Sir,

One of the recent changes in computing has been the development of more user-friendly computer environments. One specific example is the development of HyperCard (discussed in detail in Reference Nielsen,Nielsen (1990)), which provides development tools for interactive packages which are ideal for developing computer-aided learning (CAL) programmes. These have been developed for many subjects (e.g. Geographical information systems: Reference Raper, and Green,Raper and Green, 1992; Art history: Reference Saywell,Saywell, 1992; Computer-aided design, Reference Delisle, and Schwartz,Delisle and Schwartz, 1986; English literature: Reference Friedlander,, Ambron, and Hooper,Friedlander, 1988) but as yet there have been few within the field of glaciology and glacial geology. The latter includes one demonstrated by Morgan in 1992 at the annual conference of the Geological Society of America, and one produced by ourselves called “The analysis of glacial sediments” developed for the teaching of glacial sedimentology to university-level students (Reference Hart, and Martinez,Hart and Martinez, 1992). Footnote * We feel that there is great potential for creative CAL development in glaciology and glacial geomorphology, as it is easy (although time-consuming) to generate and provides a simulating learning expereince for students. In this letter, we briefly outline the stages of production of a CAL program and the students’ reaction to it, using our programme as an illustration.

There are two major proprietary systems for CAL programming; HyperCard, which runs on a Macintosh, and Toolbook (by Asymetrix) which runs on IBM PC compatibles (using Microsoft Windows). Our particular program was developed in HyperCard and runs on a Macintosh SI or above (with 2 megabytes of RAM). It includes colour images and so a system with a colour display, with at least 256 colours, is required. HyperCard (or Toolbook) can be used to make programs which are like interactive books (called stacks), comprising pages (cards) which can contain text, graphics and images.

Our CAL program uses photographs, animation and text to explain different aspects of glacial sedimentology. It illustrates graphical techniques to interpret different types of tills and related glacial sediments, and discusses in detail the different methods of till-fabric analysis. These techniques are taught through an interactive system that is, at every stage throughout the program the student is required to do something, in order to promote active learning instead of passive reading. This system is designed to be used by students at a variety of levels. Complete beginners can work through the stack progressively as there are specific “chapters” accessible through the index page ( Fig. 1). Higher-level users can use the stack as a refresher course, or work through specific sections, e.g. how to plot till-fabric data in three-dimensional projections.

Fig. 1. The INDEX card showing the five sections within the stack, and the “quit”, help (“?”), “map” and “references” buttons.

There are three stages to producing a CAL program: (i) planning, (ii) programming, and (iii) data input. The planning stage is very important as the contents and structure have to be chosen. The style of interactivity also has to be determined and this can determine how much programming will be required.

There are different levels of programming within the stack. HyperCard uses a programming language called HyperTalk (Reference Goodman,Goodman, 1990; Reference Winkler, and Kamins,Winkler and Kamins, 1990). The simplest level of programming (which is easy to learn) is the connection of one card to another, e.g. used when one presses the forward button ( Fig. 2). If the cards are displayed sequentially, they give the impression of animation; a simple example of this is the demonstration of a melting glacier and the resulting sediments. However, the more interesting interactivity requires a higher level of programming, which is more complex to learn but produces such good results that it makes the extra time-input worthwhile. At each stage of programming development, the functions have to be tested extensively to ensure that no sequence of actions can leave a card (or the users) in an unexpected state.

Fig. 2. The first card in the INTRODUCTION section, showing the general lay-out of the cards, with the card name in the top lefthand corner (card introduction 1), the forward and backward buttons and the index button. This is a colour photograph in the stack.

Acquiring the images and graphics required can be a long process. Colour slides were scanned using a Nikon LS3500 slide scanner and Photoshop. After scaling, retouching and cropping, these were compressed from 24 bit colour to 256 selected colours (by the adaptive-diffusion technique) and saved as PICT files. This is necessary, because most colour Macintoshes can only display 256 colours, but it also makes the files one-third the size and hence faster to display. Some monochrome diagrams were scanned with a grey-scale Apple flat-bed scanner, while others were drawn using a graphics program. Although the data input was the least complex of all the operations, this was the task that took the longest time. An estimate of the time involved to construct this stack is: planning 2 h, programming 8 d, data input 4–5 weeks.

The programme contains different types of interactivity to encourage active learning. The lowest level consists simply of text, photographs and diagrams ( Fig. 2), about which a student is subsequently asked a question. The next level is where the users are asked to collect their own data to work on, e.g. users collect till-fabric data from a schematic till section ( Fig. 3). The highest level is where the users have to work through various stages of plotting different till-fabric analysis diagrams.

Fig. 3. This is a schematic till section. The user clicks on the clasts in order to “collect” the till-fabric data.

Throughout the stack, the user is questioned in order to test they have understood the subject. The majority of the questions are multiple choice; this is because glacial sedimentology is such a jargon-ridden subject where each term has many equivalents and near equivalents (including out-dated terminology, etc.). This also makes it much easier to programme, as variable answers would be obtained if the student typed their own (possibly unpredictable) answers. However, this does mean that the questions have to be relatively simple and static. The programme does not allow you to move on until you have answered a question correctly. This is because the aim is not to be “correct” but to make sure the user understands the subject.

This programme was designed for second-year (U.K.) university students. The students found the programme took about 1.5–2 h to complete. They enjoyed using the programme, because it was so different to other teaching/ learning experiences they had had. However, some students seemed more interested in getting the answers “correct” rather than learning about the subject matter. Another group of users, in particular postgraduate students and other researchers, used particular parts of the stack, e.g. facies logs or plotting the equal-area projections, which they found very useful.

Many CAL packages allow the user to “browse” around the subject matter, making their own connections; however, in our package, movement is more restrictive due to its tutorial nature. Users can move around within each “chapter” (as in Fig. 1) but do not have completely free connections, as it was planned that the students approach each new subject in a logical order. However, there is scope, using CAL to make these less-structured packages, and potential for CAL in other areas of glaciology and glacial geology, e.g. ice-sheet modelling, glaciotectonic deformation and advanced glacial sedimentology. These programmes are also very easy to change and so they can both be kept up-to-date but also be built up slowly (as time allows!), and still function at every development stage.

We suggest that there is a great potential for the development of CAL packages for glaciology and glacial geomorphology, because students learn more than in a normal lecture or practical as they work through the exercises in their own time, and through an interesting medium. It also brings students closer to a fragile and inaccessible environment of which some may not have had personal experience. However, of course, there are some disadvantages, which include the somewhat static nature of the questions and, like any book, the programme is only as good as its author. These programmes are also time-consuming to produce and require expensive equipment to run on. However, since these programmes are self-explanatory, students can work on their own and therefore do not require one machine per student during a set practical session.

This is a new way of teaching the analysis of glacial sediments. The interactive nature of computer-aided learning means that it is much more dynamic than a book, and text; images (colour and monochrome) and animation can be combined together to produce an interesting and creative learning environment. Although CAL programmes can never be a replacement for examining glaciers and glacial sediments in the field, they can be a useful laboratory or desk-top tool which will hopefully encourage students to develop a further interest in studying glaciology and glacial sedimentary processes.