W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > October to December 2001

Re: Acrobat PDF & Accessibility

From: Joel Ward <ward_joel@bah.com>
Date: Thu, 20 Dec 2001 17:07:59 -0500
Message-ID: <025101c189a2$c933d890$19ab509c@BAH505131>
To: "WAI List" <w3c-wai-ig@w3.org>
> maybe I'm wrong but isn't pdf a "photograph" of the page rather than
> actual letter codeing, sort of like a fax vs teletype machine??

Yes and no.

If you scan a paper document and convert it to PDF format, it is only an
image.  From there, you can use optical character recognition (OCR) to
convert the text to actual rich-text that gets embedded into the PDF behind
the image.  However, those document are very hard, if not impossible, to
make accessible even though they may contain rich-text.

A PDF converted from Word or PageMaker or another office application will
contain real text behind the PostScript/image facade.  If you use Adobe
Acrobat 5 to create these PDFs, you can make them accessible to assistive
technology (if you follow the rules).  However, the end user still needs the
Acrobat 5 Reader plug-in WITH the accessibility features.  Older versions of
Acrobat and even 5 without the added accessibility features will not work.

I have heard that IBM Home Page Reader uses the Adobe PDF-to-text web site
to convert all links to PDF files.  Other products (like JAWS and Window
Eyes) use whatever the browser plug-in is installed, so if it's Acrobat 5
great!  If not, oh well.

But for a user of LYNX...I don't know.  You could link to the Adobe
PDF-to-text web site to view PDF files:

Ideally there would be an HTML or text version of the same document
available.  But for Section 508 that isn't required unless the PDF itself is
not accessible.  If the PDF is accessible, it will convert well using the
above Adobe site.

I personally feel that PDF is best for printing, not online viewing.
However, it depends on the situation.  There are times where having one
document is best (either accessible PDF or HTML), and times where a having
PDF, HTML, and text versions is best.

For the widest possible audience, HTML or text is much easier to deal with
that PDF, as you point out.

Received on Thursday, 20 December 2001 17:08:01 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:15 UTC