Image: https://www.domo.com/learn/data-never-sleeps-7
Aoun Jafarey, a data analytics professional based in New York City, talks about the gray areas that exist in the enormous amounts of data we produce, consume, and utilize daily.
Imagine, if you can, stacks upon stacks of iPads stretching from the Earth’s surface all the way to the moon. Or, visualize 292 Great Pyramids filled with millions of external hard drives. As a final example, try to picture around 33 Empire State Buildings packed with billions of USB sticks. According to Domo, a cloud software company based in the United States, this would have been the physical representation of the amount of data produced per year, over five years ago.
The meaning of the word “big” in the term big data is so mind-boggling that it is hard to imagine what our collective digital footprint would look like if physicalized. Though big data has always referred to information that requires advanced techniques and software to process, the scale at which it is produced is exponentially trending towards the colossal. Indeed, with over 2.5 quintillion bytes produced daily (that’s 18 zeros), all the data in existence has been produced in the last 5 years.
This scale is fueled by the ease and speed with which professionals, public officials, and everyday people can conduct their business online. It is no surprise, then, that whether it is in designing algorithms for social media, conducting market research for a specific firm or industry, or formulating public policy, large volumes of data are integral to everything we do. Data science is therefore becoming increasingly central to how we describe, model, and even predict the future of a variety of human activities.
It makes sense that the quality of the data at our disposal goes a long way in the accuracy of data analytics. And the quality of our data invariably brings up questions of how we measure social phenomenon. What is the role of measurement in data science? And how can aspects of data science, like presentation for example, affect our ability to effectively describe and model the world?
Episode 3: Gaps in the Data