Round-off errors in metrics & fontinst/finst status...

["Melissa O'Neill" <oneill@cs.sfu.ca>]

Around the start of this month I wrote:
>> [T]here is still the problem with fractional widths in the AFM file;
>> fractional widths for Kern pairs do make fontinst crash, and I have my
>> suspicions that fractional widths for characters get truncated to
>> integers.

... and Sebastian Rahtz replied:
> ah, that might be an area where fontinst needs fixes

I'm just curious as to whether whether anyone is working on fixing these
issues anytime in the near(ish) future. I've already written a Perl
script that is an `AFM rounder', but am considering some gross hacks to
either MTX files and/or fontinst itself to better support fractional
widths. I would gladly abandon such efforts if I knew that sometime soon
there would be a new version of fontinst that understood such things.

It'd also be useful to know just how much difference the current `round
to an integer' process makes -- Berthold Horn did seem pretty outraged
at the very idea, but are other people similarly upset by these
inaccuracies. Is there a noticable difference?  Similarly, when vptovf
and pltotf squark about how they had to round things, I'm left wondering
if I should or shouldn't care... (and if I am supposed to care, just what
it is I'm supposed to do).

Opinions welcome,

    Melissa.