Re: RE: Character identities

From: John Hudson (
Date: Tue Oct 29 2002 - 23:48:10 EST

  • Next message: William Overington: "Re: The comet circumflex system."

    Two very simple principles can resolve this issue:

    1. Encode text using characters that accurately carry the semantic meaning
    of the text and which enable text standardised text processing functions
    such as sorting, spellchecking and searching.

    2. Display the text by selecting a font that provides glyphs for those
    characters that correspond to your expectations, tastes and intentions.

    There are tens of thousands of fonts in the world, and we wouldn't assume
    that all of them should be able to display every text in a manner that
    corresponds to everyone's expectations, tastes and intentions. Selecting an
    appropriate typeface is the first decision in any typography, and you
    cannot bypass this decision. You have to make a choice, and if the glyph
    that corresponds to a given character in the chosen font displeases you,
    then you have to make another choice: you have to reject the font and
    select another one, or you have to hack the font and change the glyph.

    There are many fonts that can be used to display a multitude of texts, and
    do so in ways that correspond to the expectations, tastes and intentions of
    millions of writers and readers. There are some fonts that can only be used
    to represent certain texts, and in ways that correspond only to the
    expectations, tastes and intentions of a small number of people. There are
    entirely idiosyncratic fonts that display text in ways that only the
    individuals who made the fonts can read. The value of Unicode is that the
    character string behind the display is not tied to a particular display.

    John Hudson

    Tiro Typeworks
    Vancouver, BC

    It is necessary that by all means and cunning,
    the cursed owners of books should be persuaded
    to make them available to us, either by argument
    or by force. - Michael Apostolis, 1467

    This archive was generated by hypermail 2.1.5 : Wed Oct 30 2002 - 01:24:58 EST