Today e-learning is more pervasive, since trainees are no longer confined to a small training room. Here are some tips on setting up a wireless e-learning infrastructure. by Dr Seamus Phan

E-learning is big news these days, especially for human resource directors and training directors. Enterprises today are more concerned with measuring effectiveness of education not through hours and costs, but through the resultant productivity of the employees.

E-learning is not entirely new. Back in the 1980s, I designed an in-house manufacturing-related training program running across a small network, where production operators of one of the largest manufacturing plants in Asia were put through the self-paced training program. Unlike many other similar self-paced computer-based training (CBT) programs, the modules I designed ran on a network, and were completely developed in-house rather than allowing external developers to do the job. The delivery was obviously not as sophisticated as those found today, but trainees found the program engaging nonetheless.

In today's context, e-learning is more pervasive, since trainees are no longer confined to a small training room, but can easily be learning from home or any remote location with sufficient network connectivity. But the concept of wireless e-learning did not materialize until the use of 802.11b wireless LAN network became affordable and easy to manage, circa 2001 onwards.

What works with WLANs
Since 802.11 WLANs are basically wireless Ethernet-equivalent LANs, most technologies would work seamlessly on them. However, there are some considerations when designing a wireless corporate education facility. WLANs are great if you are working on an open field campus with lots of base stations across lots of green. But in Asian environments, it is unlikely you get the luxury of that much open space. More often, companies occupy tight quarters, perhaps across several floors.

Assuming you run an operation across three floors each of 2000 sq ft (total 6000 sq ft), it is wiser to hardwire the backbone of each floor by wire, since it is cheaper than creating extra WLAN grid equipment to link between floors.

Some vendors would have you upgrade to the new 802.11a technologies, which offer much higher transmission speeds than 802.11b, but for the purpose of e-learning, I would suggest that as overkill. Remember that your Internet connectivity would normally be a T1 line running at 1.5 Mbps (or less for many other organizations), and streaming media should be properly authored to work with lower bandwidths that cater to remote access employees as well.

The most bandwidth efficient and compatible data types would be RealMedia (www.real.com), since Real video and audio files can be streamed or downloaded to almost any major client platform, including Mac (only under the Classic OS 9, but not yet the OS X environment), Windows and Unix. Depending on how lossy your compression is, Real files can also be exceptionally small, although you sacrifice visual and aural quality.

If your client environments consist of Mac and Windows only, with OS X clients as well, then Microsoft Windows Media (www.microsoft.com/ windowsmedia) and Apple QuickTime (www.apple.com/quicktime) would be great, since these two formats can be played natively on OS X as well as other Mac and Windows platforms. Microsoft Windows Media and Apple QuickTime can offer similar visual and aural quality at about the same file sizes, assuming you use the Sorenson 3 codec (www.sorenson.com) on QuickTime.

All three streaming formats can either be streamed using a streaming server which often requires additional licensing fees or downloaded. For e-learning environments, it may not be necessary to spend the additional licensing fees for a streaming server, since users can easily download the files without excessive waiting. The caveat is that course and content developers have to put in due diligence in preparing and compressing the data to ensure reasonable quality at the smallest file sizes.

Another essential piece of technology is multi-party conferencing, audio or video-based. iVisit (www.ivisit.com) is one such technology that would work across Mac and Windows platforms (although it is currently still incompatible with Mac OS X). Using multi-party conferencing software, educators and learners can interact using voice and visuals as if they are in the same room. Another piece of technology, ispq Video Chat (www.ispq.com), is perhaps more suitable if you have Mac OS X clients, since it is compatible with Windows, Mac OS Classic and OS X. The likes of ispq Video Chat is not just restricted to video and audio, but also allows textual chat as well.

Many e-learning content providers also use Macromedia Flash (www.macromedia.com), which has its pros and cons. Flash is extremely fast and lightweight, since it uses mostly vector-based graphics. However, it is unwise to use Flash as your navigational system, since some Web browsers, notably those that do not accept Netscape-style plug-ins, will not understand your content. Flash is more useful in e-learning when used for interactive simulations, although you can also use Apple QuickTime for some degrees of interactivity. Use Flash sparingly.

Framing the content
There are several ways to frame the multimedia content, either through HTML, or through SMIL (synchronised multimedia integration language).

The easiest way to frame content is use variants of HTML, such as HTML 4.0, XHTML, or even XML. HTML 4.0 is perhaps the most compatible right now, since it caters to most version 4.0 Web browsers. XHTML is a transitional standard of W3C (World Wide Web Consortium) that seeks to bridge the HTML standard with the eventual XML standard. XML is not quite compatible with many Web browser applications now, and should not be used as the basic denominator until most Web browsers can parse and display XML code effortlessly.

SMIL is also a transitional standard that seeks to allow time-based media to play back in Web environments. The beauty of SMIL is that you can play back a video in one framed window, and time it to synchronize with slides. Therefore, SMIL is exceptionally powerful and elegant in the display of learning content. However, there is a big problem. SMIL is not entirely compatible with most Mac and Unix platforms, and there are playback problems, depending on how SMIL code was authored in the first place.

There is a HTML workaround to SMIL, by chopping the video segments down to discrete segments that should be synchronized with each slide on screen. For example, if slide A is displayed, only the video segment A will play in a binding frame for both elements. When the user clicks on the "next" button to go the slide 2, he or she effectively loads the next HTML page that displays slide B as well as video segment B. This is less elegant than SMIL, but it will at least guarantee compatibility with more users.

Jabber atten-hut!
Jabber (www.jabber.com) is an interesting replacement for instant messaging tools such as AIM (AOL IM), Microsoft MSN Messenger, and other instant messaging tools.

The beauty of Jabber is that besides being able to work via a dedicated Windows-only client, like AIM or MSN Messenger, it can also operate via a Web browser. This creates a pervasive platform for any client from any platform to use Jabber, as long as port 80 is available. Jabber works through XML (similar to how Apple defined the Aqua GUI in OS X), and is therefore open architecture and future-proof. Jabber even offers a 100-user Jabber server (known as Jabber Communications Platform Express, or JCP Express).

IM tools like Jabber are useful for learners who may not want to voice opinions or raise questions in front of an open class. This is especially true in Asia, where the conservative nature prevents many learners from making greater progress in their learning.

With the emergence of GPRS and eventually 3G running on cellular networks, Jabber also provides the possibility of instant messaging on cellular networks. Since GPRS and 3G phones will eventually conform to XML, learning content can also be streamed to these users-assuming that you can provide alternative displays for screens of different sizes. Learners using advanced cellular phones should not see the same display width as one using a desktop 17-inch display.

It is exceptionally important to remember that whether you learn from a remote terminal (broadband connection or analogue), a 100Base-T connected PC on your desk, or a notebook computer running 802.11b, the content you develop should be as lightweight on bandwidth utilization as possible. It is also important to cater to as wide an audience as possible by relying on either open source or widely accepted standards, rather than to adopt cutting-edge technologies without consideration for reach.

Seamus Phan is research director at KnowledgeLabs News Center (www.knowledgelabs.net), an independent technology news bureau and writes for Network Computing-The Asian Edition. He can be reached at seamus@knowledgelabs.net.