourceState'
to 'NotifyPrinterState' and 'NotifySourceURI' to
'NotifyPrinterURI'?
- doing so effectively closes the future possibility
of multifunction alert support in WIMS
HL - No!
(4) Resource schema - Should we abandon this schema?
- last week's meeting seemed against adding any
new objects except in some future PWG MFP Model
- abandoning Resources seems foolish to me
HL - Seems foolish to me too. Don't like the word abandon. Prefer "staging"
(5) Schedule schema - Should I reorganize it into
the three Monitoring, Management, and Admin groups
of Actions?
- this seems worthwhile, as it describes WIMS levels
better
HL - Yes
(6) Schedule schema - Should I delete Resource actions?
- Does WIMS WG accept the need to model Resource?
HL - No (Yes... but possibly at a later "stage")
(7) Schedule schema - Should I import 'NotifyEvents' from
the Events schema?
- this looks better, but again loses Resources
HL - Not sure... why does this loose Resources... because event schema requirments are being driven by PSI? Seems incorrect.
<PZ>I don't understand this one either</PZ>
(8) Schedule schema - Should I add the elements for
Supported[Operations|Actions|Objects] here, so
that RegisterForManagement operation works?
- the WIMS operations won't appear in any generic
PWG Semantic Model element in the forseeable future
HL - why do you say the WIMS ops won't appear in SM? Do you mean just the admin related Ops?
<PZ>This would seem to be a straight forward extension and could be added to the Schema easily, captured as a Semantic Model extension and picked up in the next version of the PWG Semantic Model spec</PZ>
Cheers,
- Ira
Ira McDonald (Musician / Software Architect)
Blue Roof Music / High North Inc
PO Box 221 Grand Marais, MI 49839
phone: +1-906-494-2434
email: imcdonald@sharplabs.com
This archive was generated by hypermail 2b29 : Thu Jun 03 2004 - 21:15:15 EDT