Character Repertories Mail Archive: CR> Minutes: CR teleconf

CR> Minutes: CR teleconfernece 11 Dec. 2002

From: ElliottBradshaw@oaktech.com
Date: Tue Dec 17 2002 - 14:52:45 EST

  • Next message: McDonald, Ira: "CR> FW: character encoding diagram"

    Minutes from PWG-CR teleconference, 11 Dec. 2002.

    Attending
    ------------
      Jim Bigelow, HP
      Mark Robb, Lexmark
      Rod Acosta, Agfa
      Ira McDonald, High North Inc.
      Elliott Bradshaw, Oak Technology

    Process and Output
    ------------------------

    We agreed to work towards these items:

    1. Implementor's guide. This should simplify the whole topic and help
    implementors find the information they need. It should have definitions of
    terms and (informative) summaries of the repertoires that are referred to.

    2. Draft standard(s) for discussion at PWG. The first is likely to
    include:
         -names and definitions for repertoires
         -minimal set of repertoires to assume in any printer
         -semantics of a query mechanism to find what repertoires are in a
    printer

    3. Coordination with other working groups:
         -SM (Ira agreed to coordinate)
         -UPnP (Elliott)
         -PSI
         -Bluetooth SIG

    Issues and (tentative) conclusions
    ------------------------------------------

    1. Bluetooth BPP has a lot of good material in this area. Things we would
    change:
         -it is not always clear which version of a repertoire is to be used,
    or where it is defined
         -they use a bit array; we should define standard names and let bits,
    numbers, etc. be part of a protocol binding

    2. Need to account for Microsoft characters. There are many options;
    more research is needed.

    3. We assume a Unicode environment. For example, we won't talk about
    ISO-8859 character sets; we'll talk instead about sets of Unicode that
    correspond to ISO-8859. However, legacy applications may need to map from
    other character sets. This is a topic we may return to in the future; in
    the meantime the user's guide needs to explain the issues clearly.

    4. The sense of the group is that a direct mechanism to query for
    individual Unicodes is a bad idea. The assumption is that if a client
    wants to print a certain type of material (e.g. Japanese), then it will be
    able to recognize the appropriate defined repertoire name(s).

    5. We may want to define subsets of some of the Asian sets.

    6. Need to make sure ASCII is accounted for. This is probably just a
    documentation issue.

    7. A client may want to know what characters are available in what fonts.
    We will treat this as a separate topic, for possible future work.

    Next steps
    -------------

    NEXT CONFERENCE CALL: Wednesday, Jan. 8, 3:00 Eastern time.

    In the meantime I will draft an Implementor's Guide.

    ------------------------------------------
    Elliott Bradshaw
    Director, Software Engineering
    Oak Technology Imaging Group
    781 638-7534



    This archive was generated by hypermail 2b29 : Tue Dec 17 2002 - 14:54:49 EST