Thursday, January 03, 2008

Coding Horror: The Trouble with PDFs

Over the past few years I've refrained from posting my own rant, tentatively titled "PDF is the problem, not the solution". I was a bit surprised to see a similar rant coming from one of my favorite developer blogs:
But I have a problem with PDF files (Coding Horror: The Trouble with PDFs)
Jeff Atwood lays out a few of his issues with PDF, and the reader comments include a more extensive and lively discussion of others'. I think most people confuse PDF with a web format, but a few of the commenters have it right: PDF is a print publishing format, but one which is reasonably friendly to web-based distribution. It is not a suitable web publishing format for so many of the reasons listed in Jeff's post and the reader comments, but it is fine as a print format for web distribution.

My rant is actually a bit different from Jeff's, coming from the perspective of commercial digital printing. PDF is a major inhibitor to the development of dynamic, data-driven print because of its static nature. (Regarding "dynamic data", what I mean is a content formatting process equivalent to what happens on virtually every web site today, where each page is generated dynamically and customized to certain context). PDF was not designed to support dynamic data, but that is the future of print publishing. PDF is good at what it does, but the market needs to be re-educated about PDF and the constraints it imposes: it is not an adequate solution to the production of high value customized digital print. Commercial print is stuck in the dark ages, perhaps the only major information-centric industry that has not been transformed by the Web.

Dynamic commercial digital print is a huge market opportunity, so you can bet this problem will be solved. But I'd also bet that today's PDF will only be a footnote in the solution.