WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: PDFs and NVDA

for

From: Duff Johnson
Date: Aug 21, 2011 12:27PM

> I assumed that NVDA would be able to list the headings, but that didn't
> happen-it was just a blank list (even though I did them properly in word).
> So I changed the reading options in Acrobat from infer reading order in
> document to tagged reading order and then NVDA happily gave me the list of
> headings.

Yes. Heading structure may be found in the tags. In PDF, there's no other proper place to look for it (I'm leaving aside a technically available but nonetheless pointless option).

> For those of you who use NVDA or JAWS, all the time, when you read PDFs do
> you need Acrobat/Reader to be set to tagged reading order?

Since tags are the only place in which to find the file's logical structure (sequence, nesting and semantics), yes, you need Acrobat/Reader/3rd party option to use the "tagged reading order".

> If this is the
> case, then what happens when there are no tags? Is it a total mess of
> confusion?

Tags are essential to accessible PDF, period. The only exception (in a purely operational sense) would be a document that is so simple that it's possible to have the raw print stream (thanks Karen!) just happen to match the document's logical structure.

As such, there is no "rule' about what to do in the absence of tags - there are merely a couple of options, both of them bad.

In the case of an untagged PDF, applications can either (a) follow the raw print stream or (b) possess the smarts to analyze the page layout and make choices that override the as-printed sequence.

An untagged PDF - even if all the content just happens to occur in correct reading order - is semantically equivalent to a single giant <P> tag containing all the text in the PDF - unacceptable for accessibility purposes.

I hope this helps.

Duff Johnson

p +1.617.283.4226
e <EMAIL REMOVED>
t http://www.twitter.com/duffjohnson
w http://www.duff-johnson.com