I just re-ran the pictograph query and got 3231 results out of ~7.5 million pages, which puts us back in the 0.04% range.
Adding a use-counter sounds like a reasonable way to see what actual usage looks like. On Thu, Aug 19, 2021 at 7:36 AM Frédéric Wang <fw...@igalia.com> wrote: > Le 13/08/2021 à 14:23, Frédéric Wang a écrit : > > > > Interoperability and Compatibility > > This feature was implemented by Apple in 2011 before the Blink fork and is > still implemented in WebKit. It has never been implemented in Firefox. > > A HTTPArchive search from March 2020 provided 1903 pages out of ~5millions > (i.e. 0.0003806%). I'm adding a user counter to actually measure when a > -webkit-pictograph font is actually resolved to the corresponding user font > setting which may give more accurate/relevant data. > > One motivation is to improve interop with Firefox and spec compliance. But > I'm also trying to refactor our internal font-family implementation that is > inherited from WebKit time and is a bit messy right now. Original generic > names like "serif" or "cursive" have web-exposed bugs ; the recently > implemented "system-ui" too but behaves inconsistently ; and we have > non-standard values like -webkit-pictograph. Once things are cleaned up, we > can consider implementing new values like font-family: emoji, math, > fangsong, ui-serif, etc without adding more problems... > > ... however, one can also argue that it's would be better to implement > "font-family: emoji" as a replacement/alias to "font-family: > -webkit-pictograph" before deprecating/removing the latter. Again, this is > possible but mean we would add more web-exposed bugs / inconsistencies in > the meantime. > So I'm not really sure about the best approach. Sending "Intent to > Prototype" for now and waiting for feedback. > > *Gecko*: Positive No support for -webkit-pictograph > > *WebKit*: No signal ( > https://lists.webkit.org/pipermail/webkit-dev/2021-August/031938.html) > This has been implemented in WebKit since 2011. In general Apple is against > removing features that could potentially break web compat or their > platform. I asked them to see if they would be happy to add "font-family: > emoji" as an alias for -webkit-pictograph, as they did for "system-ui". In > general about the current font-family implementation, Myles C. Maxfield > commented in the github PR to add WPT tests that he is aware of the issue > and doesn't think it's desirable that these -webkit-* values are > web-exposed. > > *Web developers*: No signals "font-family: emoji" was added in the CSS > fonts spec, so I guess there is interest to make this more standard. I was > not able to find the github discussion, though. > > So just to follow-up here too, > > HTTPArchive result from March 2020 from Yoav were: > > * -webkit-pictograph alone represented less than 0.04% of pages. > * -webkit-pictograph + -webkit-body + -webkit-standard represented less > than 0.1426% > > That sounded big, so I had started to prepare a use counter that would > provide finer measurement (i.e. only measure when the font family setting > is actually resolved for -webkit-pictograph) : > https://chromium-review.googlesource.com/c/chromium/src/+/2124260 > > New results from August 2021 provided by Yoav and Dominik showed that > together -webkit-pictograph + -webkit-body + -webkit-standard represent > less than 0.004% of HTTPArchive pages so it's an order of magnitude smaller. > > Reference doc: > https://docs.google.com/document/d/1nYJzL-MWQrTmf9Z-KscTWuM_5n6-IVdJeEllJ3Appro > > -- > Frédéric Wang > > -- > You received this message because you are subscribed to the Google Groups > "blink-dev" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to blink-dev+unsubscr...@chromium.org. > To view this discussion on the web visit > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/e6efc0e1-34a4-8546-a03e-53d69871ded5%40igalia.com > <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/e6efc0e1-34a4-8546-a03e-53d69871ded5%40igalia.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "blink-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscr...@chromium.org. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAL5BFfXzzmHjTpKQx_hfYdPK64ne4-hNiBVLdhpjVm2jV5tV8Q%40mail.gmail.com.
