150: SentinelOne, Amazon, Twitter, Confluent, Cloudflare, Beanie Babies, Marc Andreessen, Rory Sutherland, Tiktok, Concorde @ Mach 2, Air Travel, and a Sopranos Prequel
"use this incredibly valuable interest graph to its highest utility"
As to methods there may be a million and then some, but principles are few.
The man who grasps principles can successfully select his own methods.
The man who tries methods, ignoring principles, is sure to have trouble.
–Ralph Waldo Emerson
🧠 📝 I recently wrote (and spoke) a fair bit about note-taking systems, second brains and PKMs, whatever you call having a solid system to offload stuff from your first brain’s unreliable memory, ideally in a way that encourages connections to emerge over time, ideally when they are most useful.
On this topic, reader Matt Joass suggested this interview with Annie Murphy Paul, author of “The Extended Mind: The Power of Thinking Outside the Brain”.
She touches other concepts too, like the ways in which thinking is embodied, and we can improve decisions by paying more attention to our bodies (“gut feeling” can be pattern recognition), or by moving when we think (maybe it’s why Steve Jobs constantly went on walks while making decisions or having important conversations).
She talks about how unformed ideas can manifest first as gestures and hand movements, and how we can better communicate the intricacies of our thoughts with gestures even if they’re often not socially encouraged (screw that — I’m a hand-speaker and I lean into it when I think it makes sense… When describing rough probabilities or what I think will happen over time, I’ll often draw charts in the air or illustrate time-sequences with movements).
🐦 Follow-up on Twitter being more a platform/distribution channel that stays fairly neutral and partners with everyone (Substack/OnlyFans/Patreon/Fanhouse/etc):
Hopefully Jack & co have looked at their options, and thought something like:
“Well, as long as Apple and Google are taking these huge chunks out of every digital transaction, we can’t really be effective distribution for others, so we’ll try to build/acquire in-house and see what we can do that makes sense in that world… But in parallel, let’s work on a plan B, so that if the world changes and the A&G platforms become more fertile grounds for digital distribution businesses, we’ll have the APIs and relationships with 3rd parties already cooking so that we can use this incredibly valuable interest graph to its highest utility, which is to connect the scattered members of a million niches and turn them from lone weirdos into bioreactors of ideas/connections/creation, which is always monetizable in some way.”
Wow, my mental Jack speaks in run-on sentences… ¯\_(ツ)_/¯
💪 I’ve kept working on the “get to 10+ full pull-ups” challenge. Not like a crazy person or anything, but I’m already seeing some improvement. They’re getting easier to do and so form gets better.
Let’s keep going. Never give up, never surrender.
👶🏼 🧒🏻 Was listening to a child development expert talk about how she talks to her young kid, and I liked the approach.
She tries to avoid talking about “good” and “bad”, “naughty” and “nice” when giving her kid feedback, but focuses on “good decisions” and “bad decisions”. I like that framework.
💚🥃 Today marks the 4th month since I turned on the ability to support the continued existence of this project with a few bucks (if we were in a pub, it’d be about one drink per month, or 77₵ per edition on the annual sub).
My initial goal to feel like things are moving in a more sustainable direction is to have at least 5% paid supporters (so 95% free readers).
Right now we’re at 4.1%. Every supporter moves the needle.
If something you read here helps you make just 1 good decision per year, or avoid 1 bad decision, it’ll have been a real bargain. Join the 5% club here:
Investing & Business
‘Amazon has the highest favorability ratings of any institution in the country except for the U.S. Military 🤯’
Granted, surveys are worth what they’re worth, and a lot depends on methodology…
Via Alec Stapp, h/t Friend-of-the-show Jerry Capital
I don’t have a strong opinion on this company either way, but I have strong curiosity.
On paper, SentinelOne seems worse than Crowdstrike on almost every metric if you compare to when Crowdstrike was at a similar stage of its life — though even that is never a fair comparison, since the world around the companies changes too. But out of the gate, Mr. Market is giving it a high relative valuation.
So I see two main ways things can play out:
1) The smaller company grows faster and catches up, to some degree at least, with the bigger gorilla, and challenges it.
2) This market exhibits large returns-to-scale dynamics and the dominant player becomes more dominant over time, with its growth declining slower than expected.
For example, Crowdstrike has a lot more data than SentinelOne because they have significantly more customers and have been around longer (there’s always a question about the half-life of this data — but data isn’t the only thing, experience and expertise also accumulates over time), and quantity of data can make a qualitative difference when it comes to machine-learning models.
They also have a better-known brand, and this is a field where many purchase decisions are made by people who prioritize avoiding career risk.
Better distribution can also be a scale advantage, especially with partnerships: a lot of these where the salesforce of some other entity is selling your products are often a case where the rich get richer because they get a cut, so they want to partner with whoever has the product that is easiest to sell and will move the most volume.
Anyway, I don’t know, but it’ll be interesting to see how it plays out.
Oh, and if you want to learn more, friend-of-the-show Muji has a writeup (sub. $ required) here:
Interview: Muji on Confluent (Kafka), Snowflake, Open Source Models (vs AWS), etc
Speaking of Muji, he recently did a podcast interview with Simon Erickson:
Good stuff, am looking forward to part 2.
‘A couple dividing their Beanie Babies as part of their divorce settlement’ (1999)
This classic is worth re-surfacing every few years. It says so much about so many things…
Interview: Marc Andreessen
Good interview by Patrick O’Shaugnessy.
Agree or disagree with Marc, he’s always thought-provoking and brings to the table a bunch of specific facts and concrete proposals rather than just vague hand-waving (I’m guilty of this plenty):
I think the biggest idea is one that Marc comes back to over and over, and that our civilization should pay a lot more attention to: The divergence between certain areas that are getting better and cheaper and others that are getting more expensive and…. maybe better? Often worse? Or better on performance, but so much more expensive that it makes them worse when you take into account the price?
I always find it mind-blowing that I can get the fruits of the labor of thousands of hardware and software engineers (who are in turn building on the decades of work by thousands of other engineers), get science-fiction-like advanced parts manufactured in multiple countries around the world and then assembled precisely by humans and robots on the other side of the globe, shipped to me bundled with extremely complex software (a modern operating system + many apps) for around $1000.
But if I hire someone to dig a big hole in my backyard or build a small wooden structure, it’ll cost me multiples of that.
How can we unlock more innovation in those fields and avoid healthcare, education, and construction/real estate from eating up such a huge chunk of the production capacity of our nations with all this implies (regulatory capture, etc)..?
I also liked his history of Hewlett-Packard. It’s a company you can’t avoid while reading about other thing, but I haven’t yet really studied it head on yet.
Also, interesting was the conversation about the primary value of university to employers (“laundering an IQ test” via the admission, since companies now don’t feel like they can do IQ tests themselves, and a conscientiousness test by showing that you can follow directions and show up to do the work for years).
Of course, it’s not the only role of the legacy university system/cartel, but if the primary goal was truly how much students learn, a lot of things would be done very differently.
Trivia: Marc is one of the rare people who I have to slow down my podcast player to keep up with. For Josh Wolfe I have to slow down to 1.5-1.75x, but for Marc I have to go all the way down to 1.25-1.5x.
Interview: Rory Sutherland
I guess #150 is the podcast edition. It has 4 recs!
Here’s the eminently quotable Rory Sutherland, holder of much worldly wisdom about human psychology and behavior, and systems thinking in general:
Another great one by friend-of-the-show and supporter (💚 🥃) Jim O’Shaugnessy, who has been on 🔥 (because an image is more powerful) with his ∞loops lately.
I couldn’t write everything down because I listened to it on my morning walk, but here’s a few highlight from Rory:
the worst mistake you can make in a social setting is to appear not to understand the implicit rules of the game. [...]
quite a lot of those very subtle normative rules around behavior or language are actually nasty. If you have a school uniform, the rules are explicit. "This is acceptable, this isn't." Someone from an out-group can understand the rule. If you have hidden rules, it's a very nasty way of a group effectively practicing exclusion on their own terms. And it makes it very hard to break into that group.
On minority rule:
the fact that a small intransigent boycotting minority can actually change the
consumption patterns of a much larger group by simply refusing to consume something. [...]
we all end up drinking wine not beer at parties because 35% of women don't drink beer. So if you have a mixed-gender party, you can't serve only beer. So you ended up, well, blokes will drink anything, so if you've got red and white wine, that's everybody covered.
Pizza is popular, I think partly because nobody hates it, apart from my dad.
Some good Rory lines:
“One of the problems with the over- educated is their need for ambiguity-avoidance is too great.”
“The number of good ideas you can get to through sequential logic is a very small subset of the potential good ideas out there.”
“It’s when you’re swimming against the stream that you’re most conscious of the flow.”
“In any business, there’s an optimal amount of slack that isn’t zero.”
“We are prisoners of our models. They start out useful, but they end up replacing reality.”
This is why institutional thinking is often stupider than individual thinking, because you're forced to share the same model for purposes of mutual comprehension, and also comparative performance.
You're also forced to keep the same model in order to compare this year's performance with last year.
Where the chasing targets is easy, changing targets is hard.
(Also, I think Rory likes Canada more than most Canadians…)
Air Travel Taking Off
See, I can write one-the-nose headlines like a real journo too.
Via Charlie Biello:
The US averaged over 2 million airline travelers a day over the last week for the first time since March 7, 2020.
We're now only 21% below pre-covid airline travel levels and closing the gap fast. With business travel accounting for roughly 12% of total airline travel, we may not see a full recovery until businesses resume their pre-covid travel routine (which may not happen).
TikTok Rolling Out 3-Minute Videos
Sounds like Twitter's 280-character moment.
Probably will be an improvement. Most vids will stay short, and those that can justify the extra time will be improved (instead of being split in n-parts, which is really annoying when you’re trying to piece it back together).
I think it'll be fine mostly because it's self-correcting. People who make videos that are too long (vs what they should be) will get punished by fewer views/completions/hearts and the algo will clip their wings. Also, Bytedance already has tested this on Douyin, so results outside China will probably be similar.
Instagram is currently TikTokifying itself, so I guess we can expect that there soon.
Dyson must have some really sweet margins (at least on this line of products)… basically the price of an iPhone for a fan.
Science & Technology
‘The only picture of the Concorde flying at Mach 2’
In this extremely rare pic, a Tornado fighter jet from the Royal Air Force has captured the Concorde flying at Mach 2 (twice the speed of sound). Above the clouds and literally on the edge of space you can also see the earth’s curvature on the horizon. Over the Irish sea and rapidly running out of fuel the Tornado had a hard time catching up with the mighty Concorde.
If you’re not familiar with the Panavia Tornado multirole combat aircraft, it is rated with a maximum speed of mach 2.2.
Cloudflare “doubled computations footprint” during pandemic…
The company already had almost 200 points of presence (POPs) located around the world, close to population centers (and often located right inside the ISP datacenters where most people are getting their internet from), but it’s not standing still…
Based on this, they not only added a crapload of computation power to their network during the pandemic, but are also continuing their experiments with alternatives to x86 full-steam:
mid-year update: we've added ten new Cloudflare cities, with four new countries represented among them. And we've doubled our computational footprint since the start of pandemic-related lockdowns. [...]
2021, like 2020, has been a difficult time to be a global network — from semiconductor shortages to supply-chain disruptions — but regardless, we have continued to expand throughout the entire globe, experimenting with technologies like ARM, ASICs, and Nvidia all the way.
new Cloudflare cities: Tbilisi, Georgia; San José, Costa Rica; Tunis, Tunisia; Yangon, Myanmar; Nairobi, Kenya; Jashore, Bangladesh; Canberra, Australia; Palermo, Italy; and Salvador and Campinas, Brazil.
These deployments are spread across every continent except Antarctica.
So I guess the penguins will have to wait for now…
The Arts & History
Sopranos Prequel Film (with David Chase on the script!)
Didn’t even know this was coming, what a nice surprise. Looks good based on the trailer (but I know that can be misleading).
I kinda wish it was a series and not just a movie, but we’ll see.
Regarding the Interview: Marc Andreessen. Outstanding!!
Patrick asks a question. --->Marc 15 min monolog of amazing things.
Repeat for 90 minutes.
to books from this one: Early Retirement Extreme is a philosophy book built first around principles rather than methods. It's an annual re-read for me. Two, The Great Beanie Baby Bubble, not *great* but a good period piece if you want some early internet memories, plus Ty the creator is damn sharp, not expected given the industry.