Data is everywhere. Sensors and software are digitizing and storing it. This data is mapped , measured and analysed and as a result Big Data is producing new ways of knowing and being in the world. Datafied knowledge production is fast growing . In this flood of exponential growth of new data and datafication of knowledge, we have the challenge to understand how it is changing the already existing structures of our society. How will it influence our knowing ? Will it shun a side our need to speculate? Will it do away with theory ? Shall we simply observe and measure ? What does Big Data mean to our research and knowledge creation ? How does Big Data transform the way we position ourselves in relations to object of study , our methodologies, epistemologies , our funding sources and above all how we are perceive truth?
It appears that Big Data has emerged as a meme that speaks to and generates new ways of establishing truth. Will this mean that knowledge that cannot be dataized will be devalued? Are we moving towards positivist quantitative , scientistic paradigm of the logical positivist once again? This fears are real because data is never pure and objective. The way data is collected , cleansed and corelated is also marked by bias or assumptions that limit the kind of data is patterns are mined. Besides, commercial analysis is outpacing hard academic research . Hence, the coming of Big Data seems to narrow rather than expand our project of knowledge production. But we cannot build knowledge by keeping aside Big Data.
Datafied knowledge production is not clean revolutionary break from the past nor it is a hyped balloon of the future. One of the important constant of this datafied knowledge production is that knowledge generation, itself is hidden in an opaque black box through the operation of data analytics , algorithms, and digital operations. This triggers real fears of algorithmic governmentality. But there is always an human element in this form of datafied knowledge production. Machine learning use humanly produced data to undergo its learning . With machine learning we turn, data into valuable knowledge. What is really the case in this context is the issue of data ownership. What we need is to work to bring about data commons so that data hoarding for commercial as well as political interest does come to haunt the people who use algorithms and other data practices to produce knowledge.
Data democracy and data sharing through the data commons might be the way to go ahead even while we still have to deal with opaque processes in the production of knowledge. While we are concerned with the opacity of knowledge of datafied production, we cannot lose sight of the fact that this new mode of knowledge production has profoundly transformed the way we see, read, organize, use and dispose knowledge.
Our social activities as well as intimate private life has become data points and actively shape datafied knowledge production. Hence, the issue of the lazy reason taking charge of the production of datafied knowledge is not far fetch. Besides, the prospects of increasing the divide between the rich and power may grow with the growth of data as well as AI powered knowledge production. While on the epistemological side, we have to still deal with the authenticity of the data as well as the intellectual integrity and authorship of the persons who is producing the datafied knowledge which of course appears to be an autonomous work.
Alongside these important issues, we have the challenge to understand how with the coming of Big Data, we have epistemological shift from the causal approach to inductive and correlational paradigm. Thus, the way Big Data is used to produce and justify knowledge becomes an vital epistemological concern. This brings us to the complex relationship between the produced knowledge, knowledge producers and means of knowledge production. While we let data speak from itself without any previous knowledge, we have stepped into a new mode of epistemology and changed the condition of knowledge.
In this context of data-driven and data-intensive knowledge production , we seem to have largely outsourced our mind although most of the datafied knowledge is produced under the supervision of humans. But the human element seems to be ornamental and with AI it is going to be bear minimum. For now algorithms find patterns and human propose hypothesis that follow data. But this human element in the production of data will be replaced by AI. Will that mean that we are poised to the death of critical thinking? Are we moving to a point where the emperor will be not aware that he is naked?
The promise land of the datafied knowledge says that anyone can use data and no expertise is no longer required . If this days are going to come soon the universities that considered to be the sites of knowledge production will have to go. But with this there are fears that we might land into a chaotic juggle as we cannot assume apriori objectivity of the datafied knowledge since the bias of the data collection, cleansing and storing as well as the bias of the naked emperor will interfere. Thus, moving towards highly techno-centric approach might push humanity in new dark ages.
Subjectivity cannot be entirely ruled out from the entire process of production of datafied knowledge. While critically accessing the benefits of Big Data, we have the challenge to address the challenges it posing to epistemology. We have the challenge of assessing the rhetoric of exactitude and objectivity realistically and be mindful of the gaps that Big Data introduces and work to bridge them. It might Be too early to dump the age old methods of knowledge production and uncritically embrace the hype of data driven knowledge production.
