Prentice Hall Javatizes itself

Pop quiz, multiple choice: “Operant conditioning is another term for: a) instrumental conditioning; b) cognitive restructuring; c) observational learning; or d) classical conditioning?” Want a hint? If you were a psychology student using a Prentice Hall textbook, you’d easily be able to test your knowledge of subjects like operant conditioning by accessing the World Wide Web. Yes, hints are included. And you can submit your answers to your professor and time your own performance, all courtesy of a few simple Java applets.

Textbook publishers have been using instructional software — from floppies to CD-ROMs to Internet resources — to augment their printed materials for years, but Simon & Schuster’s Prentice Hall division is establishing a precedent in using Java to bring interactive distance-learning techniques to any student using a suitably enabled Web browser.

“This is our way of extending our printed texts on to the World Wide Web,” said Ray Henderson, director of online services for the Simon & Schuster Higher Education Division, based in Upper Saddle River, NJ. “As far as I can tell, we’ve got a fragile lead in pursuing this technology right now. At least there are no competitors that have announced that they are using Java. So, even though we’re still in the development stage, we’re very excited.”

Prentice Hall (a Viacom subsidiary) is currently implementing Java applets in three of its textbook Web sites: Psychology: An Introduction, by Charles G. Morris; Astronomy Today, by Eric Chaisson and Steve McMillan; and Physical Geography, by Tom McKnight. Each can be accessed from the Prentice Hall online publications guide at https://www.prenhall.com/pubguide. The Morris/Psychology site, for example, enables students to check out a demo study guide that features chapter objectives, homework exercises, a multiple-choice quiz, a matching quiz, a labeling quiz, Web connections for further information, a chat area, and a help utility.

Interactive homework

In the homework exercises, students are asked to write short responses to questions in a text field. An entry from the demo site for Morris/Psychology, for example, requests: “Distinguish between cognitive learning and traditional theories of conditioning. And explain contingency theory.” When all the homework exercises have been completed, students submit their answers to their own course instructors via email on the spot, again using a simple applet. Teacher evaluations, along with coaching suggestions for missed answers, are then posted to students via conventional email.

In the multiple-choice quiz, students click on a blue-box applet that presents a question (which can be regularly changed). They can then click on a hint applet in a red box to get a clue as to the right answer, if needed. After proceeding this way through the complete test, students submit their answers to a scoring applet at the bottom of the page. The scoring applet then displays the number of correct and incorrect answers, as well as a percentile grade and the time elapsed for the test. The matching quiz is similar, but asks students to match pairs of items from two rows.

“Our applets can tell us not only how a student answered a given question, but how long they worked on a question and whether they asked for hints in solving the problem,” said Rens Troost, director of Technology for North American Media Engines (https://www.name.net) of Toronto, the firm hired by Prentice Hall to design their Javatized Web sites. “This gives us tremendous insight into how students interact with the exam. By aggregating this information, we can evaluate whether the instruction and testing are doing the job, and where efforts on improvement should be focused.”

In the labeling quiz, students are presented with a graphic and a list of terms describing parts of the image. A Java applet allows students to drag-and-drop the individual terms to various parts of the graphics. The challenge is to correctly label the object being illustrated. Again, a scoring applet grades the results.

“When we began planning the interactive, Web-based version of the study guide for the Morris book, we knew we could easily incorporate traditional multiple-choice, short answer, and true-false questions,” said Peter Janzow, editor-in-chief of Prentice’s psychology textbook series. “Although Java running behind the scenes does help with response analysis and coaching feedback to student users, the really innovative — and, I think, pretty amazing — elements in the Web site are the drag-and-drop labeling and matching exercises, where Java is the force that enables us to bring the site to life. The result is not just solid traditional exercises, but also some dynamic, truly interactive activities that make learning more stimulating, more meaningful, and much more fun.”

“From a user-interface perspective, the labeling quiz is perhaps the coolest thing we’ve done so far,” said Troost. “Java is great because it allows us to go far beyond the limited user-interface tools of today’s HTML. The labeling applet gives students a natural and intuitive way to interact with the quiz.”

Why Java?

Why did Prentice Hall choose to take the plunge into Java, instead of other Web design platforms?

“When we started last year, we were impressed with the pace of change on the Web, but also aware that we needed to stay ahead of the competition,” said Phil Miller, president of Prentice Hall’s Humanities and Social Sciences Publishing Group. “Despite the fact that many of our prospective users — college students and faculty — didn’t have Java-capable browsers when we began, we were betting that this wouldn’t be an obstacle in the near future. The biggest advantage to committing to the interactivity that a Javatized site affords us is that we can provide a range of learning activities that basic HTML-based pages can’t offer. The resulting Java-enabled Web site is both technically state-of-the-art and instructionally innovative.”

“Our online applications are designed for use with textbooks that have a large circulation, in the tens of thousands in most cases, but occasionally as large as 100,000,” said Henderson. “One of our important design goals was to provide a grading facility for various question types — whether multiple choice, labeling, matching, and so forth. To do so with conventional CGI scripting is possible, but not scalable. When I realized this, we pursued Java development. There was also the realization that, for the low-end animation we hoped to serve over the Web, server push and gif89a were too limited. So offloading the work to the client using Java seemed an ideal choice.”

Caution signs

But Henderson cautions that the road to designing successful Java products is not an easy one. There are some serious pitfalls to look out for, particularly as standards are updated.

“The development process during applet creation, from last November to now, has been marred by shifting standards and the browser wars,” he said. “Trying to identify a stable platform, where the applets would run consistently, proved quite difficult. Creating a stable dwell timer, which records how long a student considers an online quiz, has been most difficult. While the timer is initialized properly, and shows the correct time during the quiz, there have been difficulties with sending accurate times to other applets in real time. The product development process relies heavily on outside reviewers, and we’ve confronted numerous browser configurations where our applets wouldn’t play. As we’ve extended this into the sales demonstration arena, it’s been very difficult as well.

“There have been numerous misperceptions about our online products due to problems with Java. Despite an opening screen that clearly articulates a path for Java and non-Java browsers, there have been many instances where someone has stared at a blank screen long enough to conclude that our product was broken. Internal education on why the product works sometimes and not others has also been a difficult process. And while we’ve created CGI versions of our content, our development time will be significantly reduced when one standard is in place.”

“Our main obstacles have been the language’s immaturity and the lack of widely deployed tools and practices, but that’s what you put up with to be first to market,” added Troost. “Compared to tools like Shockwave, Java gives you a lot more control, as well as a firm base to build from.”

The project’s leaders were understandably enthusiastic about the future of embedding Java applets within their Web-based products. “We’re designing companion Web sites for four 1997 psychology texts, and we see Java as a core teaching and learning component for our most important titles going forward,” said Janzow.

“I’m quite pleased with what we’ve accomplished so far, and the next generation will be even better,” said Troost. “What excites me the most for the near future is the tighter degree of Java integration that the Netscape 3.0 browser will provide.”

“Our future plans call for even more interactivity, and we expect our development to rely increasingly on Java in upcoming products,” added Henderson. “By July we should have several more Javatized sites online, just in time for our fall semester books shipping.”

Asked for his response to those who view products such as Javatized study materials as just slick marketing gimmickry, Henderson was adamant: “We’re deadly serious about this. The response from our customers has been overwhelmingly positive. They want us to push even further. Happily, in the educational marketplace, people really open their eyes when they see this new technology. They can see the eventual possibilities, in terms of interactive distance learning. And we want to make it as effortless as possible for them to achieve such a capability.

“Making the Web appear as software — without taxing bandwidth — is our goal. Java is simply a tool that lets us do a lot more for people with a lot less for them to worry about.”