Semantic Model Mail Archive: SM> RE: IPP> "-actual&q

SM> RE: IPP> "-actual" multivalue types issue

From: Zehler, Peter (PZehler@crt.xerox.com)
Date: Mon Oct 28 2002 - 11:44:09 EST

  • Next message: Dennis Carney: "Re: SM> Re: ISSUE 18: Or should the client be REQUIRED to support some of the Document operations?"

    Dennis,
    Comments below.
    Pete

                                    Peter Zehler
                                    XEROX
                                    Xerox Architecture Center
                                    Email: PZehler@crt.xerox.com
                                    Voice: (585) 265-8755
                                    FAX: (585) 265-8871
                                    US Mail: Peter Zehler
                                                    Xerox Corp.
                                                    800 Phillips Rd.
                                                    M/S 128-30E
                                                    Webster NY, 14580-9701

    -----Original Message-----
    From: Dennis Carney [mailto:dcarney@us.ibm.com]
    Sent: Monday, October 28, 2002 10:51 AM
    To: Zehler, Peter
    Subject: RE: IPP> "-actual" multivalue types issue

    Peter,

    I see your point, but I do feel obligated to make the following comments on
    your discussion of the example I gave:
    1) My example discussed Windows separator pages. With these, there is no
    indication given to the printer that there *is* a separator page.
    Therefore, unless the printer had additional information somehow, it *could
    not* populate the "separator-sheets-actual" attribute.
    <PZ>The Printer could however support the "PageOverrides" element based on
    its processing. The Printer would either recognize the page as a separator
    sheet or generically classify it as a page override. <PZ/>

    2) I am no expert on the "PWG 5100.3, Production Printing Attributes - Set
    1" document, but I do not see any "sides" information in the
    "separator-sheets" attribute.
    <PZ> Any template attribute can be added to the "separator-sheets" attribute
    group. It is of course an extension but that is the way your example was
    stated. <PZ/>

    Therefore, in my example, the printer *could* provide possibly sufficient
    information to the client using multivalued "-actual" attributes, but
    *could not* using single-valued attributes.
    <PZ> But the semantics for the multivalued "media" attribute is not well
    defined.
    There is no agreed upon semantics for the ordering of the multiple values or
    how
    to determine the number of each media used. It is not well defined which
    attributes are parallel attributes and must have their i'th members
    interpreted
    together. In your example "media" and "sides" are parallel attributes.
    What about something like the "finishing" attribute which is already
    multivalued. <PZ/>

    This one example might not be enough to justify the additional complexity
    of multivalued attributes, but I wanted to set the record straight on it
    nonetheless.

    Dennis

     

                          "Zehler, Peter"

                          <PZehler@crt.xero To: Dennis
    Carney/Boulder/IBM@IBMUS, "Zehler, Peter" <PZehler@crt.xerox.com>

                          x.com> cc: ipp@pwg.org,
    owner-ipp@pwg.org, sm@pwg.org

                                                   Subject: RE: IPP> "-actual"
    multivalue types issue
                          10/28/02 08:13 AM

     

     

    Dennis,
    In your example the "-actual" would contain the "sides-actual",
    "media-actual" and "SeparatorSheets-actual" attributes. The
    "SeparaterSheets-actual" attributes contains "media" and "sides". It is
    clear and unambiguous what actually occurred. The "-actual" attributes
    carry the same semantic and syntax as the base attributes. Keeping
    alignment
    between "-actual" and the base attribute will provide significant "Job
    Receipt" functionality at a low cost while avoiding additional work for the
    printer. It is easier to simply copy attributes over and change its name
    than to also remap its syntax and semantics.
    Pete

                                                     Peter Zehler
                                                     XEROX
                                                     Xerox Architecture Center
                                                     Email:
    PZehler@crt.xerox.com
                                                     Voice: (585) 265-8755
                                                     FAX: (585) 265-8871
                                                     US Mail: Peter Zehler
                                                                         Xerox
    Corp.
                                                                         800
    Phillips Rd.
                                                                         M/S
    128-30E

    Webster NY, 14580-9701

    -----Original Message-----
    From: Dennis Carney [mailto:dcarney@us.ibm.com]
    Sent: Friday, October 25, 2002 10:53 AM
    To: Zehler, Peter
    Cc: ipp@pwg.org; owner-ipp@pwg.org; sm@pwg.org
    Subject: Re: IPP> "-actual" multivalue types issue

    Peter,

    My idea was that since these "-actual" attributes are supposed to tell you
    what *actually happened*, they should cover the case where more than one
    value was used.

    An example: a job is submitted from Windows with a Windows separator page.
    The separator page prints from tray 1, blue paper, simplex; the main job
    prints from tray 3, white paper, duplex. Both the media-actual and
    sides-actual attributes would have two values. As it turns out, a
    submitting IPP client on Windows (port monitor) can find out from Windows
    that the job has a separator page, but has no way to find out from Windows
    about media or plex. Therefore, such a client can interpret the two values
    to mean the first is for the separator page, the second is for the job.
    This interpretation will be right almost all of the time.

    This proposal is supposed to address the "job receipt" shortcoming without
    adding a lot of work to the Printer or the standards body :-). With that
    in mind, I could be persuaded to get rid of the multivalued aspect. But I
    think it adds (possibly) significant functionality at a low cost.

    Dennis

                          "Zehler, Peter"

                          <PZehler@crt.xero To: Dennis
    Carney/Boulder/IBM@IBMUS, ipp@pwg.org, sm@pwg.org

                          x.com> cc:

                          Sent by: Subject: IPP> "-actual"
    multivalue types issue
                          owner-ipp@pwg.org

                          10/24/02 08:56 AM

    Dennis,
    My first comment is that the data types of the processing attribute should
    be the same as the base's. The base's definition determines if the
    semantic
    element can be multivalued. If the base can not be multivalued the actual
    should not be either.
    Pete

                                                     Peter Zehler
                                                     XEROX
                                                     Xerox Architecture Center
                                                     Email:
    PZehler@crt.xerox.com
                                                     Voice: (585) 265-8755
                                                     FAX: (585) 265-8871
                                                     US Mail: Peter Zehler
                                                                         Xerox
    Corp.
                                                                         800
    Phillips Rd.
                                                                         M/S
    128-30E

    Webster NY, 14580-9701

    -----Original Message-----
    From: Dennis Carney [mailto:dcarney@us.ibm.com]
    Sent: Friday, October 18, 2002 1:54 AM
    To: ipp@pwg.org; sm@pwg.org
    Subject: SM> ACT - IPP "-actual" attributes downloaded, version 0.1

    All,

    I have downloaded the first draft (version 0.1) of the document 'IPP: "
    -actual" attributes extension'. The document is 12 pages long, but the
    proposal is only 5 of those pages. The document is located at:

    ftp://ftp.pwg.org/pub/pwg/ipp/new_ACT/IPP-actual-attributes.doc
    ftp://ftp.pwg.org/pub/pwg/ipp/new_ACT/IPP-actual-attributes.pdf

    This document will be discussed on the October 31 Semantic Model
    teleconference. Any comments should be sent to the IPP mailing list.

    Abstract:
    This document defines an extension to the Internet Printing Protocol/1.0
    (IPP/1.0) [RFC2566, RFC2565] & IPP/1.1 [RFC2911, RFC2910] for the OPTIONAL
    "-actual" set of Job Description attributes that correspond to Job Template
    attributes defined in IPP. These "-actual" attributes allow the client to
    determine the true results of a print job regardless of what was specified
    in the Create-Job or Print-Job operation.

    Dennis Carney
    IBM Printing Systems



    This archive was generated by hypermail 2b29 : Mon Oct 28 2002 - 11:44:43 EST