XML and the Future of E-Learning
By Patti Shank

Two educators were discussing an e-learning infrastructure presentation they just saw at a conference. One turned to the other and said, "Here's what I don't get. How would I use XML?" The other thinks for a moment and replies, "I guess I'd use it on my resume--when I want to look tech savvy!" (Ba-Da-Boom!)

You've probably heard of XML but may not know what it is or why you should care about it. Well, here’s why: The promise of e-learning is the ability to develop content (about learners or for the actual course) that’s reusable anytime, anywhere, any way you want. Unfortunately, that just isn't possible…yet. Enter XML, which according to many geeks—our own Answer Geeks included—has the potential for revolutionizing the Web.

XML (eXtensible Markup Language) was developed by the World Wide Web Consortium, a standards organization that develops Web technology protocols to ensure that the Web stays interoperable. Basically, XML is supposed to simplify the translation of data in one type of document to another. Consider this: Different computer systems each have their own way of storing data, which makes it difficult to share data with each other. XML will enable heterogeneous systems to exchange data homogeneously.

The basics

A few definitions Extensibility means making it possible to extend capabilities. In XML, this relates to the ability to add new markup tags on the fly. Markup tags describe how the document is structured. In addition, they can be symbols inserted into a text file to explain how the file should look when it’s printed or shown on a screen. For example, this paragraph has blank space before and after it. Therefore, the document's markup has to explain to the printer (or screen) to leave that space blank. In a HTML document, a <p> tag puts in a blank line.

In the beginning, the Web was conceived as a place to share simple documents. Hypertext Markup Language (HTML), a very simple markup language, was developed to handle that task. For instance, if you want to make text bold, you simply add the markup tags <b> and </b> around the text. For example, the output (what the browser would show) for the HTML code <b>Make this bold.</b> is Make this bold.

For the most part, HTML is simple and easy to learn. As a result, the Web grew quickly and exponentially. Its simplicity became a problem for folks who wanted to put more complex documents on the Web. In response, browsers created new tags. Unfortunately, there was little standardization and programming would be supported by one browser but not the other. XML offers a potential solution to that problem.

HTML and XML are markup languages with several similarities. For instance, both HTML and XML use tags with angle brackets. The primary difference between HTML and XML is their intended use. For the most part, HTML tags describe how the text should be presented. Here's an example of HTML:

<h3>Bancroft PS3523.O46 A6 1993</h3>
<b>The science fiction stories of Jack London</b>
<br>London, Jack
<br>Carol Publishing Group

Obviously, this is a book listing. The output of those HTML tags looks like this:

Bancroft PS3523.O46 A6 1993
The science fiction stories of Jack London
London, Jack
Carol Publishing Group

HTML doesn't tell us anything about what the content is, however. This might not seem important for a single, isolated Webpage, but it becomes critical for using the same content in a variety of places. If XML were used to markup the exact same text, it might appear like the following:

<book>
<call_number>Bancroft PS3523.O46 A6 1993</call_number>
<title>The science fiction stories of Jack London</title>
<author>London, Jack<author>
<publisher>Carol Publishing Group</publisher>
</book>

XML categorizes the text to describe what it actually is or how it could be used, making it possible to share this information among a variety of systems. Information about this book could be used in a print-based catalog, an online bookstore, and a fiction author’s database.

Bottom line: HTML describes how the data looks. XML describes what the data is.

An everyday example

Like many people, I make a grocery list. In order to save time at the store, I try to order the list by similar products. For example, because dairy products are in the same section of the store, skim milk and cheddar cheese are listed together.

I organize my list myself, but it'd be nice if technology could help. My PalmPilot, perhaps? I could create sorting (markup) categories, such as

The following list-sorting program would organize the list for me:

if(item starts with "Canned:")
then move item towards end of list
if(item starts with "Dairy:")
then move item towards beginning of list

If I get sick and a neighbor offers to do my shopping for me, he could merge our lists for faster shopping. However, the current world of XML is fragmented. If he tried to run my list through his list-sorting application, it most likely wouldn’t work because we used different sorting categories or markup tags. His markup tags might be

Although we use the same data (skim milk, tomato soup, cheddar cheese, applesauce), it isn't marked up the same way. The sorting application will only be able to understand the data if we use a utility program to convert our formats, changing the first dairy item on my list "Dairy: skim milk" to "Wet_dairy->skim milk" and so on.

Extensibility creates another problem that XML can help solve. For example, if I have a child and add “Infant: diapers” to my shopping list, the existing list-sorting program doesn't have a category for "Infant." The existing list-sorting program becomes less useful as I need to add new types of items.

Enter XML

How might XML help? Consider the shopping example. In XML, my list would translate into the following:

<dairy>skim milk</dairy>
<canned>tomato soup</canned>
<dairy>cheddar cheese</dairy>
<canned>applesauce</canned>

Here’s how my neighbor's list would appear:

<wet_dairy>skim milk</wet_dairy>
<can>tomato soup</can>
<dry_dairy>cheddar cheese</dry_dairy>
<can>applesauce</can>

The tags are still different, but the markup structure (<tag> </tag>) is the same. The same structure will simplify data translation because XML technology includes standards for describing how to translate from one kind of tag to another. More important, because XML is extensible, new tags can be added as needed. It’s not a major concern to add <infant>formula</infant>, for example.

But how does the shopping example apply to online learning? Consider using the following markup code for a multiple-choice quiz that you develop:

Meanwhile, another developer’s code could look like:

A non-XML learning management system that understands the first set of markup tags may not understand the second set. Like in the diaper example, if the markup language only organizes multiple-choice questions, you're confronted with the additional extensibility problems if you want to add fill-in-the-blank interactions.

If you want to learn more about XML, Gibson recommends the following resources: IBM developerWorks XML Tutorial; www106.ibm.com/developerworks/ education/xmlintro W3C's XML in 10 points; www.w3.org/XML/1999/XML-in-10-points The XML FAQ; www.ucc.ie/xml/#acro. I recommend the book XML By Example (Que, 2000). It's a solid reference on a complex topic for those of us who aren't geeks!

So do LMSs currently on the market employ XML? Is having AICC compliance, for example, the same as using XML? According to Todd Gibson, a Denver-based Web programmer and infrastructure guru, the answer is no."AICC isn't extensible. It can't anticipate all possible learning interactions. An extensible LMS standard will be needed in the future." Gibson goes on to explain that XML is quickly becoming a cross-industry standard for sharing data. Indeed, because AICC doesn't cut across industries, the market will eventually need XML-compliant LMS systems regardless. In other words, XML will better allow training departments to share learner data with accounting systems to generate tuition reimbursement checks and HR systems to verify training for career ladders, and other systems. The buzz surrounding interoperability, learning objects, standards, and metadata are an effort to make design and development more resource efficient.

Today, the online learning market is still in its first generation, grappling with new systems, standards, and solutions. Because sharing data is problematic in other industries as well, XML is the likely answer. ADL, a leading e-learning standards body, agrees. ADL considers XML to be an important part of e-learning’s future and potential for growth and integration into other enterprise-wide systems. The rest of us will just have to wait and see.

 

Published: December 16, 2002