Sunday Snapshots (09/06/2020) – Uncertainty Pt. 2, A/B Testing, Publishing and e-book reading woes, Meta-layers, Mosquitos, and Caravaggio’s lighting
The one in which I get a bit angry while talking about e-books
Greetings from New York City!
I’m in the city for the long weekend here in NYC.
Verdict: definitely not done.
It’s my first time here since the end of February. It’s a different place, yet retains the energy that keeps me coming back. Much of the city has re-opened, yet there is a touch of un-easiness as people mostly go about their business. The city is certainly not dead, yet it has been re-born. I’m left wondering if this is what the great cities of the past looked like after the plague or some other disease went through it.
Side note: Taking the Acela up to NYC from DC reminded me of how objectively terrible the flying experience is. Trying to time your arrival at the airport to account for TSA lines, waiting around at the gate, the auctions when a flight is overbooked, the pressurized air, the kayfabe-laced emergency videos – I could keep going. Trains? You get there 10 minutes before your departure time, you walk up to the train and take your seat. Smooth. Frictionless. Maybe even the highest compliment — delightful. So much work to be done in terms of the flying experience. Planes are magical, flying is not.
On to this issue of Snapshots, in which I want to explore:
The Uncertainty Mindset Pt. 2 by Vaughn Tan
The power of A/B testing with fat tails
Publishing and e-book reading woes (where I get angry on the newsletter for the first time)
Meta-layer of notes, Disney mosquitos, and Caravaggio’s lighting
Book of the week
Constant testing and feedback
Back in 9th grade, my Geography teacher said something that’s stuck with me:
The most valuable thing I can do for you is not to teach you things, but to provide meaningful feedback as fast as possible on your work.
Cut to some 7 years later and Mr. McLean is providing fodder for an early morning writing session of my newsletter!
Vaughn brings up the same lesson in the context of the best R&D chefs in the world. Since the roles on these teams are less well-defined than traditional jobs, every day is an opportunity to get and receive feedback. Every job is a little test.
By getting multiple data points and course correcting over time, new chefs can avoid amateur mistakes and not have to wait for prime time dinner service to get feedback.
Same same, but different
If you’re a high-end R&D restaurant, your customers expect you to innovate. At the same time, they expect you to remain distinctly you. They may be wowed by your innovations, but they pay the premium for your style.
This actually creates some useful guardrails and constraints for what is otherwise the extremely open-ended task of creating new dishes.
This extends beyond cooking. The ability to take the same building blocks and create a new thing that is uniquely theirs is something that the top performers in any creative field are very good at doing. Writers and painters are other easy examples.
The type of people that join these kitchens will not stay long if not constantly provided with new challenges. So these teams engage in what Vaughn Tan calls “progressive overload.”
This allows them to inject uncertainty once their skills start stagnating. The point is simple, but it hit home for me.
The points I made last week about the lack of the written medium to do justice to this book still apply:
This book has some great stories. But the problem is exactly that – it's a book. These are stories that are held back by the medium with which they are told. When Vaughn describes cooking processes and colorful personalities, I struggled to keep along. If last week's book should be a Paul Greengrass–directed spy thriller, this should be a Dan Gelb–directed meta-episode of Chef's Table or a standalone film. Even a Craig Mod–styled book could work. The more fundamental question here is how to turn what you see, hear, and smell into text? It's a translation and legibility problem.
I’d like to add another piece of meta-commentary. While Vaughn does an amazing job of the showing how the best restaurants in the world are able to the innovate, restaurants doesn’t exist in a vacuum. They are a part of a complex system with many moving pieces. While you can be extremely flexible within a small team, the uncertainty emerging from that flexibility has to be absorbed by your suppliers, logistics providers, and other 3rd parties. How do you deal with them if they don’t operate under the same mindset and may have legitimate reasons not to? What happens when nimble, innovative teams like those described in the book inevitably meet complex, slow moving teams? How do they interact with each other and get things done?
Long read of the week
A/B Testing with Fat Tails by Eduardo M. Azevedo, Alex Deng, Jose Luis Montiel Olea, Justin Rao, and E. Glen Weyl
A great paper on A/B testing with one of the best abstracts I’ve ever read in terms of clarity:
Large and statistically powerful A/B tests are increasingly popular to screen new business and policy ideas. We study how to use scarce experimental resources to screen multiple potential innovations by proposing a new framework for optimal experimentation, that we term the A/B testing problem. The main departure from the literature is that the model allows for fat tails. The key insight is that the optimal experimentation strategy depends on whether most gains accrue from typical innovations or from rare and unpredictable large successes that can be detected using tests with small samples. We show that, if the tails of the unobserved distribution of innovation quality are not too fat, the standard approach of using a few high-powered “big” experiments is optimal. However, when this distribution is very fat tailed, a “lean” experimentation strategy consisting of trying more ideas, each with possibly smaller sample sizes, is preferred. We measure the relevant tail parameter using experiments from Microsoft Bing’s EXP platform and find extremely fat tails. Our theoretical results and empirical analysis suggest that even simple changes to business practices within Bing could increase innovation productivity.
Business move of the week
How to overcome publishing woes
Problem: Every author who was going to release a book in the spring/summer this year pushed it back to the fall in hopes that things will be better. Now, those books and the ones that were supposed to release in the fall anyway are now all releasing at the same time. So printing shops are not able to keep up with demand from publishers. This is leading to long wait times on popular books to the tune of 1-2 months like it’s the new iPhone. Story here.
Solution: Re-design the e-book reading experience.
Let me explain.
It should come as no surprise to any Snapshots reader that I’m interested in book publishing. What may come as a surprise to many is that despite being an early 20-something, cold-brew-chugging, Patagonia-sweater-wearing person that I am, I still almost exclusively read paper books.
It’s not because of some outdated idea of how it feels real to read the hard copy of a book or because I like having the finished book as a memento. Nope. It’s because the e-book reading experience sucks.
I’ve had a Kindle for a while. It’s slow to turn on, slow to open a book, and slow to flip through pages. Oh, and the highlighting and taking notes? Don’t even get me started. I would describe myself as a fairly collected person but adding a note to a phrase or a paragraph on a Kindle is rage inducing. I’ve just given up trying to use it.
Sometimes, I’ll read on the Kindle or iBooks app on the iPad and while the experience is a bit better, distraction is just a touch away at all times.
Amazon’s and Apple’s duopoly on the e-reading space has led to stagnant innovation. They’ve cornered the market on digital books and have no reason to do anything new.
Taking it full circle, we shouldn’t have worry about whether a printer can physically print enough books in 2020! We live in the golden age of technology. We should have amazing e-readers! The duopoly harm caused by these two companies in this space is deeply painful because at the end of the day, books are about ideas. Any friction in your ability to read a book is friction in your ability to download ideas into your head. That download speed should not be limited by poor tools.
In an effort to calm down and be constructive, here are 5 things my ideal e-book reader would have:
Fast page turning
Easy to read screen
Light to hold
Ability to hand write annotations on the margins using some sort of a stylus that get turned into text and stored in another menu along with the original context (willing to swap with fast typing experience)
Access to a marketplace of books and PDFs
Is anyone working on this? Am I being too dramatic? Do you like your Kindle? Am I missing something? Please let me know by replying to this email or by sending me a DM on Twitter.
Odds and ends of the week
Three quick links that’ll take you down three different rabbit holes:
💡 Caravaggio’s lighting: When the Nerdwriter channel puts out something, I just watch it. No questions asked. I think you should do the same.
📝 Meta-layer of notes: Julian Lehr starts with asking a simple question – how can we create the digital equivalent of the stick note in its ability to be context-aware? He ends up writing one of the best essays of the year. Read it in its entirety. Then compliment with Andy Matuschak’s Why books don’t work.
🦟 Mosquitos at Disney World: This one was equal parts entertaining and informative. It also just goes to show how much behind-the-scenes works happens in order for something to just work. Enjoy!
That wraps up this week’s newsletter. You can check out the previous issues here.
If you want to discuss any of the ideas mentioned above or have any books/papers/links you think would be interesting to share on a future edition of Sunday Snapshots, please reach out to me by replying to this email or sending me a direct message on Twitter at @sidharthajha.
Until next Sunday,