The Chomp #044

A Primer on Algorithms, Seeing Like an Algorithm, and More...

Hey Everyone,

Welcome back to The Chomp—your weekly dose of the best strategic thinking content and top emerging business trends from the internet and beyond, designed to expand your mind and get you thinking. If you’ve been sent this email and you’re not a subscriber, you can join by clicking on the blue button below. With that, let’s dive into it.

---

Quick Bite

A Primer on Algorithms and Bias: Algorithms that have a significant impact on our lives specifically need to be open to scrutiny and analysis. If an algorithm is going to put you in jail or impact your ability to get a mortgage, then you ought to be able to have access to it.

As we forge into the future it’s no secret that algorithms will have an ever-increasing role in many facets of our lives. Whether it’s an algorithm you use at work, the algorithm that powers your Instagram feed, the algorithm recommending you shows on Netflix, an algorithm determining your creditworthiness, an algorithm powering your investments, or an algorithm assessing your resume, you’re bound to run into algorithms everywhere.

Yet, despite our continued reliance on algorithms, we rarely get to see the secret sauce behind them. The majority of us have no clue how the algorithms we rely on actually work, which is pretty scary when you take a second to step back and think about it. That's not to say everyone should start firing off emails to Facebook demanding the source code to their algorithm. But I do think it’s important to have a broad understanding of how algorithms work at a high level.

This recent article from Farnam Street is a great place to start. At the most basic level, algorithms are step-by-step procedures designed to solve a problem or achieve a specific goal. They are, more or less, computer-based mathematical operations that make decisions based on the input of data. Although algorithms generate decisions, they are not devoid of human guidance. This is one of the most important points to understand about algorithms.

Algorithms will never be a perfect replacement for human decision making since humans write them to begin with. As a result of this, we often code our biases into algorithms. Humans aren’t perfect, and as such we shouldn’t expect our algorithms to be either. An algorithm is only as good as the data it’s built on, and if that data isn’t representative of the entire population or universe of potential outcomes, it will innately have a bias.

Given these constraints, it’s important to consider the limitations of algorithms. Not everything is quantifiable. With that in mind, we should view algorithms as an auxiliary tool to support human decisions rather than a tool designed to supplant them. If used the right way, algorithms can be a force for good that propel our world to a better place for everyone.

---

Deeper Dive

Seeing Like an Algorithm: James Scott speaks of “seeing like a state,” of massive shifts in fields like urban design that made quantities like plots of land and their respective owners “legible” to tax collectors. TikTok’s design makes its videos, users, and user preferences legible to its For You Page algorithm. The app design fulfills one of its primary responsibilities: “seeing like an algorithm.”

At the outset of August, I shared Eugene Wei’s fantastic piece on TikTok called TikTok and the Sorting Hat. At the time, this was the best article I had come across on explaining TikTok’s explosion. It would still remain the best article on TikTok had Eugene not followed it up with this incredible encore.

In part two of his discourse on TikTok, Eugene peels back the layers of the algorithm that makes TikTok so successful. And as the layers recede, Eugene makes it clear that it actually isn't the underlying algorithm that is powering TikTok's rapid growth. Rather, it’s the magic of the app's overall design. TikTok created a closed feedback loop that inspires and enables both the creating and viewing of videos that the algorithm then trains on.

This is incredibly important to understand, especially as the whirlwind “negotiations” around TikTok’s “sale” continue to unfold. With every passing day, there is new information on whether ByteDance will give up TikTok’s source code to Oracle (or any eventual acquirer). This source code, aka TikTok’s algorithm, has been considered to be the most vital component of the company. Without it, you have no TikTok. At least that is what we have been lead to believe.

But in reality, it’s the flywheel that TikTok has created with its design that makes the business so unique. That’s not to say that the algorithm TikTok built isn’t amazing. It is undoubtedly top-notch. Yet, technically speaking, it isn’t irreplicable. A competitor like Facebook could likely replicate the algorithm at some point given enough time. But what they can’t replicate is the flywheel TikTok has created.

The design that powers TikTok’s flywheel is what Eugene has dubbed “algorithm-friendly design”. The entire product is optimized to feed the underlying algorithm as much useful signal as possible. This, in essence, allows the app as a whole to “see like an algorithm”. As soon as you open up a video in the app, any single action that you take is registered and signals your sentiment about the video to the algorithm.

This then creates a self-reinforcing loop that continues to fine-tune your experience on the app. When you then multiply this at scale to the entire user base of TikTok, it starts to make sense how powerful this feedback loop really is. TikTok is gathering billions upon billions of signals and impressions from their users every week. With each of these signals, the experience gets better and more addictive for each individual user.

Despite my general dislike of the app itself, and my unwillingness to download it, I can’t help being enthralled by the business TikTok has built. At the moment, there is no denying TikTok has captured the zeitgeist. Whether that continues for the long term remains to be seen. But regardless of the app's future relevance, the underlying design that powers TikTok’s flywheel is here to stay.

---

Chum Bucket

• Nvidia’s Integration Dreams (Stratechery)

• The Big and the Small (Wait But Why)

• Mark in the Middle (The Verge)

• Is Oracle Biting Off More TikTok Than It Can Chew? (The Information)

• A Closer Look at Luna, Amazon's Cloud Gaming Service (Engadget)

---

Tweet of the Week

Turner Novak @TurnerNovak

Quibi would've been a $100 billion company by now if they'd have just pivoted to day trading content in March

Song of the Week

Apple Music Link

---

Books

Currently Reading

Recently Read

• The Lessons of History is nothing short of a masterpiece. Will & Ariel Durant’s achievement of distilling the most important lessons of over 5,000 years of history into 128 pages is truly remarkable. There are very few books I’ve read that offer as many interesting takeaways on a per-page basis. This is a book that I’ll repeatedly be coming back to my notes on for a long time. (5/5)

Parting Thoughts

This Week in History

• On September 23, 2008, Google and T-Mobile introduced the HTC Dream, the world’s first Android based smartphone. Android has since become the most popular smartphone platform in the world based on raw sales. (Source)

“The impediment to action advances action. What stands in the way becomes the way.”

— Marcus Aurelius

If you found something that piqued your interest this week, please help me out in expanding the reach of The Chomp by forwarding it along to a friend or sharing it with others in your network. Until next week.

-CM

This newsletter is created and authored by Cody McCauley and is published and provided for informational purposes only. The information in the newsletter solely constitutes Cody’s own opinions. None of the information contained in the newsletter constitutes—or should be construed as—investment advice.