W3C

– DRAFT –
(MEETING TITLE)

18 February 2025

Attendees

Present
Abhinav, Annalu, janina, Lionel_Wolberger, Russell
Regrets
-
Chair
Lionel_Wolberger
Scribe
janina, Lionel_Wolberger

Meeting minutes

Meething: WAI-Adapt Weekly Teleconference

date 17 Feb 2025

Introductions, Administratrivia

Lionel_Wolberger: We are not meeting March 4th

<Annalu> presemt+

Proposal for session on Well Known Destinations

Lionel_Wolberger: Thanks, Abhinav for submitting the draft.

Lionel: (read the draft)

janina: A lot of the needed concepts is in there.
… I'd like to see more of "Predictable; Predictable keywords, predictable results
… and the idea that we will be proposing reserved words for this

Symbols

Lionel_Wolberger: I am setting meetings for Zero Project, re: symbols. What should be my focus?

<Annalu> d.banes@dateurope.com

janina: We can use people's 'roladex'
… We'd like her thoughts on supporting other people's symbol sets

Russell: I see significant limits on the use of unicode

janina: People at ZP meeting can help us make this argument

Annalu: I am very interested in the increased visibility that we gain, whether it's Unicode or Bliss
… I am concerned about when the Symbol set is seen as identical to a language, and Bliss is more of a language

Russell: It's not Bliss vs unicode. If Bliss were mapped to unicode by a single character per symbol, the problem would be largely gone
… but the issue of a symbol that is not reference by the lookup-code-points remains
… let's call the missing symbols, the 'expansion problem' -- that problem remains, unless we define a static, basic set for 90% of cases

Annalu: Consider Basic English. 100 words that were selected and have a very broad power of representation
… we may benefit best from a high level discussion, how to do the minimum for maximum benefit

Russell: Take the perspective of a symbol set provider. Procedure: take the list of code points, and create a font with your symbols, one for each code point
… This is likely a compelling and simple proposition
… the problem is that Bliss does not map all code points.

Lionel_Wolberger: Revisiting the use case- people are decorating sentences with selected pictograms (they tend to be emojis today)
… WAI-Adapt is meant to allow these pictograms to be rendered in the end-user's vocabulary

janina: +1 to what Annalu said regarding limited vocabulary
… e.g. recipes, procedural lists

Annalu: I advocate returning to the use cases and sticking to them
… people who use symbols, because they have difficulty in reading and/or comprehending
… if they can comprehend the vocabulary, the will likely choose to hear it read to them
… The written, displayed symbols will be more limited. That's why I think we are looking at a limited vocabulary

Lionel_Wolberger: The Bliss registry will likely be a sufficient 'minimal' code list

Russell: I could list the 'single character' Bliss codepoints proposed in the current unicode draft

Annalu: I can send a ink to the basic vocabulary list

<Annalu> https://en.wikipedia.org/wiki/Basic_English

Lionel_Wolberger: (Demonstrates how ChatGPT is adding emojis to its standard responses)

Russell: Sample from a poem I generated, A golden crust, a cozy thing 🏠.

zaim, end meeting

Minutes manually created (not a transcript), formatted by scribe.perl version 242 (Fri Dec 20 18:32:17 2024 UTC).

Diagnostics

Maybe present: Lionel, Meething

All speakers: Annalu, janina, Lionel, Lionel_Wolberger, Meething, Russell

Active on IRC: Abhinav, Annalu, janina, Lionel_Wolberger, Russell