The research is based on infrastructures which serve the purpose of harbouring and archiving data. For our comparative and evolutionary linguistic research program, we have constructed a relational database of ≈210,000 controlled points of lexical and grammatical feature data from ≈600 languages of 18 different families (DiACL - Diachronic Atlas of Comparative Linguistics). Towards language documentation, we possess a node at the Corpus Server at Lund Humanities Lab (LUNDIC), where linguistic fieldwork data and metadata is stored. For the documentation of Tocharian, we have a text and reference corpus of ≈350,000 words (CeTOM) and a dictionary corpus of ≈250,000 words (CeDICT).