NETg

Company: KMSCustomer: NETgSubmitted by: The itpr PartnershipDeskartes, the intelligent knowledge management software from KMS, has helped leading global e-Learning specialist NETg to turn its vision of just-in-time learning into a reality.One of the major attractions of e-Learning has always been the prospect that it will enable “just-in-time” learning: training or education delivered in small chunks that precisely address a particular need or situation.NETg’s Director of Marketing Development, Jon Buttriss sums up the advantages simply: “You learn best when you need it most.

” Taking five minutes’ worth of training and then applying it immediately to a real-life situation is a sure way to make it lodge permanently in your brain.Depending on your learning style, this type of learning can either be substituted for or complement classroom training or more conventional e-Learning methods.

PAST OBSTACLES TO JUST-IN-TIME LEARNING

Just about everyone likes the knowledge-at-your-fingertips idea. But until recently, it’s been next to impossible to achieve. The major difficulty was that of enabling the user to navigate from their training need – which is not always easy to express accurately – to the appropriate piece of training material, without having to sift through a pile of less pertinent information along the way.

NETg had an advantage in this arena because of the modular way in which its courses are designed. It has built up a vast store of what it calls “Learning Objects” – 80,000 in all – mostly relating to IT training. Each of these objects is a five-minute self-contained module covering a specific task (such as printing a document from Microsoft Word), complete with an initial statement of learning objectives and a final interactive assessment to check that the objective has been met. Wherever possible both learning and assessment are done through simulations of the actual task. That ensures that a skill is genuinely mastered, in a way that no amount of instructions and multiple-choice tests can.

NETg has been developing its courseware this way for several years. As part of the courseware creation process, it prepares metadata using accepted industry standards. This metadata is used by the software engine that runs the Learning Objects, but also makes the course content suitable for processing by other applications.NETg therefore had a significant advantage when it came to delivering small learning modules on a standalone basis. However, the problem that it still had was the difficulty of pinpointing a module to match the needs of a particular individual at a particular moment.

How could users instantly find the one Learning Object in the 80,000 that would answer their question?No-one can be expected to hunt through course maps, however well-designed, when what they need is an immediate response. And creating an index of individual modules would have been extremely labour intensive, and still not resulted in a very quick way of finding what you needed, as some of NETg’s customers had discovered when they tried to do it that way.

THE QUEST FOR A KM SYSTEM

To NETg it was apparent that knowledge management technology held the key.”We knew there ought to be a system that would let users write a question and be directed to the module that they needed,” Jon Buttriss says. “Our ideal system would also let us collect feedback as to whether the answers provided met the need so that we could improve as we went along and give a better answer next time.

“All that was theoretically possible, but the right technology wasn’t easy to find. The company had been looking at the knowledge management market for several years before it encountered KMS. “I must have looked at 12 or 15 knowledge management systems over the years, and none of them fitted the bill. Some were prohibitively priced. Others didn’t have the capabilities we were looking for: a surprising number lacked the piece where the student evaluates how well the answer matches their need and the system learns from their response – a piece which we regarded as essential.” Meanwhile some NETg clients had experimented independently with knowledge management systems with a modicum of success, but none had happened on an ideal solution.

Fortunately, in mid-2000, a KMS Senior Knowledge Consultant happened to get in touch with Jon Buttriss. The consultant was acquainted with some current e-Learning offerings and spotted that what Deskartes was already doing for support functions, it could equally well do for a training provider. He comments “There’s a close similarity, in fact an overlap, between providing someone in need of support with self-service access to a knowledge base, and helping someone in urgent need of some training to find the courseware they’re looking for.”A first look encouraged Jon Buttriss to believe that Deskartes might fit the model NETg had in mind. It would not only direct users to the right piece of training in response to their questions expressed in ordinary English (or any natural language), but would also check with the users as to how appropriate the response was.

It would then “learn” from their evaluation so as to come up with a better response next time.What’s more, a little investigation showed that making NETg’s course content accessible via Deskartes would require minimal effort. He explains, “Deskartes accepts input in HTML format, so all we had to do was to get our product to output an HTML stream containing the course content and its metadata for Deskartes to read and process.”

THE PROJECT

To confirm that Deskartes fitted NETg’s needs as well as both companies hoped, KMS took a single course – on the topic of Java – and loaded the content and metadata into Deskartes. With a minimal amount of effort, the system was able to begin responding to questions about Java by delivering an appropriate chunk of learning.

The relevance of the responses could quickly be improved by having NETg training experts ask some questions and evaluate the responses, enabling Deskartes’ neural-net based intelligence to “learn” from its experiences.Based on the success of this initial two-week experiment – the results of which impressed senior NETg staff worldwide – both companies were eager to get going in earnest. After reaching an agreement about the commercial side, they did. The next step was a co-operative project, based at NETg’s office in Chicago, to create a more extensive “proof of concept” that could be shown to users. There was work to be done in defining the internal and external architecture of the system, but surprisingly little need for technical development. “Our engineers were quite disappointed, smiles Jon Buttriss.

“They were hoping for a challenge, but because the two systems fitted together so well, and because we already had well-structured material complete with metadata, it was really very simple.”The proof of concept project started in mid-July and by late August there was a usable application. The NETg/Deskartes system, called ‘Search and Select’, was rigorously tested in-house, and then went into pilot with some key users. “By that time we were convinced we had a great idea, and confident that the industry would take to it like wildfire. So it was very important to collect feedback and get the system absolutely right before we made it generally available,” explains Jon Buttriss. The system was live at the pilot sites in September.

The users’ reaction was overwhelmingly positive. One of NETg’s key customers, Bruce Mills of Honeywell, commented, “Providing appropriate employee training without impacting day-to-day productivity has been an ever-present challenge for business. The flexibility of NETg’s e-Learning solutions allow us to be very innovative in our training programmes, maximising knowledge benefits while minimising productivity impact. Search and Select is a significant step forward in achieving an environment where learning is simply part of the work process.”And so the new product was officially unveiled in November.

BUSINESS BENEFITS

In a world where everything needs to be done yesterday and many of us are constantly on the move, traditional training models can seem obsolete.

If you need to know something, you very often need to know it now, not in six months’ time when you will be able to fit in a course. And even if you have taken a formal classroom or e-Learning course in the subject at some point in the past, you often need a refresher before putting what you learned into practice.All that’s finally possible thanks to the collaboration between NETg and KMS.And, points out Jon Buttriss, it’s not only the super-busy who may prefer to take this approach. “The average person’s concentration span is just 18 minutes, so this method of learning arguably fits in better with our natural learning style than a two-day course would,” he says.

“And because it’s so convenient, it’s much more likely to make learning into a life-long habit.”Although many people will want to combine just-in-time with more conventional classroom or on-line courses, for some, it may become their preferred method.Before NETg joined forces with KMS, finding the right learning object would have meant finding out what course it was in and then searching for the right topic. For instance, If you wanted Excel help but didn’t know what version of Excel you were in, you might not get past the first hurdle. Now, typing in a plain-language question will guide you straight to an appropriate learning module, providing version number checks if necessary along the way.

THE FUTURE

What excites Jon Buttriss most, however, is Deskartes’ ability to collect information that will help NETg improve its courses. “From now on we can accumulate virtually unlimited data about how people like to take training and how they learn. We can analyse trends to make sure that we build the right Learning Objects to satisfy people’s real needs.” Not only can the objects be fine-tuned to people’s preferences but they can also start to have more intelligence built in to them. “We’ll be able to say ‘you’re doing this course much more quickly than average, is it too low a level for you?’Then we can direct them elsewhere, for example to a mentor or video.

We’ll effectively be able to start taking over some of the work of ‘learning management systems’.”NETg’s competitors have not been slow to come up with knowledge management solutions of their own. But training industry analysts have rated NETg’ s solution more highly than most.”NETg has delivered a sophisticated learning object framework that supports innovative e-Learning products and services,” says Brandon Hall, e-Learning expert and president of Brandon-Hall.com.

“The NETg Search & Select system leverages the core values of that framework to meet critical business issues surrounding employee performance.””The combination of Learning Objects and Deskartes gives us a lead which is well-nigh unassailable.” commented Jon Buttriss.