font-variant-*

[svgeesus]

Stats on which font-variant-* longhands such as font-variant-numeric or font-variant-east-asian are being used.

Also whether the font-variant shorthand is being used with the none value, to reset all the longhands.

[Crissov]

Comparative stat for font-feature-settings?

[svgeesus]

Comparative stat for font-feature-settings?

Indeed, that would be very useful, particularly where one could use either approach. IIRC font-feature-settings was deployed in browsers first, so some usage may simply reflect "that was the only thing that worked at the time".

CSS Fonts 4 has the data to map feature codes to variant values.

[svgeesus]

*Correspondences between high-level `font-properties and low-levelfont-feature-settings` features**

font-kerning

• kern
• vkrn

font-variant-ligatures

• liga
• clig
• dlig
• hlig
• calt

font-variant-position

Perhaps check if stylesheet authors try to nest this (it doesn't work)

• subs
• sups

font-variant-caps

Also depends on value of font-feature-settings

• smcp
• c2sc
• pcap
• c2pc
• unic
• titl

font-variant-numeric

• lnum
• onum
• pnum
• tnum
• frac
• afrc
• ordn
• zero

font-variant-alternates

This one was dropped from CSS Fonts 3 due to lack of implementer support. This is an area where the Open Web Platform lags behind native, which has good alternates support.

• hist
• salt
• ssxx where xx is 00 to 20
• cvxx where xx is 00 to 99
• swsh
• cswh
• ornm
• nalt

font-variant-east-asian

• jp78
• jp83
• jp90
• jp04
• smpl
• trad
• fwid
• pwid
• ruby

[LeaVerou]

Alright, so this again seems pretty straightforward:

1. We measure usage of the various font-variant properties, including usage in the font-variant shorthand and the font shorthand. We should probably keep the counts separate, to see if the shorthands are used at all (or we can just do that in #10 and not care about it here)
2. We measure usage of the various low-level settings @svgeesus posted about by measuring incidence of the code in quotes in font-feature-settings (single or double quotes).

[LeaVerou]

After consulting with @rviscomi, it appears the overlap with the Fonts chapter is too significant, so we will not have a Fonts section. Therefore, I'm closing this.