Bob,
I agree. But the subject was defining some set of operations mandatory for "compliance" with the standard. The supposed intent would be that, if a entity is "standard compliant", an interfacing entity can assume some minimum set of support. Although this is a specious argument in fact (there are very many not-fully compliant implementations of virtually every standard extant), it is a desirable goal. Because WBMM is being considered for several distinctly different types of application, there should be corresponding levels of compliance. It is unclear if providing a hard-coded 'not authorized' response to unsupported operations would constitute compliance. Or indeed, if it does, whether "compliance" is a useful objective.
Bill Wagner
-----Original Message-----
From: TAYLOR,BOB (HP-Vancouver,ex1) [mailto:bobt at hp.com]
Sent: Thursday, July 24, 2003 3:49 PM
To: Wagner,William; Harry Lewis
Cc: McDonald, Ira; wbmm at pwg.org
Subject: RE: WBMM> Queues
I can certainly think of "remote" service cases where set and queue management would be appropriate. IMHO, we should not fundamentally declare in the spec what can and cannot be done over a firewall or across a domain: this is exactly what our authentication scheme should do. Many customers may not allow set or queue ma agement externally, but many will.
As for whether NetSilicon should implement set and/or queue management for proxy solutions, if they are "required" for whatever prile of WBMM you are supporting, you should implement them - but if you are sure that your customers will never allow you to expose them to an external client, you could make them stub implementations that always returned "not authorized". This would allow you to deliver the solution your customers want without unnecessarily binding other WBMM use cases.
bt
-----Original Message-----
From: Wagner,William [mailto:WWagner at NetSilicon.com]
Sent: Thursday, July 24, 2003 12:04 PM
To: Harry Lewis
Cc: TAYLOR,BOB (HP-Vancouver,ex1); McDonald, Ira; wbmm at pwg.org
Subject: RE: WBMM> Queues
Harry,
One of the most immediate applications of WBMM, for which no standard capability now exists, is in remote monitoring. In general, any set operation (and many get operations) must be disallowed for enterprise security and confidentiality purposes. If one is producing proxy devices to support such a capability (as NetSilicon is), it would be absurd to require support of operations that could never be used. Now, it is not clear how important full compliance to a PWG spec is, but if compliance requires supporting a whole set of unusable operations, it will become meaningless.
I also have suggested partitioning this effort, although not along the same lines. I think we should encourage more discussion on this, particularly on who plans to apply what part of this capability first rather than what appears easier to do first. Indeed, since prototypes are a necessary part of the standard development, I think getting volunteers to apply these ideas may be the determining factor in the "subsetting" definition and order.
Bill Wagner
-----Original Message-----
From: Harry Lewis [mailto:harryl at us.ibm.com]
Sent: Thursday, July 24, 2003 1:39 PM
To: Wagner,William
Cc: TAYLOR,BOB (HP-Vancouver,ex1); McDonald, Ira; wbmm at pwg.org
Subject: RE: WBMM> Queues
Ouch! disallow SETs?
I agree (propose) with the concept of subsetting. I am thinking more along the lines of a "growth path"... where we start with Extranet as target and possibly limit our scope to devices, then move to include Intranet and also expand to services, then, finally include queues
But I think SET capability is needed, even at the lowest compliance level. If not, how, for example, would a remote manager take a device off-line should this be necessary... or post a message to the opPanel?
----------------------------------------------
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 at NetSilicon.com>
Sent by: owner-wbmm at pwg.org
07/24/2003 11:30 AM
To
"TAYLOR,BOB (HP-Vancouver,ex1)" <bobt at hp.com>, "McDonald, Ira" <imcdonald at sharplabs.com>, Harry Lewis/Boulder/IBM at IBMUS, <wbmm at pwg.org>
cc
Subject
RE: WBMM> Queues
Harry had brought up the notion of different classes of compliance in his last minutes. I think this is as it must go. For WBMM monitoring applications, any set operation let alone queue management must be disallowed. But that does not mean that we ought not identify and format set operations.
Bill Wagner
-----Original Message-----
From: TAYLOR,BOB (HP-Vancouver,ex1) [mailto:bobt at hp.com]
Sent: Thursday, July 24, 2003 12:55 PM
To: 'McDonald, Ira'; 'Harry Lewis'; wbmm at pwg.org
Subject: RE: WBMM> Queues
I do think doing some sub-setting makes sense - in the case of queues, not
all devices/services managed by WBMM will have queues to manage.
Once we understand the potential subsets, we can talk about which ones need
to be in WBMM 1.0, and which can follow (or potentially be done in parallel.
bt
> -----Original Message-----
> From: McDonald, Ira [mailto:imcdonald at sharplabs.com]
> Sent: Thursday, July 24, 2003 9:44 AM
> To: 'Harry Lewis'; wbmm at pwg.org> Subject: RE: WBMM> Queues
>>> Hi Harry,
>> Apparently we want to focus on WBMM device management first
> (per most of our WBMM discussions).
>> However, that doesn't do PSI any particular good, which still
> would require the box labelled "and then a miracle happens"
> to get a PSI Print Service (or Target Device) installed or
> reconfigured after installation.
>> Do we care that all PSI implementations will ship without
> standards-based management for several more years?
>> Cheers,
> - Ira McDonald
> High North Inc
>>> -----Original Message-----
> From: Harry Lewis [mailto:harryl at us.ibm.com]
> Sent: Wednesday, July 23, 2003 11:38 PM
> To: wbmm at pwg.org> Subject: WBMM> Queues
>>>> Sorry I missed the call. On the topic of queue management.
> I've no objection to adding this but wonder if it might
> warrant some subsetting of WBMM. Are you going to have to be
> able to manage queues to be WBMM compliant? Don't we want to
> focus on solving the device management problem first, and
> then move on to queue management?
> ----------------------------------------------
> Harry Lewis
> Chairman - IEEE-ISTO Printer Working Group
>http://www.pwg.org> IBM Printing Systems
>http://www.ibm.com/printers> 303-924-5337
> ----------------------------------------------
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.pwg.org/archives/wims/attachments/20030725/26200b28/attachment.html