google / fonts

Font files available from Google Fonts, and a public issue tracker for all things Google Fonts
https://fonts.google.com
18.14k stars 2.61k forks source link

filter for multiplexed fonts #5504

Open PeterConstable opened 1 year ago

PeterConstable commented 1 year ago

It's great that there's a knowledge article explaining duplexed / multiplexed fonts. It would be even better if the Google Fonts portal also had a way to filter for families that have multiplexed styles.

RosaWagner commented 1 year ago

Thanks for the feedback @PeterConstable

davelab6 commented 1 year ago

I am sure this can be automatically detected. Somewhere there's a script to confirm that kerning does not differ across grade masters in UFO sources, that could be used as a reference for this.

That should also be used as a fontbakery check on the grade axis, as part of axis registry compliance checking, to ensure grade never creates reflow.

But there are also typefaces, most monospace, and others like Asap and Recursive, that don't have a grade axis, but their weight axis - or set of static weight instances - also have the same metrics and changing them doesn't cause reflow.

Therefore I think the metadata tag could be uniwidth, for all these "whole family can shift weight without reflow" options. Since METADATA already supports multiple categories but only the final one is actually used, with such a script, we could insert (prepend) the category key with a "uniwidth" value in METADATA to get started. However, I think that's a second step and it may be premature to do so.

Finally, there also are "duplexed" families or super families, which I define as where the metrics are the same for Roman and Italic, or even serif and sans, and it would also be good to have a script to identify these, and a tag with additional metadata about what the set of styles with shared metrics is, exactly. The du in duplex is for the 2 typically unrelated things being related, and multiplexed is sometimes used for weights where there is "more than one", but this seems like a static centric thinking anachronism.