At 11:19 AM 11/9/96 -0800, Deborah A. Lapeyre wrote:
>Special spaces are, among other things, a rather old-fashioned typographic
>technique for inserting a "known" amount of space in a location. The
>advantage of em-space and en-space (and thin and the like) is that, if
>your composition device allows it, these can be both clearly specified
>and font variable. (I.E., My em-space in 36 point Bodini bold not just a
>generic 12-pt em.)
>ASCII (currently) makes a distinction between "whitespace" (spaces and
>tabs) and special characters that happen to resolve to a space on display.
>Some of these are measured amounts of space (en, em, thin, digit, .. ).
>Some of these are behaviors, such as "hard-space" or "non breaking space".
>There hasn't been a problem because SGML tools do not treat   as
>a space, and wouldn't even CONSIDER compressing "    "
>into " " although " " is compressed to " ".
>1) At least as used in the past, an author intends for special spaces never
>to go away or be replaced by a "space".
>2) How relevant is any of this to screen display? How
>relevant is any of this to Unicode now?
>3) If all spaces translate to "space", then who cares? If the special
>spaces are individual characters (a la Unicode) then they are not
>"spaces" at all. Are there any "spaces" that are not characters?
>(I'm floundering here; these spaces have tied my tongue, but I know what I
>mean. Can anyone rescue here?)
I'm pretty spacey after a day of standards meetings, but I second what I
think I hear Debbie saying here. There is "whitespace" which composition
systems, as well as SGML processors, have traditionally thrown away. Then
there is space which is special and which I don't want any system to mess
with. To Debbie's list I add the spaces that are part of the rhetoric of
mathematics, such as the one before the "dx" in an integral. And what about
italic correction? That's space that sometimes the user adds, but that may
also be added by a good composition system like TeX.
I've always been uneasy about SGML's handling of space, which seems overly
fussy from certain programming perspectives but unnecessary and downright
confusing from a user-interface perspective. (Sitting through a couple of
hours of discussions about the character-coding parts of the SGML
Declaration is good for reinforcing the suspicion that we work hard at
making things hard for ourselves. Hearing some notable parser writers'
comments on their sufferings tends to reinforce that suspicion.)
A lot of screen displays certainly aren't good enough to reproduce the old
typographers' use of space, but I don't think the issue will go away. The
mathematicians continue to use TeX because of its capabilities for
typographic expression, in spite of the rise of WYSIWYG systems. After a few
years of MacTacky, we've seen a return to concern for well-designed fonts. I
suspect that concern for spacing will return, too, as display technology
Dr. James D. Mason
(ISO/IEC JTC1/SC18/WG8 Convenor)
Lockheed Martin Energy Systems
Information Management Services
SGML Systems Development
1060 Commerce Park, M.S. 6480
Oak Ridge, TN 37831-6480 U.S.A.
Telephone: +1 423 574-6973
Facsimile: +1 423 574-0004