RE: Submission to ConScript Unicode Registry: Sylabica

From: Philippe Verdy (
Date: Mon Oct 15 2007 - 10:24:11 CDT

  • Next message: Philippe Verdy: "RE: Emoticons"

    Marcin 'Qrczak' Kowalczyk wrote:
    > It was not intended to be seriously used in practice, even though the
    > linguistic principles it employs are serious. It's an attempt to recover
    > as etymological spelling of the modern Polish language as possible
    > without hurting learnability too much.

    Well, even for etymologic reasons, nothing forbids you to use existing Latin
    letters as much as possible, and/or IPA symbols if a distinction is needed
    for its historic phonology.

    Despite the fact of studying the history of a language and the evolution of
    its orthography (that may now be too far from its etymology) is a very
    honourable and interesting work, it would certainly interest more Polish
    readers and students if they could read your notation based on their
    existing knowledge of the Polish language, its orthography and phonology
    (let's just discard, for this kind of work, its phonetic which may vary a
    lot, even today, among various speakers of the languages in various regions
    or social groups or ages, as long as they share a common phonology, mutually

    Generally, for such work, using a classic phonology notation is probably
    better, as it makes your work more easily accessible. Anyway, I find it
    admirable that its composition rules follow an arithmetic pattern similar to
    Hangul, meaning that even the composition of consonant could be supported by
    your existing OS (provided that it supports Hangul), by tweaking the
    encoding of Hangul syllables into your font... (This would nearly work,
    except that your encoding does not include a filler for a missing leading
    consonant, or filler for the only medial vowel you support, the soft sign;
    you would map your final vowels onto Hangul final consonnants).

    Then you could make two versions of your font without requiring a PUA block:
    one using your proposed glyphs for precomposed syllables, and another
    displaying the classic Polish letters or IPA symbols packed into a single
    glyph. This would allow reading your text either with your invented script
    or using more accessible letters and/or IPA symbols.

    Note that you propose using disjoiners for missing leading consonants, but
    Hangul uses a separate null consonant (filler). This is not critical for now
    given that your script is private, so using a private font is similar to
    using a private encoding agreement: in both cases it requires some user
    action to support your private script.

    But at least, for rich text documents (HTML, RTS...) that support embedding
    font styles along with the text, you have a solution to store the intended
    rendering with a specific font (other users that don't have this font will
    just see Hangul letters used in a strange way for transcribing Polish, with
    a strange orthography based on its etymology rather than the modern Latin
    orthography. This won't be an issue for interoperability, given that this
    will remain used only by those sharing your private agreement to install and
    use the specific font.

    Anyway, your fonts in the PDF are very legible, clean. The glyphs are
    creating a well organized collection with consistent style. I like your
    serif style used within the text of the document as it helps seeing the
    presence some central strokes. For larger sizes (like in your chart table),
    the sans-serif style is cleaner as it shows the fundamental core structure
    of your letters or syllables.

    I also see some visual similarities with the Ge'ez syllabary, in the way the
    base consonant form is altered into multiple truncated variants supporting
    some decorations for the vowels... That's quite smart and inventive...

    This archive was generated by hypermail 2.1.5 : Mon Oct 15 2007 - 10:28:01 CDT