[Accessibility] updated minutes Sept. 1 Access meeting

John Goldthwaite john.goldthwaite at catea.org
Thu Sep 9 06:55:51 PDT 2004

Accessibility Working Group Meeting Sept. 1, 2004 – corrected 9/9/04

Doug Beattie
Harald Fernengel
John Goldthwaite
Bill Haneman
Earl Johnson
Peter Korn
George Kraft
Janina Sajka
Gunnar Schmidt
Kris Van Hees
Matthew Wilcox

Gunnar gave a summary of the Unix Accessibility Forum at KDE Forum.
Friday was the membership meeting of KDE
On Saturday, Developer Forum began.  Janina arrived Saturday.  The KDE
accessibility developers, Gunnar Schmidt, Olaf Schmidt, and Harald Fernengel
arrived on Thursday.
IBM accessibility developers, Aaron M. Leventhal and George Kraft arrived on
Saturday evening.   Sunday afternoon was the first part of the Unix access
forum.  3 talks – Aaron Levanthal,  Peter Korn,  Harald Fernengel.-
Aaron  - Accessibility on Unix will be 3 years behind Windows.  ViaVoice
will be available in the near future.  ( George Kraft said he thought that
Aaron was speaking of the current situation- Unix/Linux accessibility is
about 3 years behind Windows)
Peter Korn- showed features of GOK and demonstrated that these features are
ahead of what Windows has.
Harald – showed preview of Qt4 with Gnopernicus.  Showed system that used
DBUS instead of AT-SPI

On Monday, several talks – didn’t think they were relevant for discussion.
Access ended.
On Tuesday, Gunnar and his brother joined the usability forum.  There were
discussions about writing guidelines for KDE developers.  Decided 3 tightly
coupled documents were needed.
1- Community integration guidelines- which fonts should be used when doing
screen shots when representing
2 – Usability guide – points the developer should look for when developing
applications and the reasoning behind them.
3 – Accessibility guideline – points for developers and the reasoning behind
4 – Checklist – point to references in other 3 documents.

KDE work summit went on until the weekend.

Harald-  The concern is that CORBA is a dead end for the desktop, both GNOME
and KDE are moving to D-BUS, which is already a freedesktop.org standard.
dependencies orbit2 and AT-SPI introduce is hard to cope even for commercial
Bill- to achieve all of are goals we’d end up with having to implement
something like CORBA.
Harald – orbit
Bill- Dbus doesn’t meet the need.  The
Harald – how to handle metadata – for situations like language switches
Bill – we need to have these discussions in the working group.  We have this
in ATK
Harald-   I suggested transferring text as XML with inlined meta-data,
Bill- that’s part of the attribution API.  In text attribution, CSS
attributes are supported by AccessibleText’s TextAttribute API, which is
also extensible to other attributes.

Bill- I hope that Aaron is incorrect,  I’m concerned about the DBUS
discussion.  The reason for having this working group is to have consensus
on cross toolkit solutions.  There was a Open Desktop accessibility working
group and they decided to adopt Gnome methods.  It is a difficult for us
working on different companies,  we have to look at the big picture.  I’m
glad to see the interest in accessibility guidelines in KDE, but why aren’t
these Open Desktop documents that could be used by all the desktops?   Is
there a reason that we can’t adopt the Gnome guidelines?  We have to work on
technical consensus

Janina- we need to have a discussion about how we do that.  It seems that
Accessibility isn’t active in Open Desktop.  Things look good for the
meeting in January where we can work on this.
Earl- what was the problem
Bill- Gunnar gave an informative report about the KDE forum.  It sounded
like the KDE forum is proposing to repeat Gnome 

Peter-  They were not trying to set up something exclusionary, that’s why
they were called it the Unix Accessibility Forum.
Bill- The current KDE development does not seem to be happening in "the"
Free Desktop space, or one like it, i.e. freedesktop.org; i.e. the
development is still focused on the KDE platform as opposed to shared
development and development of shared libraries, drivers, and standards.
Peter- how do
Harald - we need something that defines our existing guidelines.   We’re
trying to move the existing product and merge the Gnome standards to come to
a common standard.
Janina- I thought the emphasis was on how to be interoperate and not be
competitive.  How to do that will take some time.  This seems to be a
developers process within the KDE community to get specific guidance on
accessibility.  That could help us to move to a more unified effort under
free desktop.  I hope we’re not missing things in translations
Bill- once that task is done for KDE,  there is less motivation to take it
to the next step of creating a shared document.  If we try to create that it
might lead to some consensus.
Harald- KDE is creating its own Interface Guidelines, that's a community
process to define KDE. Our job is to find a consensus between the GNOME HIG
and the KDE HIG and move it to freedesktop.org

Bill- free desktop documents take effort to define, I’m not saying KDE
the guidelines have to be written.
Janina- we have much of this effort in our roadmap.  We need to have a
discussion for next week to see how we can get that.  Develop more specific
proposals on how to move forward
 names of who does what
 free desktop is a
Bill- it’s a very informal organization.  It’s a resource issue.  It’s easy
to get a project included in freedesktop.org, but no resources automatically
come with such inclusion; projects bust be self-resourcing.

Peter- my impression was that Qt4 was very much ATK based.  He’d done some
investigation with DBUS.
Harald -. 
. Would lower overhead for the AT applications, there are fewer
Bill- that may be okay in the long term but actually writing the code in
that area at this time may undermine our standards efforts and current work.

Peter – job 1 is to get this first release out and serving people.  I agree
we should be focusing on that.  (something about DBUS)
Bill- it would take a new generation of DBUS to meet our needs.
Peter- it will take the sorts of investigations Harald is doing to find out.
 -question about Aaron Leventhal’s statement about being 3 years behind
George- I though Aaron made a broad generalization.  I think Peter made a
good point about being ahead on GOK.

Bill – is he asserting that we’re 3 years behind Microsoft or that it will
take 3 years to get Linux accessible?
George – I think he was saying it will be 3 years to get all the features.
Peter-  Aaron has spent all his time on the blind issues.  I think that the
GUI desktop issues for Gnopernicus have more problems and may be 3 years
Janina- I think its devaluing the console environment.  Coming from a
Windows standpoint your it does look like .  We all are learning from each
other what doesn’t work.    I hope he spends more time working on Mozilla in
our environment.   Microsoft has announced a tentative release for Longhorn
in 2006, is it reasonable to set a goal to have accessibility in shipping
Linux distributions by 2006?
Peter-  when we get into discussion  - from the users standpoint it’s a
distribution issue but that gets into the competitive area.  May muddy
Janina- We need to get our work done far enough ahead so that firms can get
it into the distributions.  Does that sound like a possible date?
Peter- seems like a useful goal

Janina- I’d like to update on the NSF proposal.    What’s the status on the
John- I checked with Jim Zemin at FSG this morning.  He has a new secretary,
Janet Sun and he asked her to work on that last week.  He thought she had
completed it but needed to check with her to make sure; she was out of the
office this morning.  I’ll check with him late today.  Ephraim Glinert at
NSF said we need to get it in quickly so that they can fund the project with
money from this fiscal year (Sept. 30).

Janina – In organizing  for the conference, we need to go through the NSF
proposal and make a list of action items.  Then we can discuss this next
week.   We had some good meetings at the KDE forum but everyone wasn’t
there,  e.g. the speech folks.    We need to let people know that our
conference will be occurring so they can get it on their calendars.
Matthew- some people will not come because its in the US
Bill- the funding agency will only fund travel to the US, we need to look at
videoconferencing so those people can still participate.
Janina- we did budget for that and we have facilities that should be able to
handle it.  We’d be happy to have the meeting somewhere else but we’re
limited by the funding.  We’ll definitely need to have additional meetings
so we should plan on that being funded by a source that is less restrictive.
Kris- can tack onto one of the accessibility conferences in Europe next
Peter –  REHA?
Janina- there’s also LSM – next year in Dijon
I’ll go through the proposal and pull out the time line and we can discuss
it next week.
Sub teams?  Anything from the KBD group?
Earl – no vacations intervened

Janina- Daisy has recognized the need to define a standard to allow someone
to open encrypted content.  It can be done now but you have to sign a NDA.
Bill- we have the problem with MP3 also.  No way to implement a public
domain Daisy reader.
Janina- Should they leave mp3 in favor of auk?   We seem to be running into
the same problem from several directions.  We have the same issues with the
differences over the Java licenses.  Daisy is amenable to seeing if auk will
work as well as mp3.
Bill- if you look at some of the specialize codecs for voice, it should
appeal to them.  The space saving alone.

Bill- it would be possible to do a open source daisy reader if your system
has some other mp3 reader but
     The mp3 problem is a big practical

Janina- yes its appreciated within the Daisy consortium. Some of the US
interested in trying to keep thing locked up but most are interested in open
Bill- the reader examples that are on the web are all exe’s that are
self-extracting zip.
Peter- they are Windows binaries.
Janina- the move to develop new products looks like they will create
dependencies on Microsoft’s XML.   The spec is public domain like the w3c
and based on w3c technologies.  Version 1 was closed and got burned.
Version 2 is open but they had the tools developed by a propriety firm so
they are now very interested in open source for everything.
Matthew – copyright says it is public domain but then list a number of
Bill- there are jurisdictions where public domain where those restrictions
would apply.
Janina- they are trying to get funding to get that started by January.

Our next meeting is . September 8th, 18:00 UTC

John Goldthwaite
Center for Assistive Technology and Environmental Access, Georgia Tech
john.goldthwaite at catea.org

More information about the Accessibility mailing list