The paper analyzes the English computer terminology localized into Serbian between 2010 to 2020 both from the purely lexical stand-point (type and status of borrowings, as per Prćić (2005)), and from the viewpoint of pragmatic borrowings (Andersen 2014), but in a narrower set of pragmatic borrowings limited to written modes of communication: direct and indirect borrowing of interjections, discourse markers, expletives, vocatives, and paralinguistic phenomena. The chosen period saw a rapid expansion of instant communication applications with rich sets of emoticons, proliferation of cloud services and replace- ment of traditional software with software as a service. The research aims to shed light on how much the new lexis from computer-mediated forms of communication, as well as pragmatic notions stemming from the emerging trend of IT companies crafting their content to adhere to strict stylistic and audience accommodation guidelines, have been taken over into Serbian. Preliminary findings indicate that there is a huge number of both lexical and pragmatic borrowings, including interjections and paralinguistic phenomena such as names of emoticons and short codes used to insert emoticons. This influx of new, raw and, due to style guides, conditionally justified English borrowings provides a new form of evidence of English having become a nativized foreign language (Prćić 2014) of online communication, which, as a knock-on effect, leads translators and localization teams to consider raw and normally unjustified borrowings as acceptable and appropriate for the target audience that consists of more and more digital natives. The research applies a combination of quantitative and qualitative research methods by means of annotating the 2020 version of Microsoft English-Serbian Bilingual Terminology Database.