Data sets are not values neutral
Data sets are authored objects, created for a specific audience and purpose - when someone collects data there are already some assumptions and biases built into the data set. Computation isn’t value-neutral, everything is infused with choices, values and ethics. Data in itself isn’t evidence, although you can use it to form evidence. It is collected and analysed to create information suitable for making decisions.
Data sets aren’t just technical constructs, they’re also social and cultural constructs. We need to think hard about the design choices we make when deciding how to structure databases, and how to use that data as a key component in producing and disseminating knowledge. People engaged in digital scholarship need to think critically about the design choices and ethics involved in the creation and use of these data sets, not just the technical constraints and considerations of implementation.