Re: The glyph of the CAPITAL SHARP S

From: Andreas Stötzner (
Date: Fri May 18 2007 - 08:13:05 CDT

  • Next message: Philippe Verdy: "RE: Uppercase ß is coming? (U+1E9E)"

    Am 18. Mai 2007 um 06:05 schrieb John Hudson:

    > Asmus wrote:
    >> I don't know that I'd buy that as bluntly as you stated that. There
    >> are some significant differences in the basic design of some letters
    >> between serif and sans-serif, not to mention script or decorative
    >> styles.
    > Generally, the variation between different type styles is much less
    > pronounced in the uppercase than it may be in the lowercase, and the
    > relationship of typical Latin serif uppercase letters to their sans
    > serif counterparts tends to be very regular indeed. Where there is
    > variation, in either case, between the form of letters in different
    > type styles it generally reflects a variation in written forms,
    > influenced by particular tools, slant, speed, etc., which is why the
    > variation is so much greater in script and decorative fonts.
    > Simply put, I would not expect an uppercase serif and sans serif
    > eszett to vary any more in their basic form than any other letter, and
    > I think the shapes proposed in Andreas' magazine demonstrate an
    > attractive adaptability to a range of type styles and, importantly I
    > think, a parallel to existing variation: the forms in the script style
    > faces are quite different from the basic structure found in both serif
    > and sans serif types.
    >> Also, what do you mean by "model form"? If you allude to the
    >> representative glyph in the Unicode charts, the only requirement that
    >> such a glyph _must_ satisfy is that it be the most plain vanilla
    >> design possible for a Times-like font.
    > I meant model form in a more general sense, but obviously it would be
    > good if the glyph chart example reflected the basic shape that is most
    > likely to become the model for most German type design. I believe this
    > will almost certainly be something very much like the models shown
    > here
    > here
    > and in other PDFs linked from here
    > And if you are looking for 'the most plain vanilla design possible for
    > a Times-like font', it is provided:
    > I think the questions of whether the right side should or should not
    > resemble more the uppercase S, and whether the left side should be
    > rounded or squared, are within the range of reasonable typographic
    > variation of this basic form: to be decided by type designers on a
    > typeface-by-typeface basis. Might an SS ligature also become a
    > recognised and accepted variant form, maybe to become associated with
    > certain styles of typeface? I don't know, but I don't think it is
    > either likely or desirable as a basic model, because of the
    > difficulties of application to a sufficiently wide range of styles.
    > John Hudson

    Thank you, John, for pointing at this. I still think that the Dresden
    glyph would do better for the UC reference glyph than the one which is
    currently depicted in n3263.pdf. Because it is more unambiguous and
    more obviously understandable in relation to the serial logic of Latin
    uppercase glyphics.

    There’s no need and no convincing rationale so far for the right half
    of the Eszett resembling the shape of the S.

    Andreas Stötzner.

    This archive was generated by hypermail 2.1.5 : Fri May 18 2007 - 08:17:14 CDT