Welcome to Stuff I Thought About Last Week, a collection of topics on tech, innovation, science, the digital economic transition, the finance industry, complexity, and whatever else made me think last week. Please grab me on Twitter with any thoughts or feedback.

Click HERE to SIGN UP for SITALWeek’s Sunday EMAIL (please note some ad blocking software may disrupt the sign up form; if you have any issues or questions please email sitalweek@nzscapital.com)

In today’s post: retail’s margin and logistics conundrum; Netflix Shuffle; Cheetah Mode; Hot Chips; rising ransomware; portfolio construction, gardening, and risk – don’t let your narrow winners run; new business formation accelerates; a terrific new review of complex adaptive systems science; and lots more below...

Stuff about Innovation and Technology
Rootsistance is Futile
RootWave is a UK-based company that kills weeds with electricity at a comparable cost to herbicides with no known harmful side effects (unless you’re a fly on one of those weeds ⚡). IEEE reports that electric weed zapping has been around since the railroads used it over one hundred years ago, but newer systems run at much higher frequencies with smaller transformers and pose little threat to operators. Importantly, in contrast to applied chemical pesticides, weeds are unlikely to evolve high-voltage resistance and don’t need to be tilled, which keeps more carbon in the soil. RootWave partners are working on handheld wands and robotic versions that use machine vision to precisely target weeds.

Model S Cheetah Mode
Back in SITALWeek #232, I mentioned Tesla’s disappointing track showing vs. the Porsche Taycan. While fairly even on the first run, Tesla’s ability to maintain peak performance slipped in subsequent track circuits. Recently, Tesla redeemed itself in a new Car and Driver track test by matching most of the Taycan numbers. The improved results were achieved with just a software update called Cheetah Stance.

Transportable Hydrogen Surrogate
The Korean Institute of Science and Technology (KIST) is working on a way to use ammonia to transport hydrogen. Hydrogen fuel has the potential to serve a diverse array of energy and transportation needs; however, the infrastructure to move it around globally is not established, in part because the gas cannot be safely condensed for transportation. Liquid ammonia, which can be moved using existing transportation networks, can store ~1.5x more hydrogen (per unit volume) than liquified hydrogen. The KIST team’s breakthrough is in their low-cost device (a catalytic membrane reactor) for decomposing the ammonia into nitrogen and hydrogen and cleanly capturing the latter. This advance could make it easier to move hydrogen from where it can be produced with excess energy (solar, wind, etc.) to where it’s needed more. 

Fitbit Early-Detects Nearly Half of COVID Cases
Early results from Fitbit’s COVID pilot study demonstrate an ability to identify almost half of sick patients at least a day before they noticed any signs of illness themselves by using pulse, respiratory, and sleep quality data. Despite these promising early results (whose precision and accuracy will no doubt increase as more data are gathered) smartwatch shipments were flat in Q2. Apple grew smartwatch revenue market share to just over 50% in the first half of 2020 as the overall market rose 20% (dollars on flat units). That lack of growth is certainly surprising given the heightened awareness – and importance – of health monitoring during the pandemic. While Apple continues to do ok, there is no obvious contender to democratize and rapidly grow the market; but, I still believe wearables will blossom into a billion+ unit market in the coming years with a big opportunity to differentiate with fashion and luxury. Although I remain a happy Fitbit owner, Google’s acquisition of the 2.4% market-share smartwatch maker is still in regulatory limbo, so they aren’t likely to expand their market anytime soon, and Android Wear remains a small player with 10% share of wearable operating systems.

Netflix Shuffle
Who’s old enough to remember just flipping on the tube to see what’s on? Sometimes, viewers want to watch something specific; but, most of the time, everyone is channel surfing in the traditional cable sense, which is perhaps why Netflix continues to pursue its quantity-over-quality strategy. Variety reports on Netflix’s new “shuffle” feature, which automatically chooses something it thinks you want to watch based on prior viewing history. This seems more like a social newsfeed approach to content, where an algorithm completely takes over to feed our short attention spans (a key survival tactic for our ancestors in order to identify threats, so it’s programmed deep in our neurons) further diminishing the value of any one piece of content. My current split of time on Netflix is 90% looking for something good to watch, and 10% deciding which higher-content-quality app to switch to – lately, the Australian New Wave collection on Criterion or Hulu or HBO Max.

Current Ecommerce Unsustainable
Driven by online ordering, quarterly spending on groceries rose 31% to $387/person in Q2, as people ate out less and stockpiled goods, but this growth is coming at the expense of profits (see #254). Today, every retailer is trying to deliver every item to everyone regardless of whether it’s a high margin article of clothing or a low margin pack of bottled water (Amazon calls the latter items CRAP: Can’t Realize a Profit). That strategy won’t last, especially in light of the tight, under-capacity delivery networks (covered in more detail last week) and the declining network effect of UPS and FedEx in the US as they raise prices. As ecommerce continues to grow, the current irrational strategies are likely to give way to some combination of the following sustainable methods: 1) hard-to-find and/or high-margin items, for which online suppliers can charge enough to cover the ever-rising shipping costs, will continue to be fulfilled centrally and delivered for free; 2) CRAP and/or lower-margin products will either shift to click-n-collect, with a nominal fee to cover the store’s added labor, or local delivery with much higher fees than what are charged today. Amazon – now delivering two-thirds of its own shipments – is in the position to span all these buckets and continue to subsidize delivery with the Prime flywheel. Ultimately, I think the only survivors in retail will be vertically integrated platforms subsidizing the shipping component via memberships, ads, and product margins, which is a very different approach than most of retail today.

Apple Fortifies Secret Volcano Lair
Apple’s decision to retaliate against Epic games – by making it so that the company’s Unreal Engine eventually won’t work on Macs – is surprising given Apple's grand ambitions in the TV and movie industry, which increasingly relies on Epic’s gaming engine to make movies (see SITALWeek #189). I covered Epic’s lawsuit against Apple in more detail last week, but on a scale of 1 to Dr. Evil, Tim Cook is breaking the mercury. From Epic’s injunction request to the court: “Just over two weeks ago, Apple’s CEO Tim Cook was asked during a Congressional hearing whether Apple has “ever retaliated against or disadvantaged a developer who went public about their frustrations with the App Store”. Mr. Cook testified, “We do not retaliate or bully people. It’s strongly against our company culture.” But Apple has done just that. When Epic gave users of its app Fortnite a choice of how they wanted to make purchases, Apple retaliated by removing Fortnite from its App Store. Then when Epic sued Apple to break its monopoly on app stores and in-app payments, Apple retaliated ferociously. It told Epic that by August 28, Apple will cut off Epic’s access to all development tools necessary to create software for Apple’s platforms—including for the Unreal Engine Epic offers to third-party developers, which Apple has never claimed violated any Apple policy.” 

News from Hot Chips
Brinton virtually attended the Hot Chips conference last week. He reports that the rising architecture wars, as well as changes in semiconductor transistor design, stood out among all the announcements. RISC-V continues its meteoric rise, with Alibaba claiming their new open-sourced chips outperform Arm's best design from just three years ago. Startup Manticore also announced RISC-V chips for the data center. ARM continues to gain traction as well, with Marvell showing off their latest ThunderX3. On the geekier side of event news, the field-effect transistor (FET) is evolving again from planar and FinFET to gated-all-around (GAA) FET. The design comes in many flavors – e.g., SuperFin, MBCFET, and Nanosheets – and GAA will fundamentally improve the transistor once again by boosting chip performance by around 20%. However, it will also make them much more challenging to manufacture. Here’s another Hot Chips recap from EE Times.
 
Cloud to Play Bigger Role in 5G
Joe weighs in on the expanding role of the cloud as networks move to 5G: Telcos are expected to spend over $1T rolling out 5G to support current and new services such as AR/VR, gaming, surveillance, IoT, and self-driving cars. Back in SITALWeek #240, we highlighted that much of 5G infrastructure and services will be 'cloud first', so it’s not surprising that Google asked the FCC for permission to test the 6GHz spectrum, a frequency band that is likely to be used for 5G. While Big Tech dabbling in telecom isn't a new phenomenon (Amazon has been testing the CBRS band for a few years and Google is a Spectrum Access System administrator), it highlights the increasing role that the cloud computing platforms will play in wireless networks as they centrally enable a slew of new, low-latency apps to consumers on the ground.
 
Ransomware Onslaught 
The CEO of network infrastructure and security platform Cloudflare noted in this insightful interview that state-sponsored attacks on critical systems are on the rise during COVID. Indeed, the rapidly growing ransomware industry continues to drive the modernization of IT systems and software with a special emphasis on managing identity and access, as well as data encryption. There have been several, recent, high-profile attacks, such as this one at the University of California San Francisco, resolved through high-stakes negotiations with savvy hackers. The hackers locked down several UCSF servers storing critical data, which may have been tied to COVID trials, and released them for 118 bitcoins ($1.14M at the time of payment).

Miscellaneous Stuff
Harnessing Uncertainty with Emergent Engineering
Aeon published a fantastic article from Drs. Melanie Mitchell and Jessica Flack summarizing the importance of using complex systems science in decision making and planning in our increasingly unpredictable world. We took an introductory course to complex systems from Dr. Mitchell eight years ago, which heavily influenced the evolution of our investment process and resulted in our 2014 whitepaper, Complexity Investing. The article is rich with insight, and I highly recommend reading it in full. 

Heated Hexagonal Huddle
Heat-seeking penguins form fluid huddles with hexagonal patterns. This geometric shape has been long-known by scientists (and, apparently, Antarctic-dwelling emperor penguins) to allow for the most density on a flat surface. At a specific threshold temperature (which depends on ambient conditions and penguins’ body-fat stores), huddle formation is triggered. After the penguins form an initial, amorphous huddle, there is a constant flux of penguins moving away from the windy, cold side toward their warmer, leeward neighbors. As the penguins reorganize, they optimize their hexagonal grid pattern so as to (it appears) maximize the interior/exterior penguin ratio. The interior penguins can keep toasty in the huddle at 100 ℉ (38 ℃).

Stuff about Geopolitics, Economics, and the Finance Industry
Business Application Boom
After seeing the rising new business applications data in this Axios report, I went to the US Census Bureau data site to see what the actual year-over-year growth in new business applications was in the US. For the last four weeks, the average growth in new biz formation was 77% y/y (in order from earliest to most recent, weekly changes were: 84%, 83%, 75%, and 69% y/y). It appears there is a lot of energy going into starting new businesses during our current economic upheaval. No doubt the big Internet platforms are playing the role of enabler – whether it’s Shopify or Square for direct-to-consumer retail, Google with ads and productivity apps, FB with ads, Gusto for HR and payroll, etc. It’s never been easier to stand up a business (not that it’s easy by any means, even with all these new tools). This next generation of businesses will be digital first, which will cause a ripple of disruption to the status quo of business operations across the economy for years to come. 

Xinomics
The Economist has a long article explaining the accelerating shift to increase state control of more companies in China and deemphasize free market dynamics (which were integral in fostering many of the powerful Chinese companies over the last two decades). While the magazine writes that we shouldn’t underestimate this power grab (and corresponding freedom suppression) from Xi Jinping, I would suggest that the history of complex systems points to centralized control being more fragile than resilient.

Investment Portfolio Gardening
At NZS Capital, we are always reminding ourselves that no one can predict the future, even with perfect information. That notion can feel uncomfortable, and even a bit paralyzing. We do, however, know that there are certain elements of a company’s culture and products that afford better odds of future success. Chief among those are a company’s adaptability and the level of NZS, or non-zero sum, they provide – the more value they create for others, the more value they will ultimately create for themselves down the road. One way to frame a hypothesis regarding a future outcome, without having to rely too much on a crystal ball, is like this: “In five to ten years I think blank will happen, but I don’t know how it will happen, what it will look like, or who will win.” Example: In five to ten years, people will watch more video content, but I don’t know what types of devices it will be on (screens or AR/VR glasses?), what type of content it will be (scripted Hollywood, interactive life streaming, immersive metaverse gaming, casual games, etc.), or who will be making more of the content. 

We can separate these statements into broad predictions, which are more likely to happen in various futures, and narrow predictions, which are difficult to know ahead of time. Some of the broad predictions would be: 1) semiconductors and connectivity will be in more demand; 2) creativity and storytelling will always be important; 3) data – and smart use of it – will be important; 4) if interactive content rises, gaming engines will be more important in creation of content; 5) cloud computing and AI will be important; 6) people’s tastes will always be fickle and personal, and thus hard to predict. These broad predictions create a landscape in which to plant various narrow predictions and monitor growth and outcomes. Many factors will determine which predictions grow stronger and which wither and die, including, first and foremost, adaptability and NZS, but also network effects, innovation, platform dynamics, etc. If we were to look today at, for example, what content is the most adaptable and highest NZS for consumers, it’s probably video games and life sharing/streaming via social networks or YouTube. These forms of content can change their storytelling easily, provide the largest amount of content per dollar spent, and generally create win-win for all constituents. Some views will be Resilient – fairly broad and safe predictions (e.g., semiconductors will do well), and some will be Optional – narrow predictions (e.g., specific content/video game will gain share). This process is at the heart of our investing strategy and is detailed in our paper, Complexity Investing, which provides a framework for identifying winning companies and constructing a portfolio that balances Resilience and Optionality.

As you water and prune this landscape of broad and narrow predictions (based on a Bayesian approach to objective analysis of new information), it’s important not to let the narrow predictions become giant weeds. Letting your narrow prediction winners run is a false narrative that feeds off a set of cognitive biases (bias traps are also addressed in our Complexity Investing paper). At NZS, we tend to let our Optionality (narrow prediction) positions grow in the portfolio only if the range of outcomes is becoming more narrow – meaning that the prediction itself is becoming more broad (for a good visual of this, see the first diagram in Redefining Margin of Safety). If the range of outcomes has remained the same, but a company’s market value is up, then letting the position grow in the portfolio is akin to doubling down at every spin of the roulette wheel without taking any chips off the table (an obvious and risky mistake given the very narrow chance of future payout). Further, when you prune an Optionality position, you use the excess to seed new Optionality positions and/or water the ones that are experiencing a narrowing range of outcomes. This strategy allows you to maintain the overall asymmetry of the portfolio while not concentrating risk. If you let a winner run even though the range of outcomes is still very wide, then you are explicitly making a large bet on a narrow prediction about the future, which means all you have done is increase the risk in the portfolio. And, in turn, you are starving resources for new Optionality positions. Complex adaptive systems teach us that very narrow predictions are overwhelmingly likely to be wrong. Some Optionality positions end up being giant, adaptable oaks that last decades; however, the majority – including most/all of your cherished favorites – end up as weeds. Who doesn’t love a good extended metaphor?!

✌

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and is subject to change without notice and may not reflect the opinion of NZS Capital, LLC.  This newsletter is simply an informal gathering of topics I’ve recently read and thought about. It generally covers topics related to the digitization of the global economy, technology and innovation, macro and geopolitics, as well as scientific progress, especially in the fields of cosmology and the brain. I will frequently state things in the newsletter that contradict my own views in order to be provocative. Often I try to make jokes, and they aren’t very funny – sorry. 

I may include links to third-party websites as a convenience, and the inclusion of such links does not imply any endorsement, approval, investigation, verification or monitoring by NZS Capital, LLC. If you choose to visit the linked sites, you do so at your own risk, and you will be subject to such sites' terms of use and privacy policies, over which NZS Capital, LLC has no control. In no event will NZS Capital, LLC be responsible for any information or content within the linked sites or your use of the linked sites.

Nothing in this newsletter should be construed as investment advice. The information contained herein is only as current as of the date indicated and may be superseded by subsequent market events or for other reasons. There is no guarantee that the information supplied is accurate, complete, or timely. Past performance is not a guarantee of future results. 

Investing involves risk, including the possible loss of principal and fluctuation of value. Nothing contained in this newsletter is an offer to sell or solicit any investment services or securities. Initial Public Offerings (IPOs) are highly speculative investments and may be subject to lower liquidity and greater volatility. Special risks associated with IPOs include limited operating history, unseasoned trading, high turnover and non-repeatable performance.