With the move towards the sharing of linguistic data, sociolinguists are now considering, more than ever, methods for creating corpora that maintain requisite ethical principles while still allowing for data to be used by researchers from around the globe. This paper examines the guiding principles used to create a sociolinguistic corpus that would permit sharing without compromising commitments to informants, from the interview stage to transcription, verification, and anonymization. We consider and adapt theory and practice from elds such as corpus linguistics, anthropology, and sociology.