Taking XML Training to the Enterprise: A Modular, Mapped Approach
Single-source publishing has long been the holy grail of the information industries. Historically, of course, many obstacles stood in the way: content was not easily reusable, duplication was common, maintenance procedures were cumbersome, and content-management and publishing systems had not evolved to where they are today. As a result, content was often manufactured and managed inefficiently.
XML has effectively solved this problem. With its power, utility, and relative ease of use, most of the largest commercial publishers have been quick to embrace the new opportunities.
Not so with all large corporate enterprises. There are probably many reasons, but two things in particular seem to be key. First, the tools and applications we use today do not typically support XML. Microsoft, for example, may be “betting the company on XML,” but we’re still a long way from the day that everyday users casually produce parsed, granular XML files from within MS Word or any other popular office application. Yes, Office 11’s imminent release will be significant, but business users across the enterprise are most likely a long way from cookie-cutter XML. Second, it’s incredibly challenging to implement an XML-based content-management solution across a global enterprise that has, as its constituents, mostly business users.
How to Know XML
And therein lies another critical issue: how do you get an entire organization up to speed on XML and the associated technologies? The problem of disseminating XML knowledge at the appropriate levels throughout an organization is huge. And what exactly does it mean to know XML?
ISOGEN International’s XML Training Division solved this problem in early 2001 by answering that very question with its XML Learning FrameworkTM. The Framework is a comprehensive analysis of XML knowledge requirements as they exist abstractly across an enterprise. The Framework takes this abstraction and presents it in a logical matrix that acts as a roadmap in developing learning paths appropriate to needs. No such comprehensive matrix has ever existed before in the world of enterprise-level XML training.
The Unique Challenges of Enterprise-Level Training
While the XML Learning FrameworkTM helped ISOGEN XML Training identify training requirements at every level of an organization, unique challenges still existed.
One of these challenges was that training content needed to be republished to multiple formats. Content needed to be output to PDF and print for students to use and to the Web and other electronic media for presentations and in-class instruction.
Another challenge-educational content needed to facilitate truly modular curriculum building. Modular content creation is not a simple task, and it can cripple corporate training programs as they struggle to meet the full scope of evolving requirements. To overcome these issues, ISOGEN came to several critical conclusions:
- Content needs to be analyzed in terms of the concepts taught and the skills required.
- Relationships between individual training components must be articulated to create coherent curricula.
- A distinction must be made between courses and modules (this is especially necessary when it comes to sophisticated models and architectural processes).
- Content must be completely reusable.
- Content must be modeled independent of publishing issues.
The challenges identified above were compounded by another issue: The widespread acceptance and adoption of XML meant that ISOGEN’s XML training was in unprecedented demand at unprecedented levels across enterprises spanning the globe.
How to Use XML to Know XML
Given ISOGEN’s proven success in implementing XML systems and single-source publishing solutions, the answer was obvious. Rebuild the company’s training program from the ground up in XML. Use the power and functionality of XML to organize component modules into a coherent ontology that can be efficiently used to assemble logical curricula within a given business domain.
Building a modular training system in XML called for several critical steps:
- Develop a truly robust data model.
- Architect a course creation process that permits reuse and flexibility. (See Figure 1.)
- Develop a course publication process that adds further flexibility, allows one-touch, single-source publishing, and maps to the XML Learning Framework. (See Figure 2.)
Figure 1. Architected Course Creation Process
Figure 2. One-Touch, Single-Source Publishing
A critical thing to keep in mind is that you must always make a distinction between what your information is and how you want to present it. This distinction may seem obvious, but it is not an automatic distinction. It is possible to write XML that violates this principle.
With this caution in mind, XML does, like no other data language or format, enable the development of content architectures that are forward scaleable and entirely appropriate to what you need to accomplish.
As an added bonus, using XML maximizes the possibility that your content can be used by multiple software applications going forward.
Building a Modular Curriculum
ISOGEN identified three critical steps to building a modular training system:
- Develop a data model: It may sound obvious enough, but the data model must be appropriate to your information objects and your business domain. XML alone will not ensure success. In fact, developing a bad XML data model is just as easy as developing a good one.
- Architect a course creation process: By architecting your creation process correctly, you maximize content reuse and flexibility.
- Develop a course publication process: This process allows you to add even more flexibility and enables one-touch publishing from a single source. ISOGEN currently publishes its course content in two ways from the same XML repository. First, the content is output to the Web for in-class instruction and Web presentations; second, the content is rendered through XSL engines and other processes for page composition and publication.
This one-touch publishing process implements single-source publishing from a highly-structured, carefully architected data repository.
Where’s the XML?
So where is XML in all of this? In truth, it’s everywhere. The classes are obviously about XML. All the course content is written in XML. Further, all the critical metadata files are in XML. These metadata files contain information about the course data-semantic markup describing each information object appropriately, as well as information about topics and concepts. Finally, the XSLT used to manipulate the XML files are standard and widespread applications of XML.
All these implementations conform to various document type definitions (DTDs) that enforce the structure and content rules.
This structured approach to the content has the added benefit of ensuring that the assembled courses follow the intended design of the curriculum with consistency and integrity.
The benefits ISOGEN derives from its completely XML-based training curriculum are immeasurable. Don Smith is ISOGEN’s Director of Training and a principal developer of both the new curriculum and the XML Learning FrameworkTM. As he puts it, “…the reuse of our content now brings true management to our information…changing content is no longer a huge headache-the impact has been amazingly beneficial…and it is now easy for us to produce customized courses-in as little as a couple of hours.” Smith goes on to say that “the one-touch publishing (process) is a dream come true” for him and his team of instructors.
Smith cites the main benefits as including the following:
- Single-source publishing: Training courses need to change, adapt, and evolve over time and across different audiences. ISOGEN can make the change in a single place in the XML database and have it reflected in all future outputs of the material.
- Content is separate from design: This separation may be obvious to people who understand XML, but its value cannot be overstated. ISOGEN outputs content from a single repository for print, PDF, Web, and other applications.
- Content customization: Examples can be customized to make sense and be relevant to a particular audience (business users, technical users, industry, geographical location, and so on). Specific elements, such as course components, can be renamed to better fit an audience or industry.
- Modular course assembly: Courses can be assembled from a comprehensive, pre-established set of modules, with the assembled set of modules making up the individual course.
- Rapid course assembly: The assembled course file, composed of designated modules, is transformed automatically through an XSLT-based system to produce a post-process file, which is essentially a course full of appropriate topics.
- Completely granular curriculum content: XML enables completely granular reuse of curriculum content right down to component levels such as modules, concepts, even individual points.
- Ancillary file assembly: ISOGEN’s design creates a link-base file that carries information about various kinds of associated content, such as the recommended exercises for a given course or audience.
- Learning framework report: This report details exactly what a published course is teaching in terms of category and learning levels. The report is invaluable to instructors as they prepare a class.
ISOGEN’s uniquely modular approach to building a training curriculum is possible because of XML. For the same reason, a curriculum can also be easily reconfigured or remapped to the XML Learning FrameworkTM.
However, this flexibility is applicable not only to an XML Learning FrameworkTM. The principles of any matrix modeled on ISOGEN’s Learning Framework can be married to the principles of ISOGEN’s modular approach to curriculum development to build any training program. This is true whether an organization needs to train on XML or any other technology.
It strikes me that I could learn a lot from ISOGEN’s experience in rebuilding their training program from the ground up. I create dozens of complex business proposals in any given year, and invariably, I end up reusing common content for large portions of the documents. Version control is a large recurring issue. Unless I am completely hands on with every proposal, I worry that we are not communicating effectively and consistently about the company. Even if I am completely hands on, I must manually find and copy whole components from previous proposals.
About the Author