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

Re: radical thoughts

Concernant « Re: radical thoughts », Ulrik Vieth écrit :
« Perhaps Thierry could provide some references.

The difficulty here is to choose one!  virtually anything published in
France before the advent of TeX used upright shape for Roman capitals
& greek alphabets (u&lc). Constants (\e, \pi, digits e.g.) are also
set upright whereas variables & functions are italic.

Old issues of The annales de l'institut Fourier conformed to the above
rule until around 1987... Les annales de l'École normale sup. have
kept this appearance longer (using a sort of Times/MathPi
combination). You could also check the scanned pages from a book of
Henri Cartan as shown at the URL:
http://www.loria.fr/tex/fontes/maths/cartan.html
http://www.loria.fr/tex/fontes/maths/cartan-english.html

and it is still what is advized by the `lexique des règles en usage à
l'imprimerie nationale'.

However one could view many of these traditional rules as bad
consequences of the high price (and weight!) of lead founts. I suppose
that printers had one greek fount for the humanities, one math fount &
a few text founts. I also suppose that french printers had a copy of
the didot greeks (upright) while english ones had that famous inclined
greek i saw in bringhurst' book. Hence the tradition. Maybe now that
these limitations are behind us we could redesign all this on a more
rational basis (greek letters being italic if denoting variables,
upright if constants, but then greek caps should be mostly italic
too)...?

I'm certain however that the encodings/software should not decide for
us what will be our next tradition.
   Thierry Bouche.       -----       thierry.bouche@ujf-grenoble.fr
          http://www-fourier.ujf-grenoble.fr/~bouche/