Minutes from 12 March 2015 Meeting

Draft minutes from the meeting are at
http://www.w3.org/2015/03/12-html-a11y-minutes.html

A text version is below.


W3C
- DRAFT -

HTML Accessibility Task Force Teleconference

12 Mar 2015

See also: IRC log

Attendees

Present
janina, Judy, Joanmarie_Diggs, LJWatson, Liam, ShaneM, Plh, IanPouncey, JF,
SteveF, +1.617.319.aaaa, [IPcaller], darobin, Cynthia_Shelly,
Rich_Schwerdtfeger
Regrets
Chair
janina
Scribe
ShaneM
Contents

Topics
Agenda review and edits
CSUN news?
Work Plan CfC - results
http://lists.w3.org/Archives/Public/public-html-a11y/2015Mar/0012.html
Proposed Concurrent CfC Procedure
http://lists.w3.org/Archives/Public/public-html-a11y/2015Mar/0015.html
Potential Other Consensus Procedure Changes
"ARIA in HTML" Issues from PF
Summary of Action Items
<trackbot> Date: 12 March 2015
<janina> Hi, Leonie, yes, for today
<janina> It's essentially what Chaals posted, somewhat re-arranged
<janina> Sure. Mostly different order.
<janina> Additional item is "Potential Other Consensus Process Changes"
<janina> PF has been discussing whether we can move closer to HTML'as auto
publish heartbeats
<janina> I wanted to explore that a bit further. This comes up from the CfC
on "ARIA in HTML" which everyone seems to agree is FPWD ready, but which PF
wants to copublish
<janina> Thanks, Shane
<scribe> Scribe: ShaneM
Agenda review and edits

janina: Agenda is largely as Chaals proposed. A couple of corrections.

CSUN news?

Apparently nothing special to report.

Work Plan CfC - results
http://lists.w3.org/Archives/Public/public-html-a11y/2015Mar/0012.html

<darobin> Zakim. [ is me
janina: done with the list of deliverables and work statement edits.
... we need to pull the deliverables from the work statement so that they
are easily referenceable. Liam can you do that?

liam: we now have one list but it is not in a separate place. can do it.

janina: we should let the co-chairs know that all interested groups to use
the central list.

Proposed Concurrent CfC Procedure
http://lists.w3.org/Archives/Public/public-html-a11y/2015Mar/0015.html

janina: the language is designed so that we *can* do a concurrent CfC, but
sometimes it may still be useful to do separate ones just to assess buy-in.
... the groups seem to hold off on starting a CfC until things are pretty
settled.
... Note that the current process means it will take at least two weeks.

<LJWatson> +1 to the proposed change.
<Zakim> SteveF, you wanted to note that for heartbeta publications in HTML
there is no CFC needed
SteveF: Heartbeat publications in HTML don't need a CfC any longer.

janina: yes - we will discuss that next.

<MarkS> +1
<plh> +1
janina: It seems like there is agreement. We will need to do a CfC to adopt
this change.

+1

<darobin> +1
Potential Other Consensus Procedure Changes

janina: FPWD requires consensus. Moving to CR requires consensus. Heartbeat
publications do not require it in the html working group.
... the PFWG has not adopted this change yet.
... current process requires a week minimum.
... Proposal that a heartbeat would be announced in advance so that people
would have a chance to chime in. Not quite the same as HTML.

<Zakim> darobin, you wanted to mention Echidna/automated publication
darobin: W3C now has an automated publishing system. Groups that opt in can
get things automatically published as a heartbeat.
... document users have complained that there is confusion between editors
drafts and published heartbeats.
... would be nice to eliminate editors drafts altogether, instead having
frequent automatic heartbeats.

janina: we have discussed this in PF
... because of the PF horizontal review stuff frequent heartbeats are going
to make it challenging to track documents as they evolve.
... PF knows to look at FPWD, but we don't know when we need to apply time
later in the process.

SteveF: There are problems with PF documents. ARIA authoring practices, for
example, there are many URLs. Some are a couple of years old.
... when I am referencing things I want the most recent version.
... we need to ensure that content is not stale.

janina: we need to have URIs be reliability and stability.

Judy: Note that PF name change will not effect document URIs.

<Zakim> ShaneM, you wanted to ask about process rules for Echidna
ShaneM: do we need to adopt the new process rules to use Echidna?

darobin: Yes. HTML WG has adopted it for the HTML spec already.

<Zakim> darobin, you wanted to point out that Echidna still has dated specs
darobin: Echidna still generates dated versions of specs. It is possible to
get to a stable point and issue a call for wide review against a dated
version.
... this would be a point where horizontal review would take place.

janina: the issue is how do we know when changes are substantive or
editorial. The length of the diff is one way.

LJWatson: The benefit of frequent heartbeats is clear to the consumers.
Heartbeats are just small steps on the way to the milestones we use for
reviews anyway.

"ARIA in HTML" Issues from PF

SteveF: document is essentially a set of requirements for conformance
checker implementors and authors as to when and how to use aria attributes
... same requirements that were in the HTML specification in the WAI-ARIA
section.
... as part of M12N, was asked to split off this content into a separate
document.
... HTML 5 implementation requirements on browsers will be in the ARIA
mapping specification (HTML-AAM)

janina: no disagreement that this was already in HTML 5
... and that it is probably ready for FPWD.
... PF would like to be a co-publisher of these two documents because they
are a significant amount of work on the part of the PFWG.
... there has been contention in the past, and PF if concerned that we
remain in the loop so that there isn't contention in the future.
... moreover need to speak with one voice. If we go to the point of putting
this document into the horizintal review process it would change the
charater of the relationship between PFWG and HTMLWG.
... that didn't work very well. We created this task force to help ensure
things work better, and they now do.
... Particularly as things relate to ARIA, we want to help ensure work
remains smooth.

<Zakim> darobin, you wanted to point out that this has already been removed
from HTML
darobin: HTML has a vested interest in getting this published quickly. M12N
needs close coordination. It would be easy for documents to go out of sync.
... we would end up being forced to reference the editors draft instead of
a published draft if the document(s) are not published with similar
frequencies.
... I appreciate that things were problematic several years ago, but we are
no longer there. There is a much friendlier relationship.

<MarkS> +1 to healthy working relationship
darobin: we should not be constrained by things that happened years ago.

janina: the fix is this task force.

<richardschwerdtfeger> +1 to healthy relationship
LJWatson: we have indeed moved a long way. the publication of ARIA in HTML
by the HTMLWG sort of underlines the success of this task force.

SteveF: What are the issues from the PF?

janina: no disagreement that people want it published. The question is what
is the long term status of this document.
... there is a consensus developing in the PF that we would like to be
co-publishers.

<SteveF> can we have a link to the PF minutes?
JF: We are moving toward M12N. Robin said that "this has already been
removed from HTML". I am confused about what it means when something is
part of HTML or not. I thought extension specifications were supposed to be
"part of HTML".

darobin: It has been removed from the giant specification. That large
document is too hard to review, to maintain, etc.
... it has been moved to a separate document. It has not been removed from
the HTML language.

richardschwerdtfeger: A concern is that someone could create a new role,
for example. We worry about taxonomy impacts. That's the sort of thing that
the PFWG is concerned about with the ARIA in HTML document. It is about
tight coordination.

janina: and we need to keep synchronized with the other ARIA documents too.

<SteveF> note to rich, aria in html doc does not define any roles, states
or properties
<darobin> and if it did we'd hit SteveF
richardschwerdtfeger: Google has also asked us to create some way to do
things for web components with regard to ARIA.

<JF> +1 to clarity in talking points
Judy: M12N is about evolving the HTML specification. SteveF is working on
ARIA stuff as a module because it is his particular interest.
... and yes, W3C needs to get its talking points in order. We are NOT
removing things from HTML the language.
... HTML and ARIA are both evolving. SVG is going to be evolving. Unless we
are closely coordinating we are going to have a mess later on.
... there are valid concerns about timing and synchronization. We should be
able to work that out on a coordination level.

CynS: The concern is more about being part of the product design team, as
opposed to a reviewer after the fact.

<Zakim> darobin, you wanted to point out that PF can be copublishers with
the document still in Echidna (at least when Echidna gets fixed to support
that)
CynS: we want to work together, not throw things over the wall. Doing it in
PF feels like a natural way to handle that.

darobin: There is currently a bug in Echidna that makes it impossible to
jointly publish a document right now. But that will get sorted.

janina: What I want to suggest is that HTML could view this positively.
Tight coordination and synchronization should may be easier as a result of
M12N.
... SVG might need to do something similar.

LJWatson: This information as been in the HTML monolith all along. Why are
we concerned now that it is separate.
... it doesn't take any more reviewing or coordination.

<Zakim> JF, you wanted to agree with the optics taht Leonie is talking about
janina: because the subject area continues to evolve.

JF: Yes this has been taken out of the big document. I think that is
concerning.

plh: PF expecting to be an author to any document that talks about ARIA is
not going to work. HTMLWG had a similar problem with HTML extensions.

<SteveF> john note I am taking out a large section of the monolith at the
moment http://rawgit.com/stevefaulkner/elements-html/master/index.src.html
plh: at the end of the day, PF created ARIA. Now other groups are taking
and running with the idea. There will be some coordination problems. It
shouldn't mean that PF becomes a co-publisher for every document that uses
ARIA.

<Zakim> darobin, you wanted to note that the expectation is that this would
actually make things easier to review and sync, compared to the monolith
and to also mention that we are looking
darobin: monitoring the monolith was challenging. With M12N it makes it
easier to track for reviewing and synchronizing.
... We would like to remove pretty much everything from HTML into smaller
specifications.

Judy: nothing is being taken out of HTML. Things are being split into
separate, tightly related documents.
... the ARIA coordination stuff. When ARIA is getting embedded we need some
ways to ensure it develops well.
... Maybe we need to offload some of the coordination from the TF to some
off-line mechanism.

<SteveF> suggests best way to ensure it gets embedded well is people
providing technical feedback on modules
Judy: maybe PF develops a PF feature that ARIA feels is critical, and HTML
doesn't like it, what happens. Or the converse?
... maybe co-authorship isn't the right word?

plh: the task force is to help with the coordination. If some group
disagrees with what goes into a spec then that is what this is for.
... because this is on github you can get notifications on every edit on a
per-module basis.

Judy: we know its more effective to handle A11Y at the design stage.
Notifications are nice, but influence before the edits / publication is
more effective.

janina: Working together ahead of publication is more effective.

Summary of Action Items

[End of minutes]
Minutes formatted by David Booth's scribe.perl version 1.140 (CVS log)
$Date: 2015-03-12 16:02:43 $
Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/cn/can/
Succeeded: s/It should be fairly obvious when horizontal reviews are
necessary./Heartbeats are just small steps on the way to the milestones we
use for reviews anyway./
Found Scribe: ShaneM
Inferring ScribeNick: ShaneM
Default Present: janina, Judy, Joanmarie_Diggs, LJWatson, Liam, ShaneM,
Plh, IanPouncey, JF, SteveF, +1.617.319.aaaa, [IPcaller], darobin,
Cynthia_Shelly, Rich_Schwerdtfeger
Present: janina Judy Joanmarie_Diggs LJWatson Liam ShaneM Plh IanPouncey JF
SteveF +1.617.319.aaaa [IPcaller] darobin Cynthia_Shelly Rich_Schwerdtfeger
Found Date: 12 Mar 2015
Guessing minutes URL: http://www.w3.org/2015/03/12-html-a11y-minutes.html
People with action items:

[End of scribe.perl diagnostic output]

-- 
Shane McCarron
Managing Director, Applied Testing and Technology, Inc.