A PDF Progress Report --McIlroy

I happen to be a big believer in PDF. I think it's the best technology driving workflow improvement today, and that it will bring even more benefits to workflows in the years ahead. Now that's hardly a controversial statement—you could even call it a widely-held belief in our industry.

But I know there are lots of people out there, both publishers (in the broad sense of the word) and printers (ok, in the broad sense of that word, too), who, while they have a generally positive attitude towards PDF (how could they not, after all the good press we've given it?), still have some doubts, and a few questions they'd like answered.

PDF workflows are deceptive. There's Adobe Acrobat, and there's Creo's Prinergy and Agfa's Apogee, and then there's . . . what? That seems to be the #1 question surrounding PDF and PDF workflows. Is it enough just to buy one of these three systems to get a workable PDF workflow?

The answer, as some of you know, is "yes, sort of, maybe".

OK, then a follow-up question. If I buy Acrobat and Prinergy or Apogee, can I just plug them in and push print? If not, what else do I need to know?

And this is where I need to sit you down and say, "Have you got a few hours? I can give you about half of the answer in that time."

The fundamental problem with PDF is that it is, in a way, too good. It's too powerful, too flexible, too malleable. There are a great many things you can do with PDF, and literally hundreds of software tools and systems available to help you do them. I guess we all like powerful, flexible technologies. But when it comes to graphic arts prepress workflows we want our powerful and flexible tools to behave consistently and reliably, to chug through the toughest files and produce predictably perfect results.

Pedal to the Metal

With most of the prepress systems we grew up on, you'd ordinarily have to really "push the pedal to the metal" to chug through the toughest files and produce predictably perfect results. With PDF workflows it's the opposite. You grab the steering wheel tight, take your foot off the accelerator, and position it right next to the brake. Good PDF workflows take careful positioning and some restraint.

The Seybold organization has just released the results of an ambitious project to try and find the answers to some of the tough questions that surround making PDF workflows work. (I worked with Seybold 'till 2001—but not on this project.)

The in-depth "PDF Workflow Shootout & Usage Survey" ($450 from Seybold Publications) is an 84-page report that looks at two sides of the problem: what do publishers (PDF generators) want, and what do printers (PDF processors) want?

It's perhaps easier to say what the companies that process PDF files for customers want—as described above, a system that can chug through the toughest files and produce predictably perfect results. Initially, Seybold set out to test 13 systems that could create "prepress-viable PDFs from four problematic application files." Then they tested 18 systems that could output the final PDFs.

What to say about the results? Hmmm. I'm not quite sure. Oh, yes, here's what I can say, "They're interesting." I've never been a big fan of contests, whether beauty contests or prepress system contests. The surrounding drama is always fun, but they're always marred by the arbitrariness of the contest rules, and by the little things that invariably go wrong and skew the results, whether it's just a contestant having a bad day, or a corrupt judge from a far-away place.

Top scores went to one of our industry's largest suppliers, and to one of our smallest. Both Creo and OneVision have already sent out press releases trumpeting their victories, Creo, as vendor with the highest overall score (albeit for two different systems), and OneVision for the highest overall score on a single system. Congratulations to both.

But what does that really tell any of us about the two dozen other systems? That they don't work? No, that's too strong a statement. That they don't work as well as Creo's and OneVision's system? No, it only says that for this particular set of rigorous test files, with this particular set of judges, Creo and OneVision outperformed. The rest is highly speculative.

As the report's writer admits, "one conclusion became inescapable: PDF workflows are highly vulnerable to human error…we estimated that the overall scores were 2 percent to 10 percent lower than they could have been, due only to human error."

Real Value

What's most valuable in the study is the recognition that (a.) there are a lot more systems available for processing PDF files than Acrobat, Prinergy and Apogee, and that (b.) not all systems are created equally—some perform better in certain areas than others, and generally for a good reason.

For the printer, you're faced with the usual choices. Buy the name brand and get on with it, relatively confident that it will do the job you want. Or investigate the full range of choices out there, knowing there is probably another system just a little more suited to your workflow (and budget).

In my next column I'll look at the second half of the report, which examines what users are doing with PDF files.

—Thad McIlroy

About the Author

Thad McIlroy is an electronic publishing consultant and analyst, based at Arcadia House in San Francisco. He welcomes your comments at thad@arcadiahouse.com.