Three metaphors for not understanding Big Data
- Carla Calvo
- 11 sept 2020
- 4 Min. de lectura
‘Big Data’ has become a buzzword that everyone uses, but no one seems to really understand its meaning. Although it is true that there’s no static definition, its widely accepted to link the phenomenon to its size. While ‘data’ is an accounting noun that has both a plural and a singular form, when we speak of ‘Big Data’ we are referring grammatically to a mass noun. Big Data refers to those datasets whose size exceeds the capacity of traditional computer tools to collect, store, manage and analyse them. As the McKinsey Global Institute report points out, this definition is intentionally subjective and incorporates the possibility of moving the scale that measures how big a dataset needs to be in order to be considered Big Data. Due to the incredible speed of digital change, what’s considered Big Data today can easily become simple data tomorrow. Big Data is much more abstract than traditional data because of its ever growing size, but also because of the increasingly sophisticated tools used for its analysis[1]. And to make the abstract concrete we need language, and specifically, we need metaphors.
What’s considered Big Data today can easily become simple data tomorrow
As Lakoff and Johnsons[2] noted in their book, Metaphors We Live By, metaphor is not just a device of poetry and rhetoric, but a common tool we use on our daily life. Metaphor isn’t just a matter of extraordinary language, but of ordinary actions too. Lakoff and Johnsons said that our conceptual system is fundamentally metaphorical in nature, in terms of which we both think and act. That is, cognitive linguistics consider metaphor as one of the main tools that people use to understand reality. This is rooted in the premise that cognition, meaning the ability of individuals to understand their environment, is the result of a mental construction. The concepts that govern our thought, Lakoff and Johnsons pointed out, aren’t just matters of the intellect, but of our everyday realities. According to this approach, metaphors function as an analogy that allows us to understand one concept in terms of another. This process is not a matter of replacing one term with another, but rather of adding new layers of meaning. The metaphor has to expand the original concept, although sometimes this doesn’t happen. Sometimes the metaphor focuses on some specific aspects of the concept and places them at the centre of others, providing a reductionist and partial image.
Due to the high level of complexity of Big Data and its abstraction, we need to use metaphors to understand the concept. But it’s also easy end up falling into reductionist conceptions. In their study of metaphors in Big Data, Puschmann and Burgess found that the media tends to use mainly two approaches when talking about Big Data: as a force of nature or as a resource to be consumed. While conventional data was described as discrete units of information, Big Data is conceived as a seamless mass. Summoning forces that are too strong to be handled by humans is a common metaphor in the field of technological innovation since it involves the challenge of taming nature until it becomes a source of resources that we can use. This leads us to the second conception of Big Data. Once the natural force has been controlled in order to be consumed, the data becomes a resource that allows us to survive or that supplies our lifestyle. In both cases, data would be conceived as something to be consumed instead of being used knowingly. But data is not a natural source that feeds itself, but the result of human creation. Big Data appears when users make use of the social media platforms, switch on the geolocation of their smartphones, consult weather information satellites, use their now also smart fridges or any other IoT (Internet of Things) devices that collect a loads of data per minute.
This bring us to the third metaphor: the inclination to consider data as raw or aseptic. There is a widely spread assumption that when a certain amount of data is obtained, the numbers speak for themselves. Andersen’s end of theories was rooted on that idea. The chief editor of Wired claimed that the access to large amounts of data, combined with the right statistical tools, would offer a new way of understanding the world where correlation would replace causality. But, as Bowker[3] points out, raw data is an oxymoron. Data doesn’t speak for itself, but it’s people who give it a voice. Just as is people, using social media and computer tools, who create, collect and interpret them.
Raw data is an oxymoron
While it’s true that the end of theories is yet to come, Big Data, as well as the Internet, has deeply changed the way we understand reality. Many aspects of social life that had never been quantified before are now codified. Friendships, interests or even a person's location in real time have been converted into quantified data. This has given rise to new ways of analysing and understanding reality that are still difficult to understand, imagine or even conceptualise through metaphors.
Kommentare