attachment-0001

Hi Tom, To me, #2 is the issue to really focus on: is this really two semantic concepts, or one?  If any of these values are semantically really just additional enumerations for print-quality, I'd much rather we fix this than do the split it. As for #2, what "semantic meaning would be lost or mixed" if we just extended the enumeration for print-quality?  Although you can, by having these as two attributes, specify "draft" and "photo" - in practice, this is virtually never done: "photo" is usually specified in one of a couple ways: - my media type (i.e., photo-glossy paper model #C239847A) - PrintQuality (quality = Photo4800DPI) - i.e., another print-quality enumeration If it's something other than one of these, it's usually what we call in PSI capabilities a "macro feature" - i.e., it's not a real attribute sent to the device, but a feature defined as a macro combination of other features - which I don't think is a concept currently supported by IPP. Fundamentally, "refines the value specified by the print-quality" seems like a weak semantic differentiation, and I believe it is inconsistent with how extensions to the print-quality concept have been applied in at least the inkjet segment of the market. thanks, bt > -----Original Message----- > From: Hastings, Tom N [mailto:hastings@cp10.es.xerox.com] > Sent: Wednesday, June 25, 2003 1:08 PM > To: printing-driver@freestandards.org; Claudia Alimpich; ipp@pwg.org > Cc: printing-jobticket@freestandards.org > Subject: RE: [printing-driver] RE: [printing-jobticket] > Proposal to add ne w IPP print-optimize attribute > > > Bob, > > I think that the proposal to add "print-optimize" solves two > separate problems, not just a single problem of adding some values: > > 1. Doesn't invalidate the semantics of "print-quality" (which > we are treating in the same way as an enumeration in JDF, > i.e., a closed end list, in which these are the only values > that can be supported: 'draft', 'normal', and 'high'). > > 2. The Optimize mechanism isn't really just additional print > quality values, but is more specific as to what to optimize.   > Therefore, it would be wrong just to add the proposed new > values to "print-quality" as you suggest. Semantics meaning > would be lost or mixed.  (Also the "print-optimize" attribute > is like the JDF XxxDetails which is an extensible NMTOKEN > value, not an enumeration.) > > Also note that "print-quality" may be used in combination > with "print-optimize".  So you can have 'draft', 'normal' or > 'high' optimization of, say, 'photo'. > > ISSUE:  We didn't say that a Printer that supports > "print-optimize" MUST support "print-quality" as well.   > Should we, since the definition of "print-optimize" is that > it "refines the value supplied (or defaulted) in "print-quality")? > > Also this isn't a precedent that we can't add values to an > existing attribute in IPP or the Semantic Model.  It just > seems that for this one "print-quality" attribute both > reasons support not adding new values to the existing attribute. > > Tom > > -----Original Message----- > From: TAYLOR,BOB (HP-Vancouver,ex1) [mailto:bobt@hp.com] > Sent: Tuesday, June 24, 2003 17:39 > To: Claudia Alimpich; ipp@pwg.org; > printing-jobticket@freestandards.org; > printing-driver@freestandards.org > Subject: [printing-driver] RE: [printing-jobticket] Proposal > to add new IPP print-optimize a ttribute > > > I understand the desire to avoid violating the semantics of > the IPP attribute - but adding these enumerations to > print-quality does not feel as objectionable to me as > splitting a single semantic concept into two different > attributes.  If this is the precedent we take for extending > the semantic model, I'm worried that we'll end up with an > increasingly confusing and complex.  I would rather we take > the minor hit and fix the high & draft definitions in the > semantic model than create another ~equivalent attribute with > a whole bunch of special semantic rules (e.g. - what should > the service do if print-quality=high and print-optimize=save-toner?). > > bt > > --------------------------------------------------- > Bob Taylor                                         > Senior Architect                             > IPG Strategic Technology Development   > Hewlett-Packard Co.       > mailto:bobt@hp.com                         > phone: 360.212.2625/T212.2625                     > fax: 208.730-5111                 > ---------------------------------------------------   > > > -----Original Message----- > > From: Claudia Alimpich [mailto:alimpich@us.ibm.com] > > Sent: Tuesday, June 24, 2003 5:27 PM > > To: ipp@pwg.org; printing-jobticket@freestandards.org; > > printing-driver@freestandards.org > > Subject: [printing-jobticket] Proposal to add new IPP > > print-optimize attribute > > > > > > > > > > > > > > Last Tuesday during the PWG/FSG meeting in Portland we had a > > discussion about the IPP print-quality attribute and FSG's > > desire to add two new values, "economy" and "fine", where > > "economy" is lower than "draft" and "fine" is higher than > > "high". After some discussion we all pretty much decided that > > it is not possible to add these new values to the already > > existing "draft", "normal", and "high" values because of the > > current definitions of the existing values (high is defined > > as the highest quality and draft is defined as the lowest > > quality). It also seemed like what FSG wanted was a way to > > specify print optimization and not additional levels of > print quality. > > > > The FSG working group met today, and based on the input from > > last Tuesday's meeting, we would like to propose the addition > > of a new attribute, called print-optimize, that is defined > as follows: > > > >    print-optimize (type2 keyword) > > > >       This attribute refines the value specified by the > print-quality > >       attribute. > > > >       The standard keyword values are: > > > >          'image': optimize for image clarity > >          'photo': optimize for photo clarity > >          'text': optimize for text clarity > >          'text-and-image': optimize for both text and image clarity > >          'save-toner': optimize for minimal toner usage > >          'speed': optimize for printing speed > > > > We would appreciate your feedback on this proposal including > > suggestions for additional values. > > > > If this proposal looks good, we would like to propose that it > > be included in the JobX Spec. If the print-optimize attribute > > is approved by PWG by the end of August, then we can propose > > that it be added to the JDF 1.2 Spec that is being finalized > > in early September. > > > > Thank you for your time and feedback. > > Claudia Alimpich > > IBM Printing Systems Division > > Boulder CO > > 303-924-4418 > > alimpich@us.ibm.com > > > > > > _______________________________________________ > > printing-jobticket mailing list printing-jobticket@freestandards.org > > http://freestandards.org/mailman/listinfo/printing-jobticket > > > > _______________________________________________ > printing-driver mailing list > printing-driver@freestandards.org > http://freestandards.org/mailman/listinfo/prin> ting-driver > > > _______________________________________________ > > printing-driver mailing list > printing-driver@freestandards.org > http://freestandards.org/mailman/listinfo/prin> ting-driver >