Welcome to Stuff I Thought About Last Week, a collection of topics on tech, innovation, science, the digital economic transition, the finance industry, giant herds of giant sloths, 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 signup form; if you have any issues or questions please email sitalweek@nzscapital.com)

In today’s post: today is a special early edition of SITALWeek in honor of the 02022020 palindromic date and football – neither of which is a topic of today's newsletter – instead we have: power sharing power laws; Amazon’s Ring spyware; bringing depth back to digital audio; YouTube is winning with gamers; shifting power from connected TV makers like Roku and Amazon Fire to content owners highlights vulnerabilities; our Sun like we’ve never seen it before; DCF wrap up; and, lots more below...

Stuff about Innovation and Technology
Smart Toothbrushes
Wired brings us an article about toothbrushes that use RF waves to send charged molecules to teeth, optical sensors that measure film buildup, AI that identifies and grades your cleaning of 16 different zones in the mouth, and so much more. Connected apps show you your brushing time and percent tooth coverage. And, with oral disease connected to Alzheimer's, why not make brushing smarter?

MIDI 2.0
Musical Instrument Digital Interface (MIDI) was a standard introduced in 1983 to convert audio into digital and today “[MIDI] is now at the core of music making in the entire music industry, with the possible exception of classical music and acoustic based music which doesn’t interface with computers.” MIDI 2.0 has been a collaborative effort amongst the MIDI Manufacturers Association and all the major music streaming services to expand the original from 7 bit to 32 bit. That means we go from only 128 increments available to billions, and the new protocol allows instruments like strings to sound much closer to their original intended effect. If you’ve migrated to streaming for all your music listening, it’s quite interesting to go back to a high-quality original source (records, or even CDs) and hear how much more depth there is. That’s not directly related to the MIDI issue here, and I am by no means an audio snob, but we do lose nuance in streaming (don’t get me started on streaming video’s audio problems!), and it would be great to build that quality back in.

Cloud Software Stars
Okta’s 2020 Business @ Work report is out detailing the rising stars of cloud software. The report shows which companies are growing the most user logins at Okta’s customers, so it’s not direct usage growth, but fastest growing adoption amongst those adopting Okta’s identity platform (which has some other skews to the data). Companies that are accelerating away from the competition include Microsoft, Atlassian, and Zoom. In terms of opportunity to be a pervasive digital platform for companies, I thought it was interesting that ServiceNow inched ahead of Salesforce in total monthly active users in early 2019. The number of apps per customer rose 6% to 88 as cloud adoption shows no signs of slowing. Also, there are a lot of details on the security software market at the end of the report.

Content winning Tug of War with Distribution
Roku and Fox got into a carriage dispute this week in a manner akin to a traditional pay-TV blackout. Back in SITALWeek #218, I argued: “streaming apps like Netflix could/should demand higher “carriage fees” from video distributors like Amazon Fire TV, Roku, Apple, and even legacy bundlers like Comcast...The situation is due for a complete inversion of who pays who, or, at the very least, the distribution platforms' cut should be zero.” In the old world of pay TV on cable and satellite, it mattered where and how you watched video – there was value in the distribution because of the high switching costs to the user and the capital required to provide the service. But today, it's easy and cheap for consumers to access streaming apps on multiple platforms, including connected TVs, phones, etc. In some cases there is even nested access: I can watch CBS All Access on the Prime Video app on a Roku box. Connected-TV hardware platforms in the US – primarily Amazon Fire, Roku, and Apple – get a cut (generally 15-30%) of revenues when they sign up a new subscriber to streaming video app; and if it is AVOD (advertising video on demand), they can sell targeted ads as well. However, where and how I consume video content (i.e., distribution) today has materially declining value (with the exception of data the connected box makers have to target advertising). Cable has seen its video margins shrink over the years as content demands higher and higher fees, but with streaming, revenue should go 100%, or perhaps more than 100%, to the apps and content owners. The dispute between Roku and Fox (a long-time supporter that still owns 5% of Roku shares) shows that content is flexing their muscles in the tug of war with distribution; and, if the situation with cable/satellite is any indication, content will continue to gain ground. The more valuable a consumer is to a broader ecosystem, the more access fees the ecosystem could pay to apps like Disney+ or Netflix, subsidized by e-commerce at Amazon and iPhones at Apple. Standalone Roku does not have an ecosystem to leverage. Content is king more today than ever before, and it will continue to have its cake and eat it too (more on that in SITALWeek #217).

The New Eye of Sauron: One ‘Ring’ to Rule Them All
Bezos has accused Saudi Arabia of planting malware on his phone to spy on him; but, it turns out they could have just paid Bezos’ connected doorbell and security company Ring for the data. The EFF finds an egregious abuse of privacy from the app, which has been siphoning off and selling all sorts of personal information on users. In related news, several hundred Amazon employees contributed to this Medium post in an attempt to break Amazon’s stifling employee communication policies – one of them had this to say about Ring: “The deployment of connected home security cameras that allow footage to be queried centrally are simply not compatible with a free society. The privacy issues are not fixable with regulation and there is no balance that can be struck. Ring should be shut down immediately and not brought back.”

Major ESports Win for YouTube and Google Cloud
In a mega partnership, video game maker Activision has signed a deal with Google to make YouTube the exclusive home of their growing esports programming, and Google Cloud the preferred host for “Call of Duty”, “World of Warcraft” and more (curiously, Google’s new streaming game platform Stadia was not part of the announcement). Ryan Wyatt, the dynamic head of gaming at YouTube (and former esports commentator) has spearheaded a very successful campaign to grow YouTube into the premier place to watch game playing live. Notably, Activision currently hosts their games on their proprietary tech stack at a data center co-location company. So, this is a fairly big deal that such a huge, classic workload for co-located private cloud would be re-architected to GCP. It raises the broader question of whether it’s a matter of when, and not if, the large cloud platforms have 100% of computing workloads. 

2019 Phone Stats and Flattening Power Laws
In 2019, Samsung held top share in phones at 21.8% – thanks in part to the advantage of its strong supply chain in displays and semiconductors. Huawei gained 2.8% to beat Apple to the #2 spot at 17.6% (helped in part by nationalistic buying in response to US trade bans on Huawei) and Apple declined modestly to 14.5%. Interestingly, this is quite a flat distribution for smartphone market share compared to typical power laws we see with information platforms. However, if we look at these data from a software perspective, we have one large global Android winner, one large Chinese Android winner, and one large global iOS winner. In other words, we have three power-law winners in three different segments, which, in aggregate, makes for a fairly even distribution of hardware market share. This scenario reminds me a little bit of the flattening power law I mentioned in video games a couple weeks back – another market in which there are multiple, large, audience subsets that, combined, create a flatter hierarchy (several power law winners, each in their own large niche). Social networking feels like another area that has a flattened power law, with centers of gravity at Facebook, Instagram, Snapchat, TikTok/Bytedance, Tencent, Twitter, and WhatsApp (acknowledging FB owns several of these). Power sharing at the head end of a power law distribution seems like an interesting trend to watch evolve as billions of people are increasingly online and consuming digital apps and services. Notably, in streaming audio and video, it’s not clear whether we will see a power-law winner takes most or if we will again see these flattened centers of power. However, e-commerce clearly remains a classic power law in the West with Amazon running away with it. 

Mobile Spend is All Fun and Games
Gaming was the major driver behind the $120B mobile app spend last year, with subscription revenue from the top-100 non-game apps contributing a mere $4.6B (how is that so low?). Tinder was the top-grossing app, according to data from Sensor Tower. In related analysis by Mint for the NYT, Americans spent an average of $640 last year on digital subscriptions including video, music, and apps, up 7% from 2017. The fastest growing category was video apps, which rose to $170. 

Qualcomm’s First RISC-V Chip
One of the 12 embedded processors in Qualcomm’s Snapdragon 865 flagship mobile chip is based on open-source instruction set RISC-V, likely via SiFive, according to this report (link is in German, but I used Google translate).

Video Analysis a Tough AI Nut to Crack
In a Lex Fridman podcast interview with the head of YouTube’s search algorithm development, we learn that using machines to analyze video content is far harder than developing autonomous vehicles, and we are less than 25% of the way to accomplishing this goal. And, much of our progress is just in looking for predetermined triggers and patterns. It seems like a worthy goal to be able to interpret videos entirely, which means Google will need an enormous amount of computing power and/or a breakthrough in machine learning. (Spotify Link)

Miscellaneous Stuff
Stunning Solar Surface Images
From the DKI Solar Telescope in Maui comes the most mind-blowing moving image of the surface of the sun so far produced. I will never look at the sun the same again (also, don’t ever stare at the sun, that’s dangerous!). The 4-meter mirror isn’t the largest by any means, but it’s the largest scope specifically built to observe the sun. The shape of the mirror adjusts 2,000 times a second to overcome distortions in the Earth’s atmosphere. Thanks to its subject, the scope gets hot enough to melt metal, so it relies on 7.5 miles of piping and a swimming-pool-sized reservoir of ice as a heat sink to drop the Watts on the optics from ~12,000 to ~300. For telescope geeks (like me!) – DKIST is an Alt-Az, off-axis Gregorian reflector (which means there is nothing blocking the mirror).

Out of Africa, and Back Again, And Out Again
Neanderthal genes have been found in significant levels (0.3%) in African populations. It appears that humans left Africa around 200,000 years ago, migrated North, and began interbreeding with their cousins (modern European and East Asian populations have ~1% Neanderthal DNA). Anthropologists have also shown human/Neanderthal DNA mixing around 60,000 years ago. These data suggest more back and forth migrations and interbreeding than previously believed. Other recent finds, according to this detailed NYT review, suggest Neanderthals were not brutes, but in fact were more similar to humans than imagined in all of our childhood science books. 

Particle Accelerator on a Chip
Researchers at Stanford and SLAC have built a particle accelerator on a chip. The device uses infrared photons (which can pass through silicon) pulsing 100,000x/s to speed up electrons in a nanoscale channel. The team is working to get the chip up to 1000 cycles and accelerate electrons to 94% the speed of light, which would create 1MeV of energy that could be used for medical purposes (cancer irradiation) or other research. Or maybe we could make a modern-day Funken Ring? (That’s a magic gag you hold in your hand that shoots sparks several feet out from your fingers.)

Massive Gravity, not Dark Energy?
Do gravitons, the hypothetical, gravitational-force-creating particles that are assumed to be massless, actually have a mass? If so, it could explain the acceleration of the universe and solve the dark energy puzzle by not requiring it to begin with. Professor Claudia de Rham at Imperial College London believes this is the case, and she is getting some traction. While the math works, it’s a theory for now until it can be experimentally tested by gravitational wave astronomy. De Rham also has other intriguing theories, like gravity may move faster than the speed of light.

Savage PR for Spot 
Adam Savage unboxes his new Spot robot dog from Boston Dynamics on his Tested channel on YouTube. Adam will be working with the robot throughout 2020 as part of a marketing effort by the SoftBank-owned company to build a business selling the robots. The goal is to build an ecosystem of developers and sensor makers to create use cases for this amazing feat of engineering.

Stuff about Geopolitics, Economics, and the Finance Industry
PE Flippers
Median holding periods of companies by private equity, which peaked in 2014 at a little over 6 years, are now down to 4.9 years. Notably, the top quartile of holding periods (i.e., the top 25% of longest-held deals) has dropped the quickest – from almost 9 years in 2016 to 7.1 years in 2019. Most shortly-held deals have stayed pretty consistent at 2-3 years, and Pitchbook believes longer-held deals could grow in the mix as more and more long-dated funds are being raised. Brent Beshore also weighed in with some smart and balanced commentary on not vilifying private equity in this Twitter thread.

PE Goes to the Dogs
As the Economist reports that “Everyone now believes that private markets are better than public ones,” PE firm GF Sports is targeting the dog show industry in a new deal with the American Kennel Club.

ANTs
Active non-transparent ETFs, becoming known as ANTs, were a hot topic at the Inside ETFs conference this week according to Citywire. “A big topic was how significant of an impact open-end mutual fund conversions would have on the market. If asset managers convert mutual funds to active non-transparent ETFs, bringing its assets and track records into the new vehicle, the active non-transparent ETF’s popularity would skyrocket.” 

DCF Follow Up
Following up on last week’s much-shared post on the problems with DCFs (well over 50,000 engagements proves that Twitter's algorithm is truly great at surfacing silly content to the top!) I want to add some color that seemed to help people understand our points better. Many folks like to use DCFs for implied assumptions (the average market view), or alternatively they will set the discount rate to their own personal investment return hurdle rate. I understand that inclination; however, let’s say you use a discount rate of 10% today because you want all your investments to be “undervalued” today with a “margin of safety” using that return hurdle. But what if, three years from now, one or more of the following increases your hurdle rate: 1) your personal circumstances cause your risk appetite to decrease (maybe you get sick, or lose your job), 2) macroeconomic factors cause rates to rise dramatically, or 3) the company you invest in sees a significant widening of its range of outcomes due to technological, market, or other unpredictable forces. Any of these scenarios would imply that the 10% you use today should be perhaps 15%, or maybe 20%+ (or maybe 3% if these factors go in the other direction). Your margin of safety is just an arbitrary illusion, but the effort and brain cycles you put into a fancy spreadsheet with color-coded cells, etc., reinforces your own cognitive bias, and feeds a false sense of confidence. It’s about as useful as a visit from Tommy Callahan’s Guarantee Fairy. So, in a world governed by complex adaptive systems that exists in nonequilibrium, unless you are the all knowing LaPlace’s Demon, margin of safety must be redefined as a more useful tool for risk management. This is what we have tried to do with our papers Redefining Margin of Safety, in terms of adaptability and time, and Complexity Investing, applying complex adaptive systems to investing. But, we often get things wrong, and we are always learning and improving. If you’ve read our work and disagree, we would love nothing more than to debate in the spirit of intellectual honesty and truth seeking.

✌

Disclaimers:

The content of this newsletter is my personal opinion as of the date published and are subject to change without notice and may not reflect the opinion of NZS Capital, LLC (“NZS”).  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. I 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 (“NZS”). 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 has no control. In no event will NZS 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.