Thread Subject: Re: 1194.3 (e)

Note

This archival content is maintained by WebAIM and NCDAE on behalf of TEITAC and the U.S. Access Board . Additional details on the updates to section 508 and section 255 can be found at the Access Board web site.

From: Robinson, Norman B - Washington, DC
Date: Mon, May 07 2007 10:30 AM

David,

If you'll forgive me, I think we are missing the point.

While the goal of having an accessible computer, from boot time
to shutdown, is desirable, I don't think we'll get there discussing how
accessible the BIOS is - for a few reasons:

1. The terminology of BIOS is being used differently. To an
engineer, the BIOS is simply the hardware that initializes the hardware.
There is little to no expectation beyond that, which includes software
or firmware to drive audio output. I don't think that is necessarily the
end of the conversation, as we do currently have BIOS that provides
video and allows configuration using the keyboard. Even then, I think
the line is fuzzy on where the BIOS ends, and what supporting video
chipset and functions are provided on the motherboard.

2. The BIOS is simply ignored by most modern operating systems.
For instance, Linux doesn't use the BIOS anyway. It talks directly to
the hardware.

3. Think function not specifics. Related to #1, what I hear in
this discussion is the need to get computer hardware accessible to as
wide an audience as possible. HOW we do that really isn't the most
important part of that conversation. Given the role BIOS plays, I think
many manufacturers could clearly defeat our logic and show us how when
we speak directly to BIOS and only the BIOS, accessibility requirements
are a fundamental alteration. Or, turning that into how Section 508
enforcement happens, the agency itself would have to specify what the
BIOS much accomplish and I don't think that would work well at all
unless the agency in question had a need to create their own computers
from scratch.

4. So how do we do this? Well, I don't see why motherboard
manufacturers can't solve this problem as they already build everything
onto the motherboard (video, LAN, BIOS, LED status indicators, IDE,
SCSI, etc.) so haven't a text to speech chipset might already be a part
of what their soundcard or DSP chips can do years ago. They do provide
systems that activate and use network functions, but even then I don't
think you can order a "BIOS"; it is a bunch of chipsets designed to work
with their firmware. That means we need to talk about motherboards - not
BIOS.

My point is constraining your conversation to "BIOS" means
something specific to many engineers, geeks, and manufacturers and they
(in my opinion) can easily say it is not proper for a BIOS to perform
the accessibility functions we need. It can support them, and with
embedded sound for text to speech, working with onboard speakers (or
when connected), serial output, or usable systems that provide LED
status (or perhaps USB output to a specific type of serial interface) is
needed. But when we start specific conversations about BIOS we loose the
real-world logic battle. I also believe that the easiest and cheapest
way to be successful involves software; something like an accessible
version of Linux running before you "primary OS" is operational. This
also has the benefit of perhaps retrofitting to older hardware and
helping those of us who can't afford the latest version of hardware.

Finally, if you haven't had the chance to read about LinuxBIOS,
you may find it useful to our conversation or at least interesting:
http://linuxbios.org/. Wikipedia's article
http://en.wikipedia.org/wiki/BIOS is also worthwhile reading.

Regards,


Norman B. Robinson
Section 508 Coordinator
IT Governance, US Postal Service
phone: 202.268.8246


-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of David
Poehlman
Sent: Monday, May 07, 2007 11:22 AM
To: TEITAC General Interface Accessibility Subcommittee
Subject: Re: [teitac-general] 1194.3 (e)


Hi Peter and all,

There is software that allows you to write directly to the bios while
the
system is running fully. I believe that it does part of its task when
the
system is rebooted but it does allow access to the bios at a high level.

----- Original Message -----
From: "Peter Korn" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC General Interface Accessibility Subcommittee"
< = EMAIL ADDRESS REMOVED = >
Sent: Monday, May 07, 2007 10:57 AM
Subject: Re: [teitac-general] 1194.3 (e)


Hi David, Gregg,

You both are saying similar things, so I'll just quote David:

> Making the bios accessible to a screen reader is not trivial but may
not
> mean a fundamental alteration of the os or the bios.
>

My point is if (one of) the function(s) of a BIOS is to provide low
level access for configuring things like serial ports and audio devices
before the system is booted, then making that configuration user
interface talk (via that not-yet-working audio device) is in conflict
with what the BIOS' function. To do so would fundamentally alter the
BIOS (it would loose that functionality, and some "lower level still"
function would be needed for servicing the machine to enable audio.
Likewise for a serial/USB-connected Braille display.

For most software (including most AT), you need enough of the system up
and running in order for them to work. The point of the BIOS
configuration user interface it to allow someone to configure and
service a machine that is mostly NOT yet up and running.

There are some things you can do (e.g. text size, contrast setting, and
potentially optionally routing all input/output to a serial port
[something we do in Solaris on SPARC processor systems that aren't
running an x86/x64 BIOS]), but there are a great many that you can't.

Potentially a wholesale redesign of how a BIOS works could get you a lot
more functionality - but such a change might well require changes in the
software layers that sit on top of the BIOS (e.g. breaking existing
copies of Windows and Linux and ...), which again would fundamentally
alter the product, because it forces the product to become incompatible
with the hundreds of millions of copies of operating system software out
in the field that is depending upon certain behaviors. It essentially
would stop being a "PC BIOS" as we commonly think of the term - a
fundamental alteration.


To be clear - I'd like to see this technical problem solved. But it is
a very hard problem.


Regards,

Peter Korn
Accessibility Architect,
Sun Microsystems, Inc.

> ----- Original Message -----
> From: "Peter Korn" < = EMAIL ADDRESS REMOVED = >
> To: "TEITAC General Interface Accessibility Subcommittee"
> < = EMAIL ADDRESS REMOVED = >
> Sent: Monday, May 07, 2007 1:31 AM
> Subject: Re: [teitac-general] 1194.3 (e)
>
>
> Hi William,
>
>
>> And then there's the subject provision which could be construed as a
>> backdoor to avoid "inconvenient accessibility": "This part shall not
be
>> construed to require a fundamental alteration in the nature of a
product
>> or its components."
>>
>> What has been decided (and by whom) if some particular requirement
might
>> "require a fundamental alteration"?
>>
>
> I don't see how this can be other than a determination by the folks
> making the product - perhaps with some questions answer to demonstrate
> it, and with the incentive of competition from others who might
> demonstrate that they can deliver the same product, but accessible,
thus
> demonstrating that it isn't a fundamental alternation of how they
> deliver the technology.
>
> My previous e-mail about BIOS accessibility is a fine example of this
-
> it would fundamentally alter what a BIOS is if we were to insist that
it
> be accessible out of the box to someone who can't see (and likewise a
> fundamental alternate to make it compatible with a 3rd party screen
> reader).
>
> Also, fundamental alteration appropriately includes cost in the
equation
> - a redesign of a subsystem to make it accessible might entail
changing
> all of the pieces on top of it (e.g. making a BIOS work with a screen
> reader might mean that MS-Windows itself would have to be redesigned
in
> order to work with that modified BIOS). Since this theoretical change
> would mean things that used the BIOS would have to change, it results
in
> a fundamental alteration of the BIOS (and alternations of the things
> that use the BIOS - a good test if the alternation is 'fundamental')
to
> accomplish the accessibility.
>
>
> Regards,
>
> Peter Korn
> Accessibility Architect,
> Sun Microsystems, Inc.
>
>
>

WebAIM is an initiative of:
Center for Persons with Disabilities (CPD) Utah State University