See also: IRC log
<LisaSeeman> trackbot, start meeting
<LisaSeeman> agenda: this
<LisaSeeman> issues: https://github.com/w3c/wcag21/issues?q=is%3Aissue+is%3Aopen+label%3ACOGA
<LisaSeeman> We also now have a trello board to-do list What have we missed for issues page
<LisaSeeman> lol
<LisaSeeman> that was last week
<LisaSeeman> ping me if you need the passwoed for webex
<LisaSeeman> call starting now
<LisaSeeman> scribe: Mike_Pluke
For tomorrows Coga call decided to switch to "Minimize errors" and "timed events" as they are easier to get through
Moving to a strategy of having "pillars" of acceptable Coga SCs in 2.1 with a supplemental document covering other issues
Volunteers sought to work with Lisa on choosing which things should be moved to the supplement
First LV approach described by Jim Allan
User needs document created first and then chose specific SCs.
A change any element on the web page was too general - so went for lower-scoped specific SCs
Implementation directions have been looked at - CSS modification (e.g. stylish) and also looked at PDF implemtntations
<LisaSeeman> https://w3c.github.io/personalization-semantics/
Coga more ambitious
<LisaSeeman> https://github.com/ayelet-seeman/coga.personalisation
Has an approach that relies on personalization
ARIA-based and a open source plugin that reads JSON settings file and will adapt the page appropriately
Lisa screen-shared the personalization demo
Showed page simplification that showed only "Important" items and then only "Critical" items
Page authors have to add semantics to pages to identify things like "important" and "critical"
<allanj> can you post the link to the demo page
LVTF have been working on similar lines - but looking at how the JSON files can be built (the other end i.e. collecting user preferences)
<LisaSeeman> https://w3c.github.io/personalization-semantics/
Alistair queries the spec - whether attributes should begin with "coga-", rather than "aria-". Aria people though that this approach would create ARIA bloat.
Alistair believes that dividing audiences this way is a bad idea
<Jan> +1 to not dividing audiences
<allanj> +1 to Alastair
Wayne suggests that perhaps the prefix should be something like person- to indicate personalization
ARIA has been based on mapping to the accessibility API of the operating system. Coga and LV issues relate to things that are in the browser.
<LisaSeeman> https://github.com/ayelet-seeman/coga.personalisation/tree/master/json
<LisaSeeman> https://github.com/ayelet-seeman/coga.personalisation/blob/master/json/profile_exmaple_1.json
Many people assume that there will be assitive technologies involved, but for many people this is not the case (and they adjust things in the underlying settings)
Jim asks if this is created by a third party or the responsibility of authors.
Lisa suggests that authors add the semantics.
Alistair suggests pushback because of the increase of testing because of the the personalization options
Lisa suggests a way to minimize the amount of options is to restrict them to a maximum of five options
<allanj> LVTF updated the Low Vision Personalization Requirements https://www.w3.org/WAI/GL/low-vision-a11y-tf/wiki/Personalization_Requirements
Someone has made a text script - a code file in your base directory- ?????????
<LisaSeeman> https://github.com/ayelet-seeman/coga.personalisation/tree/master/inject-coga
<LisaSeeman> add the maping in a text file for the whole site or temlate
Wayne propose that the fundamental thing that both groups want for personalization is non-interference
What parameters do we need so that the material can be personalized and what semantics do authors need to address
Alistair's concern is the extensibility of this over the long-term rather than an unmanageable "hotch-potch"
LV needs "linearize" and coga needs "rearrange", but these just need to be there and it doesn't matter who they are there for (e.g. LV or COGA)
]May need LV and COGA volunteers to look at a site to see what needs to be changed.
Jim suggests that LV works off the CSS whereas COGA have invented a lot of additional things to personalize
Need to identify what additional things that are not in CSS that need to be personalized
Janina says that the need to meet the requirements of different groups has been an idea that has been kicking around for a long time
<allanj> balance user need with developer effort
INDIE UI was meant to cover personalization
Need to get people together to identify what needs to be done.
Research Issues TF think that this topic is within this scope.
A group of people can contribute to identifying the needs and categorizing the personalization
Wayne may have a site that could be used to look at and work with, Jim also identified some possibilities
Lisa: There is a small chance that we could use a small part of the BBC
<kirkwood> Lisa: I may be able to NYC.gov and/or MOPD [Mayor’s Office of People with Disabilites] on board
<LisaSeeman> john, very cool
Wayne says that we must recognize that testing will get more difficult.
WCAG 2 was based on categorical sets whereas what we are talking about a continuum of options. Will need to identify a set of test case data.
<Jan> I have to drop for another call, but want to stay in the loop on the extra meetings.
<Jan> Lisa - you had set up an extra personalization call on Mondays at 9:00 central time - I don't know if that would be an option or not, but wanted to suggest it.
Lisa will do a Doodle Poll to find a time for the joint meeting on the personalization.
<LisaSeeman> ACTION: lisa make a poll and tell everyone [recorded in http://www.w3.org/2017/04/24-coga-minutes.html#action01]
<trackbot> Created ACTION-209 - Make a poll and tell everyone [on Lisa Seeman-Kestenbaum - due 2017-05-01].
This is scribe.perl Revision: 1.152 of Date: 2017/02/06 11:04:15 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Present: allanj LisaSeeman kirkwood Mike_Pluke Pietro MichaelC janina Jan alastairc WARNING: Replacing previous Regrets list. (Old list: thad) Use 'Regrets+ ... ' if you meant to add people without replacing the list, such as: <dbooth> Regrets+ Thad Regrets: Thad Found Scribe: Mike_Pluke Inferring ScribeNick: Mike_Pluke WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 24 Apr 2017 Guessing minutes URL: http://www.w3.org/2017/04/24-coga-minutes.html People with action items: lisa WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]