Ionrock Dot Org

Jeff Atwood posted today on PDFs an expounded on what makes it a good format. I mixed emotions about that of course after dealing with PDF in a rather intimate manner. I do love the idea of PDF as far as layout and presentation. Where PDF is a huge hassle is in the fact that it is considered a binary format and not meant to be parsed for editing. This is a huge benefit of HTML and XML, it is very editable. PDFs are difficult to tear apart and put back together because the way the language is within the actual document is not very consistent. You can specify the same content in so many different ways that are all fundamentally different. When you consider this in light of editing, it means you need to consider the author that produced the PDF, which is analogous to having a browser war for anything that outputs PDF. For example, each version of FrameMaker would create slightly different PostScript, which meant different PDFs. Sure it looked the same, but the code that made it happen was definitely not the same.

Browsers deal with different HTML all the time, but they don't have to edit it on the fly in a specific way. Yes, I understand they do parse the output and create a parse tree of some kind, but they don't actually have to send their rendered parse tree out the door to some other browser. It is a subtle difference and I don't know if what I mean makes sense. The nuts and bolts of it are PDFs are now good, but challenging.