In January, 1997, the Multimedia Centre for Learning in the Humanities at the University of Toronto welcomed its first groups of instructors and students. A joint project of the Faculty of Arts and Science and the language departments, the Centre is the first network-based language learning centre established at our university, and, indeed, one of the first of its kind in North America set up expressly for language teaching and learning.
The Centre is a hybrid facility, consisting of both IBM (Pentium 166) and Macintosh-type (Power PC) computers. In all, there are twenty-four computers, twelve of each type, as well as an instructor's workstation.  The instructor's workstation, which is located at the front of the room, is connected to a VGA projector, so that the instructors can, at their discretion, lead the lesson from a central point. Each of the computers is connected to the on-site Indigo server, which is reserved exclusively for the online material of the language departments.
However, the multimedia network-based centre is really only the most recent phase in the ongoing use of technology for language teaching and learning. To appreciate the significance of this important development, it is necessary to understand it from an historical perspective. To this end, the first sections of this article present an overview of the dominant trends and tendencies in the use of technology in second language instruction in North American post-secondary institutions.
Language departments have been using language labs (LL) since the late fifties. In fact, even before the invention of tape recorders, some pioneering language instructors used phonographs to teach articulation and intonation. A reaction against unsuccessful writing-based models of language instruction, the use of audio-technology and pattern drills emerged as the fruit of what experts in the field baptised the Audio-lingual Revolution.
Then, in the eighties, due to the availability of personal micro-computers and their increasing affordability, some language departments established separate computer labs, whereas others added computerized workstations to existing LL. In addition to the development of smaller, more powerful, less expensive machines on the hardware side, more intuitive user interfaces converted many technophobes into humanities computer-users. In step with these important changes, the eighties became the golden age of Computer-Assisted Language Learning (CALL). The sum of this series of events is known as the Micro-computer Revolution.
Since the late eighties, with the rise of network technology and the Internet, another revolution has emerged as a formidable social force: the Information Revolution (de Rosnay 1996). With the birth of the World Wide Web in 1992 and the explosion of new private Internet Service Providers, the Web has quickly become a household term. Likewise, with the increasing emphasis on "authentic documents" in language teaching, language instructors recognized the power and importance of the Web from the outset. Furthermore, the term documents covers not only text, but also sound, image and video documents -- in short, multimedia documents. Meanwhile, micro-computers have evolved into powerful (yet affordable) multimedia workhorses.
On the negative side, revolutions almost inevitably have their victims, and technological revolutions are no different in this respect. The Micro-computer Revolution and the Audio-lingual Revolution, for example, are quite strictly incompatible in that the Micro-computer Revolution of the late seventies and eighties emphasized writing over pronunciation, due to the fact that sound cards were not standard issue for most PCs.  Whereas the essentially method-driven Audio-lingual Revolution was able to adapt to different audio technologies, the Micro-computer Revolution clearly made method the slave to technology. In many instances, CALL replaced the methods and philosophies of the LL. In short, even though similar drills were used, they became more visual and writing-based since monitors and keyboards replaced sounds and microphones as output and input devices. The Multimedia Network-Based Language Learning Centre (MNLLC),  I argue, is neither the slave to technology nor method. Instead, due to the sophistication and flexibility of the new technologies, the MNLLC accommodates previous applications of technology in language and CALL labs, but, at the same time, proposes an entirely new approach to "authentic documents" and resource management.
Despite its rather volatile history, the LL was one of the most large-scale, universal and enduring applications of technology in humanities teaching/learning. The history of the LL began as early as 1906 with the use of phonographs for language learning by the American Military and Naval Academy (Léon 1962). Other early initiatives included the "phonetics lab" at the University of Utah installed by Ralph Waltz in 1924, with the eventual improvements to sound he engineered with G. Oscar Russell for the "practical drill laboratory" at Ohio State University in 1931 (Léon 1962). Waltz's installations were virtually identical to modern LL, which would appear many years later in the late fifties: sounds were sent simultaneously from a central point to various listening stations; listeners could record themselves in order to identify and correct pronunciation problems (Léon 1962). With the invention of the tape recorder in the late thirties and its constant improvement over the years, LL became increasingly common in the United States. In 1949, in a survey of 28 institutions in the North Central United States, 17 were using phonographs, while 9 were using one or more tape recorders. By 1957, there were about 100 LL, 300 by 1958 (Léon 1962) and, by one expert's estimate, 15,000 world-wide by 1970 (Thomas 1988).
Curiously, with all of the changes in technology over the years, the pedagogy of the LL had remained virtually unchanged since Waltz's time. For the most part, labs were used for repetition exercises, answering questions and tests (Léon 1962). Compared to the traditional approaches to second language acquisition -- grammar, translation and vocabulary drills and the so-called "reading approach" -- the work done in LL may have seemed quite exciting. Indeed, language teachers often expressed unbridled enthusiasm for their LL and their seeming magical effectiveness. One such glowing assessment was supplied by Professor Deborah Aish Metford in her article "Happiness Is... A Language Lab" (Metford 1978). Each fifty-minute lesson (thirty-five-minute tape) began with a bell calling the students to their booths where they would listen and respond to a tape played on a central console "manned" by their instructor. The lesson was divided into four sections consisting of: 1) structural listen-repeat exercises; 2) translation from English to French; 3) a song; and 4) listening comprehension passage with questions. According to Metford, the structural listen-repeat exercises reinforced the grammar presented in class the preceding week. This approach, of course, assumed certain powers of memory and/or diligence on the part of the student, which perhaps should not be taken for granted. In the worst case, students merely went through the motions, falling into a pattern of "mindless parroting" (Taggart 1981). Worse yet is the possibility that students simply did not make the connection between sounds and meaning (Calvé 1977; Rebuffot 1981). Accordingly, they could perform rote transformations, but once put in a real-world situation, they were unable to apply their knowledge of French to the communicative situation at hand.  This criticism can be applied equally to all of Metford's exercises. In short, students who performed well in the lab may have only mastered lab-type exercises and not French.
Other criticisms applied to the methods of Metford and others include practical points such as the intrusive presence of the instructor-monitor; intimidating equipment and the dehumanizing environment, such as the ever-imposing gong used to cue students and other monotonous procedures; and, of course, the simple fact that pattern drills (structural exercises) were often too simplistic to hold students' attention (Thomas 1988: 143-5). In a survey of Swiss high school students in the Fall of 1973, similar concerns about the use of LL were identified: 1) fatigue: monotonous tasks; shorter lessons needed; 2) the presence of the teacher: the teacher seems to be part of the machinery and little real help to students; 3) lab work in general: students should be allowed to work at their own speed; repetitive, predictable, monotonous work; 4) lab exercises: one does not learn the system of the language, but rather the structure of the exercise (Weber 1976). Finally, the various responses to the preponderance of criticisms amassed since the mid-sixties resulted in a "new" LL, that is, a new use of LL.
However, before examining the so-called renaissance of the LL, it should be emphasized that most of the problems cited by critics of the LL could have been resolved without having to revolutionize its use. Both Thomas and Weber are clear on this point in their articles and, in fact, provide their readers with a series of useful, practical solutions (Thomas 1988; Weber 1976). Clearly, the problem with LL went far beyond claims of dehumanization and the monotony of dull, tiresome exercises (which nevertheless can and have been improved upon).  These were merely symptoms of a more serious problem, which was the fundamental misunderstanding of technology, its uses and limits and a deep mistrust of its introduction -- read incursion -- into language teaching.
The first casualty of the "new" LL was undoubtedly the Audio-lingual approach. Instructors no longer believed that a second language could be acquired through endless drills in the lab. Accordingly, the role of the LL was reduced to that of a support for work carried out in the traditional classroom setting. Students no longer were required to be glued to their workstations, where they would all listen to the same recording disseminated from the monitor's console. Also, in many cases, graduate students were hired to provide assistance to lab users, which relieved some of the stress associated with having one's instructor lurking about. Instead, cassette tapes were copied and made available for students to work individually and on a drop-in basis in most cases. Students were thus able to perform at their own speed, without the bothersome and stressful litany of gongs and bells associated with the "old" lab. The role of the lab as a testing facility was also given a minor role in the "new" lab, which, again, improved its image in the eyes of students. In theory, this new use provided an almost ideal learning environment for the dedicated self-learner.
Acting on the revived enthusiasm for the LL, many textbook publishers revamped the lab programme included with their texts. Many students exploited this learning opportunity with commendable success. However, over the course of the eighties, most textbook publishers were not very diligent in their updating of material. The material presented in textbooks, despite efforts to update it, often appeared out of touch with society, even two or three years after publication. For example, in one French textbook I have had occasion to use, students were asked to match up certain French figures with their respective accomplishments. The answer for Simone de Beauvoir, to the chagrin of many students and their instructors, was: l'amie de Jean-Paul Sartre. Meanwhile, other instructors still created material around their own sound documents, most often news clips and commercials taped from the radio. The value of such "authentic documents" in language teaching has been well proved (Lister 1976). However, rarely did instructors have the time to prepare constantly new tapes and lessons, which is almost certainly what was necessary. The cost of new resources also became a stubborn problem: where was the money to come from (Thomas 1988)? For all the talk about and enthusiasm for communicative teaching strategies in the eighties and nineties, method could not compensate for the lamentable lack of quality teaching resources, of lab programmes in this case. Consequently, students often reacted quite negatively to lab work, viewing it more as punishment than anything else. Indeed, I am quite certain that for whatever reason or reasons, some instructors merely offloaded what they saw as the burden of technology on their students. Thus, if there is blame to be assessed in the decline of the "new" LL, then it falls squarely upon instructors' and course administrators' indifference or inability to remedy the lab resource problem.
Nonetheless, some LL continued to grow and flourish in the eighties by supplementing their traditional equipment with various interactive and multimedia technologies. As Professor Alain Thomas writes:
The language laboratory slowly transformed itself into a learning centre: a library of media ranging from more traditional film, slide, and audio materials to commercial and off-air videos, and commercial and customized programs in CAI. All these materials could be used for autonomous learners, classroom application, reinforcement, review, and experimentation for more advanced students. (Paramskas & Thomas 1985: 4)
However, one must not exaggerate the importance of these accessory technologies because clearly, it was the micro-computer which was responsible for the most significant technological advances in language teaching/learning in the eighties and early nineties.
The relevance of the computer to language teaching and learning has been acknowledged since the early seventies. Four papers on Computer-Assisted Instruction were presented at the first conference on the role of computers in literary and linguistic research held at Cambridge in 1970 (Hockey 1987). Each of the four papers explained how mainframe computers were used to identify and isolate certain morphological and discursive elements of various texts and lexicons. However, due to the manifestly "unfriendly" mainframe user environment and the intimidating presence of these "Jurassic" computers, only a very few individuals opted for their use and, unfortunately, with very limited success.  There were also marked pedagogical weaknesses in these early experiments:
Early programs were modelled on scientific-oriented CAI [Computer-Assisted Instruction] in which there is only one right answer, and learning proceeded along well-defined and logical pathways. The pedagogical model was programmed learning, which never did attract too great a following in the language fields. (Paramskas & Thomas 1985: 8)
With the arrival of micro-computers (personal computers) in the early eighties, many of the weaknesses, practical and pedagogical, were remedied. More user-friendly programming languages such as BASIC greatly simplified the development of software packages. Yet most language instructors were still not tempted to ascend the learning curve implied by any kind of computer programming. This prompted the development of authoring-systems and templates, the equivalent of "scripting languages" today. Still another alternative was the collaborative approach where language instructors teamed up with computer programmers, an approach which produced relatively sophisticated programmes such as CLEF (Computer-assisted Learning Exercises in French; Paramskas & Thomas 1985). 
These improvements in computing environments as well as a diversification of approaches led to a steady increase in the use of (micro-)computers. A survey of 602 language-teaching institutions in the United States in 1978-79 revealed that only 10% were using computers. A similar survey conducted in Canada with 173 institutions in 1985-86 indicated that 33% were using computers, half of which were IBM-type machines. Another survey of 41 Canadian institutions carried out in 1987 reported that 44% were using computers, 66% in Ontario (Thomas 1988).
Another pan-Canadian study was conducted in 1993 in which 100 respondents from 33 post-secondary language-teaching institutions answered 13 questions about their CALL activities (Bougaïeff 1994). Although readers of the information gathered were left to draw their own conclusions, it would perhaps be useful to identify the trends in CALL suggested.
|Diagnostic testing 5||Authoring-systems 15|
|Self-learning 3||Multimedia 5|
|Tutorials 28||Dictionary 2|
|Drills 25||Writing-aids 1|
|Games 1||Word processing 3|
|Oral 5||Dictionary 1|
|Listening (Aural) 3||Writing 13|
|Grammar 28||Syntax 5|
|Vocabulary 20||Punctuation 2|
|Reading (Comprehension) 12||Spelling 5|
|Dictation 1||Discourse analysis 2|
|Culture 1||Morphology 2|
Indeed, the trends were quite distinct. Nationally, there were over twice as many institutions using IBM-type machines as there were institutions using Macintosh systems.  Most instructors with CALL in their courses used drills, tutorials and some kind of authoring-system. The linguistic content reinforced by these applications included grammar (28%), vocabulary (20%), reading (12%) and writing (13%). In fact, almost all of the skills taught pertained to reading comprehension and writing while oral and aural practice accounted for only 8% of the linguistic content. Most personal computers in our language teaching institutions were not equipped with CD-ROM drives and sound-cards in 1993 when this survey was conducted.  As a result -- and shaped perhaps by earlier experiments using mainframes -- micro-computers, like LL, became tireless "machines à répéter". Yet micro-computers found no methodological allies or alibis as LL once had in the behaviourist models of the Audio-lingual Revolution.
Micro-computers did receive an ideological push though, insofar as their use was completely compatible with the philosophy behind the "new" LL, also of the early eighties:
With computers situated in the language lab, students can practise at convenient times, using the programs for review of classroom learning, for general review, or for moving ahead of the class. (Paramskas & Thomas 1985: 9)
Yet despite the incredible leaps made since the early seventies, compared to the abilities of a real teacher, these programs remained quite unsophisticated. The fact of the matter is that most of the applications were still essentially variations and improvements on the obsessive theme of drills. Accordingly, some students saw computer use as dehumanizing, monotonous and downright boring. Others simply did not get involved with CALL applications because they were not mandatory elements of their course(s). Inadequate support and training was another key problem which may have prevented a more satisfactory integration of CALL in language learning/teaching. In some labs, students eventually decided that the best use of the micros was simply word processing. For these reasons, in various combinations, the computer remained the mysterious "other" of language teaching until recently.
As we approach the millennium, a new technological phenomenon -- a new investment in technology -- is beginning to replace the LL: the MNLLC. In the Multimedia Centre for Learning in the Humanities at the University of Toronto, for example, multimedia computers have completely replaced the traditional LL, which once occupied the same space. Yet far from seeking to forget the past -- that is, earlier CALL projects and the history of the LL -- the MNLLC challenges language teachers to maintain a dialogue with the past as well as the future. Students and teachers are free to engage in such a dialogue because of the considerable flexibility of today's multimedia computers, both pedagogically and technologically speaking. Significant advances in the areas of multimedia, network and authoring technology have dramatically enriched and simplified the operation of language learning centres.
Unlike computers from, say, ten years ago, today's multimedia workstations are equipped with a CD-ROM drive. The importance of the CD-ROM from an historical perspective is that compared to the storage media of the past -- reel-to-reel and video and audio cassette tapes, floppy disks and hard drives -- it delivers astonishing storage capacity (approx. 600 pages of text), shelf-life and extremely fast access time. This is why it provides an excellent medium for multimedia applications, allowing for the efficient transport and combination of images, sound, video and text. In practical terms, CD-ROM drives also provide lightening-fast software installations and upgrades. While CD-ROM drives are fairly inexpensive, the disadvantage of commercially produced CD-ROMs is often their cost. Nonetheless, as is almost always the case with new technologies, the prices are constantly falling. 
The development in multimedia technology which is perhaps most promising for language learning is the Internet Multimedia of the World Wide Web. Generally speaking, Internet multimedia does not require any special hardware in an age when most multimedia systems already include excellent sound and video cards, stereo speakers, earphones and microphones. Compare this situation to the "specialized box" phenomenon of the seventies and eighties where each new audio/video technology required a separate machine: the BETA VCR required a "beta box", VHS a "VHS box", audio CDs a "CD box". Today developments in Internet multimedia are often supplied by software called plug-ins that handle multimedia tasks via popular Internet browsers, such as the Netscape Navigator and Microsoft Internet Explorer. Plug-ins are widely available for download on the World Wide Web and are most often inexpensive or free. They provide sound and video on demand, in either real time (live feed) or deferred time (files must download before they can be played). Not all file formats are cross-platform compatible, but many are. "Aiff" files, for example, are sound files which can be played on both IBM-type PCs and Macintosh systems. And unlike the unhappy past of specialized boxes, once a sound, video or image file is transferred into digital code, it can be reformatted in any number of digital file formats. The universality of digital code is perhaps its most important advantage over analogue signals. As a result, material in analogue format with proven pedagogical value does not have to simply fade away, but can be digitized, recycled and reused.
The hardware which controls Internet multimedia does not take the form of a local storage device. In fact, instead of speaking of hardware, one ought rather to speak in terms of infrastructure, that is, the infrastructure of modern computer networks. This infrastructure, from cable networks to the scores of satellites encircling the globe, defers the cost of file storage and maintenance to the owners of individual computers connected to the network, the Internet in this case. Network costs are shared among the members of an international community of private and public sector network infrastructure builders. Thus, language departments using the MNLLC will benefit from improvements in speed on the Internet without incurring any direct costs.
Of course, the concept of the network is not new to language learning technology, since LL were using networks of tape recorder/players in the fifties, and computerized language labs sometimes were part of a LAN (Local Area Network). For the most part, LANs were used for practical purposes such as maintenance and backup. The LL network was used for exchanging content, but only in a local and limited manner, that is, between teachers and their students. One could not use such networks to "discover" new content, but only to passively receive information selected, prepared and broadcast by the teacher. Alternatively, the network in the MNLLC, thanks to Internet connectivity, functions as a content-rich network, a borderless and boundless source of information. Whereas it is easy to see such advances in network technology as revolutionary, in the case of language learning centres, it is perhaps more properly a case of evolution and improvement.
Meanwhile, the creation of the World Wide Web in 1992 and the Hypertext Markup Language (HTML) have spawned a new generation of authoring technology and authors. Information about this markup language and its corresponding protocol (HTTP: Hypertext Transfer Protocol) is widely available on the World Wide Web, in the many books written on the subject, in newspapers, magazines and even on television. Most of this information is free, and support on and off the Internet is easy to obtain because HTML is a tool for the planet and not exclusively for the relatively narrow purposes of language learning. These are important advantages which HTML possesses over previous authoring systems and templates used to design CALL applications. Again, because HTML is part of a world-wide movement, it evolves quickly, responding to users' needs and goals, and is, for the most part, extremely flexible. And finally, to the relief of all, authoring software packages known as HTML editors/converters are also widely available either as freeware or at a nominal cost. 
However, authoring is not mandatory since teachers may prefer to simply exploit the MNLLC as a means of accessing content-rich networks like the Internet. The Internet has ushered in a new "philosophy" for online resources and resource management. The concerns about the lack and cost of resources which loomed large during the lean nineties have been alleviated to varying degrees by so-called Internet or online resources. For the time being, use of such documents for educational purposes does not imply any costs to the end-user. Costs are rather shifted to access fees -- hardware and account fees -- and away from the purchase of the information per se. This is not to say that authors of Internet material have ceded their ownership rights, but rather that the concept of ownership is evolving. Ownership in the narrow sense only becomes a legal issue when there is copying and re-distribution involved. Hence making hypertext links to documents on the Internet for students to view in the Multimedia Centre does not appear to pose a legal question insofar as copying and re-distribution are not involved.
Resource management has also been simplified by the use of online resources -- even multimedia resources, which do not require the space and expense associated with physical storage and cataloguing. Unlike expensive CD-ROMs and video cassettes, online resources cannot be stolen, or at least not in the conventional sense of this term. Managing resources is no longer about the transformation of a limited, but expensive repertoire of audio-visual resources into museum pieces. Managing resources is all about searching and locating resources on an ongoing basis. Indeed, documents which are truly "authentic documents" exist as such due to the fact that they are spared the typical process of didactisation: discovery, appropriation, revision, duplication and re-distribution. For example, the Internet site of the French television network France 3 features a series of real-time video clips of the evening news in several French cities.  The time teachers save by not having to prepare such "ready-made resources" for classroom use -- not to mention purchase and storage costs -- can, in my opinion, be more gainfully applied to creating a useful repertoire of approaches and exercises to be used in conjunction with online resources.
Today many language teachers author course sites for their students and, sometimes, for public use.  The efforts of language instructors to create virtual space for their courses has resulted in the formation of a new genre. A course site typically consists of three essential elements: 1) Course notes and documentation; 2) Student/teacher interaction (e-mail); 3) List of online resources applicable to the course. Although students may prefer to receive hard-copies of course notes and documentation, the advantages of maintaining an archive of this material are obvious: availability for students who miss classes, easy updates, reuse. Requiring students to access the instructor's e-mail via the course site is an excellent way of reinforcing the importance of this "classroom away from the classroom". E-mail can be used for answering student enquiries and even for submitting assignments. By requiring students to read and write in the target language of the course, instructors have once again acknowledged the well documented place of electronic messaging in second language learning (Lunde 1990; Sugimoto 1993). The list of online resources is clearly preferable to placing a single copy of an article on reserve at the library, which, when available (there's the rub) students simply end up photocopying at their own expense. Online resources usually present a broader range of subject matter and points of view and tend to be more current. Generally speaking, course sites are simple and easy to maintain, thus assuring that they remain as useful to students as they remain workable for the instructor-author.
Some course sites may include another new genre: the sound-bank. A sound-bank is a standard web-page which, instead of providing links to other text or image files, provides links to sound files and, optionally, to transcriptions. This type of site breathes new life -- a second or third life -- into the use of the LL for individual oral/aural practice. Because the audio files are sometimes sourced from copyrighted documentation, the audio files of some sound-banks are password protected. Sound-banks take advantage of the multimedia environment by allowing students to listen to the sound documents with or without viewing their transcriptions. When problems do occasionally crop up, students can send questions and comments directly to their instructor whose virtual presence is useful, but not intimidating (cf. LL).
From a design point of view, sound-banks are clearly non-labour intensive. The only aspect of the process which is somewhat labour-intensive is the digitization of the sound files. However, this is a one-time requirement and, more important, once a file has been created, it can be directly copied to the server. This saves both time and space since the sound files do not need to be copied on to cassettes or the hard-drives of individual workstations. Cutting and pasting allows the designer to edit easily the content of current sound-banks or create others based on a suitable model. The advantages of sound-banks over the static, formalized oral programmes, which were so heavily influenced by the limitations of their paper documentation, could not be more clear.
However, the MNLLC is not limited to the invention of new genres, but is also compatible with more traditional humanities computing applications such as the database. Database developers have been quick to realize the potential of the web.  Instead of supplying data to a limited audience, once a database has been linked to the web via a "gateway interface", users can query the database and retrieve information. In fact, instructors have the power to decide whether access is to be limited by IP address, domain name or whether users require passwords. Users are spared the cost, care and materiality of storage media: access, not ownership. Network-based databases are extremely useful in that there can be as many users as there are terminals. Also, where the instructor is also the author of the database, the relatively light requirements of HTML allow for the integration of changes to the user interface and updating of material on an ongoing basis (cf. local storage media, e.g. CD-ROM).
Pedagogically, the emergence of new applications and approaches in conjunction with the re-emergence and renewal of older applications and approaches is most encouraging. Unlike the LL, which emphasized speaking and listening, and CALL, which emphasized reading and writing, the MNLLC is capable of servicing all of the four basic language skills. Course sites, in addition to their practical value, give students the opportunity to practice their reading and writing skills. Sound-banks allow students to concentrate on their aural/oral skills, and interactive databases sensitize groups or individual learners to the discursive complexities of language. Unlike the rigid conception of the LL as catering to group work and later to individual learners, the MNLLC is used as both an electronic classroom for language teaching and Internet training and as a support for the traditional classroom where autonomous learners drop in on a need basis to review lessons, do extra work, or to move ahead of the class. And while method fuelled and governed the Audio-lingual approach of the LL, and technology appears to have shaped the limits of most CALL projects, the MNLLC has not given itself over to either. Rather, the MNLLC is not driven by a force, but rather by forces; not by History, but by the histories of previous uses of technology in language learning; not by one group of technologists, but by humanities technologists from different backgrounds and generations; not by one school of pedagogy, but by the pedagogies with which instructors feel most comfortable and confident. It is now up to language teachers and learners to explore the remarkable potential which the MNLLC affords.
Once conceived almost exclusively as the mysterious "other" of language teaching, computers have become a familiar and arguably indispensable component of modern language pedagogy. More students and teachers in the humanities are accepting and benefiting from technology each year. One important reason for this is the proliferation of more flexible, user-friendly software and hardware. Moreover, language teachers are no longer inviting their students to take part in experiments with technologies with limited relevance to real world tasks, but are rather teaching languages and basic Internet/computer skills, the same skills which students will require to obtain gainful employment in the years to come. Another equally important reason for the growing popularity of computing in the humanities is the increased emphasis on training and user support.  The recognition by language students and teachers that one need not be a computer expert to be a computer user can only have positive implications for the humanitisation of the technologies.
 The URL for the Multimedia Centre for Learning in the Humanities is: <http://lab.chass.utoronto.ca>. Since the time this article was written, 6 new workstations have been added, and all of the computers have been upgraded to Pentium 200 MMX systems. Hardware information is located at: <http://lab.chass.utoronto.ca/hardware.html>.
 Macintosh systems were, of course, already equipped with sound cards, but as I will point out in Part 3, most Canadian CALL projects use(d) IBM compatibles.
 The University of Toronto's Multimedia Centre for Learning in the Humanities is a specific example of the phenomenon to which I refer in this paper as the Multimedia Network-Based Language Learning Centre (MNLLC). See note 1.
 See Calvé 1977. On a theoretical level, this criticism corresponds to the rejection of the behaviourist model of learning: Stimulus - Response - Reinforcement. In particular, new models at the time, such a Chomsky's generative-transformational grammar, suggested that language is much more than the imitation and generalisation of the rules behind utterances.
 See Calvé 1977. Calvé quite rightly makes the point that some so-called structural exercises are much better than others, and that the outright rejection of this type of exercise constituted a somewhat rash and unthinking response to the problem.
 One exception is the success of PLATO (Program Logic for Automatic Teaching Operations). "PLATO used specialised terminals which had high quality graphics. This was ideally suited to two applications at the University of Minnesota, for teaching Sumerian and Ancient Egyptian" (Hockey 1987: 27).
 "CLEF is an ensemble of 62 lessons covering the basic points of French grammar. It includes a presentation of the grammar point to be drilled (using color, graphics, and movement), three to five execises in several different formats, and a complex error-checking system, which can signal typing errors and different types of grammatical errors, accept alternate answers, and anticipate a great number of typical student errors. Students proceed at their own pace, and can skip ahead or go back, repeating a given segment as often as they wish. They can stop at will and return to any part of the program. They can ask for help in the form of a one-screen (or page) message reminding them of the key point of the exercise; they can also ask for translations of unknown words" (Paramskas & Thomas 1985: 8).
 Virtually all Ontario institutions surveyed were using IBM compatibles. It is also worth noting that IBM compatibles outnumbered Macintosh systems by a ratio of 4:1. Also, PC labs were generally more elaborate, such as the networked 30-position lab at the University of Calgary (Bougaïeff 1994).
 It is also worth mentioning that the recordable CD drive is quickly becoming a very afforable hardware option. Given the tremendous multimedia potential of the CD-ROM, perhaps online activities will soon be supplemented by various off-line activities, that is, CD-ROMs created by individual instructors (cf. high cost of commercial CD-ROMs).
 The freeware version of HTML Assistant Pro was one of the first HTML editors to hit the market. It combines a very straight-forward interface with a series of useful tools, such as the ISO character map, colour background picker and a preview to browser option. If you want to avoid the code all together, Netscape's "What-you-see-is-what-you-get" editor, Netscape Gold, is also available as freeware for educational use. Higher-end editors produced by Microsoft, Corel and others are also very good, but cost anywhere from $100-$250 for personal versions, and as much as $1000 for professional commercial versions.
 The Department of French (University of Toronto) already has fifteen courses with online material. The home page for online material at the Multimedia Centre for Learning in the Humanities is located at: <http://french.chass.utoronto.ca>.
 This is the case of the sound-banks authored and used by the Department of French, University of Toronto. For an example of a simple sound-bank interface, see: <http://french.chass.utoronto.ca/fsl383>.
 For example, continuing improvements to the ARTFL database (Project for American and French Research on the Treasury of the French Language, University of Chicago) have been made possible by web access to the some 2000 French full-texts in the collection. The query page is located at: <http://humanities.uchicago.edu/forms/ARTFL.form.html>. In the Department of French, University of Toronto, Professor T.R. Wooldridge has used TACT, an interactive full-text retrieval application developed in Toronto by John Bradley, to create several online databases. See FREBASE: <http://tactweb.chass.utoronto.ca/french>. For a description of TACT, see McCarty 1989: 111-2.
 Training faculty, students and staff on the use of the new technologies is an important part of my mandate with respect to the Multimedia Centre for Learning in the Humanties. Several of my colleagues and I provide seminars highlighting the pedagogical potential of current multimedia and network technologies. We also have an open-door policy as far as ongoing technical assistance and support are concerned.