On 12/29/2024 1:29 AM, Julius Ross wrote:
Mikael,
Thank you. Your reply is enough for me to understand that I am not doing anything (obviously) wrong!
I am not sure I have the expertise enough to contribute. But having a negative kern value in a font move glyphs apart feels more like a sign- error/bug than a choice/convention.
A next upload will have a better check and catch the issue.
I never use (and rarely see) prescripts in mathematics, and I only learned of their existence in TeX from your "math in context" document :-)
The opentype math spec certainly implies they were thinking of postscripts, and I just presumed symmetry. So.for a symmetric letter (e.g. V) I would expect the presubscript and postsubscript to be placed symmetrically as they would both be qualifying the variable V.
I'm not sure what they were thinking. It looks a bit like a 'maybe nice' feature. Even cambria (the reference font) is very incomplete and we noticed issues (unwanted overlaps) so in the we just ditched the whole concept as unreliable (other fonts had weird staircase kerns too) and went for a variant.
That said, it seems that other fonts have not bothered taking care with this (e.g. STIX has bottom right corner kerns on V but not bottom left). So perhaps just not worrying too much about prescripts (for mathematics) is the way to go.
Indeed, a rather inconsistent and semi-random feature. In the end it doesn't bring much to the table. FWIW, we have discussed this in articles and progress reports. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------