XML and Content Reuse Systems for Instructional Design Part III: Creating a Unified Content Strategy

The previous two articles in this series discussed the basics of XML, and the taxonomies, processes, and tools used for generating and modifying content. XML is the organizational methodology used to construct repositories of learning objects that have been created with taxonomy.

The learning taxonomy is used to classify information according to a set of rules. Repositories for these derived learning objects represent the culmination of a gradual process that began with simple file sharing. The reason for entering into all this extra work is to allow the learning organization to be capable of responding more quickly and flexibly to the needs of students. Different tools can be used and none is perfect. Proper training and supervision in how the tools are used will have much more impact on the finished product than will tool selection.

This concluding article of the series examines how the repository content can be used as part of a unified content strategy aimed at creating measurable return on the system investment.

Implementing a unified content strategy

Figure 1 represents the four basic component areas of a content reuse system (CRS). The specific applications illustrated could be included in any CRS that resulted from a unified content strategy.

Figure 1: The XML Content Reuse System has four elements.

Let’s examine the four elements:

• Content System — This is the repository. The repository is comprised of the database engine and may also include a version control system. In Figure 1 on page 2, the database component is shown as Oracle 11i, which is only one of many possible choices. The most important consideration in choosing the database application is to ensure that it will have the capacity and throughput to handle your anticipated use. ClearCase and Perforce are both examples of version control systems.
• Parser, or Query Engine — This is the set of methods that enables users to get content back out of the repository and to use it efficiently. In the diagram, Stored Procedures, XSLT, Perl, and Python are all examples of different means of serving complex queries to users.
• Authoring Content Management — This represents the user applications you have identified for authoring new content. The examples listed in the diagram are FrameMaker, Epic Editor, Dreamweaver and Word.
• Delivery Outputs — Your delivery outputs encompass the output formats (paper or online) as well as the means provided to students for accessing the current instances. Examples of delivery applications include Adobe Document Server, Adobe FrameMaker Server, or a LMS.

Each piece of this larger system is associated with specific benefits and costs. Although each element must be evaluated on its own merits, as well as in the context of its performance within the content reuse system, system considerations can easily outweigh the benefits of any individual application choice. If the favored application, for example, does not play well with others then it will be of little use in the system. As discussed in the second article in this series, Microsoft Word is a good example of an application whose system behavior makes it a difficult choice. It is precisely because so many organizations are adopting content reuse strategies that Microsoft has intensified its efforts to make Word (and other Microsoft applications) XML friendly. The future will certainly bring successive versions of Word that integrate better into a unified content strategy.

There are basically two ways of achieving a content reuse system: build your own from available components, or buy one that does most of what you want and then customize it. If your organization has many specialized requirements and diverse processes, along with considerable expertise and experience developing, implementing, and maintaining software solutions, you will probably not save any money by customizing a proprietary solution. If, on the other hand, your organization has much more general requirements for training, fewer audiences and simpler outputs, purchasing an off-the-shelf system may be a better solution. A vendor-supplied solution may also be in your future if your organization lacks in-house technical expertise and you normally contract out such projects.

Build your own

In order to devise your own content reuse system, you need to have some specific areas of expertise available:

• Database Architect (DBA) — The DBA creates a data library that exactly matches your document type definition (DTD). The library consists of data tables that are optimized to perform well with the most common search routines. The architect should be experienced with hardware and with the network configurations appropriate to your organization’s needs.
• Database Interface Designer (DID) — The DID is going to organize your query engine and make sure that all the routines necessary to input and output data to your authoring and delivery environments operate properly.
• Configuration Engineer (CE) — The CE configures and maintains the version control repository. This person should be an expert in the software you have selected (ClearCase, Perforce, etc.). Many DBA’s think they can do this job, but very few can. Configuration engineering is very important to making the whole system reliable and expandable.
• Template Designer — You will need one of these for FrameMaker and another one for Dreamweaver, if you use these products. Many organizations contract this task. Contracting is an acceptable alternative, as many excellent consultants exist in this field.
• LMS or Server Engineer — This is an expertise that is generally provided (for a fee) by the software vendor that supplies the LMS or server platform. As noted before, Adobe has a wide range of supporting and training services for their enterprise server products.

You’ll only get the full value of your analysis and planning if you apply the results of that research by developing your own solution. Any other approach compromises your results. You’ll also build a core competency in developing and delivering learning objects.

The principle requirement for success when developing your own solution is buying from top management. There must be commitment and a requirement to achieve a workable system in a modest time frame for a realizable cost. Successful completion of a system in-house results in the biggest gains in productivity and largest reduction in cost per training hour. Any organization that has a sincere commitment to providing quality training programs, especially one that aims to increase the percentage of e-Learning in its training offerings and that has more than 10 training content designers, should consider creating its own system.

Some of the main advantages and risks of developing your own content reuse system are shown in Table 1.

Buy existing system

The principal advantage to purchasing a system off-the-shelf is that someone else claims that it will work for you, and they guarantee that they will support your implementation of their software. As with any vendor, you are negotiating a relationship of mutual benefit. Always spend more time researching the company and their references than you spend listening to the sales pitch.

Things to avoid when shopping for a system:

• Being the first customer, or being that vendor’s first “big” customer.
• Buying a solution you do not understand — or one that the vendor is unwilling or unable to explain so that you can understand it.
• Becoming a client of a company whose primary goal in software design is to lock you into their proprietary framework. This can be very dangerous, especially if the company disappears in five years.
• Purchasing a product that does not do some of the main things you require it to do, on promises that the company will customize it to do exactly what you want. At this point you might as well make it yourself.

Some of the main advantages and risks of buying an off-the-shelf content reuse system are summarized in Table 2.

The deciding factor in whether to buy a vendor product, as opposed to creating a custom solution from other components, is resources. Getting any new system implemented is going to require resources. If the resources are not going to be available within your own organization, then you will need to purchase those hours from external vendors. Creating your own custom solution is going to require many more hours of development than implementing a vendor solution. If your training department is small, or your organization does not have the budget to spend on developing future capabilities at the expense of deliverable training hours today, then you may have insufficient resources to properly design and implement your own system.

No system that has insufficient development resources allocated to it can compete with an off-the-shelf product. In developing that solution, the vendor can amortize development costs across many different clients. Continuing development and maintenance costs are similarly shared. Many organizations have a cultural bias toward purchasing turn-key solutions, even if they do not perform as well as custom applications. Regardless of the technical benefits bestowed by one kind of system or another, it is often better to pick the right solution that matches the business realities of the enterprise.

Implementation scenarios

The following scenarios represent three different approaches to implementing a unified content management strategy. Though each is based upon a concrete case history, some details have been specifically altered to avoid the identification of the businesses or their employees.

Scenario A — aerospace

Business A is an aerospace company with a very large and capable IT organization. It has a history of developing very complex, highly customized solutions that meet exacting business and regulatory requirements.

When Business A went out into the content management marketplace, they did extensive research of many different vendors with competing products. They had a tendency to “study a product to death.” The IT and engineering organizations generated thousands of pages of conflicting and contradictory requirements, which no vendor was able to meet.

Business A purchased an off-the-shelf product, which the vendor promised to customize to fit the needs of the enterprise. The IT organization fought the project tooth and nail from start to release. When eventually implemented, the system was largely ignored by many of the divisions of the organization, despite having been specifically tailored to meet their needs. The Director of Information Services and Communication then used this software as a club to bring each of the disparate organizations into line — to streamline their procedures and to regularize their methods for producing documentation and training for each of their markets on five continents.

Though training productivity suffered initially, after all was said and done, the system achieved a 40% increase in training hours per designer. The resulting training was consistent, won numerous industry awards, and was instrumental in creating a truly global training organization.

Scenario B — manufacturing

Business B is a large manufacturer of consumer products, with a relatively small and under-appreciated IT organization. It regularly purchases software solutions and maintenance contracts that provide for the special needs of specific user communities within the organization.

When Business B went out into the content management marketplace, their aim was to find a state-of-the-art product that they could purchase to perform a limited set of very specific tasks. They concentrated on vendors having associations with their existing vendors and very quickly narrowed the choice down to two competing products.

Business B hired a team of three consultants to work with every division to develop a customized solution from open source components. In the process of analyzing the communication and training needs across the different divisions, the team discovered large pockets of inefficiency and waste. During the three-year development cycle, the development program cost the organization approximately $17 million.

In the ensuing five years, the resulting system consistently produced higher quality training deliverables throughout the enterprise and contributed significantly to lowering the training costs for new employees by 38%, resulting in an average cost savings of $12 million per annum. By selecting this solution path the company identified training as one of their core competencies.

Scenario C — retail

Business C is a major force in retail, with both corporate and franchise operations world-wide. Their stated aim in adopting content reuse stemmed from dissatisfaction with the results of their training programs. They felt that they could achieve better, more consistent training outcomes by creating better and more consistent training content.

Business C quickly selected a content management package from one of their existing vendors and implemented it on a trial basis in a single division whose training outcomes were dead average for the organization as a whole. Although the system did result in economies in the production of training content consistent with the vendor’s promises, the training outcomes did not improve.

The trial implementation was written off to experience and a new vendor with a different content management offering was selected. The results of this trial in a different but equivalent division produced approximately the same economies and the same mediocre training outcomes.

Leaping to the correct conclusion that garbage in results in garbage out, Business C conducted another trial of the second system within the organization that had the best track record for producing positive training outcomes. To their great surprise, the training resulting from this trial was as indifferent to the technology as the others had been.

An expensive consulting firm was brought in to study these three trials and to find the silver lining in having apparently wasted several million dollars. Six months later, the consultants returned their verdict: Business C was attempting to solve the wrong problem with the right solution. The consultants recommended that the organization implement the second vendor’s solution across the entire enterprise. This would produce economies in production of training, but more importantly it would save tremendously in localization costs for training materials.

The consultants concluded that the poor outcomes from training indicated that training was being used inappropriately as the cure for problems that did not arise from a lack of good training. By reducing their focus on training as a cure for all ills, the company was able to concentrate on better internal communication. The same content management system that was adopted for training was the perfect solution for most of the new communication initiatives.

Comments

Login or subscribe to comment