This Isn’t Working

| 2 Comments

I hear “This isn’t working” from frustrated customers on a fairly regular basis. Most of the time it’s because a particular PDF file isn’t behaving the way that they expect it to. So, with the recent release of the “Compatibility Advisory Regarding Adobe Reader plug-in and Acrobat plug-in with Safari 5.1“, I thought I’d address a few broader compatibility issues that may arise when working with PDF files.


But first… Yes – I’m biased – I work for Adobe. So I’ll try to stick to just the facts in this article and let everyone come to their own conclusions. I’m also not going to cover the obvious compatibility issues that are inherent with PDF Portfolios, Multimedia and other Flash dependent objects in PDF files, but will, instead, limit my discussion to the core of PDF that is in use my most people out there and not the cutting edge stuff that I talk about in my developer blog.

There are three scenarios that are the root cause of most issues around use of PDF files and I’ll address each separately but they can… and do… occur in combination with each other.

The PDF was created by something other than Adobe Technology:
While the PDF file specification is a published ISO Standard, not all PDF creation tools follow it as completely as they could or, even worse, they follow the specification correctly but take shortcuts that make the resulting PDF file essentially useless.

Adobe Acrobat preserves the fidelity of virtually any document that can be printed. You’d think that’d be the case with any PDF software. It’s not. Some non-Adobe tools will render the text on the page as a low-quality bitmap image when they cannot correctly interpret a font that was used in the source document. PDF files created in this manner are not searchable, making it difficult to archive and retrieve the document in the future or to repurpose it for other devices and formats. The image below illustrates the problem. You can see the bitmapped text behind the failed search dialog.


In the image below, you can see that Acrobat successfully finds the word “Local ” because PDF creation using Acrobat supports a wider range of fonts than the other tools.


Acrobat X also preserves the smoothness and sharpness of vetor shapes. The image below is a segment of a CAD drawing that has been converted to PDF and viewed in Adobe Acrobat. As you can see, Acrobat preserves the smoothness and sharpness of the shapes. Compare that the non-Adobe PDF creation tools on the right hand side – as you can see, the circle around the N has been turned into a polygon, and the quality is diminished. In the fields of architecture, engineering, construction and manufacturing, not being able to accurately communicate design intent reliably with others throughout the document lifecycle can cost time, money and business.


Another important detail to consider is whether the documents you create are accessible and usable by all. Acrobat provides the option to create a tagged PDF for accessibility and reflow purposes when it creates the PDF document from Microsoft Office applications. Whether you need to follow accessibility guidelines and laws or not, this is a very important consideration, this option can also ensure that the document can be reflowed and read correctly on desktops, laptops and mobile devices.

It is these tags make PDF documents truly accessible so you need to be sure that the tagging information is thorough and valid. In the image below, you can see on the left the tags in Acrobat for a tagged PDF that was created with Acrobat’s PDFMaker. The original structure of the source document is correctly preserved, including correctly tagging the text, images and tabular data. On the right are the results of tagging a document that you might see when using a non-Adobe tool. It’s clear here that the tagging is not correct in any way – everything is just marked as a paragraph; the image has not been tagged, and the text runs in to itself without any spaces. This may be a tagged PDF document that conforms to the standard, but it is not necessarily an accessible one and is essentially useless for anything other than printing or searching; copying or otherwise trying to reuse the content won’t work very well either.


But even if the PDF file has been created perfectly, by Adobe tools or otherwise, you still may have trouble using it if you are not using the Adobe Reader. Which brings me to the second scenario.

You are viewing the PDF file in something other than Adobe Technology:

Third party PDF viewing tools typically cut corners when it comes to displaying PDF files specifically in areas around fonts, image types, transparency and output intent. Users of Photoshop, Illustrator, InDesign and even Microsoft Word and PowerPoint may find that when PDF files are exported from these applications. They don’t render properly in anything other than Adobe Acrobat and Reader. For example, The PDF specification and Adobe layout tools like Illustrator and InDesign support 17 different modes of transparency; so do Acrobat and Reader. Most non-Adobe PDF viewers support one or just a few. This means that PDF files created with common tools may not look right on some people’s systems.

In the examples below. A test file created by the Printing Industries of America shows the differences in how different viewers render the same exact PDF file. In a perfect viewer, the blue and gray bars should be the same flat color all the way across. The test strip is actually composed of various combinations of transparent and opaque color blocks that represent the different transparency modes that are supported in the PDF specification. The top row shows that Adobe Reader renders the file correctly. As you can see in the bottom two, non-Adobe viewers don’t do so well. The problem is, as the recipient, you don’t know what the file should look like so, unless you are using and Adobe viewer, you can’t be sure you’re seeing it correctly.


But it’s not just the rendering of PDF files that the non-adobe tools get wrong, they also can’t manage to get the most basic of forms functionality right. One of the more popular viewers on the Mac platform can fill and save PDF forms. Great – Unless of course someone with a different viewer tries to read the data. The tool in question doesn’t generate what are known as form field appearances correctly so when a user with Adobe Reader opens the filled in form, it appears to be blank. That’s right – blank. The data is actually there. In fact, you can tab into each field individually and see what was entered, you just can’t see it or print it. Which brings me to the third scenario.

A perfectly good PDF file was corrupted by something other than Adobe Technology:

Adobe added the ability for the Reader to save changes to PDF files that are part of common workflows like filling forms, digital signing files, getting approvals and comment and review cycles. Non-Adobe tools that also let you save changes to PDF files often corrupt the special key that we put in these files and break the workflow. What’s truly unfortunate, is that the people who are participating in these workflows using non-Adobe tools are never informed that they are corrupting the file. It’s the next person in the workflow, who may actually be using the Adobe Reader, that now has to deal with a broken file and may no longer be able to save the file even though the original author enabled the file for form fill-in and save. In essence, the inexpensive PDF editing tool has crippled the file so that the free tool (Adobe Reader) can’t be used.

How do we deal with this problem?
If you are a heavy user of PDF files on your web site and you depend on Adobe Reader, either in general or a specific version, you may want to consider adding a plug-in checker to your pages that have links to PDF files. The badge in the upper right of this post tells the user if Acrobat or Reader is installed as well as the version. You can detect the Adobe viewer and then warn users if it’s not found. You can read how to add this to your own site in The Web Designer’s Guide to Acrobat.