Meeting minutes
Agenda Review
Action Items
<addison> https://
addison: action-1183, we scheduled a call but person didn't show up.
atsushi_: action-1196 next meeting is next Tuesday.
r12a: will merge actions 1203 and 1173
addison: What email address to use for Markus Scherer(sp?) at Unicode? Didn't get response.
<r12a> https://
r12a: Markus is the only chair, but there are other people in the group, see ^^
… You can try writing to all of them.
<addison> close action-1203
<trackbot> Closed action-1203.
<addison> action-1212?
<trackbot> action-1212: Addison Phillips to Merge specdev pr 80 and copy glossary items to glossary -- due 2022-10-13 -- OPEN
r12a: I sent some editorial comments on the PR (re: action-1212)
<addison> action-1214?
<trackbot> action-1214: Richard Ishida to Finalize and publish font styles and fallbacks doc -- due 2022-10-27 -- OPEN
<addison> close action-1214
<trackbot> Closed action-1214.
r12a: I did action-1214.
Info Share
addison: action-1215, will schedule after the DST time change.
<r12a> The Script Links page has been updated to support Unicode 15. Also some addition information added (esp. language use), and bugs fixed.
<r12a> https://
addison: I updated the calendar with more iterations of this call. Time will change next week and after for some people, as the time is fixed in Europe.
r12a: I updated some pages for Unicode 15, see ^^
RADAR Review
<addison> https://
addison: I filed trace-context issues to awaitinf comment resolution.
RADAR Review
Pending Issue Review
<addison> https://
addison: Some incoming issues.
<addison> https://
r12a: I took a look at it.
… This test ^^ shows different in Firefox and Edge. Scrollbar left or right.
… But that is tangential.
… Question is if user could switch the side, based on being left-handed, e.g.
addison: Would that be a CSS thing or a browser setting?
r12a: It is about a Media Query for it.
… I can respond.
<addison> https://
addison: Somebody wants to mark up language fragments inside an attribute. Richard already responded.
xfq: Should remove the "pending" label.
<addison> https://
r12a: I think we should all look at these issues as they come in.
addison: I talked at TPAC with Ian Jacobs and Stephen McGruer about the language to show a payment request in. They added it to their spec, that's why it is marked "close?". OK to close?
xfq: Looks good to me.
<addison> https://
addison: About gloassary and the term natural language vs human language.
… We use natural, WCAG uses human.
… Do we want to change? Or add a term?
JcK: Natural language is the term everybody understands. Might create confusion if we change. Or add a footnote.
… Code is a human language, too.
<xfq> https://
addison: Who wants to own that issue 16? I'll take it.
unicode-range issue
<JcK> @Bert: actually two problems there. One is "code" as in "programming language" and the like. The other is "code" as secrecy, with encryption being a subset of that category.
<addison> https://
<addison> https://
addison: CSS has a unicode-range descriptor. It accepts code-point ranges.
… I pointed them at ICU and Unicode regular expressions.
… Do we add a label needs-resolution or just track?
r12a: Probably just track at the moment.
… About your comment in the issue whether greek is equiv to ISO15924 Grek: I hope not.
<addison> https://
atsushi_: I think some languages are unified.
xfq: Many punctuations and symbols are shared by scripts.
… Author needs to specify script explicitly in the style sheet.
addison: When it says "arabic", does that mean also all the compatibility characters, e.g?
xfq: Code-point syntax hard to read. Maybe put the characters themselves?
addison: How useful is the feature?
r12a: There is debate. It doesn't always do what you expect.
… Some presentation forms for Arabic are often used. But even if you can write the character instead of the code point, you still have to find out what the first character in the range is.
xfq: More useful for single characters.
r12a: But if you put characters from rtl scripts in the value, the value may actually become harder to read.
addison: CSS doesn't currently have Unicode regular expression, does it?
… JavaScript does.
xfq: There is a long-standing issue, but there are performance problems.
<xfq> https://
xfq: Example of a style sheet I've written with unicode-range ^^
addison: Anything we want to do with it? Let CSS tussle with it for a while?
Non-ASCII characters in CSP policy
<r12a> Using unicode-range in font-face in CSS https://
Non-ASCII characters in CSP policy
<addison> https://
<r12a> Using unicode-range in font-face in CSS https://
<addison> https://
JcK: Rational would be to make it correspond to A labels, but that needs a lot of testing.
<xfq> something like this: Content-Security-Policy-Report-Only: script-src 'self'; report-uri https://
addison: Could the browser do the work of encoding, assuming the value is in UTF-8?
JcK: Differences in edge cases between UTR-46 and IDNA.
… WhatWG only looks at what browser does. But other things count, too, such as DNS.
addison: We also looked at email recently.
JcK: Email is extremely precise.
I18N-glossary
addison: I asked input events to define "grapheme cluster"
… What is our policy going forward? We said glossary would remain informative.
… But there may be definitions we want to point people to.
… Marcos suggested they could be put in the Infra spec.
r12a: We need a "W3C glossary".
… People have worked on it in the past.
addison: Normative?
r12a: At least authoritative.
… I might ask Dom what happened to it.
… We now have tools like ReSpec that make it easier to update the glossary.
addison: In any case, we still want our glossary.
r12a: Would we? If we can easily update a W3C glossary...
addison: We have definitions like for grapheme cluster. I'm not unhappy with them, but they are not normative.
r12a: SpecDev is not going to be a Rec either.
addison: No, but it says what to write in your own spec.
… We have a normative definition in Charmod, but it makes me itch.
AOB?
addison: Watch your calendar for the right time of this telcon next week.