Ý

In this paper we discuss our use of the web for teaching and research during the past few years. Although each of us is from a different discipline (physics, philosophy, computer science) we have collaborated on many projects during that time. Our collective web experience began in 1994 when one of us set up the first web server on the Indiana University Southeast campus. Before the end of the spring of that year the three of us had our own servers running from desktop computers in our offices. During the summer we ran a workshop which introduced the web to over one hundred faculty and staff. By the end of the year we had set up a main web server for the campus.

Web use has increased on our campus during the years since that time, but even today very few faculty make serious professional use of the medium in authoring web pages. Our recent investigation of faculty web use indicates that about forty percent of the faculty on our campus serve personal web pages, but only about twenty-five percent of us use the web in teaching, research, or service. Moreover, only about sixteen percent of us serve web pages with any serious educational content; and most of the faculty comprising that sixteen percent appear to be using the web primarily for course administration -- as a repository for text materials normally distributed to students on paper (syllabi, assignment lists, study guides, etc.). We have not investigated web use among faculty elsewhere, but we suspect that our faculty is typical of others in this respect. One indicator of this is the fact that in most disciplines there is real difficulty in gaining professional recognition for web-based authorship. Web pages themselves are not commonly peer-reviewed, and are of little value for promotion and tenure decisions unless subsequently published in more traditional media. If most faculty do not see the web as a serious forum for academic work in their fields, that has important consequences -- particularly for web-based education schemes heavily dependent on the web for content delivery. For until the web is viewed by faculty as a viable medium, one cannot expect that they will make much use of it.

It is hardly surprising that faculty who are interested in maintaining and increasing academic quality in teaching and research in their disciplines see the web in this way. Before they can be expected to take web authorship more seriously it will have to be shown that there are things one can do by using the web which cannot be done, or cannot be done as well, in traditional media such as print and in class discussions and lectures.

We think that a few of the projects we have worked on since 1994 show that there are indeed such things. Web pages combine world-wide ease of accessibility with a great potential for client-initiated interactivity, and some of our projects utilize these features to do some teaching and research which would be difficult or impossible to do by other means. We start by listing some things we have used the web to do which offer relatively small benefits to our academic work, then we go on to discuss more ambitious projects.

Ý
Ý

Web use began for us, as it seems to begin for most, with the posting of text material such as syllabi, handouts, old tests, and coded grades. Eventually materials for most of our courses, ranging from introductory to senior-level, were placed on the web. Posting such materials is a help for students who tend to loose that kind of material in hard copy, and it is easily done: modifying text documents to be web documents can be done nowadays on most word processors just by saving them in HTML format. Syllabi and assignments posted on the web can be kept up to date as the semester progresses -- if you miss a day due to snow you can fix the assignment due dates and students may get this information before the next class. Of course this constant updating requires more work, but not much of it.

We have also posted supplementary textual materials such as notes and study guides and links to sites offering more of the same. Posted notes and study guides can be quite helpful with thoughtful use of hypertext. For example if in week three students are reading your notes on energy and see the word "Joule" but have forgotten what it means they can click on it and jump back to the previous day's lecture notes where it was defined. This obviously can require lots of work, particularly if you need to use special symbols (such as math symbols) which hardly exist yet in HTML or if you need diagrams or charts. Also, getting students to study such supplementary textual materials on the web is not less difficult than getting them to study them in traditional media. And if one makes web material on other sites required reading one must keep checking to make sure it's there -- even very good sites disappear surprisingly often.

Two of us have posted grades on a web site -- with secret codes, of course, to protect privacy. In the classes where this is done there are weekly quizzes and/or homework and lab reports. The grades are cut out of a spread sheet and pasted onto a web page once a week. Students can see how they are doing and compare their performance with other students in the class. This appears to offer some motivation for keeping up in the course and helps everyone keep track of what has or has not been turned in and recorded.

Also, two of us began to use web pages for presentations during lectures and class discussions. Using such pages is in some ways superior to using overheads: the smooth forward and backward scrolling is superior to flipping to the next overhead, and the ability to drag over and mark an area helps focus attention.

In our experience the main significance of web-based text material was convenience and ubiquity and timeliness of access. These advantages are not to be sneezed at, but they do not constitute major improvements in our professional efforts in teaching or research. None of the changes described above has had a radical effect upon how we teach classes or how students interact with us or with each other. Our teaching was essentially no different from what it had been before -- except that paper handouts had been replaced or supplemented by the web pages.
 

Our more recent work has involved attempts to increase the level of interactivity through the use of form pages using CGIs, JavaScript, Java Applets, interactive video, web based laboratory experiments, and collaborative databases. We began with form pages using simple CGI programs to store textual inputs.

Creating such a form page is more involved than creating HTML documents but still not terribly time consuming or difficult. What has to be done is to create the form page itself (which requires looking up some HTML commands peculiar to forms in a reference) and a small program (called a CGI) which will respond to the form page by placing all the incoming information into a text file on the server or somewhere the instructor can get to it. A knowledgeable web master should be able to supply the CGI to get the form page working in less than 30 minutes. Our experience indicates that the increased interactivity possible through such form pages greatly increases the web's educational potential.

Perhaps one of the simplest examples of this is the use of such form pages as student evaluation tools. This way of providing for student criticism has significant advantages over the more usual paper-and-pencil-at-the-end-of-the-semester procedures. For one thing, it virtually eliminates the amount of class time that needs to be devoted to these evaluations. More importantly, rather than students having to wait until the end of the semester to criticize any aspects of a course, a form page provided for this purpose means that they can provide criticism at any time during the semester. The instructor can then make mid-course corrections or address the problems identified promptly. It is not difficult to set up an evaluation form page which provides anonymity so students feel free to be frank with no fear of recriminations; and it is also easy to provide limited access to the page using a class password so that only students enrolled in the class have access to the page. Two of us have been doing this for two years and have received some very helpful and timely comments and criticisms.

One of us uses a form page as a math review required for students in an introductory course. The students fill out answers to review questions (both multiple choice and fill in the blank) on their own time. Once submitted, the answers appear in a text document on the server where the instructor can grade them. Students may take the review at anytime (and so avoid taking up class time) including before the course has really started so that they know immediately if they have the skills required for the course. The form page includes hyperlinks to review material if students get stuck or are rusty. We think that this method has the potential to be used in a variety of ways. For example, in practice tests or required pre-class quizzes (which would have to be filled out and submitted before a particular class meeting) so that students have to do some reading and preparation before coming to class. One could also extend this method to present a "teaching-diagnostic test" where, when wrong answers are entered, students are sent to remedial pages which are keyed to the various types of wrong answers. There they would be given more examples, hints, and easier questions, and be then sent back to the original test so that the student would eventually get the right answer to the original question. Scores can be based on how deeply students have to go into the stack to understand enough to get the right answer. Scores also can tell the instructor which concepts are causing the most difficulties. The exact questions a student is asked can be set up to be random so that doing the test a second time gives new questions. An advantage of having this at a web site is that the site could eventually create a baseline score profile which could become a diagnostic center for testing differences between various teaching methods. Teachers wishing to try new teaching ideas could verify the outcomes of their students' studies under the new methods.
 

When Java became available it offered possibilities for much more interaction between the user and the web page. One of us took advantage of this in developing a web-based computer simulator designed to give beginning students in a computer architecture course a tool with which to control and visually follow the internal operation of a computer at the fundamental level of executing a program.

There were several pedagogical purposes which the simulator served: (1) to provide a concrete visual example that demonstrates a concept, (2) to provide students some range of control so that they can develop intuition about how the concept works, and (3) to give students a means of repeated experimentation to test their understanding of the concept. Computer aided instruction was originally intended to accomplish similar goals and a large number of hardware and software systems have been developed and discarded, such as the Plato system of the seventies that ultimately proved too expensive for widespread use due to requirements for special terminals, communications, and large, central computers. The web offers capabilities comparable to those of such specialized educational systems but on relatively inexpensive, widely available equipment and communications. In addition it provides access to an enormous amount of educational resources, and is adaptable to new technologies.

The topic which the simulator was designed to treat was traditionally handled in the standard lecture format by explaining the basic computer operation, giving an example of a simple computer executing a few instructions, and making a homework assignment. The text provided a good explanation and several completed examples. But each example spanned several pages, consisted of very detailed diagrams, and was, not surprisingly, hard to follow. Program execution is a simple concept, but due to the large number of details required to track during an instruction execution, it is difficult and tedious for a student to have much success in following the process by hand that is normally handled by the computer itself. Even those who are inclined to make the attempt often find the experiment discouraging due to the low probability of success.

The simulated computer was intended to supplement the text and lecture discussion. The simulator gives students a concrete, visual example of the simple computer's operation and an experimental environment in which they can test their understanding of both the concepts and details of program execution. Though the simulator is designed to follow the text closely, it has several important advantages over any static medium. The text is limited to several worked examples while the simulator can work through any program the student cares to supply. Another limitation of the static text presentation as compared to the simulation is the lack of movement as a visual cue. To a reader, following the text presentation is much like attempting to watch a cartoon by moving the eye from one still frame to the next. By comparison, the simulation appears movie-like, allowing the eye to detect any visual changes as cues to important events. To carry the movie metaphor slightly further, the student can set initial conditions that include entering a computer program and stepping through the computer operation to observe how things work, much like playing a movie in slow-motion to witness the fine details. And, as with a movie, the process can be precisely repeated, as many times as the student finds useful. The simulator was demonstrated once in class via a computer projector and used to explain the operation of the simple computer, students asked questions, and a homework assignment was made that included several questions that could be answered with or without the simulator.

At this point, since similar computer simulators have been available for years, nothing new or innovative had been done. But the web added one important element -- convenience. It meant that including the simulator on the homework assignment created no additional effort for students to use it: any student who accessed the web page immediately had both the assignment and tools needed to do the assignment.

The simulator was implemented as a Java applet, which allows it to be started on a web page and run directly by a web browser. Java, being a general purpose programming language, can be used to implement almost any application. However, these sorts of web projects can be time consuming: though relatively simple, the simulator program took the author approximately twenty hours to complete. In this particular case, the effort was worthwhile, since there was no good alternative way to implement the simulator. And, though no quantitative measures have been collected for comparison, the advantage of an active simulator over paper and pen seems obvious.

In contrast to simulations, two of us are currently working on a web-based laboratory which will permit students to control real experiments via a web page. Many laboratory experiments require precise control of apparatus and speedy collection and analysis of large amount of data. For this reason, such experiments are often interfaced with computers. Our web-based laboratory extends the interface to the web.

The expected advantages have mainly to do with greater access since students will be able to do the same experiments from home or in the lab at any time of the day. Usually only one piece of laboratory equipment will be required for the entire class: since the equipment can be left connected and the data are collected over a very short interval, there is rarely any waiting by one student for another to finish and move on. Less obvious advantages are that the equipment might not be available because of time restrictions on the student or the instructor, location of the equipment, or expense of duplicating the apparatus. Also, since the computer will collect the data, long running experiments are feasible. Web-based delivery will also offer a means to conveniently package a complete laboratory that includes written guides, data collection, and data analysis tools, whatever the location of the student or the experiment. Finally, because the software to monitor the experiment and collect the data do not change from one experiment to the next (only the student instructions and the experiment do) the development costs can be low.

We currently have two experimental sites set up for the purpose of providing access to real experimental apparatus via the web. The first site has a web server connected to video equipment and a Java applet which allows motion analysis. When a student accesses the site they can activate the capture of live video of a moving physical situation (car moving, pendulum swinging, etc.). Once the video is captured a Java applet allows the student to analyze the motion seen in the clips (for example find the x and y velocity, acceleration etc.). It's well established in the literature that this kind of analysis is a pedagogically sound means of teaching the relationship between motion and motion graphs, but seldom does one have the resources to provide every student the experience of using this equipment. Our setup will allow a single piece of equipment to be accessed by multiple students in a cost effective way.

The second remote lab exercise involves web access of a commercially available computer interfacing device called a ULI (Universal Laboratory Interface). This device is controlled by a computer and allows the computer to control a laboratory instrument and retrieve data from that instrument. One use of the ULI is to take many data points over a very short period of time; for example to analyze a sound bite from a musical instrument which might necessitate taking 15000 data points for a 1 second sample. We have written a control panel in Java to access any ULI attached to any computer connected to the internet. In this way all the data collection and experimental control capabilities of the ULI can be accessed from a web page. Currently we only have a sound collection device connected to the ULI but we plan eventually to allow full experiments where students can start, stop, gather data from and otherwise control scientific apparatus at a remote site via the web.
 

We have also developed three projects to take advantage of the potential for collaboration offered by the web. Each of the three projects explores this potential in very different ways.

Two of us have used form pages for student peer review in a team-taught history and philosophy of science course which required that students write several papers. We required that the papers be submitted electronically, then removed the names from the papers and posted them on a web site with a form page for comments. For each of the papers the other students in the class were required to provide critical remarks. These comments were posted at the end of the paper on the web and the author could then re-write the paper taking the comments into account for a possibly better grade. Our experience suggests that this idea is a good one, but a difficulty arose that needs to be taken into consideration: students didn't always criticize in intellectually responsible ways, and one or two authors' feelings were hurt to the extent that the experience was not productive for them. We plan to try the idea again, but within a framework which will reduce the "flaming." One way to do this might be to grade the criticism as well as the papers criticized.

One of us developed a web-based class project using student collaboration to help them learn a difficult topic. The goal was to develop intuitions about the fairly abstract concepts of concurrency and serialization necessary to understand multithreaded computer languages. A series of graphical models was used to augment classroom discussion. Each model in the series presented a graduated, progressively more complex concurrency problem. Students could control the behavior of each model by changing key parameters, visually observing the difference in the behavior of the model, and reviewing the program code that implemented the behavior.

The format of the class exercises in some ways paralleled that of a workshop. The instructor gave a short (15 minute) lecture over each new concurrency topic, demonstrated the simulation, and discussed the program code. Students were paired to run the simulations, review the simulation code, and work on a series of questions. The problems and answers were then discussed. Since this was a class format new to the students, they had to be encouraged to work together.

The results were encouraging. Many more questions were asked than in previous class sessions, and students that had never asked or answered a question did so in this exercise. Perhaps this was due to greater confidence in a collaborative answer. Covering a set of related but progressively more complex exercises one at a time in class, having the students manipulate the simulations, and immediately applying the concepts seemed to produce positive results in that the students could correctly answer most questions at the end of each exercise. Moreover, about the same amount of material was covered as in a pure lecture format even though less time was available for lecturing. The class project's use of the web was at least very helpful and possibly critical in delivering the material to the students complete with models and questions combined. Also, because of the web format, the same work covered in class was accessible to students outside of class to experiment further or review for a test. One drawback to the class project was the large amount of preparation time -- approximately 25 hours to develop the Java applets and other materials for a single class session.

Since the Spring of 1995 one of us and a student of his have been serving a collaborative bibliography of works by women philosophers in a searchable database. The web site from which the bibliography is served takes advantage of possibilities inherent in the web to do some sorts of academic work more effectively than can be done in other media. Large parts of the bibliography are produced not by us but by the users themselves -- by appealing to clients at the site and providing easy means of response we have successfully induced many of those who visit the site to collaborate with us by forwarding errors, omissions, additional entries, and other helpful information. In this way we expect that as the bibliography continues to expand a reference work will develop which is larger and more accurate than bibliographies produced in traditional print.

The bibliography now has over one hundred and thirty contributors, several of them well-known in the field, and altogether they have been responsible for thousands of entries and corrections. Checking all such contributions for accuracy before adding them to the database has uncovered very few errors -- almost all submissions have been accurate and helpful. The number of entries in the bibliography has increased at a gratifying rate, and shows no signs of slowing down. At present, it contains over 15,000 entries on over 5,000 authors. According to our analysis of the server logs, the bibliography is searched over one thousand times a month. It has been regularly accessed by students and faculty in over fifty countries outside the United States, and from almost every major university and college in this country.

Also, we have received a large number of responses from philosophy students and faculty at other institutions (here and abroad) thanking us for providing the bibliography and indicating that it has helped them in their work. And we have also received queries from librarians about the search engine and how they might go about setting up similar projects. The American Philosophical Association web page provides a link to the site, as do numerous philosophy departments, links lists, and library web pages maintained at major universities around the world. Last year we received an unsolicited offer from The Philosophy Documentation Center to publish the bibliography in book format. We accepted the offer and the book appeared in print in July of this year. Recently a Spanish publisher has expressed interest in publishing an edition in Spanish, and negotiations are underway. We have also noticed that since the site began to attract attention its lead in utilizing the potential of the web to produce reference works by collaboration with users has been followed by others.

The benefits of this project for the student editor have been manifold: She is a co-editor of a book and a coequal participant in a successful experiment with a faculty member. She has picked up a vast amount of experience in scholarship. She has, through extensive correspondence with other students and faculty, learned how to deal with a wide variety of critical comments (some tactful, some not; some accurate, some not) in productive ways. We intend to set about securing the services of competent assistant editors in (at least) Europe and Asia with the language skills necessary to help her check entries in a wide variety of languages for accuracy. We expect to succeed in this with the help of several of the collaborators who are located in those parts of the world.

The success of the bibliography shows that it is possible to bring the combined expertise of the entire community to bear on creating information about a particular subject. We know of no other medium as suitable for doing this as the web, and the possibilities for similar collaborative projects in other areas are almost endless. A curriculum-related database containing information solicited from colleagues around the world could be composed via an interactive form page. The continually updated database could include such things as information about scientific demonstration devices, collections of instructors favorite test questions, favorite teaching techniques, searchable archives of images, video clips and sound bites related to a particular subject, and the like. Ratings for educational student software, lab equipment, and other teaching-related products could be developed in a similar fashion: input forms could solicit opinions of various items which would be stored in a searchable database. A list of pros and cons for each entry could make the reader aware where opinions are divided. A list of ideas and suggestions for improvements on equipment and software could be also be maintained so vendors could get ideas for improvements and learn what we really need. A textbook information database could be maintained, in which information about various text books could be found and commented upon. Users could make complaints, offer corrections or suggestions and authors and publishers could respond.
 

We have cited the combination of ubiquitous and asynchronous access with a high degree of interactivity as indicating that the web has the potential to be very valuable for academic work. As evidence of this potential we have offered some projects of ours. However, we also argue, the academic value of the web is at this time far more potential than actual. Regarding web use, the short answer to the question raised in the topic of this collection (How Does it Change What Faculty Do?) is, "Not much, yet."

But if we are right about the web's potential, an analogy with the printing press may not be out of place. The printing press made communication of information and ideas much more convenient. At the time of its invention one might have made the mistake of supposing that it merely made the communication of information and ideas more convenient, and that coming up with ideas worth communicating was just as difficult as it had always been. Of course with the advantage of hindsight it is not difficult to see that this way of looking at the importance of the printing press is mistaken. For by making the communication of ideas more convenient, the printing press made it possible for people to use other people's ideas and information more effectively to build upon in science and scholarship. But it took some time to learn the potential of the printing press and to take full advantage of it to produce the improvements in science and scholarship that it supported through enhancing communication. Similarly, if the web is what we think it may be, it will take some time for us to learn to take full advantage of it to do our work better.