Big Data. Biggish Data. Now dark data. The idea plays on the silliness of the dark Web; that is, it is information that is “there”, but you don’t know about it. Well, get with it, pilgrim. Datameer used this term in “Shine Light on Dark Data.”
Here’s the definition:
At every organization neglected data sits overlooked in log files and archives accumulating digital dust and incurring costs. But as more organizations look for ways to become better, stronger and faster, they’re digging into this “dark” data and uncovering a gold mine of business intelligence.
Now how do you shine light on dark data? Great question. I will not probe the logical aspects of this concept. There are, according to the article, five steps to take. These are—unsurprisingly—the same steps a prudent and informed manager takes to figure out just plain old data.
Words to marketers make all the difference. I am not sure data has an opinion.
Stephen E Arnold, October 3, 2014
Phi Beta Iota: There are two forms of “dark data.” The first is useful data badly presented — PDFs, for example, that are neither fully-indexed by mediocre search engines including Google, nor translatable on the fly by online translation services (among which Google offers one of the best). The second is simply useless data collected for the sake of collection, using old models and flawed assumption, unprocessable for varied reasons, and therefore never likely to impact on anything — simply a waste of time, energy, and money.