I was recently asked my thoughts on the Accessibility Support Documentation for PDF. Reading through the document is quite reassuring, with every single success criterion (even the AAA ones) either supported by Adobe, or the responsibility of the document author.
It’s only when one reads the Appendices that it becomes apparent that all is not as it seems. Adobe PDF does fail in some serious ways, it just seems to have escaped the author of the Accessibility Support document.
But firstly, let’s look at the actual testing undertaken.
Testing was incomplete
Testing was conducted on the following:
- Windows XP, JAWS 9, IE 7
- Windows XP, JAWS 9, FF 3
- Windows XP, WindowEyes 7, IE 7
- Windows XP, WindowEyes 7, FF 3
- Windows XP, Zoom Text 9, IE 7
- Windows XP, Zoom Text 9, FF 3
What about other operating systems?
It all seems a little Windows-centric doesn’t it? What about Mac users? In fact the Mac operating system has a very large range of assistive technologies and is often used by people with disabilities in preference to Windows.
What about other browsers?
IE 7 and FF 3 are not the only browsers out there. There’s other versions of these products and there’s Safari, Opera and Chrome. And that’s just the popular browsers.
What about other assistive technologies?
JAWS and WindowEyes are not the only screen readers around. What about NVDA, BrowseAloud and the Apple’s inbuilt screen reader? What about Read Out Loud- Adobe’s inbuilt screen reader? And testing on only one version of each assistive technology seems short-sighted in the least.
What about other disabilities?
This is my biggest concern, and I’ve left it to last. What about other assistive technologies that are for people with disabilities other than those with vision impairments? It is generally accepted that PDFs can be used by vision impaired users, but they are not the only people with disabilities out there. Firstly, PDF was not tested with other assistive technologies such as an onscreen keyboard, joystick, touchscreen or thumb switch (and that’s only some of the other assistive technologies out there). Secondly, from the testing that was conducted, it seems that the success criterion were not reviewed in relation to how they affected other people with disabilities, such as cognitive disabilities (the largest group of people with disabilities on the web), physical disabilities and hearing impairments.
In the next post
In the next post I will be talking about exactly why PDF isn’t as accessible as HTML, and which success criteria in WCAG2 PDF does not meet.
When using JAWS, it is recommended to open a web-based PDF in Reader rather than the browser. By using Reader you get the benefit of JAWS scripts for PDF. Small point perhaps, but one worth noting.
James
Gian,
The focus of the documentation for accessibility support was to help WCAG 2.0 reach Release Candidate status, and the requirements for that step did not include documenting every possible combination of assistive technologies. You may note, for example, that we reported on six configurations whereas HTML reported on four. So, does this reflect the totality of supported assistive technologies? Of course not. Does VoiceOver on OSX support accessible reading of PDF documents? Yes, but that isn’t discussed in this document. The absence of this information doesn’t mean anything with regard to capabilities or intentions.
You say that Adobe PDF does fail in some serious ways. I’d like to first point out that “Adobe PDF” isn’t really an accurate name to use, since PDF has been an ISO standard since 2008 (ISO 32000-1). Second, you didn’t point out any specific issues related to accessibility support. You did highlight that the testing doesn’t include all disabilities, and that is accurate, but also expected since the concept of “accessibility support” in WCAG 2.0 relates to interoperability with assistive technologies and not all accessibility concerns are mediated by assistive technologies.
Your largest concern also relates to the focus of the document – it is certainly important for assistive technologies other than those used by blind or visually-impaired users to function properly, but since this report was not attempting to be comprehensive other assistive technologies were not included.
One small correction – you mention that VoiceOver is Adobe’s own built-in screen reader in your post. I’m sure this was a simple typo, but the reading tool in Adobe Reader is called “Read Out Loud” and it is not a full-fledged screen reader. We do of course test with it, but screen readers like JAWS, Window-Eyes, Hal, NVDA, and VoiceOver provide more robust support for users who are blind or low-vision.
Thanks,
AWK
Hi Andrew
Thank you for your comments- I really appreciate someone from Adobe responding to my post.
I was not aware of the issues surrounding the release of the document (as you know, I haven’t been on the WCAG Working Group for awhile!), and now certainly understand why it was not as comprehensive as I would have liked. Is a more comprehensive document something that Adobe will be considering? Because I am very interested in how other assistive technologies interpret PDFs and how other groups of people with disabilities are affected.
As for the specific issues – I do have them, and I will be releasing a second blog post soon. I am very interested to hear your thoughts when I do.
Thanks for picking up the VoiceOver typo – don’t know how I missed that.
Cheers,
Gian
[…] A few problems with the concept of accessible PDFs […]
[…] Wild muses on a few problems with the concept of accessible PDFs and there’s even a reponse from […]
Great post and conversation. Nice to see Andrew jumping in from Adobe as well:)
Gian makes some very good points. The challenge is that whether we like it or not, PDF is the standard for electronic documents and statements. In the case of finance for example, institutions have to legally store and provide exact copies of printed statements in which case HTML will not suffice.
Keep up the momentum, the best we can do is articulate the issues and do our best to make PDF as accessible as possible!
[…] firstly, I looked at the lack of testing, and make sure you read the comments because there is one from […]
[…] A few problems with the concept of accessible PDFs […]
[…] Part 1: A few problems with the concept of accessible PDFs […]