Pauline McInnes in <[log in to unmask]>:
>Dear All,
>I wonder if any of you know about this? Our University website like most has a
>lot of PDF files on it. We want to try and make these accessible to disabled
>students, but this is proving rather difficult. Providing a URL for each PDF
>with a connection to the adobe site is not a good option as many of our PDFs are
>on restricted sites and so this wouldn't work.
>
>However, there is a plug in for adobe writer called Make Accessible which will
>provide a tagged PDF file. What we're not sure about is which screen readers
>would be able to access this information? Our Computer team are a bit concerned
>that only the very latest screen readers could do this. Does anyone have any
>information on this and how prevalent the browsers are can support this plug
>in?
I am afraid that it is mainly the later screen readers (Jaws 4.0, Hal
5, and Windoweyes 4.x (can't remember exact version)) that will read
the pdf files. Furthermore, the PDF reader should be Acrobat Reader
5.0, and the full version should be used, which is a couple of mb
bigger than the standard version which does not contain the MSAA
hooks. This is where a lot of people fail, as they have the later
screen readers but not the later versions of the reader installed.
As regards older screen readers, there are a few choices: 1. Use the
Acrobat reader version 4 and access plugin which _may_ still be
available on the Adobe access sites. This is by no means as
convenient as the later versions, and sometimes can produce
unpredictable results, but is addequate for most purposes. 2. Use
one of the publicly available conversion servers (or install your own)
that will convert via an HTTP interface the file and either email it
back or give you the file on the web page. As I said if you have a
private restricted site you can install your own conversion software
by doing a bit of configuring of the Ghostscript daemons on most *nix
flavours. I have set up such a system which worked via email, and
there are a number of public servers out there that will accept email
and HTTP requests; Adobe used to run one, they may still run it, I
don't know unfortunately.
Then we come to the design of the files: I have come accross some
sites offering pdf journals etc, but these have been secured to the
point that nothing can get access to the files, thus the screen
readers will not read them. There is a company out there that
provides software to break the encryption on the files, I am a
customer and use the program regularly for such files, and make no
secret of the fact. The other major problem is that sometimes
publishers make up their PDF files by using a set of scanned images,
which look no different to a sighted user, but of course there is no
text to get out of the file, and so no amount of manipulation will
yield anything useful. For this I use Kurzweil 1000 and its virtual
printer driver to print the pdf file to the Kurzweil program, where it
can be rekognised like any other scanned images, with the usual
problems this can cause. If the author of the document has disabled
printing access to the file, again the program above can break this.
Hope this helps,
Andrew.
--
Andrew Hodgson, Bromyard, Herefordshire, UK.
Email: [log in to unmask]
|