[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: More radical thoughts

Concernant « Re: More radical thoughts », Laurent Siebenmann écrit :
«      Which brings me to my most radical thought.  Is a complex new
« encoding with twice as many symbols (at least) going to inhibit
« creation of new math fonts?  I hope not.

Well, the slight extension in the number of glyphs from iso-latin to
unicode (or, more effectively, WGL4) did not frighten designers too
much! Moreover they never feel forced to fill in all slots if they
don't have an application in view. On the other hand, many new symbols
are simply prebuilt compound symbols that only require a few minutes
to be assembled and properly placed, I think that these symbols are
worth the (small) effort in the designer's mind because they will look
better in print.

I also think that the french style is very balanced: essentially, what
tells you it's maths rather than text is spacing, not letter
shapes. Honestly, (co)homology groups look much more fundamental when
H is upright;-}

« I find
« italicized theorem statements set in the German-American
« style hard to read because the mainly italic math hides in a forest
« of italic prose.

this is addressed somewhere in the TeXBook, where it is suggested to
use slanted style for text emphasis, so that italics are only used in
maths. However, mechanically slanted letters are ugly, introduce yet
another slope. That scheme is good for preprints or copy-editing, but
not really usable in professional publications (it is used however by
the Ann. Inst. Fourier...)

«     Yes it is a tradition worth reviving, but I doubt revival
« will succeed without the help of some distinctive fonts...

using virtual fonts, it is rather easy to obtain the appearance of
pre-TeX Gauthier-Villars publications (using Times+MathTimes, MathPi
for Bbb and greek). I've done it when preparing a note for the CRAS
(my idea was that when seeing this note, they would be ashamed to have
lost this superior style wrt what they make now;-) [the CRAS is now
published by Elsevier]

However, designing a brand new font set (type 1) covering T1, TS1 and
M* encodings would be a major achievement, that could be supported by
user groups and/or math societies?

   Thierry Bouche.       -----       thierry.bouche@ujf-grenoble.fr