Why Fraktur is irrelevant (was RE: Fraktur Legibility (was Re: Response to Everson Phoenician)

From: Peter Constable (petercon@microsoft.com)
Date: Tue May 25 2004 - 12:42:36 CDT

  • Next message: E. Keown: "Re: Response to Everson Ph and why Jun 7? fervor"

    > From: unicode-bounce@unicode.org [mailto:unicode-bounce@unicode.org]
    On
    > Behalf Of Dean Snyder

    Dean:

    You've done an interesting, controlled test. I think you made things in
    overly difficult by using all uppercase, but I'm not really interested
    in critiquing the test methodology.

    I'm interested in whether discussing Fraktur has any relevance for the
    issue of encoding PH.

    Let's say that you have adequately demonstrated that Fraktur text is not
    legible to most Latin speakers. (This can be disputed since there is
    some measure of legibility -- all of your subjects did recognize some
    portions of the text. But I'll assume it's demonstrated.) All that this
    demonstrates is that *glyphs* may not be recognizable. It does not
    demonstrate that characters are distinct. For *any* script, one can
    create glyphs that the average user of the script will find illegible.
    I've seen Thai typefaces that I certainly had difficulty reading, but
    that does not mean that the characters are not Thai. The point is that
    *some* people can read such text, and they recognize those characters as
    Thai, or Latin in the case of Fraktur.

    The important question for character encoding is whether the needs of
    users of Fraktur are best met if they are encoded as distinct
    characters. For mathematicians, distinct Fraktur symbols have been
    encoded. But for people using Fraktur style for German (or whatever
    language) text, they will use the characters in the Basic Latin block.
    That is what will best meet their needs. (If any of them *wants* to
    encode text using the math symbols, that is their prerogative, but that
    is not necessarily what others might recommend to them for their
    particular application.)

    The fact that people who do not regularly use Fraktur find it illegible
    has no bearing whatsoever. The people who *do* use Fraktur (apart from
    the mathematicians) consider it Latin and are best served by encoding
    their text as Latin.

    For the case of PH, we can ask questions about legibility, but there is
    no need to look for analogues with Fraktur. Some people who know Hebrew
    find Hebrew-language text set with PH glyphs legible, some (the vast
    majority) do not. The same is true for German and Fraktur. But all that
    tells us is that *perhaps* PH can be unified with square Hebrew
    characters. It is not the only consideration in whether we do unify
    characters, however, and the situation wrt Fraktur does not tell us
    anything additional that pertains to that issue.

    Can we **please** discontinue this discussion of Fraktur in relation to
    making decisions about PH.

    Peter Constable



    This archive was generated by hypermail 2.1.5 : Tue May 25 2004 - 12:44:26 CDT