Happy Birthday XML


      Bookmark and Share

BEST PRACTICES SERIES

As anyone with kids—or a good memory—knows, when you cross the "double digits" birthday threshold, it’s a big deal. This year, XML crossed this threshold on Feb. 10, and this got me thinking about questions that I might ask this 10-year-old in order to gain perspective on its past and future. I know I’m late, but XML is nothing if not flexible. It assured me that even a belated party is better than none, especially if I invited Alexander Falk, founder and CEO of Altova (its flagship product XML Spy is one of our favorites) and a real XML aficionado (www.xmlaficionado.com).

I began with this: "XML, you’re much more famous than your parent SGML, and your sibling HTML 4.01 was deprecated in favor of an XML standard, XHTML. Techies worldwide have heard about you, and mighty standards battles have been waged about you. However, growing is easy the first 10 years, but soon you’ll have to be able to point to practical accomplishments. What have you been up to lately?"

Confident as any precocious 10-year-old, XML replied: "First, I’ve inspired more than 38 core recommendations, everything from Canonical XML 1.1 to XQuery to XSL Transformations (XSLT), and many others. And, of course, the list of standards built on XML is enormous, including SOAP, DITA, and XBRL. And don’t forget: The World Wide Web consortium fosters only the development of basic XML standards. Other organizations such as OASIS and XBRL International have built many practical XML applications on top of the core recommendations. I’d say I’m off to a pretty good start."

I then pointed out that one of XML’s siblings may be making a comeback as HTML 5. Given earlier deprecation, one wonders why we would need a new version of HTML. One reason, according to the W3C, is that "new elements are introduced based on research into prevailing authoring practices." This sounds a lot like backsliding. XML countered: "Well, you know about sibling rivalry. And HTML 5 may be a long time in getting approved, if it ever is. And I might point out that even its authors admit it isn’t a complete replacement for XHTML. You see, I’m so flexible that I can do just about anything." Falk, in defense of XML, said, "I’m afraid that the reality is that a lot of HTML is still created by hand. Tools (such as DreamWeaver) have been very slow to enforce XHTML compliance, and people continue to generate sloppy HTML pages." Without wanting to spoil the party, I noted that, as ever, practicality triumphs. If sloppy webpages work, standards take a back seat.

I agreed about the need for practicality and moved on to my next topic: Microsoft Office’s OOXML for Word. I see this as essentially a replacement for its proprietary "rich text format," a way of displaying text on pages. Since XML has the separation of look-and-feel from content in its DNA, I asked: "Isn’t OOXML for Word’s use of XML pretty superficial, since you can’t really do much more with it than you could with rich text formats?" Sensing a rhetorical trap, XML replied, "Yes, I personally prefer OpenOffice’s ODF standard since it takes better advantage of other XML standards. But you can’t argue with Microsoft Office’s success, and it may become an ISO standard (ISO/IEC 29500) just as ODF is. Not only that, but this change forces Microsoft to document and manage future changes in an open way." That was true—XML seems wise beyond its 10 years—but Microsoft won’t support that standard until Office 2014, which means at least another 6 years hence.

However, since this was a party, I moved on to my last, least contentious question: "What do you think will be the biggest surprise use of your standard in the next 10 years, XML?" I was amazed to hear the response: "XBRL! This will transform public corporations and the financial services industry, affecting investors, processes, and data. It will level the investment playing field." I was impressed to hear XML’s familiarity with capital markets and investing. I was even more surprised when Falk agreed: "Altova is working on support for XBRL in the next major software release, v2009, and plans to have XBRL-specific features, including XBRL validation, taxonomy editing, and data mapping." This statement from a longtime XML vendor is a serious commitment indeed, and it can only support widespread acceptance of XBRL.

Differences of opinions aside, XML’s next 10 years look bright indeed.