See also: IRC log
<trackbot> Date: 04 September 2014
<janina> Meeting: HTML-A11Y Task Force Teleconference
<scribe> Scribe: David
Need PF conversation with HTML PF considering topics along with media sub team
PF nervous about Encrypted media, we're not finding anything but still nervous...
want to identify where we can try testing
<MarkS> s/Encrypted Media/EME/G
RS: key things happening SVG...
it's becoming more in sync with HTML... SVG might subclass HTML
element, came from accessibility conversations
... working on interoperability spec, both referencing core
specifications, same in core ARIA for base semantics
need to pull name computation from core spec, for things like alt vs title amd desc in SVG,
RS: Will pull out q section for
name computation and SVG and HTML will call that.
... working on taxonomy for grqphics, like parts of a chart,
lines, axis, etc... applicable to Javascript and other
languages... lot's of interlock. would be good to have a
meeting with all 3 groups
... Digital publication,s we can pull this into text
books...
JS: epub may introduce extensions
to aria, to coordinate...
... html.next might be on toac agenda also
... event enumerations on event handlers on list
<paulc> SVG, PF and HTML WGs all meet on Thu and Fri Oct 30-31
<paulc> http://www.w3.org/2014/11/TPAC/#details
Paul: SVG, HTML and PF meeting on same two days at tpac
<chaals> RESOLUTION: At TPAC it would be useful to talk about event handler enumeration in DOM
<chaals> RESOLUTION: At TPAC it would be good to talk about taxonomy for graphics
<chaals> RESOLUTION: At TPAC it would be good to talk about roles and validation for Digital Publications
<chaals> RESOLUTION: At TPAC it would be good to talk about name computation and the like
1) event enumerations on event handlers on DOM 2) SVG Taxomony 3) dPub role 4) HTML Next 5) interoperability of mappings accross svg html aria
6) EME review
JF: EME will probably be short...
Paul: I'll prep for EME
DM: WCAG...
JS: we'll take a look at that
<chaals> [Along the lines of discussing alt calculation - with WCAG, as well as HTML/SVG - the question of whether a longdesc is appropriate on an image with a null alt]
JS: Candidate response, ORCA
implementation in Linux environment... they provided
feedback.... 5 items... we accepted 1 and made none substantial
editorial
... are we ok with the changes?
CM: Their concerned regarding a reference... I made a fix and will see how its accepted
JS: have copy for us?
<chaals> proposed text: "Note that while in some cases this will allow user agents to present the description, there will be cases where user agents can not or do not restrict the information presented to the container element."
proposed text: "Note that while in some cases this will allow user agents to present the description, there will be cases where user agents can not or do not restrict the information presented to the container element.
JS: yeah let's keep it simple as
this, don't want to give best pracice in spec
... proposed dispositions ok? before CFC
... James C. feels we addressed a secondary issue, not the main
issue... from ORCA
... others?
... CR worked well... we got good feedback from the
implementors
JB: this group went into CR with fairly good implementation prior...
JS: don't here any recs to clarify our disposition, or more about James comments
CM: some people on list said they like the additions
MS: made it clearer
JS: would like to move to TR,
will involved director looking at formal obj.
... we feel the archived record is sufficient, no new issues in
formal Obj. should we move forward
... what is between us... little dospositions on fragment
IDs... don't think we need to respond to formal Obj... and 3rd.
have an online pointer to adjusted spec. as per decisions
JS: html.next access key
CM: old folks here, remember
accesskey... useful idea, but not well used... currently JS
hijacks keyboard, causing unexpected results
... twitter on my mac... keyboard shortcut on system
Accesskey was designed to do what JS does these days... html advice bad for browsers
<JF_> conflict resolution? discoverability? internationalization?
CM: No browser should be able to
highjack things I've got using on my system
... my russian keyboard... latin script useless for access
keys
... it was put back in HTML5 but we have to be clear that what
makes it work is that user has control of which key to use
JF: problems woth Accesskey.
conflict resolution? discoverability?
internationalization?
... can conflict with AT
s/conflick/conflict
JF: lot's of problems, devil is in implementation
Cynthia: there are NO keys are left, when considering all broawers.
CS: the word web app might take over a browser feature ... which is good reason
<Zakim> chaals, you wanted to say the browser is responsible for discovery (since it may have re-mapped the key)
CS: are you aware of Keyboard event spec?
CM: Keyboard event interferes
with your system in uexpected ways, access keys is easy to
do,
... discover ability critical, responsibility of browser, it
may choose to remap for language... cyrillic etc...
... OPERA had one key to find all the accesskeys for a
website,
<janina> I think handlers will be put over
CM: it put in notification into
the chrome... Accesskeys available... and let keys be
reassigned
... regarding speciallized cases ctl-p to print... not related
to accesskeys, ... can rrdirect standard function, which is an
ongoing discussion
<Zakim> liam, you wanted to mention indirection by function-name
<chaals> s/rrdirect/use web activites/wishes (né Web Intents) to redirect/
<ShaneM> it was an element
<liam> [access key at rong level of abstraction, should be using abstract names of functionalities associatd with keys]
JF: tag along on LIAM... way back in the day... <access> element or attribute, left open... I think it's a good directions
<Zakim> chaals, you wanted to talk about access
Liam: [access key at wrong level of abstraction, should be using abstract names of functionalities associatd with keys]
<paulc> (I have to depart to get ready for the HTML WG meeting)
CM: if you are writing a web page
and have to come uo with a conceptual map it is more complex
thqn just adding attribute to things thqt need to be
shortcutted... easy to make a browser interface thqt could
manage this
... extend it to hold a word, ie for a voice rec... system
JS: we're doing overview of items that we want to get into detail....
I'd like to take up a footnote idea
This is scribe.perl Revision: 1.138 of Date: 2013-04-25 13:59:11 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/DRM/Encrypted media/ FAILED: s/Encrypted Media/EME/G Succeeded: s/thqt/that/ Succeeded: s/eoub/epub/ Succeeded: s/SV/SVG/ Succeeded: s/eme/EME/ Succeeded: s/Candidate/JS: Candidate/ Succeeded: s/Lenox/Linux/ Succeeded: s/wkll/will/ Succeeded: s/problames/problems/ Succeeded: s/conflick/conflict/ FAILED: s/conflick/conflict/ WARNING: Bad s/// command: s/rrdirect/use web activites/wishes (né Web Intents) to redirect/ Succeeded: s/ rong/ wrong/ Found Scribe: David Inferring ScribeNick: David Default Present: paulc, janina, Liam, MarkS, David_MacDonald, John_Foliot, ShaneM, Plh, Judy, Rich_Schwerdtfeger, Sam, Cynthia_Shelly, [IPcaller], chaals Present: paulc janina Liam MarkS David_MacDonald John_Foliot ShaneM Plh Judy Rich_Schwerdtfeger Sam Cynthia_Shelly [IPcaller] chaals Found Date: 04 Sep 2014 Guessing minutes URL: http://www.w3.org/2014/09/04-html-a11y-minutes.html People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]