A Conversation with ... KEVIN DUFFY, XyEnterprise

Thinking in abstractions gives me a headache. That’s why I was never good at algebra. And THAT’s why I went into journalism...no math requirement.

So imagine how refreshing it is to find a plain-speaking person who can explain one of the most abstracted of all recent subjects—the automated, component-based formatting of content for authoring and publication through the many channels available today—to a literal-minded, J-school student with an abstraction problem.

As most of you are aware, the “MLs”—XML and before that, SGML—are pioneering standards that add consistency and structure to plain text with extra information. “Element names” indicate what kind of text is included (e.g. I am a “para” or I am a “step”). “Attributes” are value-added information (e.g. author = “Andy Moore”). These are just some of the “extras.” There are many more, but XML is considered pretty simple to understand and use. In fact, software can be taught to process and format it without human intervention. And that’s just the beginning of our story.

My tutor is Kevin Duffy. Kevin is the president and CEO of XyEnterprise, a company which can only be described as one of the earliest movers out there focused on solving the challenge of publishing complex content, in variable formats, to various channels, in the easiest way possible. Not an easy task. But Kevin and Co. were thinking about it a long time before you or I ever did.

“Twenty years ago—when there was no XML or SGML—our first publishing product had some of the characteristics that now exist in some form in the prevailing standards such as XML. Called XPP, we applied a generic mark-up onto the text structure, which divorced the formatting and presentation information from the content,” Kevin explains.

“This was 1983-’84, and we were already promoting that people move away from typesetting and proprietary mark-up and start thinking about labeling the structure of the content—a head is a head, a paragraph is a paragraph, a bulleted list is a bulleted list...

“Once you had this,” Kevin continues, “you could then set up style sheets that could automatically change from A4 (Europe’s standard “letter” paper size) to US 8-1/2" x 11" , or from portrait to landscape. All those kinds of nuances that define the document’s presentation difference can be separate from the actual content.” This is one of the fundamental theories underpinning XML, and XyEnterprise was onto it 20 years ago.

But merely (merely!) simplifying the presentation of content is not the entire story. “The second piece is the high level of automation possible,” says Kevin. “The idea is to keep human beings out of the mix as much as possible, and let the system do its thing algorithmically. For example, in the legal market, where you have gnarly elements like footnotes and cross-references, being able to avoid having an individual sit and handcraft pages is a HUGE cost-savings. When the look of the document is fairly predictable, you can instruct software to do all the things that, theretofore, a person had to do,” says Kevin.

I take this opportunity to ask another typical (i.e. dumb) question: “Sounds like a good solution for big companies with complex technical documentation and ‘gnarly’ formatting problems, all right. So what’s it mean to the little guy?” I thought I was cornering him. But I didn’t.

“It’s not just large companies that have to wrestle with very dynamic product schedules,” says Kevin, more politely than I deserve. “It’s no longer acceptable to update products every two years... you’re probably updating every six months. And you probably have five variations on the theme of each of those products. So even small companies have more dynamic and complex products. Thus the cost of creating and maintaining the ancillary content (the manuals, the sales literature, the online help) becomes a larger percentage of their costs, so they need to find a way to do it more cost-effectively.

“Furthermore,” Kevin continues, “even smaller companies are doing business globally. It’s not just the Global 1000 that need multiple-language content. As soon as you need multiple languages, you exponentially increase the complexity and
potential cost of the support materials.” Translation, as it turns out, is one of those unexpected cost centers that tend to sneak up on the less experienced companies. And XML excels at making language translation a heck of a lot cheaper. Thanks to its “component” nature, content in XML is in chunks; when one of those chunks changes, you only need to translate that chunk. It saves you from re-translating an entire chapter or section when you only need to fix a paragraph or two; it’s a “changes-only” strategy that makes eminent sense to me...an avowed non-abstractionist. “The challenge is that now even small companies are wrestling with things that only the largest companies had to wrestle with only a few years ago,” says Kevin.

XML isn’t the only advancement making life easier for those facing publishing challenges. The Darwin Information Typing Architecture (DITA) is an XML standard that pre-defines content granularity as small, self-contained “topics.” DITA also provides a means of extending the standard —sort of like taxonomies— for various vertical markets and business functions; vendors and open source developers are building on this DITA foundation every day. (That explains the “Darwin” part—it keeps evolving.)

XyEnterprise—and Kevin Duffy himself—are very big in the DITA movement, and evangelize the concept as well
as anyone can. Read more about it in these pages.

Let me just say that the XyEnterprise story I learned is many-faceted, no doubt. It leverages content management systems (CMS) to control and keep track of content components, who uses them and when they were most recently changed; it uses XML to simplify the publication of that content in its many ultimate forms; and it increasingly is welcoming the DITA (and S1000D...I didn’t even get to that one!; you’ll have to look it up) topic libraries to bring the power of the “component content concept” to the masses.

I think you owe it to yourself to see where any or all of these facets can play to help with your organizations’ content demands.

Special Advertising Section

KMWorld Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues