Pronunciation Task Force Teleconference

27 Mar 2019


irfan, Roy, Dee, paul_grenier, janina, SteveNoble, JF, Sam
Dee, Dee_Dyer


<Roy> ACTION: Roy to create a track

<trackbot> Created ACTION-1 - Create a track [on Ruoxi Ran - due 2019-04-03].

<Roy> https://www.w3.org/WAI/APA/task-forces/pronunciation/track/

<Roy> trackbot, close action-1

<trackbot> Closed action-1.

<Roy> trackbot, status

<irfan> ACTION: Mark-Hakkinen update the gap analysis document

<trackbot> Error finding 'Mark-Hakkinen'. You can review and register nicknames at <https://www.w3.org/WAI/APA/task-forces/pronunciation/track/users>.

<irfan> trackbot, close action 2

<trackbot> Sorry, irfan, I don't understand 'trackbot, close action 2'. Please refer to <http://www.w3.org/2005/06/tracker/irc> for help.

<irfan> trackbot, close action-2

<trackbot> Closed action-2.

<irfan> ACTION: Markku update the gap analysis document

<trackbot> Created ACTION-2 - Update the gap analysis document [on Markku Hakkinen - due 2019-04-03].

<Roy> https://www.w3.org/WAI/APA/task-forces/pronunciation/track/

<Roy> https://www.w3.org/WAI/APA/task-forces/pronunciation/track/users/35712

<irfan> ACTION: Steve update the gap analysis document

<trackbot> Created ACTION-3 - Update the gap analysis document [on Steve Noble - due 2019-04-03].

<irfan> ACTION: Christine update the User Scenarios document

<trackbot> Created ACTION-4 - Update the user scenarios document [on Christine Loew - due 2019-04-03].

<Roy> scribe: Dee

<irfan> scribe: Dee_Dyer

Way to work, google doc or GitHub?

<Dee> Irfan: Steve Noble raised this point.

<Dee> Irfan: Some are not familiar with github. So we will talk more about this.

<Dee> Steve: Experience with githun from programming point of view. Github does not seem to be the best for documents. If a very small number of editors are involved, could that be done in whichever edition environment that is convenient.

<Dee> Steve: would like to avoid initial drafts in github.

<Dee> Irfan: Understood.

<Dee> Janina: Final form should be HTML. Choose document creation accordingly.

<Dee> Paul: markdown is a native syntax in github

<Dee> Paul: all issues are structured that way

<Dee> Paul: can copy it to HTML and then manipulate.

<Dee> Irfan and Janina agree with Paul

<Dee> Irfan: offer training on github, would this be helpful

<Dee> Janina: Training would be useful. However, there is more than one interface.

<Dee> Steve: literary document does not have much markup

<Dee> Paul: code samples would not be well formatted in Google docs

<Dee> Janina: Again output needs to be in HTML

<Dee> Steve: example, 3 editors working simultaneously on a document, simple to do in something like Google docs

<Dee> Janina: command line in github is easier for me

<Dee> Steve: doesn't use an HTML editor to create a document

<Dee> Irfan: If you want to work in Google docs, work with Irfan or Roy to update HTML

<Dee> JF: We can work in whatever format we want. The advantage of github, multiple people working on it. Other working groups use different tools, such as a Wiki. In the end, it needs to be HTML. Github has proved to be effective.

<Dee> Irfan: Does that make sense?

<Dee> Steve: whatever anyone wants to start with appears to be fine.

<Dee> JF: Whatever we create in github would be a working document.

<JF> +1 to have a Primer course on using GitHub

<Dee> Irfan: Roy has offered to help with gfithub.


<Dee> Irfan: topic raised by Roy

<janina> I expect to be there

<Dee> Irfan: discussin a face-to-face at TPAC

<Dee> Janina: expects to be there. APA will have usual 2 days. Can make space for this task force during APA sessions.

<Dee> Janina: Time given depends on interactions with other W3C groups.

<Dee> Irfan: We will have more information as we get closer to Sept.

<Dee> Irfan: 2 docs, user scenarios and gap analysis

New Members

<Dee> Irfan: 3 new potential members, 1 from Rutgers, 1 from MIT, 1 from UK Police(?)

<irfan> Robin Ferguson from UK police

<irfan> http://www.pnct.apps.police.uk

<irfan> Sunish Gupta from MIT

<irfan> M. Ketty Ombadykow from Rutgers

Gap Analysis

<irfan> https://www.w3.org/WAI/APA/task-forces/pronunciation/track/

<Dee> Irfan: I have created some action items and assigned tasks.

<Dee> Irfan: Mark and I are working on the gap anlayis.

<Dee> Irfan: Tentative due date: April 30

<Dee> Irfan: In previous call we talked about various approaches.

<Dee> Irfan: Any additions to the previous discussions?

<Dee> Paul: I added all of the examples to issues in github. So far no comments.

<Dee> Irfan: Issues are on the agenda.

<Dee> Janina: can't resolve gap analysis until we know requirements

User Scenarios

issue list from wiki

<irfan> https://github.com/w3c/pronunciation/issues

<irfan> https://github.com/w3c/pronunciation/issues/10

<Dee> Paul: issues in reverse chronological order

<Dee> Paul: They follow the same format

<Dee> Paul: Summary, Issues from a Technical Standpoint

<Dee> Paul: Understanding our user scenarios important. Issues can be seen on a web page. Links available for testing.

<Dee> Paul: It's on Glitch

<Dee> Paul: need to understand use cases to evaluate implementation approaches.

<Dee> Paul: very simple one use, same phoneme (pecan variations)

<Dee> Janina, examples may be appropriate for gap analysis

<Dee> Irfan: we can have more examples. I agree with Janina

<irfan> https://github.com/w3c/pronunciation/issues/7

<Dee> Paul: custom element, details to be determined

<Dee> Paul: ruby, based on existing standards

<Dee> Paul: SSML, browser treats it like a div

<Dee> Paul: aria-details came up last week

<Dee> Paul: only ssml is in there without content

<Dee> Irfan: can you add more details to ruby, microdata, and custom element?

<Dee> Irfan: add to issue itself

<Dee> Paul: if you have comments or suggestions, I will be editing to incorporate them

<Dee> Janina: aria-details, supposed to be what we move to instead of using longdesk, you can throw many things into aria-details

<Dee> Janina: full use of aria-details not fully realized

<irfan> https://github.com/w3c/pronunciation/issues/8

<Dee> Janina: details display on demand, could be a configuration setting

<Dee> JF: presented as a drawer type of thing, expanding and contracting div

<Dee> JF: no robust support in all browsers

<Dee> JF: goal to provide pronunciation support, do we want to expose that to sighted users who struggle to pronounce complex words?

<Dee> Janina: fundamental questions, do we want this to be a general support?

<Dee> JF: any solution could work beyond AT

<Dee> JF: tooltip implementation to show how to pronounce the word

<Dee> JF: In terms of thinking about it, keep this in mind.

<Dee> Janina: user scenario needs to be added

<Dee> Paul: this is why I started looking at microdata and microformats

<Dee> JF: They have support in search engines, but not browsers

<Dee> JF: not W3C standards

<Dee> Paul: can provide microformats example

<Dee> JF: Microformats simply a Wiki. Was of interest a while back, but not a lot of interest now. Personalization task force walked away from it.

<Dee> Janina: we need broad publishing and assessment support

<Dee> Irfan: good to have as part of analysis

<Dee> Irfan: Can we address all use cases using SSML?

<Dee> Irfan: data-SSML platform dependent. Only works on MAC?

<Dee> JF: data- attributes need a helper app. Point was to allow experimentation.

<Dee> Irfan: The way SSML is implemented by Apple, it is different from the rest.

<Dee> Paul: JSON and data-ssml, non-standard

<Dee> Paul: issues contain specific information

<Dee> Paul: no other attributes take JASON, it is something that frameworks do.

<Dee> Irfan: we need more information. I will look at the issues.

<Dee> Irfan: request everyone to look as well

<Dee> Irfan: and provide comments

<Dee> Irfan: JASON approach - taken due to validity and ease for AT and browsers

<Dee> Irfan: there could be several way to write

<Dee> Irfan: idea to come up with a common approach acceptable by AT and browsers

<Dee> Irfan: using more than one property at the same time

<Dee> Irfan: we can refine the approach

User Scenarios

<Dee> Paul: has a conversation with a PhD game designer

<Dee> Paul: retrofitting games so that players could use QR codes to check pronunciation and do it all with web technology

<Dee> Paul: QR code would take you to a web page with encoded SSML, browser updated, work is read aloud, no additional plugins.

<Dee> JF: QR codes can link to MP3 files.

<Dee> Paul: could be a fallback.

<Dee> JF: would love to see multiple ways of exposing this data

<Dee> JF: as we think trough solutions, we need to be as robust as possible

<Dee> Janina: may need to lay out a roadmap

<Dee> Janina; we may not come up with one single solution

<Dee> Irfan: thank you Paul, for the issues.

<Dee> Irfan: add any other scenarios.

<Dee> Irfan: add use cases

<Dee> rssagent, make minutes

Summary of Action Items

[NEW] ACTION: Christine update the User Scenarios document
[NEW] ACTION: Mark-Hakkinen update the gap analysis document
[NEW] ACTION: Markku update the gap analysis document
[NEW] ACTION: Roy to create a track
[NEW] ACTION: Steve update the gap analysis document

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.154 (CVS log)
$Date: 2019/03/27 15:02:43 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/John:/JF:/
Default Present: irfan, Roy, Dee, paul_grenier, janina, SteveNoble, JF, Sam

WARNING: Replacing previous Present list. (Old list: (no, one), irfan, Roy, Dee, paul_grenier, janina)
Use 'Present+ ... ' if you meant to add people without replacing the list,
such as: <dbooth> Present+ irfan, Roy, Dee, paul_grenier, janina

Present: irfan Roy Dee paul_grenier janina SteveNoble JF Sam
Found Scribe: Dee
Found Scribe: Dee_Dyer

WARNING: 0 scribe lines found (out of 192 total lines.)
Are you sure you specified a correct ScribeNick?

Scribes: Dee, Dee_Dyer
Found Date: 27 Mar 2019
People with action items: christine mark-hakkinen markku roy steve

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)

[End of scribe.perl diagnostic output]