Language Learning & Technology External links valid at time of publication!
Vol. 7, No. 3, September 2003, pp. 38-41

Paginated PDF version

Language Learning Online: Towards Best Practice
from the series Language Learning and Language Technology

Uschi Felix (Ed.)

ISBN 90 265 1948 6
EUR 65.0 / US$ 65.00
221 pp.

Swets & Zeitlinger
Lisse, Netherlands

Uschi Felix: Language Learning Online!

Review by Steve Bird, Universiti Brunei Darussalam

The Foreword to this volume provides a clear problem statement for the book as well as computer assisted language learning generally: "Technological innovation is driven by many factors, but not one of them concerns a pedagogical imperative. There is no dialogue between teachers and technologists about what kind of technological innovation learners need: neither side knows how to begin the dialogue" (p. ix). Language Learning Online: Towards Best Practice attempts to begin the dialogue, focussing specifically on language learning materials and programmes accessed via the World Wide Web. Best practice is here loosely defined as "using the most appropriate tools to their best potential to achieve sound pedagogical processes and outcomes" (Felix, chapter 1, p. 8-9). Achieving this requires informed teachers, good research, and dialogue with developers. The book aims mainly to inform teachers about the world of online learning, its current solutions, problems, and potentials.

One repeatedly stated opinion from various contributors to the book, and summarised by Felix in chapter 1, is the strong need to encourage a software development environment where pedagogy drives the technology rather than the other way round. This is not uncommon. Most software development must cope with the problem of technology-dominated design, having to spend much time and money compensating for the fact that software developers are not end users/customers. Historically, software design decisions have too often relied on developers' faulty intuitions and a love of bells and whistles instead of accurate information and guidelines on user preferences and needs. When it comes to language learning software development, this problem is exacerbated because of the complexity involved in language learning, the specialised expertise required in designing any language teaching materials, and a lack of clear theoretical and research-based guidelines for developers to work from. All of this requires in part an expert knowledge of how people learn, which few developers have. Whether most teachers and researchers themselves can tell the developers what constitutes best practice is not entirely clear, though, since language teaching in general remains very much a context-specific trial and error world. But the authors of the present volume generally seem optimistic that there is a best practice towards which we will move so long as technology-aware teachers and language learning researchers are driving development.

The book includes sections on design, tools, and pedagogy. We read in the design section (Hémard, chapter 2) that the field of human computer interaction (HCI) has for a long time assumed that best practice in design must begin with detailed knowledge of the end user. Essentially, the designer must understand the user's preferences and needs, then translate these into system requirements and finally a detailed product specification prior to programming. Hémard describes some of the well established data collection techniques, including questionnaires, focus groups, interviews, discussions, surveys, user walkthroughs, and think-aloud protocols. In think-aloud protocols, for example, users are given interactive tasks to perform and are asked to comment on how they are conceptualising the problem, thus giving the designer clues about preferred navigation through a learning system. The aim is to ensure the final product corresponds to users' preferred mental models, making the product easier to understand and use. For language learning purposes, ease of use is obviously a crucially important factor which, if dealt with correctly, increases the likelihood that the user will stay with the system and actually learn.

One very important methodological problem only briefly touched on in chapter 2 is the limited external validity in the results of this sort of study. There will be a great deal of variation in users' mental models, depending on their expertise with computers and the analogies they tend to make between the current system and other systems they have used (both computer based and non computer based), and even the user's cultural background. Since a great deal of this research is qualitative in nature, the designer learns much about individuals and less about large populations. One solution to this problem is simply to test many subjects, but as Hémard points out, this generates huge amounts of data that are expensive and time consuming to analyse. The result is that testing must be curtailed and the end product works well only for a certain kind of user.

Godwin-Jones (chapter 3) focuses on what he sees as ways of optimising Web course design for language learners. The chapter provides more of a summary of the authoring and design structure options available to teachers than a clear view of how best to optimise the design. There is a short review of some of the basic assumptions underlying the pedagogical side of Web design and a brief discussion of theoretical constructs such as "comprehensible input." However, on the whole the chapter leaves the reader feeling that Web design is currently reliant more on trial and error exploitation of available tools than any optimal approach based on solid evidence that one or another design in fact facilitates learning more than others.

The section on tools provides a well balanced look at the technical possibilities and constraints in online software development (Arneil & Holmes, chapter 4), as well as arguments for how current development could better exploit feedback mechanisms to maximise learning (Bangs, chapter 5). For example, Bangs argues that intrinsic feedback (i.e., as close as possible to the kind of feedback one receives in real life learning/communication environments) is preferable to the typical extrinsic, artificial feedback (e.g., "Well Done!" when correctly answering a question) found in most current software. Bangs provides no evidence for this claim, however, and this leaves the reader wondering what happened to the earlier advocacy of various data collection techniques for determining pedagogically sound design choices. Again, though, one wonders how it would be possible to defend the notion that intrinsic feedback is better than extrinsic feedback for all learners. Such learner preferences will probably be affected by many cultural and variable educational practices as much as basic psychological factors in human learning. It will be hard to make such general claims when we are dealing with a large consumer audience. In this sense it may be a mistake to assume that there is a best practice beyond very general design guidelines, unless one is aiming design at very specific audiences.

Shield (chapter 6) and Svensson (chapter 7) describe the language learning potential of the multi user domain -- object oriented (MOO; chapter 6), a kind of virtual world, and virtual worlds (chapter 7) in general. MOOs are traditionally associated with online games (e.g., Dungeons and Dragons), but Shield argues that they hold great promise because of their sophistication relative to chat rooms. Essentially a MOO is a public database in which users can not only chat synchronously with one another, they can also construct their own "rooms" (i.e., simulated three-dimensional spaces) and other objects such as Web pages. The environment therefore allows collaborative and constructive learning between individuals and groups who are geographically distant. Svensson (chapter 7) points out that one of the main attractions of virtual worlds generally is that they open up the classroom to a much wider world. Svensson argues that use of virtual worlds stimulates creativity and motivation, allows students to construct knowledge, and encourages better understanding of state of the art media. Both chapters provide interesting overviews of some recent projects and realistically assess the practical limitations (e.g., technical breakdowns) involved in setting up a virtual world project. One is inclined, however, to wonder whether collaborative learning within the classroom might still not be far more accessible and valuable for language learners since the language of the virtual world is mainly text-based. Students still need perhaps more than anything to be working on spoken language with people who are literally rather than virtually in the same room, and they of course always need to obtain some corrective feedback from an expert, such as a teacher. Neither chapter provides much more than anecdotal evidence that learners benefit from constructivist learning in the virtual world, and one wonders where the usability testing and learning outcome testing came in during these projects. Moreover, there is little in the way of specific design guidelines for virtual worlds; the chapters are more of a survey of available technical options and some speculation about what we might try to do next with them for the benefit of learners. Still, virtual worlds could well be useful for the reasons Svensson outlines. There are many opportunities for learning in the virtual world, and the expansion beyond the confines of the classroom to other worlds and other learners shows great promise, assuming teachers are able to come to grips with the technology and incorporate it into lessons, and the learners are motivated to try and work within the target language rather than simply play around in an unsupervised environment. In the end, as with all language learning materials, it comes down to who the learner is and how hard s/he is willing to work.

Felix (chapter 8) begins the section on pedagogy with a standard question in SLA theory literature: "what are the principles of good practice in teaching?" (p. 148). Felix seems optimistic that at least some of the principles are known (citing Chickering & Gamson, 1987, as the exponents of some of the correct principles), and these include fairly common examples of good practice found elsewhere in the SLA literature -- such things as authentic information gaps and the need for meaningful communication. Felix sees these as desirable and achievable in the online domain. At the same time, though, Felix points out that there are many different learning theories driving much online learning design. Essentially, best practice seems still to be more of a hope than a reality. Chapter 9 (Hampel & Baber) informatively discusses applications of internet audio-graphic and video conferencing in language learning. Davies (chapter 10) concludes the book by describing various teacher training initiatives (e.g., TALLENT, aimed at educating practitioners so that they can fully exploit what is now available on the Web. Such programmes include not only an introduction to the available learning systems but also instruction in how to evaluate learning systems on the Web.

In sum, Language Learning Online: Towards Best Practice presents an informative, wide-ranging overview of the evolution of language learning online, both the technical state of the art as well as the pressing pedagogical issues. While in many ways the reader is left somewhat unclear as to what best practice really is in online learning, this is almost inevitable since developing teaching materials of any sort is as much an art as it is a science, and artistic creativity in all domains resists notions of best practice. Nevertheless, we read that best practice at least in the form of helpful technical standards is being adopted more and more by the influential companies in software development (chapter 4), and that educational institutions are devoting resources to educating practitioners about the existing and potential power of online learning. It all looks very promising in many ways. One gets the feeling, though, that things remain somewhat chaotic -- that the technical side of things rather than education expertise will for now largely determine design, and that the technology is still at a point where using it and advising on its design will require perhaps more time and effort than many teachers are willing to put in. Things are apparently getting better and we may be moving towards something like best practice, but there is some way to go before online learning really comes into its own.


Steve Bird is a Lecturer in the Department of English and Applied Linguistics, Universiti Brunei Darussalam. He is also founder and chairman of APL Systems Ltd., a language learning software company based in Cambridge, UK.



Chickering, A. W., & Gamson, Z. (1987). Seven principles for good practice in undergraduate education. AAHE Bulletin, 39(7), 3-7.

Home | About LLT | Subscribe | Information for Contributors | Masthead | Archives