Re: Compression through normalization

From: Peter Kirk (peterkirk@qaya.org)
Date: Fri Dec 05 2003 - 15:02:36 EST

  • Next message: Jungshik Shin: "Re: Unihan kKorean pronunciations"

    On 05/12/2003 10:03, Mark Davis wrote:

    >>OK. So it's Mark, not me, who is unilaterally extending C10.
    >>
    >>
    >
    >Where on earth do you get that? I did say that, in practice, NFC should be
    >produced, but that is simply a practical guideline, independent of C10.
    >
    >Mark
    >
    >
    >
    Well, of course "unilaterally extending C10" only in the very specific
    sense defined by Doug, who wrote:

    >Any approach that ... ignores the Composition Exclusions table, ... is NOT of interest. That amounts to unilaterally extending C10
    >
    I did rather turn this round: Doug's idea of "unilaterally extending
    C10" referred to allowing more transformations than C10 does; but my
    idea of it was to further restrict the allowed transformations e.g. by
    specifying that transformations should be only into and not away from a
    normalisation form. Ignoring the Composition Exclusions table is of
    course irrelevant to C10, as this table relates not to canonical
    equivalence but to normalisation forms, and C10 has nothing to say about
    normalisation forms.

    You have now made it clear that what you proposed is a practical
    guideline, but as originally presented in your "too brief" note

    >I would say that a compressor can normalize, if (a) when decompressing it
    >produces NFC, and (b) it advertises that it normalizes.
    >
    it did look more like a proposed rule for a compressor, and one which is
    more restrictive than C10.

    -- 
    Peter Kirk
    peter@qaya.org (personal)
    peterkirk@qaya.org (work)
    http://www.qaya.org/
    


    This archive was generated by hypermail 2.1.5 : Fri Dec 05 2003 - 16:03:14 EST