In her 1996 paper, The Rage to Master: The Decisive Role of Talent in the Visual Arts, Ellen Winner presents a concept she calls, well, the “rage to master.” The idea is that intellectually gifted children have a natural inclination to focus on a subject and immerse themselves in it until they reach mastery.
With proper support, the “rage to master” creates a positive feedback loop. Their interest combines with their gifts, enabling him or her to more easily grasp a topic than a more average individual. This provides a feeling of satisfaction, reinforcement that encourages the child to continue mining the subject. Using the initial knowledge as a springboard, the cycle repeats itself, creating an outward-spreading spiral of knowledge.
Data Harmony has something in common with that gifted child: the feedback loop in its indexing. The software knows nothing at first, but when it is fed content, its subject of choice, and is given support and encouragement in the form of taxonomy building and editorial analysis, it can start the learning process.
With one piece of content, it can only learn so much. It grows with each new piece, the next feeding off what came before, but it needs consistent and diligent editing of those results. Given that, the software can become progressively smarter.
Just like with the gifted child, though, who can never learn everything about the given subject, the feedback loop that indexing software can create won’t last forever. Eventually, progress will slow down. There’s a big difference between the highly accurate search results it delivers and perfectly accurate search results, an unattainable goal.
Voltaire’s aphorism, “Perfection is the enemy of the good,” applies well here. The “rage to master” in the gifted child depends on progress and satisfaction. Attempting perfection undermines both. Progress will slow to a halt, denying the child the satisfaction that was the driving force in the first place.
Of course, we’re talking about software here, so feelings and stuff like that don’t actually apply. Where it does apply is with the user, though, who “motivates” the software by feeding it content. They are the impetus for software’s education, giving it new material while honing and fine-tuning the output. All of this delivers accurate results and the user gets the feeling of satisfaction.
Indexing software has the “rage to master” content because it was built to serve that purpose. It can’t do anything alone, though. It takes a dedicated team of editors to feed it that content and interpret the results. The responsibility is on them to understand how to leverage the results into valuable commodities. Without that side of it, the software achieves very little.
The emergence of Big Data has made this increasingly vital to business in industries of any stripe. The amount of data is growing at an astonishing rate and shows no signs of slowing down. If it was difficult to collect and analyze large amounts of content manually a few decades ago, imagine the struggle today with the glut of tablets, phones, and computers collecting and transmitting data every moment of the day.
There is so much out there that even a large team of editors can struggle to sort and analyze it with much effectiveness or insight. But this is exactly where the feedback loop created by indexing software can change the game. The software speeds the process, facilitating the analysis, but it can’t make decisions on its own. The editors are absolutely crucial to the accuracy of the software’s output. It starts with an analysis of a single batch of content, but with their guidance, that analysis builds on itself with each new batch. Before long, patterns start to emerge.
Now, the people who would have had to endure the tedium of slowly going through the data by hand can work with these emergent patterns instead. This is a far more meaningful way to interact with data and enables new ways to look at the results. Now, people can more quickly and easily identify and react to trends in their industry.
In publishing, this means understanding how users search for content and potentially directing them to content they may not initially have found valuable. Using Data Harmony, the publisher has a controlled vocabulary that narrowly and accurately directs searches, but it also allows them to observe and analyze how the user searches and what else they search for, which gives them tools find patterns in their customer base and tailor future initiatives to their specific needs.
The mountain of data in this world is only going to continue to grow, so while large-scale analysis is important today, it will be even more important tomorrow, next week, and in a year. Who knows what the landscape will look like in a decade, but we can safely speculate that the positive feedback loop that emerges from software like Data Harmony will enable organizations to handle it, no matter how massive it may have grown.