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

Re: preliminary EuroTeX paper for review

Berthold wrote:

> It looks great, I only have some trivial comment regarding section 9.2:

> This might benefit from addition of the UNICODE numbers for the
> symbols mentioned, for added precision.  Greek Chi is 03A7.  The alternate
> lowercase Greek beta is 03D0. The archaic Greek numerals are Stigma (03DA),
> Digamma (03DC), Koppa (03DE), and Sampi (03E0).  

Thanks for your comments.  Although I agree that this information
might be useful, there's a problem in that Unicode doesn't cover all
the variants of the archaic Greek numerals.  The Mathematica fonts
have uppercase and lowercase versions while Unicode has only one slot
for each.  Furthermore, implementations like Yannis' OmegaTimes (see
TUG'96 proceedings) even have multiple variants of some of these.

> By the way, I might just mention that fontinst is but one way of generating
> metric files for fonts.  It can't handle math fonts --- as far as I know, 

Not sure what you mean.  To me it appears that fontinst can handle
math fonts, at least as far as the TeX specific features (nextlarger
and varchar) are concerned.  It does however take some effort when it
comes to adjusting glyph widths and italic corrections for tuning
subscript and superscript placement.  These have to be specified
manually in the .mtx files, but I suppose the same applies to the
virtual font file of MTMI.

> and I never use it, even for text fonts.  We generate all of our TFMs ---
> including those for math fonts --- using AFMtoTFM.  Of course, since this
> isn't widely available it wouldn't be appropriate to mention it.

I suppose so.

Cheers, Ulrik.