The first provocation “Automating Research Changes the Definition of Knowledge” is of interest to me because it is very relevant to the “new entity” at MIT now know as IDSS. Some of the things we want to know about require interdiscplinary work, so that the scholarly traditions of one discipline may clash against the norms and protocols of another. I think this is likely to happen for awhile as people attempt to make sense of everyday life from Big Data stemming from popular sources – is this sociology? huamnities? something in between?

A lot of the way Danah Boyd expresses her point seems more urgent than understandable to me, but here’s what I get out of it: Once you start using really transformative research methods and tools, you need to rethink what it is to “know” something. When you move from cause-and-effect claims to statistical claims, knowledge has shifted a bit. When we start using methods associated with Big Data, what counts as scholarly rigor, what has to be transparent about methods and data collection, for something to count as knowledge? Her point about we need to be able to describe and understand the limitations of our data sources (twitter, google, consumer footprints on the web, etc) seems pretty straightforward as an exhortation – settling on norms will take some time.

Not all data are Equivalent. When you move something from a more ethnographic, individual realm, to a very public source – meaning shifts. Her examples about behavioral and articulated networks, as compared with personal networks, is a good example. What I can even do in a massively connected virtual environment is so different than what I can do when limited by physical geography, that we need to be very careful not to conflate concepts across smaller and Big data.

Enter text in Markdown. Use the toolbar above, or click the ? button for formatting help.