Recently, the instructional demand for multimedia applications--any application that uses multiple media types such as text, still images, audio, digital video, and generated media such as animation and synthesized music--has positively impacted Computer-Based Instruction (CBI). CD-ROM and the Web are two popularly used media to deliver CBI today. However, the use of specific media always affects the quality and quantity of information presented to the learner. Bandwidth plays a key-role in determining which forms and qualities of media are available to the learner in a CBI program. High bandwidth audio and video are most often delivered on CD-ROM; however, new technologies are enabling the delivery of these media types more efficiently on the Web. The quality of the media delivered is also a very important factor in delivering instruction to the learner. By means of a sample CBI module delivered as CD-ROM based instruction (using Macromedia Authorware as the authoring program) and as Web Based Instruction (using HTML and Java-Script). This case study investigates the strengths and weaknesses of these two formats by comparing and contrasting their abilities to efficiently deliver a gardening module employing various media types. The criteria for contrasting CD-ROM and Web delivery of the different media types includes quality, media bandwidth, and the degree of user-friendliness. The results of the study demonstrate the advantages and disadvantages of both environments, from the media delivery standpoint.
Key Words: Computer Based Instruction, Multimedia, Media, Educational Technology, CD-ROM based Instruction, Web-Based, Bandwidth, Instructional Technology, Multimedia Authoring, Authorware, Media Quality, Streaming
The demand for computer-based instruction (CBI) has recently increased drastically in accordance with the growth of computer-based training, distance education, and instructional technology. Technically, multimedia applications require dynamic handling of multimedia elements (text, still images, audio, video, and generated media such as animations). From an educational standpoint, multimedia usage should be interactive, reinforcing the assumption that interactive multimedia is a technology-based stimulation of a process that takes place between a learner and a collection of subject matters (Reynolds & Iwinski, 1996). Today, multimedia applications in education are delivered in two major formats: CD-ROM based (CDB) and Web based (WB).
In the past, professionals preferred CD-ROM delivery of multimedia instruction. The CD-ROM is designed to contain large amounts of data.. Its storage capacity enables the delivery of multimedia applications that contain a lot of high quality media. The World Wide Web (WWW) has become a major medium for delivery of information in the past few years. While multimedia applications are among the information delivered on the Web, the majority of these applications lack interactive multimedia capabilities. As Gillani and Relan (1997) explain, "Web sites that are multimedia-enhanced generally included small sound and video files that had to be downloaded and individually played" (p. 231). The delivery of media on the Web has always been limited by the bandwidth, the carrying capacity of communication lines or channels (Reynolds & Iwinski, 1996). Development in this field is happening in two directions: faster connections and communication technologies that are altering the capacity of the communication channels and new multimedia technologies for the Web, such as streaming audio and video, shockwave, and others that are allowing for better delivery of media on the Web.
Media delivered on CD-ROM and on the Web raises the following questions: "What are the differences in quality of media delivered on the Web compared to the same instruction delivered on CD-ROM? What are the differences in "user-friendliness" on the Web vs. the CD-ROM?"
I constructed a "Create a Home Garden" to teach novices the basics of gardening. The module was delivered in both environments—a CDB format and WB format. Sample media (text, images, audio, and video) were used in both formats, after making the required conversions. Building the module led to several other discoveries that will be discussed in the conclusions of this paper.
The rest of this discussion will cover the following topics: section 2 presents the CD-ROM module implementation; section 3 describes the Web module implementation, including the media conversion made to the original media; section 4 discusses the comparison results; and the final section covers the conclusions and general suggestions. This paper concludes with my assertion that even with the existence of bandwidth limitations, it is possible to design multimedia integrated WB programs that have advantages over CDB programs, with a higher degree of user friendly and satisfactory quality. An abridged version of the CDB program is available for downloading (for demonstration purposes) at: http://tblr.ed.asu.edu/mhabash/naweb98/cd, and an abridged version of the WB module is available for demonstration at: http://tblr.ed.asu.edu/mhabash/naweb98/web/.
The "Create a Home Garden" module is an interactive hypermedia application that provides a friendly front-end to permit the user to move through the information, based on words or objects of relationships that different pieces of data have with each other. The CD-ROM module consists of about 100 icons, including 7 video and 15 audio icons. Display icons contain text and images. All icons were imported into libraries before being placed on the flow line.
The CD-ROM module was implemented using Macromedia Authorware, an icon-based multimedia authoring system (Fig. 1). Authorware provides the tools for creating high quality educational multimedia pieces that use digital movies, sound, animations, text, and graphics. Graphics were processed using Adobe Photoshop, and videos were digitized using the Media 100 system and Adobe Premier.
Authorware was ranked as one of the best multimedia authoring programs (reviews at http://www.macromedia.com/software/authorware/productinfo/reviews/ main.html), and many CDB multimedia applications are built using Authorware.
Fig. 1: CD-ROM Based Sample Screen
The media used in the program were formatted as follows:
Text: The text in the program is implemented in two ways: pure text and graphics-embedded text. Pure text is a text that is typed and formatted directly into an Authorware icon. In this case, system fonts were used for cross-platform font compatibility. The graphics-embedded text was done using Photoshop and saved as images. This made it possible to use any font available and to use different languages (non-Latin character sets). Obviously, the text that was done directly in Authorware was editable, while the graphics-embedded text was not.
Graphics: The graphics were implemented into the Authorware icons in two different formats: PICT and GIF. Image file sizes were between 120 KB and 250 KB.
Videos: The videos used were digitized and saved as QuickTime movies, with 320 x 240 resolution. Intel's Indeo compression (12 % ratio) was used.
Audio: All audio files were imported into Authorware icons from 16 bit, 11 kHz WAV files. Later, some of the audio files were converted into videos in order to have control over start and end points of the file. This made it possible to have more than one audio in the file that could be saved externally (since it is a video).
Quality of Media: There was no need to reduce the image sizes by performing a lossy compression. The Indeo technology used to compress videos applies multiple "lossy" and "lossless" compression techniques (Andleigh & Thakrar, 1996.)
User-friendliness is an important aspect of any instructional program. Authorware provides built-in functions and variables (i.e., tracking access variables and evaluating data) a for anticipating user responses and user interaction and provided a variety of learning, practice, and feedback. The program had a 640 x 480 screen resolution enabling it to run on machines with screen resolution. When the program was tested on a PC (Pentium 200, with 8 MB of memory), videos didn't load, and there were delays in displaying pages on the screen. The best performance of the program was on a Macintosh G3 machine with 64 MB of memory.
In the "Create a Home Garden" module, the first design issue considered was how user-friendly the module should be, especially in regards to the use of frames for navigational purposes. In their WB instructional model, Gillani & Relan (1997) presents a design for interactive WB instruction based on appropriate instructional theories and the impact of Vygotsky's social cognitive theories on the findings of neuroscience. In this model, the browser window is divided into four distinct viewing frames. Robertson and Solomon present and evaluate a frame-based module where they have looked at the advantages and disadvantages of using frames, and they have conducted a survey that included comparisons to traditional education programs, analyzing the students' attitudes regarding the frame-based module and quality of media. Their results indicated that using frames had many educational advantages and that a balance between bandwidth limitations and image quality was necessary (Robertson & Solomon, 1997).
The "Create a Home Garden" module implemented Gillani & Rellan's (1997) 4-frame model (Fig. 2). I added a control panel bar in the presentation frame that looks similar to the one on the CD-ROM module. This control panel is supposed to eliminate the need for scrolling and give the user the ability to move forward and backward in the presentation more dynamically. The lower-left frame contained the index (navigation bar) for the current unit, and the upper-right frame contained the instructional model (Gillani & Relan, 1997)
Fig. 2 Gillani & Relan (1997) Model, Web-Based Instruction pp. 233
The module had three option of navigation: using instructional model, unit, and presentation control panel. The top two frames take up the first 90 pixels and the lower-right frame takes up 75% of the screen width. The navigation frame to the left takes up the rest, 18% of the screen width (Fig. 3). The navigation bar displays the index for each item. There is a presentation for each item in the index. A JavaScript is used to do rollovers, and cookies are used to keep track of the learner's achievement. The interaction between the learner and the teacher is implemented using CGI (common gateway interface) scripts.
Fig. 3: Web-Based Module sample screen
In WB instruction, there is no place to separate quality issues from bandwidth issues, and as mentioned before by Robertson & Solomon (1997), "there is a need to balance bandwidth limitation with pedagogical needs" (p. 256). In my module, I have tried to attain such a balance by looking for optimized techniques that keep the quality and reduce the high bandwidth demand, as follows:
Text: The text was implemented in the module using the font "Verdana," for cross-platform compatibility. Using a variety of fonts and different languages was possible through embedding text in graphics only. Hypertext links were available inside the text in many places.
Images: All images used were converted into JPEG format (except animated GIFs), with the quality value of 3 and progressive value of 3. The average size decrease in the images was approximately 68% for PICT images, and 50% for GIF images. Since JPEG compression of lossy, there was some loss in the quality of images, however, it was not remarkable, except in some high-color images. It was not possible to have the 640 x 480 background perpetual image, since its size was large. Further, it was necessary to make some images transparent, which is not possible for high-detailed images. Rollover images and some other images were embedded in the first screens of the program with 0 x 0 dimension. This enabled downloading and caching of images before the user was able to use them.
Audio: Audio files were converted from WAV format to real audio format using the Sound Forge program. Conversion was done for streaming these audio clips on a 28.8 Kbps modem with 20 Kbps bandwidth (Lodd & O'Donnell, 1998). There was no problem in playing them later on machines with a 28.8 Kbps modem and with a T1 connection. Sound quality was excellent, and there were no delays.
Video: The initial idea was to convert the QuickTime video into real video, and stream them over the network; however, this was not possible since compression was used previously on these videos. The conversion had to be made from AVI (Video for Windows) format, using the real publisher plug-in for Adobe Premiere, for a 28.8 Kbps modem. The quality of the converted videos was good when tested locally, but when tested on a 28.8 Kbps connection, the video took about 5 minutes to start playing, then it needed 30-40 seconds of buffering every 20 seconds. When played over a T1 connection, videos played smoothly, and the buffering time was less than 7 seconds. When played over a 56 Kbps-modem connection, the videos played with delays, and dark blurs appeared sometimes during playing. The overall quality of videos was not satisfactory.
Many researchers have made comparisons between CDB instruction and WB instruction. The points for comparison included control, language and cultural issues, accessibility, data tracking, time dependencies, data size, refreshment, ability to update, and legal and copyright issues (Hedeberg, Brown, & Arrighi, 1997).
Media delivery issues were among those issues that were researched and written about separately for each of the two media. The two modules I have presented in the previous two sections represent a case study in this field that may lead to useful results. In comparing the two modules, the main comparison points are the bandwidth factor and its effects on the quality of the media delivered, and the amount of user-friendliness each module provided. A look at the media delivered in both CD-ROM and Web modules shows the following:
Text: The authoring system, Authorware, enabled manipulation of the text as an element in the program. Care was necessary while moving the program between Macintosh and PC platforms. Later, fonts available on the machines were used, including fonts that had non-Latin characters. These fonts had to be included in the packaged run-time version of the program. On the other hand, it was necessary to use a generic font for the WB module. The selection of the Verdana font came from the fact that it is available on both Macs and PCs. Using different languages on the Web was not possible without having a plug-in for it. This problem was solved by embedding the text into graphics. Many overseas newspapers do that in order to make them readable around the world.
Images: It was easy to import graphics into the Authorware icons, with no limitations on image sizes, given the authoring system support of all major image formats. In the WB module, graphics conversion and compression was necessary. This process caused some loss in the quality; however, it was minor compared to the two-thirds reduction of the size of images. Image sizes were between 15 and 68 KB, which takes less that 20 seconds to download on a 28.8 Kbps modem connection. Image alignment was easier in the Web module, since it was calculated alignment, while visual alignment was required in Authorware.
Audio: The streaming audio in the WB module had good quality and provided the user with more control to stop, play, and replay audio, since it uses an external Real Audio player. The audio in the CDB module was static, and it was up to the developer to include mute, stop, play controls or not. Having audio files in video formats gave more control to the user and the developer. It also enabled having more than one audio in the same file.
Video: The quality of videos in the CDB module (QuickTime) was better than that in the WB module under a 28.8 Kbps and 56 Kbps connection, the streaming video. This problem is still in consideration and some new technologies are being developed, such as the G2 player, which is available in a beta version. Knowing that a good quality, 320 x 240 video needs about 10 MB of storage per minute, we can see that a CD-ROM does not have large storage capabilities, with 640 MB capacity. A whole CD-ROM can take less than one hour of video. In a situation where we have a 3-hour video, it is better to have a video server with a 9 GB hard drive and to stream the video to the users since they don't need to worry about having enough storage or buying 10 CD packs!
In short, the WB module delivered the media more efficiently than the CDB module most of the time. Further, it provided flexibility to the media, an element missing in the CDB module. The delivery problems faced in the WB model came mainly from external sources, not from the design itself. A quick look at the rate at which modern technology is progressing impels us to believe that most of these problems will be solved soon.
The degree of user-friendliness that the WB module provides is better than that of the CD-ROM. This is due to the fact that frames increase the degree of user-friendliness by presenting information in a more flexible and useful fashion, making documents more aesthetically appealing and facilitating display of different types of information side by side (Robertson & Solomon, 1997). The CDB module does not provide the user with this flexibility automatically; however, the authoring tool provides the developer with all the necessary tools to upgrade its degree of user friendliness.
In both modules, the care that was taken to make the programs more user-friendly played the key role in the development of the product. It is up to the developer to make a user-friendly WB or CDB programs, using tools that are available. What makes the difference in this issue is how much money, time, and effort are spent on the work.
Given the conditions the media was viewed under, the quality of the media delivered in the WB module was good, except for the videos. The WB module had the advantage of giving a higher degree of flexibility to the data delivery. The problems encountered were mainly ascribed to bandwidth limitations. The WB module also proved to be more user-friendly than the CDB module; however, it is important to notice that the efforts exerted by the developer and the amount of money or time spent on the production have important roles in this issue.
Emerging technologies are creating an "in-between" module that has the advantages of both WB and CDB delivery. Currently, some authoring tools that create multimedia-enhanced applications come with Web delivery support (such as compression, shockwave, Java and hypertext support). Other tools enable creating such multimedia-enhanced applications with dynamic content and enhanced flexible media delivery (such as using streaming audio and video). A new version of Macromedia Authorware has Web support, compression for the Web delivery, hypertext support, and an ability to play streaming audios and videos.
Allen, M. (1996), Authorware Academic User’s Guide. New Jersey: Prentice- Hall,Inc.
Andleigh, P., & Thakrar, K. (1996), Multimedia Systems Design. New Jersey: Prentice-Hall, Inc.
Gillani, B., & Relan, A. (1997). Incorporating Interactivity and Multimedia into Web_Based Instruction. In Khan, B. (Ed.), Web-Based Instruction (pp. 239-244). New Jersey: Educational Technology Publications, Inc.
Hedberg, J., & Brown, C. (1997). Interactive Multimedia and Web-Based Learning: Similarities and Differences. In Khan, B. (Ed.), Web-Based Instruction (pp. 47- 58). New Jersey: Educational Technology Publications, Inc.
Jonassen, D., ed (1996), Handbook of Research for Educational Communications and Technology: Project of the Association for Educational Communications and Technology. New York: Simon & Schuster Mcmillan.
Khoshafian, S., & Baker, A. (1996), Multimedia and Imaging Databases. San Francisco: Morgan Kaufman Publishers, Inc.
Lodd, E., & O’Donnell, J. (1998), Using HTML3.2, Java, and CGI. Indianapolis, IN: Que Corporation.
Macromedia Website:
http://www.macromedia.com/software/authorware/productinfo/reviews/main.html
Mason, R. (1995), Using Communications Media in Open and Flexible Learning. London: Kogan Page Limited.
Porter, L. (1997), Creating the Virtual Classroom: Distance Learning with the Internet. New York: John Wiley & Sons, Inc.
Reynolds, A., & Iwinski, T. (1996), Multimedia Training: Developing Technology- Based Systems. New York: McGraw-Hill.
Robertson, J., & Solomon, A. (1997), FrameBased, Image-Oriented Instruction. International Journal of Educational Telecommunications, V. 3 No. 2,3, 237-259.
Acknowledgement: My special thanks to those people for their infinite knowledge: Veena Mahesh, Lutfi Hussein, Tiffany Winman, Dr. Sukhee Kim, and Pambos Vrasidas.
Presenter's name and address
Mohammed Ali Habash
mhabash@asu.edu
Educational Media and Computers Department
College of Education
Arizona State University
Tempe, AZ 85287-0111
Tel. (602) 965-3651
Fax. (602) 965-7193
Copyright: © 1998. The author, Mohammed Ali Habash, assigns to the University of New Brunswick and other educational and non-profit institutions a non-exclusive license to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The author also grants a non-exclusive license to the University of New Brunswick to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers, and for the document to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the authors.