The company’s autonomous vehicles have driven 5 million miles since Alphabet began the program back in 2009. The first million miles took roughly six years. The next million took about a year. The third million took less than eight months. The fourth million took six months. And the fifth million took just under three months. Today, that suggests a rate on the order of 10,000 miles per day. If Waymo hits their marks, they’ll be driving at a rate that’s three orders of magnitude faster in 2020. We’re talking about covering each million miles in hours.
But the qualitative impact will be even bigger. Right now, maybe 10,000 or 20,000 people have ever ridden in a self-driving car, in any context. Far fewer have been in a vehicle that is truly absent a driver. Up to a million people could have that experience every day in 2020.
2020 is not some distant number. It’s hardly even a projection. By laying out this time line yesterday, Waymo is telling the world: Get ready, this is really happening. This is autonomous driving at scale, and not in five years or 10 years or 50 years, but in two years or less.
Whether users are OK with this is a personal judgment they make, or at least should be making, when using the services. In open and democratic societies, perhaps users are less worried about what large corporations, who can be secretly compelled to hand over data to the state, know about them. Users are protected by the rule of law, after all. If they are going to see advertising in exchange for content, storage and functionality, then they would rather see relevant than irrelevant advertising alongside their web pages, emails, photos, videos and other files. Most citizens are not criminals and not concerned about what the state knows – they just want to share their holiday photos and chat with each other and in groups via a convenient platform, knowing that Facebook can mine and exploit their data.
But in authoritarian states such as China which control what their citizens can see and which lack a reliable rule of law, such networks pose a bigger threat. Tencent, for example, with its billion active accounts, knows the social graph of China, who your friends and associates are, where you go, what you spend (if you use their payment app) and what you say to each other and in groups on the censored chat platform. Similarly Sina Weibo. The state security apparatus has access to all of this on demand, as well of course as access to data from the mobile phone operators. So even if you stay off the Tencent grid, if you use the phone network then the state will know a lot about anyone you call who is a user of these platforms, as well as being able to profile you based on your repeated common location with other users. All of this data is likely to be accessible to the state in China’s forthcoming Orwellian Social Credit System, a combination of credit rating with mass surveillance. Knowledge is power. No wonder then that China won’t allow Facebook into the game.
The new chip, NVSwitch, is a communication switch that allows multiple GPUs to work in concert at extremely high speeds. The NVSwitch will enable many GPUs – currently 16 but potentially many more – to work together. The NVSwitch will distance Nvidia from the dozen or so companies developing competing AI (artificial intelligence) chips. While most are focused on their first chips, Nvidia is building out highly scalable AI systems which will be difficult to dislodge.
[Nvidia CEO] Jen-Hsun [Huang] is very clever in that he sets the level of performance that is near impossible for people to keep up with. It’s classic Nvidia — they go to the limits of what they can possibly do in terms of process and systems that integrate memory and clever switch technology and software and they go at a pace that makes it impossible at this stage of the game for anyone to compete.
Everyone has to ask, Where do I need to be in process technology and in performance to be competitive with Nvidia in 2019. And do I have a follow-on product in 2020? That’s tough enough. Add to that the problem of compatibility you will have to have with 10 to 20 frameworks [for machine learning.] The only reason Nvidia has such an advantage is that they made the investment in CUDA [Nvidia’s software tools].
A lot of the announcements at GTC were not about silicon, they were about a platform. It was about things such as taking memory [chips] and putting it on top of Volta [Nvidia’s processor], and adding to that a switch function. They are taking the game to a higher level, and probably hurting some of the system-level guys. Jen-Hsun is making it a bigger game.
Nervana’s first chip didn’t work, they had to go back to the drawing board. It was supposed to go into production one or two quarters ago, and then they [Intel] said, ‘We have decided to just use the Nervana 1 chip for prototyping, and the actual production chip will be a second version.’ People aren’t parsing what that really means. It means it didn’t work! Next year, if Nervana 2 doesn’t happen, they’ll go back and do a Nervana 3.
Apple’s decision to switch away from Intel in PC’s wouldn’t have a major impact on the chipmaker’s earnings because sales to the iPhone maker only constitute a small amount of its total. A bigger concern would be if this represents part of a wider trend of big customers moving to designing their own components, he said.
Apple’s custom processors have been recently manufactured principally by Taiwan Semiconductor Manufacturing Ltd. Its decision may signal confidence that TSMC and other suppliers such as Samsung Electronics Co. have closed the gap on Intel’s manufacturing lead and can produce processors that are just as powerful.
Ticket prices are at record highs. Service fees are far from reduced. And Ticketmaster, part of the Live Nation empire, still tickets 80 of the top 100 arenas in the country. No other company has more than a handful. No competitor has risen to challenge its pre-eminence. It operates more than 200 venues worldwide. It promoted some 30,000 shows around the world last year and sold 500 million tickets.
Though the price of tickets has soared, that trajectory predates the merger and is driven by many factors, including artists’ reliance on touring income as record sales have plummeted.
Live Nation typically locks up much of the best talent by offering generous advances to artists and giving them a huge percentage of the ticket revenue from the door. Why? Because it can afford to. It has so many other related revenue streams on which to draw: sponsorships for the tour, concessions at venues, and, most of all, ticket fees. The fees supply about half of Live Nation’s earnings, according to company reports.
Critics say enforcement of the consent decree has been complicated by what they call its ambiguous language. Though it forbids Live Nation from forcing a client to buy both its talent and ticketing, the agreement lets the company “bundle” its services “in any combination.” So Live Nation is barred from punishing an arena by, say, steering a star like Drake to appear at a rival stop down the road. But it’s also allowed, under the agreement, to redirect a concert if it can defend the decision as sound business.
That’s far from the only ad inventory Roku has access to. The Roku Channel offers free-to-watch popular movies, which Roku sells ad time against. Many of Roku’s “free” channels are ad supported, with Roku having access to all or some of the ad time on many of those channels (not all of them).
While selling ads is the biggest piece of the company’s Platform business, there are some auxiliary sales as well. See those Netflix, Amazon, Pandora, YouTube, etc. buttons on your Roku remote? The company was paid to put them there. Additionally, some TV brands have licensed the right to include Roku OS right into their television set, another source of revenue.
All told, Platform revenue is 44% of total sales, and growing rapidly. In fact, it more than doubled in 2017, and has increased more than 3-fold over the past 2 years. Even better, Platform revenue carries a gross margin near 75%, meaning that already it makes up 85% of Roku’s gross profitability. Completing the trifecta of good news, Platform sales are far more recurring and reliable in nature than hardware sales, giving the company a firmer footing from which to expand their business. Bottom line here? Roku is not really a commodity hardware maker. It is more of a consumer digital video advertising platform.
There is no shortage of ways to get streaming content. And all of them are fighting tooth-and-nail for users. Google and Amazon practically give away their devices to get users into their ecosystem. Against that lineup, it really has very few competitive advantages. There is no meaningful lock-in to the platform. It is really quite simple and painless for a consumer to switch from a Roku to a competing offering. Getting new customers is even more of a dog fight.
Netflix makes up over 30% of streaming hours through Roku’s platform, but the channel provides essentially no revenue back. Same for Amazon, Hulu, and the most popular ad-supported video network in the world, YouTube. Roku relies on monetizing Roku Channel and other, less prominent content channels. However, there is nothing stopping those other channels from switching to a different ad provider, or (if they are large enough), building out their own.
Alibaba leaned heavily on its long-time ally SoftBank — an early backer of Tokopedia and Grab — to get the Tokopedia deal ahead of Tencent. That’s despite Tokopedia’s own founders’ preference for Tencent due to Alibaba’s ownership of Lazada, an e-commerce rival to Tokopedia. SoftBank, however, forced the deal through. “It was literally SoftBank against every other investor,” a separate source with knowledge of negotiations told TechCrunch. Ultimately, Alibaba was successful and it led a $1.1 billion investment in Tokopedia in August which did not include Tencent.
While the Cas9 protein is involved in cutting and correcting DNA, the Cas4 protein is part of the process that creates DNA and genetic memory. CRISPR evolved from a bacterial immune defense system in which bacteria destroy viral invaders. Now we are beginning to understand how bacteria detect the invaders and remember the encounters. With Cas4, bacteria can record these encounters in their DNA, creating a permanent ledger of historical events.
Our understanding of Cas4 is rudimentary, but its potential applications are provocative. Not only will it timestamp key events, but it should be able to monitor how an individual’s body works and how it reacts to different kinds of bacteria. A Cas4 tool should be able to fight antibiotic resistance, an important use case addressing a significant unmet need.
Our research is not alone in reaching this conclusion. A 2013 study of US equity markets found that in the month after the US enters conflict, the Dow Jones has risen, on average, by 4.0 percent—3.2 percent more than the average of all months since 1983. A 2017 study found that volatility also dropped to lower levels immediately following the commencement of hostilities relative to the build-up to conflict. During the four major wars of the last century (World War II, the Korean War, the Vietnam War, and the First Gulf War), for instance, large-cap US equities proved 33 percent less volatile while small-cap stocks proved 26 percent less volatile. Similarly, FTSE All Share and FTSE 100 volatility has historically fallen by 19 and 25 percent over one- and three-month horizons following the outbreak of conflict.
Missing a bull can be even more detrimental than taking part in a bear. Following the two huge bear markets we’ve experienced this century, many investors decided it was more important to protect on the downside than take part in the upside. Risk is a two-way street and I’m a huge proponent of risk management, but investors have taken this mindset too far. Missing out on huge bull market gains can set you back years in terms of performance numbers because you basically have to wait for another crash to occur, and then have the fortitude to buy back in at the right time. I have a hard time believing people who missed this bull market because they were sitting in cash will be able to put money to work when the next downturn strikes.
In the last 50 years medical schools subtly shifted teaching away from treating disease and toward treating patients. That meant laying out of the odds of what was likely to work, then letting the patient decide the best path forward. This was partly driven by patient-protection laws, partly by Katz’s influential book, which argued that patients have wildly different views about what’s worth it in medicine, so their beliefs have to be taken into consideration.
There is no “right” treatment plan, even for patients who seem identical in every respect. People have different goals and different tolerance for side effects. So once the patient is fully informed, the only accurate treatment plan is, “Whatever you want to do.” Maximizing for how well they sleep at night, rather than the odds of “winning.”
Everyone giving investing advice – or even just sharing investing opinions – should keep top of mind how emotional money is and how different people are. If the appropriate path of cancer treatments isn’t universal, man, don’t pretend like your bond strategy is appropriate for everyone, even when it aligns with their time horizon and net worth.
The best way to talk to people about money is keeping the phrases, “What do you want to do?” or “Whatever works for you,” loaded and ready to fire. You can explain to other people the history of what works and what hasn’t while acknowledging their preference to sleep well at night over your definition of “winning.”