Because innovations in media technologies affect all aspects of our daily lives, multimodal and computational literacies are becoming increasingly important as a way to empower English as a Foreign Language (EFL) learners in their second language practices. Content-sharing platforms, such as YouTube, provide instant access to authentic language environments in a wide variety of contexts. While many learners operate emergent software services intuitively and with great ease, their potential benefits for language learning remain widely unexplored. This paper examines multimodal and computational meaning-making practices in EFL contexts from a theoretical and empirical perspective. More specifically, it focuses on how competencies in multimodal meaning-making and software programming can be combined with the lexical approach to language learning when repurposing YouTube videos to teach English idioms and phrases. Based on a prototype language learning website designed and implemented by students at the University of Klagenfurt and improved by students from Bundeswehr University Munich, the paper identifies fundamental literacy skills required for a critical and self-determined application of software services in English language classrooms in terms of four emergent digital meaning-making practices: multilinear choices, multimodal meaning-making, data curation and procedural reasoning. The prototype language learning website that evolved as part of this project can be accessed at www.lingo.farm.