Technology and language teaching — So far, I have mainly talked about the language lab and the TV, which were a big part of my own language learning. In my teacher training, the technologies I had learnt about and was tested on in the late 1980s were the slide projector, overhead projector, record player, and a few others. It was only when I began to teach at the Manchester Metropolitan University in England that I wanted to use computers in language teaching. I was tasked with teaching a course on Computing for Language Students in the early 1990s. For a short while, this class was the first opportunity that some students had to operate a computer. The first CALL program – the old word for app – I used with my students was TUCO II, created by Heimtraud F. Taylor and Werner Haas at Ohio State University (Taylor, 1979). It had fill-in-the-blank and multiple-choice questions for German grammar learning and came on 3½–inch floppy disks. Some students proudly told me that they had done all grammar exercises that the disk had to offer. For this work, students sat in the Tandberg computer lab and worked on large desktop PCs whose keyboards and monitors were wired together via the teacher’s desk. I controlled them from there by sending mine or a student’s screen to the entire group, monitoring an individual student, and by disabling students’ keyboards and monitors when I needed their attention for the next phase of the lesson. Tandberg (Wikipedia contributors, 2026c) was a Norwegian company that produced the first language laboratories, essentially creating a new language teaching methodology.
Figure 6: ToolBook By SumTotal Systems, Inc. – CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=52493913
To help the students study in their own time and to integrate the different CALL programs into the syllabus of different language courses and to link them with their textbooks, I wrote the CALL Guides (Schulze, 1994, 1995) to help them orient themselves with the digital resources. All this software was text-based and often ran under DOS (Wikipedia Contributors, 2026a) and did not make use of Windows (Wikipedia Contributors, 2026b). The use of multimedia was still in its infancy. When my colleague Chris Jones took ill, I completed the coding for Gertie (Jones & Schulze, 1995) – a multimedia vocabulary learning program – with Toolbook (Figure 6). Chris had conceptualized Gertie’s chapters so they were based on one picture, for example of a bathroom at home. Each of the items in the picture was clickable. For example, if someone clicked the towel or the tub, the German word would be shown and a sound file with the spoken form could be played. After this presentational learning phase, the vocabulary item could be practiced using the same pictures and hyperlinks.
This is an excerpt from an early draft of a book chapter. The book will be about language teacher education and GenAI. I am posting these in smaller (mostly) self-contained posts. The posts are numbered consecutively. After they will have all come out, I will link them with each other.
At Manchester Metropolitan University in the early 1990s, the prescribed textbook was German: A structural approach (Lohnes & Strothmann, 1973), which was bundled with CALIS – Computer-aided Language Instruction System (Humanities Computing Laboratory, 2026) – another authoring tool that employed regular expressions (Computer Science Field Guide, 2025) used for providing individualized and contingent corrective feedback on sentence-based exercises. Regular expressions have well-defined special characters and a syntax that makes them look like mathematical functions, so that they can be used to match sets of character patterns. I must have been one of very few language instructors who loved scripting feedback loops using CALIS and its regular expressions. In a way, this was a step towards natural language processing (NLP), which is a branch of artificial intelligence research. I used different computers with increasing processing speed and storage space for these projects.
Figure 7: Soviet stamp depicting Valentina Tereshkova, the first female cosmonaut
Excursion into space — June 16, 1963: Valentina Tereshkova (Figure 7), already wearing her space suit, was taken to the launch pad in Baikanur by bus. After she peed on the tire of the bus – in Yuri Gagarin’s, the first cosmonaut’s, tradition and as the first woman ever – Valentina orbited earth in Vostok 6 for three days. This happened in the year I was born. Vostok 6 did not have an on-board computer. Its mechanical and electromechanical systems were analog and hardwired. Only in 1980, the crew of the Soviet Soyuz T-2 was the first to have such a digital computer on board. Argon-16 (Figure 8), their board computer, weighed 70kg and had 3 x 2 kilobytes of memory (Штейнберг & Чесноков, 2026). Today, a graphic card that is used for creating large language models, the NVIDIA H100 (Figure 9), has a high-bandwidth memory of 80GB, about 14.3 billion times as much as the Argon-16. Here high bandwidth means about two terabyte per second. Two terabytes are equivalent to about 2 trillion characters, 367 billion six-letter words, or 73 million 5,000-word essays. With an LLM – for an accessible introduction to LLM, see (Wolfram, 2023) – the H100 (Figure 9) can process up to 5,500 tokens per second (Gupta & Kiely, 2024) using the many calculations in an artificial neural network. The exponential growth in the storage space and processing speed of digital computers has been one important factor also in the recent developments of GenAI and was an impactful developmental factor throughout these sixty years.
Figure 8: Argon-16 board computer (https://www.computer-museum.ru/english/argon.16.htm)
Figure 9: PNY NVIDIA H100 Graphic Card –80 GB HBM3
References
Computer Science Field Guide. (2025). Formal languages. 15.4. Regular expressions. Retrieved 2025-01-27 from https://www.csfieldguide.org.nz/en/chapters/formal-languages/regular-expressions/
Gupta, P., & Kiely, P. (2024, 2024–02–06). Unlocking the full power of NVIDIA H100 GPUs for ML inference with TensorRT. https://www.baseten.co/blog/unlocking-the-full-power-of-nvidia-h100-gpus-for-ml-inference-with-tensorrt/
Humanities Computing Laboratory. (2026). WinCALIS. Computer-Assisted Language Instruction System. Retrieved 2026-05-26 from https://www.humancomp.org/wincalis.htm
Jones, C., & Schulze, M. (1995). Gertie. A multi-media vocabulary-learning software for French, German, Italian, and Spanish. Manchester Metropolitan University.
Lohnes, W. F. W., & Strothmann, F. W. (1973). German: A structural approach. Second edition. Norton.
Schulze, M. (1994). Computer-Assisted Language Learning. Students’ CALL Guide. Manchester Metropolitan University.
Schulze, M. (1995). CALL Guide: Grammar. Manchester Metropolitan University.
Taylor, H. F. (1979). DECO/TUCO: Students’ reactions to computer assisted instruction in German. Foreign Language Annals, 12(4), 289–291.
Wikipedia Contributors. (2026a). Disk Operating System (DOS). Retrieved 2026-05-25 from https://en.wikipedia.org/wiki/Disk_operating_system
Wikipedia Contributors. (2026b). Microsoft Windows. Retrieved 2025-05-25 from https://en.wikipedia.org/wiki/Microsoft_Windows
Wikipedia contributors. (2026c). Tandberg. Wikipedia. Retrieved 2026-05-26 from https://en.wikipedia.org/wiki/Tandberg
Wolfram, S. (2023, 2023–02–14). What is ChatGPT doing … and why does it work? Stephen Wolfram Writings. https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work
Штейнберг, В. И., & Чесноков, В. В. (2026). БЦВМ «Аргон-16». Retrieved 2026-05-26 from https://www.computer-museum.ru/histussr/13-3.htm
