Re: Forming an Accessibility Steering Committee

["Peter Parente" <parente gmail com>]

> Choice quote: "we have succumbed to the tyranny of the popular, the
> typical, the average, or the norm"...  but let's not!  We should be able
> to go so much farther and faster in FOSS land...

Simply getting together to talk in an open forum helps spark new
ideas. Some of the best brainstorming sessions I've ever had have been
with my thesis adviser after reviewing other research papers. It's
amazing how many intriguing ideas can result from reviewing what other
people are doing to push the envelope.

Of course, collaborating and working to resolve the problems here and
now is still the primary concern. But it would be nice to see some
time and cycles spent freely discussing projects that may appear a tad
"crazy" at first. Even if this small community of accessibility
developers doesn't have the time or resources to tackle them, getting
the ideas out there for other people to think about might entice them
to join the effort (the "Wow. That's cool. I can do that." effect).

Here's the kind of "crazy" I'm talking about:

Humanware makes a product called Trekker which provides spoken GPS
navigation aid for people with visual impairments. Their solution is
unique among competitors because it uses a standard PDA (the Dell
Axiom X51 I believe), overlays a custom rubber keyboard on top of the
touch screen for input, and provides speech output. The Trekker can be
bought as part of the Maestro package which also provides speech
displays and button input for applications including a text reader,
contact list, agenda manager, etc. The price is not listed on
Humanware's page, but other sites list it somewhere between $1500 and
$2000. While it seems high, it's actually cheaper than other, custom
manufactured devices suited to one and only one user group.

So what could we do to push the envelope under the same premise, that
is, reuse existing, commodity hardware and platforms? The Nokia N800
line of Internet tablet products with Maemo comes to mind. It's
GNOME-based, supports open development, and has an API for
communicating with Bluetooth GPS receivers. The cost is around $300
for the N800 and roughly $100-$300 for a separate GPS receiver.

There isn't much documentation on doing speech synthesis on the Maemo,
but it looks like at least a few people have tried. I have to believe
it's possible using flite or espeak. Granted, input is also a problem
since interaction is mostly driven by the touch screen. However, the
N800 does have more than a few physical buttons and the N810 has a
fully qwerty keyboard. A custom navigation app could make use of
either for stepping through directions en route. The user could plan
the route on their desktop computer where I/O is more full-featured,
and then sync it with the mobile device where a less interaction-heavy
UI could speak directions as the user follows his or her route.

This setup seems ripe for doing something new. Speech-based
way-finding is just one possibility. I'm sure with some time spent
thinking on it, we as a community could come up with many, many more.

I believe it's this kind of thinking the committee should promote and
document for others to see. Again, we as individuals may not get time
to work on implementing every concept. But I bet if we use an
idea-shotgun, and blast enough into the public-awareness, someone will
grab them and run.

Pete