(Required)

DLE was created to close the gap between formal systems (such as learning management systems and performance support systems) and informal information gathering tools, such as search engines. It serves knowledge workers (such as technical professionals at IBM) who need to continuously learn in the context of their job.

In the basic Dynamic Assembly mode, users input a query (e.g., “authentication”), as well as some preferences – how much time they have and whether they would like an overview or an in-depth course. They can optionally specify other preferences, such as level of difficulty desired.

Taking into account the information and preferences provided by the user, the system automatically assembles a course – a set of learning modules organized in a logical progression.

Users can begin learning immediately, using the system course player, or replay courses later, suspend and resume at any point. They can also edit the course, by removing or re-arranging modules. They can bookmark courses and share them with others through e-mail notification.

In the Manual Assembly mode, users have more control over the course assembly. In response to the users’ query and preferences, the system returns a list of relevant learning modules. Users manually select modules of interest. The system recommends additional modules to complement the users’ selection. When the manual selection is done, the system arranges the modules into a coherent learning progression.

The assembled course is a sequence of learning objects, each containing short, modular Web content, such as a few slides of a presentation or a section from a reference book with associated assistance in the form of learning objectives and questions.

DLE uses a patented method for arranging the learning modules in a logical progression, based on a selection of content organization principles. The current implementation uses two principles: a topical ordering -- lessons are grouped by topics arranged in a logical learning progression – and a rhetorical or instructional ordering principle -- modules that are about motivation come first, followed by concept definitions, and then more detailed content, such as implementation procedures, or code samples, for IT-related content.

To enable DLE, learning objects are tagged with the following metadata: title, description of content, topic, intended use, duration, and difficulty level. Topics are related to one another through a topic graph that provides many possible dynamic learning progressions through the content.

DLE can generate in-depth or overview courses based on the topic graph. In-depth courses cover a single topic. Overview courses cover a set of topics connected along a path in the topic graph.

A DLE demo can be accessed at https://antemural.watson.ibm.com/eldev/wps/myportal/. Log in as “tester1” through “tester6” with password “dletester”.

Note that due to the firewall, some links in the learning module may not be accessible from the Internet.

The DLE tutorial is available at https://antemural.watson.ibm.com/repos/docs/DLE-Portlet-tutorial.doc.

(Required)

This entry is (check one):    In production (being used today in a live work setting)         In a formative stage (prototype, proof-of-concept, introduced a sample of its intended users)                   

(Required)

Busy professionals often cannot attend formal courses, even when offered online, because of time constraints. Also "one-size-fits-all" courses are not appropriate for serving the needs of many.

DLE allows for just-in-time learning, in the context of daily tasks, offering learners a high degree of control over the course contents and options to customize the content to fit individual preferences.

Users may need access to content before it is found in classroom or online courses designed by professional instructors. DLE provides a fast way to assemble reference material authored by subject matter experts into custom courses.

2. Establishes, or aids in establishing, goals.

Experts or team leads use DLE to help guide novices in establishing goals and focusing on them. In some scenarios we have observed, experts assemble courses on new or complex topics, and then send them to more junior colleagues to guide their learning.

3. Minimizes terminology translation or interpretation.

4. Provides access to supporting and learning resources.

5.  Focuses on task(s), processes, and the natural flow of work.

6.  Reduces or eliminates the need for training/learning.

7.  Supports performance FIRST, and learning only as a secondary consequence of doing.

The ability to suspend and resume the learning process also provides a mechanism for more flexible learning. Users can easily access their personalized courses and pick up where they left off.

8.  Is innovative with respect to supporting performance.

DLE provides a simple but effective interactive approach to organizing information for learning. It puts the responsibility for finding relevant material on the learner, but moves the responsibility for organizing this material to the system.

Our evaluations of DLE suggest that the system could be useful for improving performance on a wide range of tasks requiring time-constrained comprehension, such as those typically faced by technical professionals when studying for a comprehensive certification or preparing a detailed technical presentation. Users noted that many job groups could benefit from this approach, such as consultants, who often need to find out about a wide variety of topics quickly, or sales people, who need to find out about competitive offerings.

The results also indicate that the system would be useful for project teams. Many users specifically liked the custom course sharing feature, as a way of having a project team communicate and develop shared understanding, especially since teams are often global and team members are remote from each other. Finally, other users suggested that custom courses would be quite useful when someone changes jobs and has to come up to speed quickly on a new product or product family.

Users also reported that they were often using the system in a “multi-tasking” or “interrupt-rich” environment and that the system was responsive to this situation.

DLE is quite flexible as an e-learning delivery system. As the business changes, new learning objects can be seamlessly added to the existing repository. Because of the small level of granularity, only a few learning objects may need to be updated. Learning object repositories can be organized around domain areas with associated topic graphs, reducing duplication and tearing down the traditional boundaries between on-line help, performance support, and training.

(Required)

While shorter 30 minute to 4 hour courses are sometimes available, courses usually last three to five days.

(Required)

As a result of implementing DLE in several internal pilots, we were able to expand to three groups of additional users in a production mode: 400 IBM Software Solutions consultants for the WebSphere product line, 130 first-tier business partners world-wide, and 1000 IBM Global Services consultants.

(Required)

The results of this limited experiment were very positive. Users of similar experience levels performed significantly better after using DLE compared with users of the search engine. Based on measured behavioral differences, users of DLE were more focused on learning while the search engine group was more focused on finding quick answers. The results suggest that actively engaging users in assembling their own learning content provides an effective personalized structure for users to integrate their knowledge for better performance.

In addition to the controlled experiment, we conducted in-depth one-hour interviews with about 20 users and continue to collect user feedback through the feedback mechanism implemented in DLE. Users reported highly favorable experiences with the system. Users reported that the system helped them quickly come up to speed on their selected topic of interest. Without exception, all of the users thought the concept behind the system was sound.

Users universally expressed their desire to take responsibility for their own learning and apportion their time to topics according to their needs. The system provided the necessary support for rapid knowledge acquisition by being available anytime and anywhere thus representing a true learning-on-demand environment.

Participants found the system “user friendly” and easy to use. Not one user reported having difficulty figuring out how to use the system’s basic features, such as searching for and displaying modules, and assembling modules into custom courses. They also found the system to be fast and very flexible. They were able to learn about information related to their task at hand or a special topic of interest according to the role they were playing in a customer engagement. Each person’s precise needs seemed to be unique, but the system adapted well to their highly variable learning requirements.

Users saw the system as a single integration point for diverse content. They liked the ability to assemble together small but meaningful “chunks” of information. They felt that key information would normally be hidden in the middle of a long e-learning course or huge technical manual, so the ability to easily access and integrate information from the middle of these sources was a tremendous benefit.

Users were very excited to find that they could limit the selection of modules by their available time. Most of those interviewed consider their time and attention to be scarce resources. Users were delighted with the ability to reorder modules in a personally-meaningful sequence.