In June of 2007 Ecma TC46 took over development of the XPS specification from Microsoft. After two years of hard work and fixing over 200 issues the new Ecma standard for OpenXPS (Ecma-388) was approved yesterday. Like all Ecma standards it can be freely downloaded: get it HERE.

Also approved yesterday was a formal JFIF technical report. The JFIF specification defines a file format for a JPEG-compressed image, and was written quite a while ago by an informal group lead by Tom Lane, and has been referenced and implemented very widely, but it’s never had any official home; you just had to trawl the net to find a copy (probably ending up with one unofficially hosted by the W3C). TC46 worked with Tom, and liaised with ISO/IEC JTC1/SC29 (the home of JPEG) to bring out a technical report that is technically identical to the informal specification to enable other standards to reference it in a more robust way. JFIF is published as TR/98 and available HERE.

Congratulations to all of TC46 on a job well done.

There’s more information about TC46 HERE.

And more on XPS, both the format and Global Graphics’ technologies HERE.

Martin

Ian Douglas at the Telegraph has written a blog post complaining about PDFs because they often don’t work well when read on computer monitors and especially not on mobile devices like an iPhone. I’m generally a PDF advocate, but I have to say he’s right … sort of.

First, look at the life-cycle of a document; it needs to be managed in an authoring format at early stages. Allowing free and complete editing needs a “flow format”; something like a Word file or ODF that allows text to run over line ends when you insert or delete a word. Later in the cycle most documents benefit from translation to a distribution format, which is less susceptible to accidental change and (more or less) guaranteed to accurately reflect the author’s intentions.

But the proliferation of publishing opportunities today (in print, on the web, on mobile devices etc), combined with different requirements for fidelity can make selection of an appropriate distribution format difficult. In some cases all that matters is that the words are in the right order with some identification of headings and emphasis. In others the precise design matters. In between those extremes the layout can be a vital adjunct to a stream of words by enabling references to a specific clause, as would be the case for legal texts.

If the page design is important then PDF (amongst a few others) is an extremely good distribution format. But the need to respect that formatting can then cause annoyance if the formatting is inappropriate for the way that you view the file.

Most computer users have landscape format screens; most documents are written on portrait format pages; both are for historical reasons. Reading a portrait page on a landscape monitor can be a very frustrating experience, especially if it uses multiple columns. I’m lucky enough to have a rotatable monitor that lives almost permanently in its portrait position and I would be loath to lose it. Large tables that can’t be fitted into view at a sensible scale are also difficult to read.

Document designers and publishers could alleviate matters hugely by designing pages appropriately. If most of your readers will be using a computer screen rather than printing out your documents could you design on a landscape page? Can you sensibly split up your big tables, or re-format them so that each segment as you move through them makes sense on its own? What really is your goal in distributing the document; is it to provide just the words, or does the design matter as well?

Of course, if you use gDoc Fusion to read files it doesn’t matter whether somebody sends you a Word doc or a PDF … but the way the pages were designed will still play a role in how easy it is to read!

I have very mixed feelings when I see recommendations like this one from RIM: “Prevent the BlackBerry Attachment Service from processing PDF files in a BlackBerry Enterprise Server environment” (link). Does this, and similar recommendations on other platforms, including mainstream desktop operating systems, mean that PDF has failed as a preferred document distribution format?

No, of course it doesn’t. It shows that PDF is ubiquitous and important enough that the bad guys see it as worthwhile to target. If nobody was using PDF they wouldn’t bother because they couldn’t gain any advantage from doing so.

And for PDF vendors? We’re in the front line now; we have to code well and avoid all those buffer overflow issues and other vulnerabilities so that PDF usage can continue to grow and be successful.