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

Re: Future of the `Karl Berry Font Naming Scheme'...

Walter Schmidt writes:
> The problem is not just "real font names" vs. "8.3".  The 8.3
> restriction is going to become obsolete soon, but there are other
> restrictions, like the 31 chars on the Mac or certain symbols
> not always being legal in file names.  And most operating systems 
> (except for Un*x) ar not case sensitive.  Finally you cannot use 
> arbitrary file names on CD-ROM, even with specifications beyond
> ISO.  So there are obviously  sensible reasons for having a 
> _portable_ font naming scheme.
> IMHO a mapping file is not a suitable solution, as long as this
> is popular with Web2C implementations only.  So why not keep the 
> Karl-Berry-scheme, somewhat modified, explicitly _without_ the
> 8.3 restriction?

I'm certainly not proposing using `real font names'; in fact my goal
hasn't been to push for anything in particular. Mostly I'm trying to
evaluate the current status-quo and try to understand why we're doing
things the way we're doing them.

It seems to me like we've had several parallel developments since the
KB scheme was conceived. These include the VFAT filing system for DOS
that eliminates the requirement for 8+3 names on that platform, the
NFSS mechanisms for LaTeX (which have their own nomenclature for fonts
and add a level of indirection into the font selection process), the
addition of a `texfonts.map' file for web2c TeX, etc. Given these
things, it only makes sense to me to ask whether it is still needed.

I suspect that the answer is that no, it's not really needed, but it is
a standard, and one somewhat anachronistic standard is better than none.

Still, I can't help but wish that LaTeX's names for my fonts, like 
`T1/pkp9/m/n' had a clear, simple, obvious mapping to the TFM file
name (which in this case would be pkpr9d.tfm). I pine for a name like

Opinions welcome,