SITALWeek #340
Welcome to Stuff I Thought About Last Week, a personal collection of topics on tech, innovation, science, the digital economic transition, the finance industry, avatars, and whatever else made me think last week.
Click HERE to SIGN UP for SITALWeek’s Sunday Email.
In today’s post: minor modifications of the way we perceive reality are beginning to condition us to a major transformation in computing; 3D maps may be the fabric of augmented reality – as important as hardware and operating systems; mission critical companies shirking multi-factor login as White House calls for action; vehicle-2-home charging creating lock-in for car companies; and a new theory of memory formation.
Stuff about Innovation and Technology
Reality 2.0
I’ve been using Nvidia’s Broadcast app for the last few months for video calls. The tool operates as an audio/video input for Zoom, Teams, and other video call apps, with features like background noise/echo cancellation (both for your microphone and audio from other call participants). It also acts as a director, keeping participants in frame as they move around during calls (which people tell me makes them a little seasick when I can’t sit still!). Using a DSLR camera about five feet away in front of a 75-inch screen has virtually eliminated “Zoom fatigue” for me, even on long days of meetings. The Broadcast app operates as a testing ground for various AI tools Nvidia makes available as SDKs to developers for their own apps (you can find the app here, but it does require an advanced Nvidia GPU to function). The suite of video-call AI SDKs is called Maxine, which Nvidia highlighted last week at their GTC user conference. First announced in 2020, Nvidia also has a tool that re-renders your face and eyes in real time, so it looks like you are always looking at the camera. While this feature isn’t yet in Broadcast, I am looking forward to trying it out. Nvidia has other tools that allow for real-time speech translation (also not yet in the Broadcast app). If we examine this set of technologies at a macro level, it's clear we are starting to alter reality more and more in real time. These minor tweaks are conditioning us to accept more profound transformations of our visual reality as we enter the age of spatial computing. At some point in the not-too-distant future, it will seem unreal if something in the real world is not augmented. Last week, Zoom announced avatars that allow you to appear in real time as various animals. This might seem trivial or silly at first, but our perceptions of people and objects are going to change dramatically. To paraphrase the old New Yorker cartoon “On the Internet, no one knows you’re a dog”: In augmented reality, no one knows you’re a human. As this cosplay goes virtual with avatars, our entire concept of human identity is likely to evolve significantly.
Fabric of Spatial Computing
The first IBM mainframe computer was introduced in 1964. Personal computers became common two decades later, in the early 1980s. The age of smartphones began with the iPhone in 2007. While the old technology platforms never die – mainframes and PCs remain large markets today – the successive platforms are orders of magnitude larger than previous generations. We’ve gone from thousands to millions to the smartphone’s billions, and the next platform for connected devices of all types could easily reach into the trillions. In its limited history, computing platforms have shifted around every two decades. We are on the cusp of the next shift to spatial computing, comprising various forms of augmented/virtual reality and context-aware hardware and applications. Prior platforms have shared a common, virtuous circle of hardware innovation, new software, user growth, network effects, and market creation. Historically, winners have not maintained their leadership positions into the next platform, largely due to the bureaucracy, paralysis, and bias that keep big companies from disrupting themselves and evolving. These same inhibitors, and a host of other challenges, will keep companies in all industries from maintaining leadership in their sectors as the economy transitions from analog to digital and from one digital platform to the next. After watching so many companies fail in the tech sector, today’s megatech platforms are well aware of these pitfalls and are keen to invest tens of billions of dollars each in new technology to avoid missing the next big thing. The advent of cloud computing and the massive app ecosystems, with their data lock-in and increasing-returns network effects, will assure that, just like mainframes, today’s platforms will be around for decades to come in some form or another. One theory is that spatial computing will be built on today’s platforms, with apps being more important than hardware. I view the hardware for spatial computing as a jump ball that’s still up for grabs by new entrants. As I noted in Meta-mess:
Over the next decade, I expect we will see some great new platforms emerge for spatial computing, all of which will likely rely on the smartphone as a hub. For the foreseeable future, those phone hubs are controlled by the iOS/Android duopoly, barring some huge regulatory shift that transforms those companies into utilities like power or telecom companies. That means anyone that wants to have a shot at building a spatial computing platform needs to either be subservient to Apple or Google, or they need their own phone platform and end-to-end developer and hardware ecosystem. Android, being open source, offers the best way to jump start a phone platform. Microsoft has been experimenting with Android phones, but we would need to see a complete phone platform from a company like Meta, Epic, or Roblox to take on Apple and Google. I don’t know that a gaming-centric approach to spatial computing will ultimately solve for enough use cases, but if someone were to make a killer gaming-based phone and spatial computing platform, like an Xbox phone, it might have a shot at becoming a third platform. It would be inspiring if the next computing platform was more open, more creator driven, and more distributed, but odds favor incumbents controlling spatial computing with the same potential problems we experience today.
One of spatial computing’s key elements is that people and objects interact with the real world in real time, which means we need real-time maps and context of every object, place, and person. I’ve previously noted (#271) Niantic’s work to create a global map with games like Pokémon Go. Snap is another company leveraging a large user base for creation of a 3D map of the world using existing smartphone hardware. The company’s 250,000 lens creators are being tasked with creating geospatially-anchored AR context for landmarks, shops, restaurants, and more. We will likely see more companies trying to control the hardware and software for tomorrow’s computing by leveraging today’s technology to build out 3D world maps using compelling apps to amass data. It’s far too early to know, but it’s possible this idea of a context-aware, real-time, 3D map of the world will be the fabric of spatial computing. If Apple and Google were to render their hardware and smartphone operating systems proprietary, they could shut out the competition (e.g., as with the rise of AirPods and the fall of competitors like Bose). Full access to native hardware sensors and specialty AI processors is increasingly important as computing moves off the screen and into the world. If there were to be a limiting of access, it would surely force germination of alternate hardware platforms at an accelerated rate.
EV Charger Lock-in
One of the more exciting developments over the next few decades will be the steady decentralization of the power grid. As more homes and buildings have solar, batteries, and connected vehicles, we’ll be creating more flexible, efficient, and resilient systems. But, it’s a long path from here to there with many unknowns and a lot of investment. IEEE Spectrum covered the desires of some of the big auto OEMs to enter the bidirectional charging market. Bidirectional charging allows you to use your EV to power your house during peak demand times and power outages, known as vehicle-to-home charging, or V2H. It’s unfortunately not as easy as plugging your car in, as it requires specialized hardware that connects to your breaker panel. There is a risk we get an Apple-like lock-in, where cars require specialized systems that increase the switching costs from brand to brand when you want a new car. Tesla has been quiet about when they will have bidirectional charging, but it’s a natural fit with their solar and Powerwall systems and virtual utility aspirations. Ford has dangled the prospects of a $1310 charger for the Lightning truck capable of V2H, but it may require other hardware and professional installation (not to mention the truck itself). I think this rollout of decentralized power generation and storage is particularly key for slowly weaning ourselves off natural gas and other legacy fuels.
WH: Multi-Factor, Please!
The White House is encouraging companies to modernize their IT security amidst the ongoing war in Ukraine. Many are speculating as to why Russia has so far failed to effectively launch cyber attacks. People seem to believe that it’s either going to happen eventually, Russian hacking has been aggrandized, or, more likely, it’s something else entirely. The number of recent hacks that leveraged third-party vendors as the attack vector into corporate networks argues for a more centralized enterprise identity and authentication platform spanning multiple companies. Much like having only one personal email address, username, and cell phone number, having one identity that is secure and verifiable for access inside the companies we work for – and across outside apps and other companies we interact with – could make permissions easier. One of the keys is to turn on access only when needed and scrub old profiles of their credentials and access. While it may seem like Microsoft would be an obvious vendor, an independent platform may make more sense. In the broadest view, it argues for a winner-takes-all identity authentication system, but that’s a complex task and unlikely to happen. Centralization can also create fragility and single points of failure. In the meantime, companies should focus on modern identity tools, multi-factor authentication, governance, and privileged access tools. I was shocked to learn that Nvidia, with its critical position in technology around the globe, was not enforcing multi-factor authentication, leading to their recent hack. How is that possible? There are still a lot of basics to cover for almost every company out there, and I would not be surprised if we see mandated, modern IT security in order to do business with any government agency in the future. Our very own Joe Furmanski discussed zero trust on our podcast last October.
Miscellaneous Stuff
Making Memories with Pavlov’s Zebrafish
Scientists have been able to image memory formation in the neurons of a fish using a conditioned fear response paradigm. This raises many questions, such as, how do you scare a fish? In this case, researchers at USC trained larval zebrafish (which are conveniently transparent and have an amygdala analog, the pallium, amenable to imaging) to associate a light with an uncomfortably hot laser. The fish that learned to associate the light with the laser (responding with a flick of their tail) showed significant neuronal repatterning in their brain’s “fear center”, whereas control fish did not. While it’s widely thought that memories form by increasing and decreasing the strength of synapses, this latest study actually shows removal of connections and creation of new ones: “Contrary to expectation, the synaptic strengths in the pallium remained about the same regardless of whether the fish learned anything. Instead, in the fish that learned, the synapses were pruned from some areas of the pallium — producing an effect ‘like cutting a bonsai tree,’ Fraser said — and replanted in others.” While this discovery may not necessarily map to mammalian brains or formation of other types of memories, it underscores how little we know about the inner workings of the brain.
SITALWeek will be on break next week on Sunday April 3rd.
✌️-Brad
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 an informal gathering of topics I’ve recently read and thought about. I will sometimes state things in the newsletter that contradict my own views in order to provoke debate. 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.