Remarks on April 1st, 2030 AD at the tenth anniversary celebration of the first Analog Summit on April 1st, 2020.
[The following is a work of fiction. The companies and people referenced had no input into this article, nor does this reflect their actual products or strategy. At least not yet…]
Remembering the Dark Ages of the 2020s
Back then it was inconceivable that most companies would have automated pipelines that unified critical financial, operational, and data science metrics into a single real-time dashboard. Today’s incoming generation of QuantOps specialists probably experience physical revulsion at the idea of manual financial closes, copy-and-paste of unversioned data from spreadsheets, and being forced to perform repetitive Extract-Transform-Load tasks. They take it for granted that they will not only be able to leverage the same cutting-edge tools and processes as their peers in software engineering, but to earn the same level of respect and compensation as well.
To help those newcomers appreciate just how much we have accomplished in the intervening decade, it is worth remembering what the world used to look like. Back in the dark days of 2020, organizations considered Finance, IT, Operations, and Data Science four completely separate departments. Microsoft Excel was still considered state of the art for managing datasets, using the same paradigm of caching central data they had pioneered (actually copied from Lotus) twenty years before. Reconciling financial books — never mind auditing them — was a tedious process performed only once a quarter, and you were considered a superstar if you could do it in less than a week.
Treating Data as Code
Data Science was the first discipline to break away from the tyranny of Excel. Quilt Data (at the time just another hungry young startup) pioneered the practice of managing data like code, through automated end-to-end pipelines that properly versioned both data artifacts and the tools used to produce them. Their key conceptual breakthrough was realizing that this was actually an “Analog to Digital” (A-to-D) process, which required an “analog” human to iteratively compare the digital output against the desired business and social impact.
Quilt recognized that the most important metric is decreasing cycle time to actionable insight: how fast the human in the loop could either approve data to send to the next stage, or else identify which of the inputs or operations to tweak in order to debug and fix the process. This led to a unified model of curating data for use by both Artificial and Business Intelligence (A+BI) that quickly became the gold standard for data-driven organizations.
What Quilt did for Data Science, NetSuite was attempting to do for Finance through its skunkworks “SuiteClose” moonshot. The goal was to enable “single-click” closing of financial periods by automating all the manual checks and fixes, so that the responsible humans could quickly scan for anomalies and select or edit the appropriate fixes; with their actions fed into a machine-learning engine to help optimize the next close.
It turned out the key to pulling this off was implementing a Shadow Reconciliation System™, which allowed implementors (and eventually the AI) to rapidly and safely experiment with changes to both the business logic and financial data. This enabled accountants to prototype and visualize changes in a secure sandbox, and then to do the equivalent of a “git merge” (or revert) on the append-only system of record.
Meanwhile, the new Chief Resilience Officer at the Long-Term Stock Exchange (LTSE) was torn between exhilaration and despair. She now knew exactly what was needed to evolve capitalism into being more sustainable. She just couldn’t imagine how to build it.
Her epiphany came from a podcast by Scott Looney of the Mastery Transcript Consortium (MTC). MTC was developing standards for a rich, interactive portfolio that better reflected and incentivized students’ diverse achievements than a narrow report card. She realized that this was precisely the “killer app” LTSE needed: replacing static financial statements with a comprehensive data repository incorporating all the metrics that resilient companies cared about, such as environmental impact, diversity and inclusion, employee and stakeholder engagement. LTSE could define standard schemas, making it easy for analysts and activists to visualize and compare performance across different businesses. As someone once said, “Information creates Vision creates Mobilization creates Transformation.” Best of all, it would enable companies to align external stakeholders around a public subset of the exact same metrics they used to manage the business.
That was the exciting part: a future where the tools of extractive capitalism were repurposed to serve humanity. The challenge was that she couldn’t figure out how to get there. She could barely even imagine the herculean effort required to collect, audit, and publish those mountains of data. No company she knew of had the bandwidth — much less the skills — to take on the additional work. It was just too much to ask of corporations already struggling with GDPR and SOX compliance, never mind the innovative startups she hoped to attract to the the LTSE.
The Quantifier’s Dillemma
Everything finally came together at the first Data Business Congress, held in San Jose on February 3rd, 2020. Tim O’Reilly was doing an impromptu lunchtime talk, riffing on his WTF thesis that financial capitalism is the first example of “rogue AI,” enslaving humanity to a flawed objective function. He started by showcasing the numerous benefits and great flexibility that come from our ability to quantify and digitize every aspect of modern life. The wild applause quickly became muted as he shifted to examples of how those same tools had enabled people to more efficiently oppress, discriminate, exploit, addict, and destroy.
Tim expanded this to a historical critique of Aristotle’s claim that “Man is a Rational Animal” (MIARA). He traced this “idolization of quantification” through Descartes and Taylorism, showing how it was both incredibly powerful and extremely dehumanizing. Tim claimed that the great challenge of the next century is learning how to harness digitization of the economy and society in the service of humanity, rather than to exploit it.
He ended by making the bold counter-claim that “Humans are Reflective Social Animals” (HARSA). Specifically, “rationality” is just an extremely valuable side-effect of our ability to thoughtfully communicate with others — not the primary marker of humanity, or even of intelligence. He called on attendees to reinvent our institutions, philosophy, and culture around that richer understanding of what it means to be fully human — before it is too late.
The Beer Summit
As it turns out, there were representatives from Quilt Data, NetSuite, and LTSE in the audience. Each of them resonated powerfully with Tim’s message, because it provided a larger context for the challenges and opportunities that were central to them taking their business to the next level.
The representatives ended up mobbing Tim with questions, to the point where they were forcibly ejected from the room to make way for the next session. Along with a handful of other hanger-ons they adjourned to the Gordon-Biersch beer garden, where they commandeered a table for the next several hours. They spent of the time in almost confessional testimonies, as they shared their personal and professional frustrations with the seductive power of digital abstraction.
At one point, Tim remarked that “The core problem seems to be that our digital abstractions have become divorced from analog reality.”
A beaded Indian gentleman on the fringes remarked: “Well, maybe it is time for them to get remarried.”
This led to a flurry of debate and brainstorming about what it would take to reconcile the digital and analog into a “more perfect union.” In the end, they managed to boil it down to four words, which they described as the Motto of the Analog Revolution:
“Digital must serve Analog”
This motto captured their core belief that digital techniques and measures must never become ends in themselves but rather be efficiently monitored by people, to ensure that they serve broader humanistic and social goals.
They finally adjourned at nearly 10pm, after having scheduled weekly conference calls to figure out next steps.
The April First Analog Summit
Those conference calls eventually led to twenty-seven people meeting in an all-day conference on April 1, 2020. The conference was held at the San Francisco offices of the Boston Consulting Group (BCG). BCG had also been building a practice around better measures of human flourishing, and quickly jumped on board once they heard about the Analog Revolution.
The main focus of the conference was figuring out how to formalize, operationalize, and institutionalize the core values of the Revolution, to ensure that they didn’t become diluted or coopted. Importantly, the first step was coming up with the right analogy or metaphor to ensure that they were all talking about the same thing.
Surprisingly, the key turned out to be realizing that digitizing anything is actually a form of violence, like performing surgery or killing an animal. In a very real way, when you reduce someone or something to a set of numbers, you are effectively stealing their soul because you can then determine their destiny at no emotional cost to you. Data is like blood: a little bit can tell a lot about you, and giving away too much can destroy you.
Being able to represent and manipulate people (or a planet) with data is a great power, which implies an equally great responsibility to use that for the greater good.
Because civilization and humanity’s progress now depend on digital data, it is impossible to completely renounce the practice of representing humans with numbers. Instead, we must professionalize that act of violence — like modern societies do with doctors, and ancient ones did with priests. This enforces a moral and social obligation on practitioners to use their craft humbly and wisely, in recognition of the larger realities they only imperfectly understand yet are sworn to serve.
After much debate, the attendees voted to use the broader term “quantification” to signal that this was about more than just digital technology. With the blessing of Ally Miller they adopted the term “QuantOps” for this new class of professionals responsible for creating and using the tools of digitization to serve analog ends. They also drafted a Quantifier’s Oath and founded a Quantifier’s Guild, to help those professionals remember that they serve a calling larger than whatever business, government agency, academic institution, or non-profit they currently work for.
The day ended with a formal signing ceremony, where the attendees pledged their “lives, fortunes, and sacred honor” to make the Analog Revolution into reality.
Inventing the “A” Corporation
This led to a flurry of behind-the-scenes activity as the nascent revolutionaries prepared for a public coming-out party at TED 2020 on Earth Day. Angie Coleman led the packed event, the centerpiece of which was a template for a radically new kind of company, dubbed the “A” Corporation. They also announced that the very first A Corporation would be the “Analog Benefit Consortium” (ABC), the organization chartered to promote the Analog Revolution.
An “A” (or Analog) Corporation extends the public benefit commitment of a “B” Corporation to cover its internal processes and culture, not just the triple-bottom-line. In addition to the expected polices around employee diversity, stakeholder capitalism, and data ethics, the bombshell was that all A Corporations would provide financial, compliance, and social impact data over OpenAPI using Hypermedia CSV (HCSV). HCSV is a strongly-typed, easily-parsed reimagining of compact CSV data files which adds support for formulas and relationships. All the tool vendors present pledged to support HCSV in the next version of their products, enabling a new data commons, which holds the promise of supplanting the narrow focus on financial capitalism that has dominated the past century.
Reflections on the past Decade
The rest, as they say, is history.
Of course, it was hardly an easy or straightforward road. There were numerous legal challenges, character assassinations, and at least one fistfight at the SEC.
But the end result was the world we live in today. Where it is taken for granted that corporations exist as part of a fragile ecosystem — involving people, markets, and the planet — that must be continually monitored to maintain a healthy balance. Where all self-respecting organizations publicly share operational data in a way that enables them to continually improve their performance while minimizing negative impacts. And QuantOps professionals enjoy the same prestige as medical doctors did in the 20th century, using tooling and infrastructure more sophisticated than DevOps introduced at the beginning of the 21st.
So on April 1st, 2030, let us tip our hat to those brave visionaries from a decade ago, who bucked conventional wisdom to imagine a better, fairer, and more sustainable — and then brought it into being.