I am innately distrustful of buzzwords.
I recall in the early 1990’s that the hot management trend was “self-directed teams.” Its main attraction to executives seemed to be as an excuse to lay off lots of managers, because with the right training, a group of employees could manage themselves (I had forty-seven direct reports at one point thanks to this brilliant idea!) “Self-directed teams” is not a terrible construct, but its promoters didn’t stop there. Instead, they proclaimed the wonders of “teaming,” and that “teaming is goodness.” Ugh.
I can’t help thinking of a book I treasure, The Elements of Style by William Strunk, Jr. and E. B White, specifically White’s contribution, An Approach to Style. His sage words about “finalize,” which he cites as an example of the language of business, apply to any buzzword I’ve had the misfortune to grapple with during 36 years in financial services:
“Finalize, for instance, is not standard; it is special, and it is a particularly fuzzy and silly word. Does it mean “terminate,” or does it mean “put into final form”? One can’t be sure, really, what it means, and one gets the impression that the person using it doesn’t know either, and doesn’t want to know.”[1]
Fast forward to 2015-2016, when I started hearing about “data curation,” especially from consultants. I don’t mean to imply anything negative about consultants, as I’ve been one myself, and have benefitted often from their guidance and insight. But consultants often pounce on the latest buzzword. They promoted data curation as the greatest thing since… well, fit-for-purpose data.
I remember recoiling, viscerally, from this concept. I associated “curation” with art or history museums, and curators who spend their professional lives deciding which artwork and artifacts best suit a particular exhibit. The definition of curator is: “a person who oversees or manages a place (such as a museum or zoo) that offers exhibits.”[2] Handling data like paintings in a museum is the exact opposite of how I think about data governance. Data is alive, involved with real-life business transactions and analysis. It’s not locked in a museum cabinet for passive study.
Data curation definitions offered by notable organizations in Dataversity’s article What is Data Curation? reinforce my belief. For example: “Data curation, as defined by The University of Illinois’ Graduate School of Library and Information Science: ‘is the active and ongoing management of data through its life cycle of interest and usefulness.’”[3]
The same article offers a software company’s definition: “In practice, data curation is more concerned with maintaining and managing the metadata rather than the database itself and, to that end, a large part of the process of data curation revolves around ingesting metadata such as schema, table and column popularity, usage popularity, top joins/filters/queries. Data curators not only create, manage, and maintain data, but may also be involved in determining best practices for working with that data. Data curators often present the data in a visual format such as a chart, dashboard, or report.”[4] What could bring data to life more than this?
But there is one definition in this article that resonated with me: “The process of ‘caring’ for Data, including to organizing, describing, cleaning, enhancing, and preserving data for public use.”[5]
This description, from Inter-university Consortium for Political and Social Research (ICPSR),[6] isn’t so different from the others, except for the beginning: “The process of ‘caring’ for Data.” It’s a term one usually applies to living things— caring for a person, or a pet, or a tree— and not something as inhuman as data.
It happened that I read this article at the same time I had begun to re-read Thomas Moore’s Care of the Soul for the first time since the early 1990s. When I picked it up again this spring, I found I had underlined many sentences which must have arrested my attention then, as they did now. I remembered how Moore’s writing stirred me deeply with his references back to Greek and Roman mythology and how they spoke to the soul.
I also could see where I first became fascinated with etymology, long before my pandemic binge of the Tolkien podcast The Prancing Pony (which still serves up the best pints of word nerdery this side of Bree!) Right at the start, there on page 5, Moore explains the Latin source of the word “curate”:
“Cura, the Latin word used originally in ‘care of the soul,’ means several things: attention, devotion, husbandry, adoring the body, healing, managing, being anxious for, and worshipping the gods.[7]
First, this is a remarkable assertion, that one word can mean so many things. I began with E.B. White’s observation that buzzwords often mean nothing. We recognize “managing” from the various “data curation” definitions, but what of these others— Attention? Devotion? Adoring the body?
Let’s break it down. I’ll leave aside “adoring the body” and “worshipping the gods,” in favor of the more readily relevant meanings. Paying attention to data is clearly important. As my late and much-loved college music theory professor, James Sellars, told his students: “The secret to life is paying attention.” Attention also implies observation. Moore notes that “care of the soul begins with observance of how the soul manifests itself and how it operates. We can’t care for the soul unless we are familiar with its ways.”[8] Reading this reminded me of a new buzzword in the data management space, “data observability.” Databand’s definition associates data observability with the idea of caring for the soul of data: “’Data observability’ is the blanket term for understanding the health and the state of data in your system.”[9]
Devotion? Consider the Oxford English Dictionary’s definition for non-religious use: “The quality of being devoted to a person, cause, pursuit, etc., with an attachment akin to religious devotion; earnest addiction or application; enthusiastic attachment or loyalty.”[10] This sounds like a great quality for a data steward— devotion to data.
Husbandry? The applicable definition here is not so much “the science or art of farming” as “careful, thrifty management; thrift; frugality.”[11] Careful, thrifty management of data has many benefits.
Healing? Fixing poor data quality. And finally, “being anxious about data” is totally relatable and requires no further explanation!
Now, all this is metaphorical, right? Yes, I am using the metaphor of caring for the soul to illustrate a different perspective on data management. But that doesn’t mean it has no practical application.
Let’s do a thought experiment with the “care of the soul of data” approach. When we data professionals receive business’s call to remedy a data quality issue, we ask for a few examples, dive into a root cause analysis, then produce remediation plans as quickly as possible. We finish by rushing in a tactical fix. Voila, job done!
Instead, following a more soulful approach, we choose to manage our anxiety about the data issue and focus our attention on the process where the issue occurs. We devote several days or even weeks to observing the data flows, the inputs and the outputs, taking note of the quality of the data over this period. We refrain from jumping into root cause analysis and “solutioning” (there’s a buzzword I can absolutely do without!) Instead, we approach healing the data with husbandry, looking for the most efficient and frugal means to improve the quality while not disrupting the data flow.
There is so much more in Care of the Soul which is relevant to data governance and management, enough for several columns. Here is one example from Moore’s introduction: “The word care implies a way of responding to expressions of the soul that is not heroic or muscular.”[12] How well this quote ties into the DataOps view of the downsides of the “heroic” efforts of a data team:
“Data engineers, scientists and analysts spend an excessive amount of time and energy working to avoid these disastrous scenarios. They work weekends. They do a lot of hoping and praying. They devise creative ways to avoid overcommitting. The problem is that heroic efforts are eventually overcome by circumstances. Without the right controls in place, a problem will slip through and bring the company’s critical analytics to a halt.”[13] For now, I will leave you by paraphrasing Moore’s wonderful expansion on paying attention. To care for the soul of data, we must pay attention to form as well as function, to retention as well as invention, and to quality as well as efficiency.[14] Words a data curator can live by.
[1] Strunk, William Jr. and White, E.B., The Elements of Style, Fourth Edition, Needham Height, Allen & Bacon, 2000, pg. 85.
[2] https://www.merriam-webster.com/dictionary/curator
[3] Knight, Michelle, December 25, 2017,What Is Data Curation? – DATAVERSITY
[4] Ibid.
[5] Ibid.
[7] Moore, Thomas, Care of the Soul, New York, NY, HarperCollins Publishers, Inc., 1992, pg. 5.
[8] Ibid, pg. 5
[9] https://databand.ai/data-observability/
[10] Oxford English Dictionary online, last updated June 2022
[11] https://www.collinsdictionary.com/us/dictionary/english/husbandry
[12] Moore, pp. 4-5
[13] DataKitchen, May 31, 2017, https://medium.com/data-ops/the-seven-steps-of-implementing-dataops-step-7-work-without-fear-c9959574543e
[14] Moore, pg. 277
