LATEX-L Archives

Mailing list for the LaTeX3 project


Options: Use Classic View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Frank Mittelbach <[log in to unmask]>
Thu, 10 Apr 1997 20:17:34 +0200
text/plain (59 lines)
Johannes Kuester writes:
 > This is to point at a shortcoming of NFSS2: (I don't know whether
 > this has already been discussed when designed NFSS)

this has been discussed both during and after development of NFSS2

 > There is such a thing as a slanted/italic small caps font
 > (see Sebastian Carter's "Typographers on Type" for an example:
 > it contains a page showing different fonts of the Zapf Renaissance (1986)
 > family, including such a font).

agreed, calling small caps a shape is a simplification. similariy for
making, say "outline" a shape

 > So it doesn't seem appropriate to classify small caps as a shape,
 > rather, I think, it should be considered as a case. Thus resulting
 > in control sequences like
 >   \textcaps{...} and {\capscase ...}
 > or the like, analogous to uppercase and lowercase.

again, agreed in principle; having "case" as a further axis to the FSS
would be a valid possibility

 > Unfortunaly I can't think of any way to incorporate this easily
 > in NFSS and especially in its current interface.

it would be easy enough to integrate it in the general model, but
there are downsides of it as well:

 - no way to integrate it into the specification interface without
   either a horrible syntax or incompatible changes

 - nfss is designed to model major font use faithfully, adding an
   additional axis that is only sparsly populated would result in a
   large number of additional cases where font substitution rules need
   to be applied. that in turn would either blow up substitution rules
   ro would result in badly rendered documents.

so as far as latex2e is concerned, there will be no modifications
(Robin sumarised the reasons for that quite well). I do have plans for
a different fss that deals more intelligently with encoding issues and
with substitution. for that implementation a fifth "case" axis will be
considered (and perhaps even be implemented :-) or perhaps even a more
general scheme with arbitrarily many axis', although the latter i do
find still questionable.

to answer Robin's question:

  There are many potential axes that NFSS could have incorporated.  I
  don't remember enough of the original papers to be sure off hand how
  the present set were arrived at.

they where derived by emulating common usage taking into account
available fonts (both TeX and commercials) to arrive at a flexible
scheme that does work in most cases (while still being implementable
and manageable