The glyph of the CAPITAL SHARP S

From: Michael Everson (
Date: Thu May 10 2007 - 10:19:30 CDT

  • Next message: Marnen Laibow-Koser: "Re: The glyph of the CAPITAL SHARP S"


    At 14:13 +0200 2007-05-10, wrote:

    >I personally hope that the use of a capital ▀
    >code point will be limited to supporting the
    >correct transformation from upper to lower case
    >and in historical context.

    That is the intended use of the character.

    >So we agree to the technical justification in the proposal. The conversion
    >from lower uppercase SS to ▀ has always been a nuisance.
    >>Your idea is fine, but at the end of the day it
    >>falls to me to draw the glyph that goes in the
    >>charts, and it is expected to harmonize with the
    >>rest of the glyphs. And the UCS is an
    >>International Standard, not just a DIN standard.
    >No argument here, but it should be recognizable
    >as the capital form of 00DF. A discussion on
    >what this glyph should look like will not
    >produce a unanimous decision and we do not have
    >a consensus even within Linotype.

    I understand. Currently the glyph at 3f in
    is the favourite "vanilla" glyph.

    >The contemporary use of a capital ▀ is
    >considered to be evidence of low literacy and we
    >teach our children not to use it. If we had to
    >implement the glyph in our fonts we would put an
    >SS ligature into 99.98 % of our fonts.

    *That* solution would be very bad indeed, in my
    opinion. It would offer opportunities for
    spoofing (since CAPITAL SHARP S and CAPITAL S
    CAPITAL S would look the same). Implementers of
    the UCS should not choose "bad" glyphs because of
    their views about literacy. If a person *wants*
    to use a CAPITAL SHARP S, the glyph a vendor
    supplies should be a "real" one if the vendor
    supplies a glyph for U+1E9E in a given font.

    Please see the original DIN proposal at

    Asmus Freytag also wrote a long essay about this
    which is also worth consideration.

    Best regards.

    Michael Everson *

    This archive was generated by hypermail 2.1.5 : Thu May 10 2007 - 10:26:17 CDT