I'd add one additional comment about PDFs. PDFs do not deal with actual
letters and characters, rather they deal with sequences of glyphs. PDF
documents may not contain the actual text. Nor may the fonts contain well
formed cmap tables. This will impact on accessibility of the document.

Probably not a concern with Microsoft Word where lots of the
OpenType specification is non-functional, but will be an issue with
applications and documents that make use of the OpenType GSUB tables in
some fonts and use alternative glyphs or ligatures within a document. Esp.
if the \ToUnicode or \ActualText data is not present in the PDF.

Andrew

On Sat, February 23, 2008 4:57 am, Steve Green
wrote:
> PDFs can be more accessible than was previously possible
but there are
> lots
> of gotchas, and it's way too big a
topic to cover in this reply. Note that
> by default, PDFs are not
tagged, so they are only marginally more
> accessible
>
than before. Maximum accessibility is obtained by tagging them, but
that's
> where it gets hairy.
> 
> To start with a
lot depends on what application your document was authored
> in.
If it was authored in Word (and a small number of other applications),
> any semantic structure that you have implemented by means of
paragraph
> styles will be carried over to the PDF. If it was
authored using graphical
> applications such as Quark and
Photoshop, not only is there no semantic
> structure, but the
reading order is likely to be wrong because these
> documents are
frequently made up in layers with no consideration of how
>
they
> will linearise.
> 
> Tagging can be
conducted automatically, using the Autotagger, but the
> results
are highly variable. I have never seen a document that could be
>
used
> without manual adjustment of the tags. In many cases it is
actually easier
> to manually tag the entire document and not use
the Autotagger. In fact we
> always do this, but that may be a
reflection of the complexity of the
> documents we get asked to
handle.
> 
> If the source document does not contain
semantic structure, the Autotagger
> uses heuristics to make a
best guess. This sometimes comes close but often
> doesn't. There
is a tendency for it to mark up multiple columns of text as
>
tables, and background colours or images can confuse it.
> 
> Both Acrobat 7 and 8 are pretty buggy, and the Accessibility
Checker in
> Acrobat 7 sometimes reports faults when there aren't
any. In our opinion
> the
> user interface in Acrobat 7 is
much better than version 8 so we often use
> both for different
parts of the job.
> 
> Forms are a lot of fun. The user
cannot save the document with the data
> they
> enter in
the form, but it is still worth making forms accessible so they
>
can
> be filled in and printed. There are ways to submit the forms
online but
> there are significant licensing issues (it's ages
since I looked into
> this,
> so I can't recall the
detail).
> 
> Even after making your best efforts, a PDF
will not be as accessible as a
> properly marked-up HTML page with
the same content. There are some
> resources
> out there
but my recollection is that they all deal with the ideal
>
scenario
> that you are creating new documents in Word or
InDesign. There are some
> resources below, and they link to other
useful stuff.
> 
>
http://www.brucelawson.co.uk/2006/accessible-pdfs/
> 
>
http://alastairc.ac/2006/07/the-four-levels-of-pdf-accessibility/
> 
>
http://alastairc.ac/2007/08/comparing-tagged-pdfs-from-office-and-acrobat/
> 
> 
> For those who want someone else to take all
these problems away, we offer
> a
> PDF accessibility
service - http://www.labscape.co.uk/accessible_pdfs.htm,
> and we
also provide training for mad people who want to do it themselves.
> 
> Steve
> 
> 
> 
>
-----Original Message-----
>
From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> Behalf Of McLaughlin, Gail G
> Sent: 22 February 2008
16:57
> To: [email protected]
> Subject: [WSG] PDF
Accessibility
> 
> I have a client who discourages the use
of PDF forms and files on their
> website because they believe
that they are not accessible.
> 
> Researching this on the
Web, it appears that this may have been true
> several
>
years ago, but that Adobe has made an effort to make PDF forms and
files
> accessible in Adobe Acrobat 7.0 and 8.0.
> 
> Are PDF forms and files created with Adobe Acrobat 8.0 truly
accessible? I
> assume that certain protocols must be followed to
assure accessibility,
> just
> as there are protocols to
assure accessibility of HTML files.
> 
> Can anyone direct
me to a good resource detailing what protocols one must
> follow
to assure accessibility of PDF forms and files?
> 
>
Thanks much,
> Gail
> 
> 
> 
>
*******************************************************************
> List Guidelines:
http://webstandardsgroup.org/mail/guidelines.cfm
> Unsubscribe:
http://webstandardsgroup.org/join/unsubscribe.cfm
> Help:
[EMAIL PROTECTED]
>
*******************************************************************
> 
> 
> 
>
*******************************************************************
> List Guidelines:
http://webstandardsgroup.org/mail/guidelines.cfm
> Unsubscribe:
http://webstandardsgroup.org/join/unsubscribe.cfm
> Help:
[EMAIL PROTECTED]
>
*******************************************************************
> 
> 


-- 
Andrew Cunningham
Research and Development Coordinator
Vicnet
State Library of
Victoria
Australia

[EMAIL PROTECTED]


*******************************************************************
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
*******************************************************************

Reply via email to