Personalization Task Force Teleconference

01 February 2021


becky, CharlesL, janina, JF, LionelW, LisaSeemanKest
Lionel, LionelW

Meeting minutes

<sharon> Charter for APA (due end of the month) - https://raw.githack.com/w3c/apa/charter-2021/charter.html

<LisaSeemanKest> https://www.w3.org/WAI/PF/wiki/Teleconference_cheat_sheet

Reminder to all: autocomplete is working, start typing a person's name and their full alias will auto-complete.

<LionelW> Agenda: Item #1 Review draft to I18N (Lisa)

<LisaSeemanKest> -https://lists.w3.org/Archives/Public/public-personalization-tf/2021Jan/0017.html

Sharon: This is in response to 145 (which has joined with 144)

John: The purpose of the spec is to create machine readable tags
… the spec should be agnostic to the language of the content. Content is person to person, the spec stipulates machine to machine communication.
… What is the answer?

Lisa: Add an editor's note saying: we are supplying syntax for only the purpose of the element. Authors should be aware that the default formats are specified in language date and time specifications [XXXX]. When using a format that is not the default, authors should ensure it is specified. For example, by using [XXX].
… The editor's note sends the reader to the proper specification
… If a18n is OK with this, done. If not, then we need a call.
… XXX is defined as relevant different Lang, Date and time semantics and specifications for different locations , and XXX is the default specifications.)
… Let a18n tell us the right specification.

Becky: Agree.
… John says we must be machine readable. We only said, 'date/time'. If you need to know how to express date/time you need the right specification.

John: Example: I cannot write simplified chinese, but I can write the coding language, the markup, e.g. <table>

<CharlesL> I agree with John and Becky

Janina: agrees with John and Becky

<JF> Our "CODE" is language agnostic

Janina: we defer to a18n for which particular specification to point to

<Zakim> JF, you wanted to ask how and where (and why)

John: I do not see a good place to put this note. In fact, putting the note drags us down a 'rabbit-hole'

becky: Why didn't they object to WCAG 2.1, which added things as values?
… It seems, they want us to use HTML, which is fine. But the non-Gregorian calendar systems is another issue.

Action: JF to draft a response to i18n

<trackbot> Created ACTION-77 - Draft a response to i18n [on John Foliot - due 2021-02-08].

Draft response to issue 76 (John)

<JF> https://github.com/w3c/personalization-semantics/issues/76

Sharon: Looks closed and done.

John: It seems that people and engineers do not really understand what we are trying to do.
… education and outreach partnership with __ would be helpful
… Education and Outreach (E&O)
… to communicate what we are trying to do.

to johm, yes we hear them all

Janina: We get a complaint, the name of the TF also possibly misleads, 'personalization' doesn't fully and accurately communicate

Charter for APA (due end of the month) - https://raw.githack.com/w3c/apa/charter-2021/charter.html

Becky: Add another non-normative one?

Janina: We don't have to be too specific about non-normative, it's the normative that needs to be well-named

Janina: We encountered, 'does our QTF need to be specific about the areas that it will look into?" and the research group found it can do whatever non-normative documents support the grup's scope
… so we do not need to be too specific about the non-normative
… so charter is not a main concern and can be closed.

Janina: Lisa, take a look at COGA status, it is a candidate for being dropped.

Review other remaining Actions - https://github.com/w3c/personalization-semantics/wiki/actions

Sharon: can we clarify the action item, "CFC on WD of Explainer"
… Is someone working on, "add to tr explainer - user needs from email to atg" ? Is that tag (not atg)?

Sharon: John reaching out to Steve?

John: Not a great time right now for Steve (personal issues)

Sharon: look over other open issues?

Sharon: The rel attribute

<sharon> https://github.com/w3c/personalization-semantics/issues/172

Issue #172: There seems to be some overlap between the destination attribute and the existing rel attribute. What would be the difference between data-destination="help" and rel="help"? Would authors be encouraged to use both?

John: rel is short for, 'related to'
… so there is not much difference between these two, just as aria-label is really deep down the same as alt
… current tooling is not taking advantage of destination

<LisaSeemanKest> fyi, i have a request from implemtor, so we should probably respond to that

Janina: I agree, this sounds right

Sharon: So John will respond, after double-checking 'rel' making sure it can take string text, while we are using a fixed taxonomy.

Becky: MDN says rel does have fixed values, see https://developer.mozilla.org/en-US/docs/Web/HTML/Attributes/rel

John: Hmmm, if an author uses both, we will need a conflict resolution path. Comparing again to aria, there is a clear order of preference
… so we will need to specify here, which has the stronger semantics and over-rides. So here, data-destination should over-ride rel=

Becky: Think also of 'pervious' and 'next'

Janina: 'alt' remains a broad purpose tag, 'aria' is devoted today to screen readers. (not the original intention)

Sharon: Look at module 1 and module 2

John: Need some more research

Becky: Note that rel is context sensitive and functions differently depending on element (anchor, ...)

Action: JF to research @rel versus @action and @destination and report back to TF

<trackbot> Created ACTION-78 - Research @rel versus @action and @destination and report back to tf [on John Foliot - due 2021-02-08].

Charles: What is their list for rel? Is ours even needed anymore?

John: Remember, even though alt was pre-existing, we added ARIA

Lisa: We were thinking that rel would be limited to only certain elements.
… suggest we expand it (liks with tab-index and ARIA) or do it ourselves.

<LisaSeemanKest_> need to step away

<LisaSeemanKest_> should be back in a minuet

John: Also, rel seems to be little-used (like kbd)

Sharon: Agenda:

<sharon> https://github.com/w3c/personalization-semantics/issues/170

<JF> JF reclaims his comment: @rel is used *ALL THE TIME* to link stylesheets to documents

Becky: (discussion of reduced motion, distraction values)

Lionel: Clarify content module 1, 2, 3 please.

<JF> https://www.w3.org/TR/personalization-semantics-1.0/#modules

Charles: Link from the explainer


<sharon> https://github.com/w3c/personalization-semantics/issues/160

Sharon: To read it, "In building design we have the concept of marking an accessible route. Is this something you have been considered within the web space? ....

Janina: Sounds like WCAG 3 performance scoping.

John: The goals are close if not the same.
… our rubric is, "given three paths, which is the most effective path.
… "take path three, do steps 1,2,3,4,5

Janina: But, "accessible for who?" is the question

Becky: reminder, she wrote "worth discussing even if the end decision is not to pursue it."
… we can respond, we plan to revisit this when we work on add'l modules

John: and invite rachael!

Summary of action items

  1. JF to draft a response to i18n
  2. JF to research @rel versus @action and @destination and report back to TF
Minutes manually created (not a transcript), formatted by scribe.perl version 127 (Wed Dec 30 17:39:58 2020 UTC).


Found 'Agenda:' not followed by a URL: 'Item #1 Review draft to I18N (Lisa)'.

Maybe present: Charles, John, Lionel, Lisa, Sharon