My own work is in science laboratories and how scientists communicate to build scientific theory. Certainly, repetition and "re-use with modifications" (Charles Goodwin 2018) are a key component of this process. The infrastructure scientists use (journals, email, University departments, banks of computers/servers) to achieve their goals are largely invisible to them and they often don't realize how the direction of their theory depends on the contingencies of infrastructure. Question: should the infrastructure be 'visible' or can it only do its work enabling productivity when it is invisible?
