A language corpus is one of the modern Internet technologies, on the basis of which students can develop the skills of written foreign language interaction. Under the language corpus we propose to understand the various types of arrays of texts combined in a single electronic system for a specific typological basis. The most common typological types are: type of linguistic data, tags, language, register, purpose of the corpus, chronological sign, indexing, character marking, purpose of the corpus, mode of existence of the corpus and online access. Based on the analysis of methodological literature on the use of modern information and communication technologies in teaching a foreign language (blogs, wiki, podcasts, linguistic corpus, social service Twitter), we identify the didactic properties of the language corpus. These include the following: multilingualism, contextuality, dynamism, the ability to search in the corpus according to a given pattern, the ability to select functional text types in the corpus, sorting of the search results and network access. Each of the didactic properties corresponds to a certain methodological function, consideration of which is necessary when identifying and justifying a number of psychological and pedagogical conditions, as well as developing a methodology for teaching students written foreign language interaction based on the linguistic corpus.