Re: U+3004 JAPANESE INDUSTRIAL STANDARD SYMBOL

From: Asmus Freytag (asmusf@ix.netcom.com)
Date: Mon Aug 10 2009 - 22:58:08 CDT

  • Next message: verdy_p: "Re: Hangul leaves the nest..."

    On 8/10/2009 7:43 PM, James Cloos wrote:
    >>>>>> "Asmus" == Asmus Freytag <asmusf@ix.netcom.com> writes:
    >>>>>>
    >
    > Asmus> But claiming that both new and old version of some logo, no
    > Asmus> matter the difference in their design, are "glyph" variants
    > Asmus> of each other would assert that what is encoded is the owning
    > Asmus> organization. If fonts were created on that principle, then
    > Asmus> existing documents created with the "old" logo could one day be
    > Asmus> opened on another system and show the "new" logo. Or vice versa.
    >
    > Interesting.
    >
    > Asmus obviously read the earlier post as suggesting that fonts sould
    > just put the new logo at the cp the old logo currently occupies.
    >
    > I read the earlier post as suggesting that <U+3004><U+FE00> should be
    > used to specify the new glyph.
    > (That is <JAPANESE INDUSTRIAL STANDARD SYMBOL><VARIATION SELECTOR-1>.)
    >
    > I wonder which of those the earlier poster actually meant?
    >
    Jim,
    adding the VS to the mix changes less than you might think.

    The rule of the game is that the FE00 may be ignored. (In fact should be
    ignored unless it is explicitly supported). So you can get the same
    effect of older systems/fonts showing the old logo for new documents.

    For certain very minor tweaks in a logo, it might be acceptable to use
    such "fallbacks", but at least to me, the redesign that was shown on the
    list here crosses the threshold where I would still be comfortable
    asserting that the two logos are the same *character*.

    I see a recent tendency by people discussing encoding proposals to treat
    a VS1 as a supposedly "cheap" way to get another symbol encoded by
    sidestepping the unification question. I find that disturbing, because
    looking for a cheap way to get the "glyph" doesn't properly define the
    underlying _identity_ of the character.

    A./

    PS: The usual caveat applies: the entity under discussion is a logo,
    logos are considered out of scope for the standard - I'm ignoring this
    for now, because much of my argument holds for large classes of symbols,
    not just logos. It's easier to discuss the concrete case, rather than
    transpose the argument step for step so it's expressed for the wider
    case of symbols in general. But that's the place where these arguments
    would ultimately be applicable.



    This archive was generated by hypermail 2.1.5 : Mon Aug 10 2009 - 23:00:48 CDT