279: Mark Leonard's $1bn Thermal Oil Idea, Twitter Fires Kayvon, Elon Musk on Everything, Trade Desk Q1, Ted Weschler, Germany's LNG, and the UK's Nuclear
"Perhaps the web is not a microscope but a blender"
To be ignorant of what occurred before you were born is to remain always a child.
—Marcus Tullius Cicero
❓🙋♂️🙋♀️❓Hey, I’m curious about you
Please fill out this anonymous survey. It’s a quick thing, I promise. Thank you! 🧡
🎧🎙 My new podcast with Cedric Chin (💚 🥃 🇸🇬) is here. It’s a fun conversation, IMO.
📚📖 Gurwinder Bhogal has a really good piece about avoiding being trapped in the various echo chambers and bubbles that are an emergent property of the online world, where groups all read the same things and end up thinking alike.
Most of what you see online has been seen by countless other people, and since everyone is being fed the same small range of inputs, they're being conditioned by the same stimuli, resulting in the same small range of outputs (thoughts). Even if everyone is not thinking the same things, they're very often thinking about the same things. [...]
All in all, viewing the web as intended causes you to lose your uniqueness and become more average, more predictable, and more manipulable.
Perhaps the web is not a microscope but a blender, taking a set of unique humans and liquidizing them into a homogenous, infinitely malleable gloop.
In which case, how do you stop your personality from being puréed?
The simplest way is to stop behaving like everyone else. [...]
Instead, wander the web's backstreets. Read ignored texts. Examine the news stories in the bottom corners of the page. Watch last decade’s podcasts. Click on the 20th search result. Regularly break your own browsing habits to shake the algorithms off your tail. Occasionally, switch off your phone, spend whole days in the real world. Go to the library, plumb the past for forgotten ideas. Step outside the zeitgeist so you can see it with fresh eyes.
This is a bit similar to what Ted Weschler said in the podcast I share in today’s edition. I add more of my thoughts on this topic there, including how I try to inject some new ideas and randomness into your life through this paddleboat.
💚 🥃 If you are not a paid supporter yet, hopefully this edition makes you go:
“Hey, I think I want to support what he’s doing here.”
Liberty’s Highlights is reader-supported. To support my work, consider becoming a paid subscriber. 🤘
🚜🛢🌏 A Word From Our Sponsor: Doomberg 🔋⚡️
Ads are usually written by whoever is sponsoring, but I asked Doomberg if I could write this one from the point of view of a reader.
I tend to focus more on the positive, because it’s in my nature. But it’s important to keep an eye on problems, because it’s the only way to avoid them, if we act in time, or to fix them, once it’s too late…
Doomberg is my favorite source for this, because it is the smart version of the genre, with realistic solutions contributed to the discourse, instead of the hysterical doom & gloom of so many others.
Here’s Doomberg in their own words:
At Doomberg, we pride ourselves on early pattern recognition, connecting dots between seemingly unrelated subjects, and articulating the essence of complex topics in a way that makes them more accessible to the broader financial community. We are particularly proud of our early warnings about the coming supply chain, energy, food, fertilizer, and water crises, as well as our ability to expose the insanity of our current policies around energy, natural gas, and nuclear power.
Investing & Business
🔋 Elon Musk and JB Straubel on Tesla’s beginnings & “the future of the car” + where Twitter is going 🐦 + SpaceX & Mars 🚀 (1h20min)
Elon on the early days of Tesla:
It was an insane nightmare, basically. We screwed the pooch six ways to Sunday and made so many mistakes, it’s embarrassing. Basically, almost everything about the first design of the Tesla Roadster was wrong and wouldn’t work. First of all, we had no idea how to build a car…
What stands out was both how difficult it was to pull off, and how much of a variant perception it was at the time to think that electric cars could not only be good — but better than internal combustion vehicles. It seems obvious in hindsight, but carmakers were cutting back on EV projects and hybrids were the butt of jokes.
Battery chemistry wasn’t what it is today, and even getting to a decent range for a price that wasn’t astronomical seemed *barely* possible…
🛢 Mark Leonard’s intriguing ˜$1bn ‘thermal oil’ almost-investment 🤔
Friend-of-the-show Sleepwell (🛏) transcribed this bit from the AGM:
Q: For non-VMS potential investments, there's a lot a lot of competition out there, can you give an update on how far you've gotten? Have you taken a close look on anything? Been outbid? Mark Leonard: Took a hard look at a thermal oil situation, looking close to a 1 billion investment and it was tax advantaged, at a time the sector could not get financing and unfortunately the price of oil ran away on me. I was trying to be opportunistic on a sector that was incredibly beaten up. Q: Does it fit any sort of template of business characteristics you would be looking for outside of VMS software? Mark Leonard: Very large tax assets because we pay an awful lot of cash tax and it's going up because of new regulation in the states. Would have the return through dividends from a public company and would have flown through us in a tax deferred way, so from a tax perspective incredibly attractive. One of the characteristics there, was complexity. Where there's a troubled situation with a peculiar circumstance and there's lots of complexity I think we can compete better than the average investor, particularly where people are willing to take capital forever. Q: If large VMS investments prove attractive and fruitful would you still pursue other markets outside VMS? Mark Leonard: Of course not.
Too bad it didn’t happen, it would’ve been fascinating to see how Leonard would’ve structured a “complex” deal outside of his usual hunting ground…
It certainly shows that Constellation Software is really looking outside of software for opportunities to deploy capital. The VMS venture fund isn’t the only thing.
📺 The Trade Desk Q1 Highlights 📊
I’ve been thinking about Jeff Green lately. His prediction from many years ago about Netflix adding an ad-supported tier is being proved correct, and it does make me wonder what else he may be seeing on the horizon…
I had a look at Q1’s transcript, here are my highlights:
this year, the macro environment was challenging with the ongoing global pandemic, war in the Ukraine and the higher rates of inflation around the world. Despite those challenges, we again exceeded our own expectations. I believe we are now firmly established as the default DSP for the open Internet and that we are very well-positioned to grow and grab share regardless of the macro environment. We now have over 1,000 customers, representing tens of thousands of advertisers spending on our platform across the open Internet.
This is really the bet that Green has made.
That there’s going to be the big walled gardens on one side — big Jupiter-sized planets — and that there’s going to be the open internet on the other, like the asteroid belt and smaller planets and moons… lots of smaller celestial bodies that, when aggregated together, amount to a pretty large thing.
TTD wants to be the best way for advertisers to access this constellation of ad inventory.
As many of you know, I have spent many of the last 10 years publicly predicting that Netflix and nearly everyone else would eventually show ads.
Netflix recently announced that they are likely to make ads a part of their future. This and so many other great things are happening in CTV. In fact, I can't think of a time that the TV landscape has had more positive changes in a short period of time than what has happened in Q1 of this year.
It wouldn’t be worth mentioning if it had been the received wisdom, but almost nobody else believed this would happen — including Reed Hastings at the time! So this one is worth mentioning.
The next big prediction/bet that Green is making is on online identity. It’ll be really interesting to see if he turns out to be right on this one too:
first, identity. The Internet has been operating on a quid pro quo since it became commercial, but the Internet has been a suboptimal Internet. And by that, I mean, we've used cookies as a makeshift technology to enable relevant advertising. But due to a series of events and choices, cookies are going away. Currently, Google has the majority of the browser market share around the entire Internet, outside of China.
So they decided when this transition is going to take place, and they have currently targeted the end of Q1 2023. This has created a very unique opportunity to upgrade the Internet. Indeed, without those many circumstances, this opportunity wouldn't exist. We are upgrading from an opt-out Internet, to an opt-in Internet. The open Internet is scrambling to coordinate and collectively upgrade.
Many different opt-in IDs are being created. Some of them will scale and survive, some won't. However, those that scale will be distributed, encrypted and interoperable. We don't believe IDs can or will scale if they don't upgrade the experience for consumers. With this in mind, we developed and launched UID2. [...]
UID actually does enable consumers to set preferences and restrictions for each web publisher, mobile app and CTV app. They choose who they trust and can change their mind.
Their preferences are tied to an encrypted, hashed, salted ID based on e-mail address or phone number that a consumer can take with them.
How has it been going? What kind of scale of adoption are we talking about?
In terms of the number of unique IDs, we are now measuring UIDs on billions of devices. With every passing month, we are posting new all-time highs.
That comes as more advertisers activate on UID2 and as more publishers adopt it. [...]
MediaVine has now adopted and deployed UID2. Their publishers have seen CPMs increase more than 100%. With UID2, they can create privacy-safe identifiers for their readers, pass those on safely to advertisers, who can then serve relevant advertising without ever knowing anything directly identifiable about the reader. Better opt-ins and more sign-ins, including new lightweight SSOs, is what a post-cookie Internet is going to look like.
Sounds like a good deal compared to what we used to have.
Many people have been using the privacy proxy of “stop tracking!!!” for so long that they can forget that the goal should be privacy, and not bad, un-targeted ads creating a worse experience for both consumers and creators, ie. we don’t want *more* ads because CPMs are lower, seeing the same ones over and over again, and they’re not interesting because the system has no idea what you like and what you’ve seen, and creator can’t earn a living so they make less stuff online. How does that help?
If you can have well-targeted ads without invading people’s privacy, that seems like the best of both worlds.
Back to connected TV/streaming, here’s a history lesson on premium channels, bundling, and ads:
For consumers, on-demand content is just better. As content owners and the streaming apps are competing for more subscribers, it has become very clear, providing consumers with choice is the only way to win. This is a great setup for Netflix.
By not having ads to date, Netflix has protected an amazing user experience, but it required Netflix to keep raising prices so that they could keep making and buying content. At some point, price becomes an issue for nearly all consumers. Consider what we can learn from TV of the past. Even in the days of cable bundles, very few consumers pay for every premium channel, SHOWTIME, HBO, Cinemax, NBA League Pass, et cetera. In the past, nearly all channels have had ads.
Now once again, consumers will have something better, choice. Netflix will very likely setup a clean option for consumers, see ads and pay meaningfully less or pay slightly more and avoid ads altogether. Consumers will continue to be given more choices. I predict that with this move, Netflix will continue to be something of a pace car in TV innovation. They were the first ones to really nail the subscription model for CTV. [...]
As Netflix explores advertising options, they will be unburdened by legacy processes that some of their competitors are working through. For example, they will be able to structure their advertising operations so they don't have sales channel conflict. They can be data-driven from the start in everything that they do, using data to ensure that they maximize yield on every ad impression. In all of these dimensions, and likely many others, I believe Netflix will continue to innovate and set the pace. And as with any market, others will have to adapt accordingly. [...]
This messaging from Netflix, Disney+ and HBO Max is requiring everyone to embrace biddable environments and move away from legacy models, like upfronts and even programmatic-guaranteed, where advertiser choice is limited.
On the dynamic at play for streamers:
CTV needs persistent identity to have effective and high CPM ads, and they need high CPM ads to fund that content. It's economics that's driving that process. As more CTV leaders embrace advertising, they want to ensure that they create as much addressability as possible because that's the only way that they can maintain high CPMs. And an advertiser will pay, say, a $12 CPM if they know the viewers are watching the latest hot reality show, but they will pay 3x that if there's a reasonable chance those viewers are interested in their product. And that's why they will be among the pioneers of the new identity framework or the open Internet.
🐦 ‘Twitter Product and Design Chief Kayvon Beykpour Says He Was Fired’ 🤔
I’m leaving the company after over 7 years.
The truth is that this isn’t how and when I imagined leaving Twitter, and this wasn’t my decision. Parag asked me to leave after letting me know that he wants to take the team in a different direction.
Is this because they knew about Musk putting the deal on hold (and is that a way to negotiate a lower price or to back out of it altogether)? I guess we’ll find out soon…
🎧 Interview: Ted Weschler 🪑🍔
A surprisingly good interview from the Nebraska Furniture Mart podcast, of all places!
Ted is fun to listen to. And if you squint with your ears (?!) a bit, he sounds like a younger Buffett at times, with that damaged-vocal-chords kind of quality to his voice.
But most importantly, he’s thoughtful, level-headed, good-natured, and smart.
What he says about reading widely but not the same things that everybody else is reading is key.
This is what I try to do, and share with you here. Everybody’s so specialized, most don’t have time to look across other fields and explore randomly, so injecting a bit of that in your life through the Serendipity Engine helps!
📈 ‘Stonks go up, stonks go down.’ 📉 ¯\_(ツ)_/¯
Noah Smith: “The wipeout in medium-big tech stocks means that a lot of engineers who thought they got paid $400k a year just found out they actually get paid $180k a year...”
Science & Technology
🔭 First image of supermassive black hole at the heart of our galaxy
Astronomers have unveiled the first image of the supermassive black hole at the centre of our own Milky Way galaxy. This result provides overwhelming evidence that the object is indeed a black hole and yields valuable clues about the workings of such giants, which are thought to reside at the centre of most galaxies. The image was produced by a global research team called the Event Horizon Telescope (EHT) Collaboration, using observations from a worldwide network of radio telescopes.
Wait, isn’t a black hole totally black since even light cannot escape it’s gravitational pull?
Although we cannot see the black hole itself, because it is completely dark, glowing gas around it reveals a telltale signature: a dark central region (called a “shadow”) surrounded by a bright ring-like structure. The new view captures light bent by the powerful gravity of the black hole, which is four million times more massive than our Sun. (Source)
☢️ Nuclear in the UK 🇬🇧
Nuclear's share of energy in the UK is currently about 16%, however almost half of the country's current capacity is due to be retired by 2025 and all but one of its reactors will retire by 2030.
A new government body, Great British Nuclear, is being set up to bring forward new nuclear projects at a rate of about one a year this decade, including the Wylfa site in Anglesey, the strategy says. [...]
The UK currently has one nuclear power station under construction, at Hinkley Point C in Somerset in south west England. Construction began in December 2018 for the plant which is composed of two EPR reactors of 1630 MWe each. The start of electricity generation from unit 1 is expected in June 2026, with unit 2 following in 2027 with a projected lifespan of 60 years. Negotiations between the government and EDF are on-going for a replica project of Hinkley Point C at Sizewell C in Suffolk in eastern England. There is also a project under way to extend the life of Sizewell B by 20 years to 2055.
They’ve got a lot of work ahead of them if they want to replace and expand the fleet!
I hope they’re looking at expanding the life of some existing plants, because as previously mentioned, a lot of these are Ships of Theseus where every part has been refurbished and/or replaced multiple times, so their “age” can be misleading; what matters is that they are safe and effective.
It usually costs less to extend the life of an existing plant than build a completely new one…
✂ 🟥 ‘Germany is set to become a liquefied natural gas powerhouse as it fast-tracks new import terminals’
Germany’s Federal Cabinet gave the green light to draft legislation to cut the approval process for such facilities to a 10th of the usual time. The government is planning four floating terminals, allowing it to replace at least 70% of Russian gas imports and marking a significant u-turn in energy policy after years of resisting costlier U.S. LNG. [...]
With fewer hurdles to clear, the planned floating terminals, which would normally take years to build, could all be installed by next spring (Source)
When there’s a will, there’s a way!
Not saying any of this will be easy and will automatically work out, but a lot of red tape can be cut when the costs of sclerosis becomes apparent and incentives change.