Don’t Forget To Test Your PDFs

The Adobe Acrobat PDF (Portable Document Format) file has become ubiquitous on the World Wide Web (WWW), and a lot of times Web site and content creators don’t give it a thought when they plop files out there on the Web site or generate them dynamically. Since they’re out there, you need to make sure that they’re correct and proper.

So what should you look for when you’re reviewing PDFs?

For starters, you should check the layout, grammar, and whatnot within the document. Of course, most places don’t tend to run the copy and the creative by the QA department; most places suck, and the two might correlate more than most places think. So when you’ve got a PDF file on your desk, take a look to make sure that the “writers”–who will either be crumb-paid former English majors happy for a job or semi-literate executive types who can’t be bothered to communicate effectively while they’re paid to lead from the safety of some golf course–haven’t misspelled the company name or anything. You should look to the written document as a printed document and proofread accordingly, making sure that fonts are visible, headers and footers say the right things, and that the page numbers line up.

That last bit is important, as some applications repaginate base on printer driver settings, and if your content creator used Microsoft Word with old Adobe Acrobat print drivers, it is possible that Microsoft Word automatically reformatted the document before generating the PDF files. Sometimes, this had the effect of “printing” the document to pages shorter than standard size, so that the bottoms of pages would carry over into the next page. This would often “orphan” headings or regenerate page numbers without regenerating the table of contents or indexes so that those two directories didn’t line up.

Okay, you’ve looked at the contents. Now, look at the document properties available from the File menu. When your experts generate the PDF files, the process carries forward some metadata information from the source document and from the user who generated it:

Adobe Acrobat Properties
Click for full size
You want to make sure that these properties don’t carry with them your company’s name instead of your client’s name, a template name that doesn’t mesh with the document’s goals, or your expert’s user name on the corporate network.

If your company, for some stupidity, transmits Microsoft Word or other source documents, it’s doubly important to check those file properties as well and to flush any functionality in macros out before making the documents publicly available, but that’s a whole nother story.

Okay, now that you’ve got the document properties cleaned up (after three or four rounds with the content generators, no doubt), you’ll also want to check the backward compatibility of the PDF file.

Because as a document that requires a user-installed plug-in to appear correct, your “experts” have a later version than most of your readers and they, your experts, no doubt want to include gee-whizzery in the document that will look like crap with older versions or that might not work.

So you’ve got to open the document in a couple versions of Adobe Acrobat Reader to make sure the file is backward compatible. You don’t want to subject your reader to things like this:

Adobe Acrobat does not support
Click for full size
Even if they wanted to upgrade, some of them cannot due to security practices at work and whatnot.

And don’t zero in only on this message. Any warning from Adobe Acrobat Reader, such as missing fonts or missing language packs, should be squashed before you subject the world at large to your important and meaningful communique.

That’s just the basic PDF files, friends. If your organization introduces hyperlinks or other gee-whizzery into its PDF files, you need to test those things, too; if your organization generates dynamic PDF files using customer-entered data, you need to not only test the above, but you need to test to make sure the data is transferred to the PDF correctly and that the document is formatted correctly.

For example, one such application I know of failed a backward compatibility test because the line break characters showed up like this in Adobe Acrobat Reader 6.0:
Adobe Acrobat does not support

That’s the result of an unrecognized character, of course; Adobe Acrobat 8 recognized it, but it looked like carp in 6. What, you don’t find the center indicative of a mouth and the lines going to the corners are representive of gills?

Regardless, the common and often overlooked PDF file represents one more way that your organization can look amateurish and foolish unless QA steps in to save the day. Again.

Comments are closed.

wordpress visitors