<lisa> logs: https://rawgit.com/w3c/personalization-semantics/master/semantics.html#logs
<lisa> https://rawgit.com/w3c/personalization-semantics/master/semantics.html#language-type-support
<lisa> https://rawgit.com/w3c/personalization-semantics/master/semantics.html#reminders-and-messages
<lisa> scribe: Beckyaa11y
<lisa> scribe: Beckya11y
<lisa> scribe: Becka11y
Fiirst item - updates from TPAC
<clapierre1> https://www.w3.org/2017/11/10-aria-minutes.html#item05
minutes from Personalization / ARIA meeting late Friday afternoon at TPAC
started talking about prefixes but need to decide if going to be accomplished with RDFa/Json
discussion about if we want these attributes or not. Or, focus on microdata - that seemed to be what people were leaning towards
group seeking pros and cons of different implementations; wondering if we really need a prefix
Charles brought up timeline isses, group felt we (personalization) wouldn’t be able to be complete before reschartering
Want smaller group to present a realistic timeline and multiply by 3; it is fine if this pushes out into the recharter
Lisa pointed out that group did look into RDF and microdata and Lisa did create some example techniques
Lisa doesn’t see why we can’t go ahead with publishing working draft
Lisa is concerned about not using a prefix - not using may make it more complicated
Can be used via microdata but need it to be easier to implement for authors and AT
Lisa stated that could have used mircodata and RDF for ARIA but didn’t because of more complicated implementation. Why would we want to use for cognitive?
Andy backs up Lisa on RDF and microdata increasing complexity. Sees prefix as a “brand” that helps people understand what it is for/about
Charles indicated that ARIA group wanted to know if we had researched alternatives - and it appears that we have so we can present that to teh larger group to get closure.
Joannie is asking for montly updates from aria sub-groups
Charles had side meeting with John Foliot - he has been talking to schema.org about having personalization as a vocabulary. Schema.org felt that vocabulary is at a higher/document level
Lisa mentioned that we did discuss this with schema.org at TPAC last year. We did have a section in spec about why we were not using RDFa but it was removed in latest draft - perhaps we need it back in a FAQ page
We are going the ARIA/attribute route so that is easier to implement. Charles did bring this up but it was not well recieved by group
Some of the attributes have an implied role: example - action implies button role
Charles agrees that FAQ would be helpful - esp. for people that don’t understand RDF, microdata as well
Charles - we need to split out what is at page/document level and lower levels
<lisa> ACTION: lisa ask joanmair and mihal for coodnation call as soon as possibe
and split out the two
Andy comments that people aren’t used to the vision of the user needs
and we need to help them think that way
<lisa> ACTION: charles make faq page with resions for commen desisons with why we need a prefix sections
group agrees that FAQ page will help us to explain the “mission” of this group and explain the rationale behind our decisions
<lisa> https://rawgit.com/w3c/personalization-semantics/master/semantics.html#reminders-and-messages
Lisa has been reviewing document and feels that there are sections that need more work
<lisa> https://rawgit.com/w3c/personalization-semantics/master/semantics.html#logs
may need a better data model for the reminders and messages section
example Help: more help - not everyone will want 3 different types of help button, would be better for user to be able to select the type of help they want. How do we let people switch between help types to get the one they want?
so need to think about alternative version - for example don’t want to say numberfree for each item of a train schedule - just want an alternative. how to select the alternative type?
and how to specify what alternatives are available
maybe we should no be calling things logs but rather history - more like breadcrumbs.
but need a language to represent that in a way that is machine interperable
Lisa is looking for someone to take a piece and propose a rewrite for that piece
Andy: systems without a built in history can be very frustrating - for example, how to find the BBC piece I listented to 3 days ago? pointing out that it is not just for cognitive
Lisa would like to assign out the different pieces to people and have them come back to the list with a proposal
Charles: we can do this via git and issue a pull request for the changes so Lisa and Charles will see them
Becky and Sharon to send Lisa times to meet for Lisa to help get them started
Lisa will provide an orientation call for Becky and Sharon
<lisa> ACTION: lisa to codinate with becky and sharron for orentation and assign items
simplification; Becky provided some proposed modifications
Lisa - earlier the group had proposed this as importance but changed it to simplification. She is concerned that people will have issues with the name (simplification) and the 3 levels we have selected
<lisa> https://rawgit.com/w3c/personalization-semantics/master/semantics.html#simplification-0
Lisa is concerned that we haven’t had enough people look at this one
Charles - there are concerns with simplification and that it will be marked incorrectly. For example if something is not correctly marked as important, a user may not see it until they try to submit - now they get an error and perhaps it is confusing to figure out how to resolve it
Lisa - we should put into the spec that anything that is required must be marked simplest
Lisa - we need to provide guildelines for implementation; for example to always include a More button
<lisa> ACTION: Lisa to make guidnece for implemention - more button should alwasy be availible
<lisa> how can a user alow specifc fields to have higher importance
Charles gives example of use case that may need more. For example, most people don’t need an address2 or address3 line but this should show up for people who do need this
Lisa - context applies to the address example
context implies what the user may be interested in
Lisa - are people happy with the word simplification? Are we happy with the values of simplest, medium, and low
Charles - perhaps a two tiered system? Do we need 3? Would a developer understand how to use 3 levels? We would need really good examples.
Andy +1 for Charles comment about a 2 tiered system
Lisa - we did go down from 4 to 3 levels
Charles - what if we flip it over and document the complex alternative
Lisa - there are some examples in the spec; send is absolutely necessary but reply all may not be
Gregg - when talking about whether or not something/some option gets shown or not shown - you probably need less subjective terms
<lisa> greag, this isnt just forms
required and all are eaiser to understand but not sure what simplest will mean to others
<AndyHeath> +1 fdoes it mean lower complexity (i.e. less) ?
need to continue this discussion for next call
<AndyHeath> the “+1 was accidental)
<lisa> add comments into the survey
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: AndyHeath clapierre1 Roy lisa Becka11y Regrets: Thaddeus Found Scribe: Beckyaa11y Found Scribe: Beckya11y Found Scribe: Becka11y Inferring ScribeNick: Becka11y Scribes: Beckyaa11y, Beckya11y, Becka11y WARNING: No meeting title found! You should specify the meeting title like this: <dbooth> Meeting: Weekly Baking Club Meeting WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth 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: charles lisa WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. 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]