Comments on: Why PDFs suck!

By: Texx Smith

Texx Smith — Mon, 28 Jul 2008 19:13:08 +0000

One thing that is hardly ever mentioned in “PDF’s suck” discussion is that .pdf’s crash lots of older computers and bog down even newer modern computers.

There is almost NEVER a need for a pdf!

By: Henny

Henny — Fri, 11 Jul 2008 13:27:17 +0000

That’s great, thanks Andrew. I’ve also updated the resources list in the post as well.

By: Andrew Kirkpatrick

Andrew Kirkpatrick — Fri, 11 Jul 2008 12:57:48 +0000

I’ll also add that our screen reader user’s guide is currently available as an accessible PDF file, but it is also being produced in HTML (doesn’t seem too fair to teach people how to use PDF files in a PDF file…).

By: Andrew Kirkpatrick

Andrew Kirkpatrick — Fri, 11 Jul 2008 12:52:52 +0000

Great post Henny - I’d like to chime in with a couple of things:

1) We have a 2-page reference card for creating accessible PDF from Microsoft Word. This doesn’t cover all topics on accessibility for a PDF document, but highlights 5-6 very important things that anyone can easily do in a Word document that will have immediate impact on the accessibility of the PDF file produced. If you follow the advice on this sheet most documents that most people produce in Word will be highly accessible. The reference card is at http://blogs.adobe.com/accessibility/assets/WordToPDFReferenceCard_v1.pdf

2) We also produced a screen reader user’s guide for accessing PDF files. It focuses on what users need to do when encountering different types of PDF files (untagged, scanned, a form, and well-tagged). This is available at http://www.adobe.com/accessibility/products/reader/ where you can also download Adobe Reader 9.

AWK

By: Henny

Henny — Wed, 09 Jul 2008 16:29:55 +0000

I agree with you Joe in that the quality of what goes in, and the tool used, has a huge impact on the quality of accessible PDF that is produced. There will always be times when retrofitting a PDF isn’t the way to go and especially so with complex ones. As with HTML an informed decision needs to be taken to establish if in fact PDF re-design is needed. When it comes to creating accessible PDF using MS Word style sheets is definitely a technique that we advise whilst avoiding coming anywhere near Quark which is really not our friend here. As I’m sure you yourself know this in itself is a challenge when the publishing industry have invested heavily in software’s like Quark and designers are only familiar with it. This is why I believe that making PDFs more accessible is as much a culture and process change as it is an understanding of what must be done to create an accessible PDF.

We do already do a lot of work both with Adobe and with various groups here in the UK on promoting the creation of accessible PDFs. Hugh Huddy especially has represented many groups with an interest in accessible PDF to Adobe. He’s worked to engage with disabled people, content authors and the tool makers to gain an holistic understanding of the challenge and this work continues.

By: Joe Clark

Joe Clark — Sat, 05 Jul 2008 14:14:35 +0000

Simple documents are easy to make accessible in PDF *up front*. So are several kinds of complex documents if you use modern software like InDesign or Framemaker (yes, from Adobe) that can produce tagged PDFs automatically. I have had no trouble creating semantic multicolumn PDFs – with images, no less – that fly through the Acrobat accessibility checker without problems.

Now, if you’re creating a document with complex tables, that’s another story, but PDF opponents need to be honest about the fact that comples tables are always difficult to make accessible. Footnotes and so on? Well, at least there are actual defined tags for them in PDF, unlike in HTML. Mathematics? Tricky right now, but MathML will be required in PDF/UA.

What isn’t being honestly discussed here is the fact that most of the “problems” people face with PDF come about during retrofits. Somebody hands you a crap PDF and you’re expected to fix it. Well, of course that’s going to be difficult. It isn’t like HTML, where you can engage an orbital-nuke scenario and just remove all the markup and start over. PDFs are databases of nearly anything you want, and trying to untangle a database of objects isn’t going to be easy.

If people want a larger number of accessible PDFs, then they’re going to have to learn to use stylesheets in MS Word and, finally, for the love of God, get with the program and quit using Quark Xpress. Do those two things and you solve a large part of the new-PDF problem. The retrofit-PDF problem is rather more difficult to solve. Maybe RNIB should work harder on the relatively easy win of making new PDFs accessible.

By: Henny

Henny — Thu, 03 Jul 2008 15:30:07 +0000

Thanks for your comments folks, all useful stuff. I’ve added in a couple of additional links to the resources section that provide further guidance on creating accessible PDF.

By: Hugh Huddy

Hugh Huddy — Tue, 01 Jul 2008 09:12:02 +0000

there are two reasons why RNIB chose not to attend the PDF UA discussions:
1. as a not-for-profit we focus resource into areas where they’re really needed and the PDF UA working group is already a strong mix of experts.
2. related to the point above about the 115 page long Adobe guidelines, we have fed back to the relevant people on several occasions that these guidelines are swollen by workarounds to compensate for what information producers tell us are inadequacies in the PDF authoring tools, so we have invested the time that PDF UA has been running in understanding these issues at ground level.
I believe the ‘PDFs suck’ situation whether you fully or partly agree with this statement cannot be recovered as long as the actual people out there in the field are faced with extensive guidelines and confusion over using authoring tools no matter how well PDF UA specifies and defines accessibility.
I do think we as accessibility experts have a primary duty to understand what it’s like at ground level because that’s where accessibility is failing.
I am sure Henny’s posting here will help content authors understand their responsibilities and increase demand for technologies that actually work!

By: Joe Clark

Joe Clark — Mon, 30 Jun 2008 17:58:53 +0000

I really don’t think it’s accurate to state, rather baldly, that “most PDFs… are inaccessible.” You have to separate the PDF from the viewing application. Recent versions of Acrobat are more or less capable of making untagged text-only PDFs accessible as soon as you open and start reading them. This example should not be minimized, as it is a favourite tactic of e.g. press-release writers to issue a few hundred words of barely-marked-up text in PDF.

In other words, the software can overcome the inaccessibility of many documents. It isn’t going to help you with images, which require intelligently-written alt texts, but even in some cases a document can be functionally accessible without alt texts (e.g., a press release with a company logo at the top that also mentions the company name in the press release).

I think you should expand your list of links a bit. Among other things, on the WCAG Samurai site there is now a very solid list of PDF formats that should never be posted on the Web as PDF. You need to be a bit less wishy-washy on this point. It isn’t true that people should ask themselves if a document should be PDF; for online distribution, many documents should never be PDF (or, more accurately, should never be available solely as PDF).

Also: How come nobody from RNIB is ever on the PDF/Universal Accessibility phone calls? We’re almost done writing V1.0 of our spec, you know. And it’s pretty good, if I do say so myself.

By: simon gray

simon gray — Mon, 30 Jun 2008 14:09:18 +0000

It’s all very well suggesting people read the Adobe guidelines on creating accessible .pdfs - but when that document itself weighs in at 115 pages, who is expected to bother ? I’m quite sure there isn’t really a whole 115 pages worth of actual necessary information needed to digest in order to do the job properly…