W3C

- DRAFT -

TPAC Personalization Breakout - Symbols

18 Sep 2019

Attendees

Present
CharlesHall, stevelee, DavidClarke, CharlesL, Roy, JamesC, Markku, Lisa, Janina
Regrets
Chair
CharlesL
Scribe
Steve

Contents


<LisaSeemanKest> i cant get into the meeting?

<LisaSeemanKest> can somone message me the password

<LisaSeemanKest> thanks

scribe stevelee

scribe+ stevelee

Charles: working on for p13n for a few years

<Roy> scribe: Steve

Charles: looked at AUI but found a few conflicts
... so using data-* attributes
... try a few alternatives and waiting to reach consensus and critical mass
... the symbol attribute provides for extra symbols and takes several parameters

Lisa: you don't change the specific different symbols in the content
... they are in personal preferences
... many symbol sets are proprietary and users have paid to use on the device (not licenced to use on the web)
... for symbols user symbols are their 'lnaguage'

Charles: deck is on the personalisation wiki
... time for a demo.
... once chrome extension is installed a new UI item allows personalisation settings to be altered
... [demoing setting ARASAAC symbols which appear on the page]
... when I select "personalize page" the symbols appear.
... if I change the profile then the Bliss symbols appear.
... can also say if appear above or next to words
... there is also option ot have symbols appear on [pointer] hover (AKA tooltips)
... need to set focus 1st
... can select another profile that only shows critical portions of a page
... we demoed symbols but there are several other personalisation
... eg numbers should be replaced with relative terms
... eg rather that 90 degrees say HOT

janina: useful for dyslexia and discalculia etc

Charles: we want people to get use to this and find out whicj are most popular
... and then try to get wider support (say browser)

<CharlesL> https://github.com/w3c/personalization-semantics/blob/master/TPAC/W3C-Personalization-2019-TPAC.pptx

link to slides ^^^

<LisaSeemanKest> i cant hear the question

<LisaSeemanKest> maybe type it into the chanel

charles: {in response to question] we have 3 levels of distraction setting in the preferences

<CharlesL> reader mode in browsers

Juan: Reading mode in browsers does simplify content but not support symbols

?: is there a move for browsers to standardise?

<CharlesL> always on reading mode

hadleybeeman: i personaly find low distraction modes very useful

?me thanks

<Zakim> hadleybeeman, you wanted to ask about privacy

hadleybeeman, is all this happening in the browser or server

<LisaSeemanKest> we have it as an extention and server side

<LisaSeemanKest> can you unmute me

<LisaSeemanKest> done

charles: lisa yo seem to have dropped of webex

<LisaSeemanKest> i can still here

<LisaSeemanKest> thanks

<LisaSeemanKest> can u here me?

<LisaSeemanKest> u cant here me, but if anyone is interested in the code or what we are doing please be intouch

<LisaSeemanKest> lisa.seeman@zoho.com

charlse: do you think this might be usefull to students in exams needing special considerations?

?: this could be more disruptive than what we've seen

<LisaSeemanKest> we very much want more implementions

<LisaSeemanKest> and we need more implementions\

<Roy> FYI, See the Personalization Overview at https://www.w3.org/WAI/personalization/

<Zakim> CharlesHall, you wanted to discuss element type of the symbol

<LisaSeemanKest> up to the implemetion I think

<LisaSeemanKest> ori is also on the webex

Charles: what the is the render element that contains the symbols?

jania: we could do with some code samples.

<LisaSeemanKest> +1

<LisaSeemanKest> and also some people just know a symbol that is familure to them like a hand set for phone

janina: saying some people need the symbols as it is their language

<LisaSeemanKest> but a younger user may not know what that is

charles: looked at code and is using png symbols in img element
... also use the old aui-attribute rather than data-*

<LisaSeemanKest> so it lets people with dementa etc use the old symbols, and designers use modern ones without losing an audience

charles: the client process is to got through content and inject symbols where ether the attribute appear

<Zakim> aboxhall_, you wanted to ask about implementations

aboxhall_, is the intention to publish the extension?

charles: leaving up to developers to decide how to make it available

stevelee: ideal eventually in the browsers

janina: user numbers are quite large

<aboxhall_> https://en.wikipedia.org/wiki/Blissymbols

<Roy> https://github.com/w3c/personalization-semantics/wiki/About-Data-Symbol

<LisaSeemanKest> ARASAAC are open free symbols

<Roy> http://www.blissymbolics.org/index.php/

https://mulberrysymbols.org/ also CC BY

<chrishall> http://www.arasaac.org/pictogramas_color.php

<mhakkinen> is there a source for the WordNet Symbols? Can't find anything.

<LisaSeemanKest> yes. they are conceprts

<LisaSeemanKest> but http://conceptnet.io is more useful

<LisaSeemanKest> wordnet not symbols

<LisaSeemanKest> very english and not all parts of spearch

<LisaSeemanKest> bliss is just a starting point

<LisaSeemanKest> wordnet is a speach thing

<chrishall> http://conceptnet.io/?

<LisaSeemanKest> no

<LisaSeemanKest> http://conceptnet.io/

<LisaSeemanKest> we have mapping between a few words a a single symbol

<LisaSeemanKest> yes

? is there work on machine translation words to symbols eg google translate

<LisaSeemanKest> but not reliable

<LisaSeemanKest> ibm concept coder

<LisaSeemanKest> we are testing it.

ori: no, lisa: yes butnot reliable

<LisaSeemanKest> lots of errors. very difficalt for this user group

<LisaSeemanKest> to work out what is needed

<LisaSeemanKest> what was intended

<LisaSeemanKest> ai not reliable enugh

<LisaSeemanKest> at this point

tomayac: client side machine translation by avoid potential security issues

<tomayac> tomayac is Thomas Steiner, Google

<jcraig> https://en.wikipedia.org/wiki/Ruby_character

<LisaSeemanKest> thank YOU1!

<tomayac> Advertisers could embed fake annotations to see if a "load" event is triggered

<LisaSeemanKest> sory about the tech problems

<CharlesL> Summary: Presented an overview of the prototyping data-symbol attribute for personalization of web content. Showed a demo using a Chrome Browser Extension that could switch depending on the users personal preferences between using BLISS or ARASAAC symbols which would show up as images above the text they represent. Also shown was how the personal preference could show only the critical sections of the authored page and not ren

<CharlesL> der those portions of the page below a certain threshold also set by the user's "data-distraction" level of which there are three choices (low/med/critical).

Summary of Action Items

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: 2019/09/18 19:01:36 $

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/?/Juan/
Succeeded: s/?/hadleybeeman/
Succeeded: s/concept.net/http://conceptnet.io/
Succeeded: s/?/tomayac/
Succeeded: s/ARASEC/ARASAAC/
Succeeded: s/arrosac/ARASAAC/
Present: CharlesHall stevelee DavidClarke CharlesL Roy JamesC Markku Lisa Janina
No ScribeNick specified.  Guessing ScribeNick: stevelee
Found Scribe: Steve

WARNING: No "Topic:" lines found.


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report


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]