297: Facebook's Metaverse R&D, Amazon Prime Creation, PayPal Deep Dive, TSMC Roadmap, Russian Hackers vs LNG Facility?, and Sriracha Shortage

"Outsiders often have more perspective than natives"

I accidentally stumbled on a dynamite question to ask someone when a debate is going wrong: "what do you think I believe?"

It prevents strawmanning, clarifies the argument, and forces them to adopt your headspace.

—Tom Morgan

🤔 🍸 If alcohol was discovered today (as a drug you can drink, I mean, not for industrial processes), how would we look at it differently as a society?

Isn’t it interesting that perceptions largely get grandfathered in and has a ton of inertia, on both the positive and negative sides (ie. see also Marijuana, psychedelics).

🐦 📝 Twitter is testing yet another new thing, Notes.

It’s a way to do long-form blogging within twitter. This is what it looks like.

My first question when I saw this was, how is this going to be discoverable? Are you going to spend hours on a long-form Note only to see it quickly buried in the flow of tweets and impossible to find on your timeline?

Twitter writes they will create:

• A Notes tab on your profile that holds your published work.

Ok, that could work, but things are getting cluttered… Spaces, Communities, Lists, Moments, Revue Newsletters, and whatever else I’m forgetting.

A Notes tab somewhere in there may get lost in the shuffle. We’ll see, but I’m skeptical.

🛀 📏 Nothing ruins a good movie more easily than expecting it to be great.

Nothing ruins a great movie more easily than expecting it to be sublime and transformational…

And so on.

Learn to manage your expectations and keep them well-calibrated, you’ll enjoy life a lot more.

🏘 David Perell (✍️) has a good piece on the city of Austin, his new home.

I’ve never been to Austin, and I didn’t know much about it except for a few clichés, but I now feel like I know it a little better.

Reading his piece made me wish that more people would write about where they live, especially those who didn’t grow up there, but moved as adults. Outsiders often have more perspective than natives, and can more easily notice things that are taken for granted by the locals.

💚 🥃 I'm thinking of it kind of like we're sitting in a pub somewhere, and I'm talking about various things that interest me or that I've learned about recently.

If you like it, once in a while, you send me a scotch to keep me talking and show appreciation. That’s what becoming a paid supporter is. All very casual & civilized. And a jolly g'day to you too, mate.

---

🧩 A Word From Our Sponsor: Heyday 🧠

Do you have 100+ browser tabs open right now? 😬

Give your memory a boost with Heyday, the research tool that automatically saves the content you view, and resurfaces it alongside relevant Google search results. 👩‍💻

It’s like cheat codes for your memory. 😲💡

🧩 Give your memory a boost today 🧠

---

🏦 💰 Liberty Capital 💳 💴

I wish I could see Facebook’s itemized spend on Reality Labs R&D 🤔

Mostly Borrowed Ideas @borrowed_ideas

@MoatsLikeKodak $META bought Oculus for $2 Bn in 2014. In my estimates, they already burned $25 Bn in FRL in last 5 yrs. My guess is another $30-40 Bn cash burn in the next 5 years is on the cards before being profitable or significantly scaling back. Total unwinding is unlikely.

I just can't comprehend these sums of money for what is mostly R&D... I wish I could see the line items of what this is spent on, that would be interesting.

Successful R&D doesn't tend to be done by armies. Once you have something, you can iterate on it by throwing lots of bodies at it, but typically in earlier phases, too many people hinder rather than helps.

People forget that the iPhone was created by a relatively small team in a relatively short period of time, not by 10,000 people spending $10bn a year for a decade... Same for the original Mac, or iPod, etc.

The question is, is Facebook trying to create the killer product, or do they think they already have the killer product and they just need to brute force the tech forward to make it get traction?

Because if you have the product — say the iPhone — you can atomize every aspect of it and then put larger teams on just the camera, just the screen, just the SoC, just the biometrics, just the battery, just the app store, just the messaging app, etc. But until you have The Thing™️ nailed, it’s much harder to play scale games, IMO.

For great books on effective R&D, I recommend Skunkworks by Ben Rich and Creative Selection by Ken Kocienda.

📦📦📦 Amazon Prime: How fast was it built? 🏎

• Amazon started to implement the first version of Amazon Prime in late 2004 and announced it on February 2, 2005, six weeks later.

This is via Patrick Collison’s list of fast projects.

This oral history of Amazon Prime has all the details you could ever want about how Prime came to be.

🛒 PayPal Deep Dive 💳

Most of what I knew about PayPal was historical (from the excellent book by Jimmy Soni about the early days of the company — check out my podcast with him about it), but thanks to this deep dive by my friend MBI (💎🐕), I feel like I have a better sense of the modern incarnation too:

• PayPal: Beyond the Button (inexpensive $ sub. required)

He makes a good case that while PayPal has heavily diversified beyond the checkout button for which they are best known (mostly through lots and lots of M&A), that it’s still the most profitable part of the business and there are increasing competitive threats that could — over time — make it more ‘just one more button’…

🎧 Request for Podcast: Let’s re-learn to do ambitious, difficult projects fast 🚜👷🏻‍♀️🪚🏗🧱⛓🛠

Best idea I've heard this week, from Max Olson:

Request for podcast series: Every episode is one of Patrick Collison’s Fast entries.

Radio-show-style storytelling series with interviews, narration, good sound production. Some episodes could compare the fast case with a modern-day slow case

Here is Patrick’s webpage documenting “examples of people quickly accomplishing ambitious things together” (clearly he’s trying to learn from history to build Stripe, but also to help amplify these oft-forgotten ideas).

I’d LOVE to hear that podcast, and I think it would be a good way to reach more people and spread the ideas. Stripe Press should finance production. It also overlaps with what Jason Crawford has been doing at Root of Progress with his study of human progress.

A good example of a very very very large scale and very very very difficult project that was accomplished in record time is the Manhattan project. I really enjoyed friend-of-the-show David Senra’s (🎙📚) podcast about how Leslie Groves and J. Robert Oppenheimer ran things:

• #215 The General and the Genius: Groves and Oppenheimer—The Unlikely Partnership that Built the Atom Bomb

Makes you wonder about how many other 'Leslie Groves' are out there and never got recognition for their incredible organizational leadership.

WWII in general probably had a bunch of them, since so many unprecedented things happened. The book Freedom Forge is about this. Before the war, the US had the 18th biggest army in the world and very little of the industrial capacity was dedicated to the military.

In a matter of months, they converted factories, designed and built new fighter planes and bombers, tanks and ships, making hundreds and hundreds of units of each... And an endless quantity of all kinds of ammunition.

Today it feels like we can't do anything quickly! Just shipping a few howitzers to Ukraine is taking forever… 😔

🏴‍☠️🇷🇺 ‘Did Russian hackers blow up a Texas LNG facility on June 8?’ 💥🔥🔥🔥

I found this one via John Arnold (who knows a thing or two about natural gas), but I remember wondering when I first saw the news if this was sabotage just based on the cui bono (Latin: who benefits).

Arnold wrote: “I'm the guy always disparaging conspiracy theories but I find this one that Russian hackers blew up a Texas LNG facility to be plausible and logical enough to merit serious attention.”

Basically: Europe finally gets its act a little together to try to be less dependent on belligerent Putinistan, the US is going to send a bunch of LNG over to help out… And there’s an explosion at a LNG terminal that knocks out about 20% of the US export capacity, making European gas prices spike.

Could be a coincidence — we may never know if it was actually an attack because the implications of admitting it are not good — but it’s not implausible:

According to two sources, around the time of Russia's late February invasion of Ukraine, a cyber unit of Russia's GRU military intelligence service again conducted targeting-reconnaissance operations against a major U.S. liquefied natural gas exporter, Freeport LNG. [...]

On June 8, Freeport LNG suffered an explosion at its liquefaction plant and export terminal on Texas's Quintana Island. The damage suffered means the facility is not expected to resume major operations until late 2022. The June 8 disruption had an immediate impact in spiking already soaring European gas prices and has reinforced Russia's ability to hold gas supplies to Europe at risk in retaliation for the European Union sanctions imposed on Russia over the war in Ukraine. U.S. LNG futures have fallen significantly since the explosion.

How would this work?

Named XENOTIME by researchers, the unit has utilized boutique TRITON/TRISIS malware developed by the Russian Ministry of Defense's Central Scientific Research Institute of Chemistry and Mechanics. That malware is designed for the seizure of industrial control systems and the defeat of associated safety systems. In 2017, GCHQ (Britain's NSA-equivalent signals intelligence service) outlined the need for network compartmentalization to protect safety systems against this malware better. In March 2022, the FBI warned that TRISIS malware remained a threat.

XENOTIME is assessed by the U.S. and British governments as a critical infrastructure-focused, advanced persistent threat actor. The unit's modus operandi involves targeting industrial control systems and supervisory control systems in order to effect unilateral control of a network. XENOTIME has caused specific concern in Western security circles for its targeting of safety systems that would otherwise mitigate threats to life during a cyberattack. XENOTIME's activity has escalated in 2022.

Source.

This week in supply chains: Red jalapeño shortage means no more Huy Fong Sriracha for a while 🌶🌶🌶🌶🌶 🌮 🫙

Huy Fong Foods, a company based in Irwindale, Calif., that produces one of the most popular sriracha sauces in the world, confirmed that it was experiencing an “unprecedented shortage” affecting all of its chile-based products, which also include chile garlic sauce and sambal oelek.

In a statement by email, a company representative said that the issue stemmed from “several spiraling events, including unexpected crop failure from the spring chile harvest.” Huy Fong Foods generally goes through 100 million pounds of chiles each year, the representative added.

The company had foreshadowed the sriracha scarcity in an April letter to customers announcing that unfavorable weather conditions had resulted in a “severe shortage” of chiles.” It said that all orders placed after mid-April would be paused until September. (Source)

NŌŌŌŌŌŌŌŌŌŌŌŌ!

Good thing I don’t believe in just-in-time supply for hot sauce and I always have a multi-bottle buffer…

---

🧪🔬 Liberty Labs 🧬 🔭

Source.

🕯 Let there be light! 💡

When technology makes something cheap enough, we tend to take it for granted and forget about it.

But if you’re forced to go without that thing, you suddenly remember just how useful it is (ie. I’m thinking of recent multi-day power failures in my area).

Lighting is a good example. It’s now so cheap that it’s basically free (too cheap to meter?), but for most of humanity’s existence, it was a luxury, and even for those who had plenty, it consumed a significant fraction of the average family’s income.

Think about that.

TSMC’s Roadmap 🗺📍

As fabrication processes get more complex, their pathfinding, research, and development times get stretched out as well, so we no longer see a brand-new node emerging every two years from TSMC and other foundries. With N3, TSMC's new node introduction cadence is going to expand to around 2.5 years, whereas with N2, it will stretch to around three years. 

While full nodes won’t happen as often as they did before, TSMC plans to create a bunch of intermediary steps that will keep things improving in between the big shrinks.

If you want way more detail, AnandTech has a great overview of TSMC’s roadmap for the next few years.

China’s surveillance-state dystopia and the technology behind it 🇨🇳 📸

Video produced by the NYT showing what their investigation of over 100,000 Chinese gov’t documents reveals about how they’re trying to track everyone:

• China’s Surveillance State Is Growing. These Documents Reveal How

---

🎨 🎭 Liberty Studio 👩‍🎨 🎥

🔥 Doom + 25 years of improvements and mods (Warning: Extremely violent video game) 🔥

The *one* video game that defines my early teen years is Doom by id Software.

I spent countless hours playing it, solo and with friends (LAN parties in my parents’ basement, with everyone bringing their computers so we could make a local network while eating pizza pockets). I created my own maps and fiddled with all kinds of mods, it was great!

The source code of the game was open-sourced in 1997, so ever since, a large community has been upgrading the rendering engine and adding capabilities to the game (larger maps, new weapons & monsters, etc).

The above shows what the game looks like with a few modern mods (you can see the list in the Youtube video description — my own personal favorite is ‘Brutal Doom Platinium 3.0’).

If you’ve only ever played the original in the early 1990s at 320x200 resolution on a 486 33mhz with 8 megs of RAM, this is going to be a shock! 🤘