Challenges to the Optimal Delivery of
A Training Program via the World Wide Web
By Russ Williams
Last Update: 6/23/98
Note 1: This paper has been accepted for presentation, at the Assocation for the Advancement of Computing in Education's WebNet '98 Conference, after rigorous refereed review.
Note 2: This paper contains information and addresses challenges that will change as World Wide Web (WWW) technologies evolve and mature. Given that the changing nature of these technologies impacts certain points made within this paper, its contents will be periodically updated to reflect the changes, improvements, and advances made in WWW technologies.
The increasing popularity of the use of the World Wide Web in business has brought about a number of very useful applications for companies looking to improve business processes. One of the most promising of these applications is Web Based Training, commonly referred to as WBT. The benefits of Web Based Training are well documented, including (but not limited to):
Despite the increased attention WBT has been receiving from the Training community, and the increasing number of firms offering Web based courses and WBT software, many organizations are reluctant to commit to WBT until more information becomes available.
This paper attempts to address a few of the challenges facing educators, trainers, and others when developing a training program for delivery over the Web. This paper will examine the following three challenges to the optimal delivery of training programs delivered over the Web:
Users of the Web connect to it via a multitude of modems operating at a variety of speeds. According to Georgia Tech’s Graphic Visualization and Usability Center’s (GVU) 8th WWW User Survey most Web users (55%) are connecting at modem speeds 33.6K or less. The study also notes that 21.3% of the respondents were unsure of their connection speeds. Almost half (44.54%) of those respondents who were unsure of their connection speed describe themselves as novice users. Assuming a small percentage of these novice users are connecting at slower speeds, a reasonable case can be made that more than 60% of Web users connect at speeds of 33.6K or less.
The GVU study found that 13.68% of users indicated that they connect at speeds of 128K or greater (T1, T3). Another 10.3% of Internet users connect at speeds ranging from 56K to 128K. As in the above example with slower connections, it is likely that some of the 21.3% of respondents who were unsure of their connection speed were connected at these higher speeds. Using a generous assumption that ½ of the uncertain users are connecting at speeds of 56K or greater, we estimate approximately 33% of users connecting to the Internet at moderate or high speeds.
How users connect to the Web (a visual representation):
Users connecting at slow modem speeds |
(33.6 kbps or less)
|Users connecting at high modem speeds||~33%|
The fact that 60% or more of Internet users connect at slower modem speeds brings into question the wisdom of using certain media and Web technologies in WBT:
Video – The use of video on the Web to supplement a training program is not as much of a challenge as it had proven to be in the past with the improvements being made in video streaming technologies. Leaders in the development of streaming video solutions include Real Network’s Real Video, Microsoft’s Net Show and VDOnet’s VDOLive. The use of non-streaming video clips creates a pause while the trainee waits for the file to download, interrupting the flow of the training. The use of streaming video over the Web is much more desirable.
Java – Java is held in high regard by many in the Internet community. Java’s promise of sharable, re-usable applets written in cross platform code (also known as "write once - run anywhere") rightfully creates enormous excitement with Web developers. Initially the hype was so exciting that many web sites introduced Java interfaces or they spiced up their home page with a Java applet.
That excitement has faded slightly as Java experiences growing pains. Many of the same Web sites leading the charge in the use of Java have since removed their Java interface and/or applets in response to user feedback. It seems the use of Java interfaces and applets on Web sites were causing users long delays while they waited for the Java to load. While Java holds much promise for the future, the current limitations in bandwidth mean it is likely to take a year (at least) or two until these limitations are adequately addressed.
The author is not Java phobic, rather quite the contrary. I see unlimited potential in the technology, however it is future potential that should be integrated into future plans. Using Java today, simply for the sake of using the technology, may be wrong for a training program. It must be used in a way that does not interfere with the goals of whatever training program one is designing. The goal of the training program is what is important; yet sometimes exciting new technologies make that easy to forget.
There are potential solutions to providing increased Internet connectivity speeds under development. The telecommunication companies are trumpeting Asymmetric Digital Subscriber Line (ADSL) and the cable television companies are developing high speed cable modems for accessing the Internet.
Despite the hype surrounding both of these solutions, neither is ready for widespread use, nor does it appear they will be any time soon. The telecommunication companies are hard at work getting ADSL ready, however they need to decrease the cost of both the service and the ADSL modems in order for it to become a widely accessible solution (remember ISDN?). The installation process is also complicated for this type of service. Companies such as NetSpeed, Rockwell, and others are working on products to reduce the complexity of installing ADSL, however more work remains to be done before they will be readily available.
Many of the Cable companies need to upgrade their cable networks to enable two way Internet traffic. Unfortunately this is expensive, and the cable companies do not have all the money in hand that would be required to upgrade quickly, thus slowing down their potential. The modem technology that Cable companies will use is also under development, however there are prototypes available and tests have begun or are ongoing in certain parts of the country. Like ADSL, more work remains to be done, and it is unclear when this technology will be widely available.
These and other potential solutions (utility companies, new satellite systems) are often
discussed, yet none of the services are widely available, nor are they likely to be any
time soon. It is probably a safe bet to say that it will be at least two years until
substantial progress is made in any of these competing technologies.
Couple the fact that users primarily connect at low speeds with information about response times in computer mediated environments (CME or CMC - computer mediated communications) and we see how important it is to understand how a training audience is accessing materials provided via the Web.
Here are the three important limits in response times according to Nielsen (1997):
Source: Jakob Nielsen’s Alertbox – Response Time Overview
At the current connection speeds at which users access the Web, it’s highly unlikely many users are experiencing page response times of 1 second or less. The author has previously worked on a Pentium - 90 desk-top computer, connected at a fast (T-1) connectivity speed, and there was a noticeable delay (from a couple to a few seconds ) when randomly surfing the Web. In order to truly develop a working environment where the pages are returned in less than a second, all users would need extremely high bandwidth. As mentioned earlier, this is not an option that will be available to users any time soon
According to Nielsen, after 10 seconds a user loses concentration and will think of other things aside from the task at hand. During a training program, a distracted trainee is likely to lose his concentration numerous times, if she must continually wait for pages to load. Working within the framework presented by Nielsen, one can see why users rejected the use of Java applets on Web sites. It took so long for the applets to load, the users were becoming increasingly frustrated, and eventually began to look for something else to do (which often meant finding another Web site to visit).
Based on our discussion regarding modem speeds (discussed earlier in the paper), low optimal response times are difficult to achieve. The delivery of training on the Web is probably not going to be effective if trainees are required to wait at various points during the training for pages to load. Delays may cause a loss of focus, and if the trainee’s focus is repeatedly lost, the intended objectives of that training may not be achieved.
It could be argued that users are conditioned to the fact that waiting for information on the WWW is normal. However, to develop a training program while assuming such a user mentality is not the answer. Consideration must be given to the expected response times of Web pages in a training program delivered online. This in turn is impacted by how users connect to the Web. All of which impacts how one should design training programs for delivery on the Web.
Most existing Web based training programs utilize a modular organization of pages with the content divided into short pages of information. This is a good start – but one has to examine the transition which takes place between each of these pages as the user works her way through the training program. In order to ensure effectiveness of the program, page transitions must be optimized in every way possible. In order to achieve optimization of page transitions, one must design for the lowest common user denominator, and that is most likely to be the user’s connection speed.
In designing for the slower user connection speeds, here are a few ideas to improve optimal page transition: 1) If video will be used, use streaming video. 2) Keep the graphic file sizes to a minimum, and be smart about how you use graphics. Don’t add pictures for the sake of adding them. Make sure they serve a specific purpose or get rid of them. 3) The use of Java in web pages should be kept to a minimum. If you are sold on using Java, try to write small applets, as this will help to reduce the amount of time users wait for it to load.
There are a number of companies with high speed computer networks where employees of the firm may access the Web at high speeds. In many companies, the use of an Intranet also provides for excellent user connectivity (since the information is local) which may improve response times. In these cases, the use of bandwidth intensive technologies (Java, Video) may have less of an impact on optimization. However, unless you are designing for a specific audience that connects at high speeds, it is important to understand how response times may impact your training program.
Re-Purposing Training Materials
Re-Purposing Training Materials
It is common that training content is developed in one medium and then reformatted to be delivered over the Web. It may be a computer based training (CBT) program that has been compressed and is available for download over the Web. Or it may be a presentation that has been saved in a format which makes it viewable via a Web browser. Both are examples of "re-purposing" training materials. Re-purposed materials are defined here as training programs which were developed for a particular medium, then reformatted to be accessible via the Web.
Re-purposed training programs are not Web based, rather, they use the Web as an alternative delivery medium. Programs such as these fail to make use of the communication tools available to training via a dynamic medium like the World Wide Web. Unlike most CBT, Web based training may be integrated with discussion software for trainees to interact with one another and an instructor. This ability to communicate through the Web is sometimes overlooked when one talks about interactivity online, yet conversation between trainees and instructors is an integral part of learning via the Web.
Effective and meaningful interactivity can and does take place when people are communicating with one another online. If a trainee is confused she may ask a clarifying question, or maybe add some comments that may clarify something another trainee appears to be confused about. These are powerful additions to any training program – the power of interactivity and the ability to share in the experiences of others. It is this type of interactivity that improves upon the previous models of computer based training, and is an important aspect of Web based training that should not be overlooked.
The two models of communicating over the Web, asynchronous and synchronous, offer a flexible approach as to how this communication will be structured. Often it is beneficial to offer an asynchronous conferencing area which allows the trainees to participate as their schedule allows. However, the synchronous option allows trainees to have scheduled conversations, which sometimes may be a better fit with an instructor’s particular training goals.
Web based training has been receiving more attention lately as companies discover the benefits of using the Web as a delivery vehicle for their training. The attention is rightly deserved – the WWW platform will have a significant impact on the future of training. In order to fully realize the potential of the medium, it is also important to understand its current limitations. The three obstacles to effective Web based delivery of a training program mentioned in this article (User connectivity, Response Times, & Re-Purposed Material), are all challenges with potential solutions. Until these challenges are addressed however, they must be considered when designing a training program for delivery over the Web.
Catalog | Events | Community | Company | Demo
Thank you for visiting The Training Place.
Contact Us. We respect your privacy.
© 1997-1999, The Training Place, Inc.