[Irl-dean] Accessibility of PDF Format Resources?

[Barry McMullin]

On Tue, 21 Nov 2006, Gerry Ellis wrote:

> Here are a few further thoughts to consider when deciding if
>  PDf is a suitable format for document presentation.

Hi Gerry -

Thanks for all these comments, which were very helpful and
informative.

The situation with PDF documents incorporating forms is
particularly interesting and challenging.  On the one hand, there
are some use cases where these might be very beneficial, and
where it would be very clumsy, if not impossible, to use HTML in
the same way.  On the other hand, this functionality frequently
relies on "controlled" features of the so-called "free" Adobe
reader; which are unlocked only on payment of a fee to Adobe
(normally levied on the document publisher, rather than the
user). This is not quite a secret, but is certainly not something
that is very widely publicised...

I appreciate your mention of DIASY; and I too would like to see
wider deployment of it.  On the other hand, I think I would be a
little more constrained in my enthusiasm about this, for the
following reasons:

- Just as HTML and PDF "support" accessibility, but cannot
  enforce it, DAISY is just another *potentially* accessible
  format.  It is perfectly possible (and I have seen examples
  already!) to produce resources "in" DAISY format, which are
  thoroughly inaccessible.  Thus we must be careful not to
  promote it as some form of "magic bullet".

- Like PDF, DAISY supports, but does not require, the use of a
  DRM/encryption layer.  However, the practical reality is that
  the vast majority of DAISY resources currently available *do*
  use this layer (which is the opposite situation to that of
  PDF).  This encryption layer is basically a (faustian?) bargain
  between groups representing users with disability and major
  publishing houses. It has the great advantage that it has
  delivered participation by publishers in making available
  digital editions of many books. But there are also significant
  disadvantages.  The current encryption technology is
  (deliberately) not public and is available only to
  organisations which sign a non-disclosure agreement.  The nett
  effect is that competition in developing DAISY readers is
  significantly hampered.  Indeed, it is currently impossible to
  have an open source implementation of this encryption
  layer. Further, in certain cases, the adoption of DRM-based
  DAISY has actually resulted in a *reduction* of service to
  users.  Specifically, while the US-based "Recording for the
  Blind & Dyslexic" has a mechanism for individual membership by
  people outside the US, such "members" are no longer allowed to
  borrow from the full collection: all DAISY books are excluded,
  and only legacy, audio cassette-based books are still offered.
  As far as I am aware this is not due to any *technical* issue,
  but is part of the DAISY "bargain" with the publishing
  industry. Faustian indeed.

Having said all that: even if it is imperfect in some ways, DAISY
is a real, functioning, useful technology.  For many of the
resources that are currently published in PDF, a DAISY version
would definitely offer significant advantages to the (growing)
number of people who have access to, and are familiar with the
use of, DAISY readers.

Further, in the same way as I believe it should be possible to
more or less completely automate conversion between "accessible
PDF" and "accessible HTML", it should also be possible to
automate conversion from either of these to (non-encrypted!)
"accessible DAISY".  So, if an organisation does invest the
(modest) work required to produce "accessible" documents in *any*
format, then the marginal effort to support *multiple* accessible
formats, including DAISY, should, in many cases, be negligable.
So, although the tools to support this are still pretty immature,
as they become more widely available and deployed, I would
certainly encourage publishers to provide users with choices in
formats, and to specifically add "accessible DAISY" into that
mix.  (However ... I would be frankly horrified if anybody
proposed adopting DAISY distribution *exclusively*; but I assume
that is not what Gerry was suggesting?)

Best regards,

- Barry.