137: Anthropic Principle for UFOs, Nvidia GeForce Now, AMD + Tesla, Depth of Knowledge, Business Pixels, Diderot Effect for Companies, Counterpoint Global, and David Foster Wallace

"it expands the surface area of your ignorance"

A hero is one who knows how to hang on one minute longer.

—Norwegian proverb

⛵️ ⚓️ When it comes to knowledge and learning: the farther you get from shore, the deeper the water gets.

At first, you think you know how deep it goes, that you know a lot, but the more you learn, the more you realize there's a lot more remaining to learn, and that you know a lot less than you thought…

It’s the whole: “The larger the island of knowledge, the longer the shoreline of wonder.”

Which we could adapt as something like “the more you learn, the more it expands the surface area of your ignorance”…

That’s because learning increase your absolute level of knowledge, but every thing you learn usually also raises multiple open questions, so the number of those grows faster.

If you want to look for people who know what they're talking about, look for people who seem unsure¹. It's a paradox, corollary of Dunning-Kruger effect.

🐦 A bird built a nest right in the middle of my backyard, in the corner of our gazebo.

I finally was able to see mama feed the younglings. I think there’s at least two of them up there…

I wonder if animals ever get bored. Does a bird sitting in a nest without moving for hours ever get sick of it, or do they feel satisfied doing whatever instinct compels them to do?

If boredom a side-effect of a certain level of intelligence, a way to get you seeking novelty and do stuff with a less specific mechanism than instinct, then which other animals get bored? Dolphins? Great apes? Whales? Octopuses (octopi, for the fancy Latin-lovers reading this)? ¯\_(ツ)_/¯

🛀 Secret Capital on Twitter shared this observation:

The word 'fascinating' is a polite way to say

'this is kinda weird but I'm intrigued'

In my opinion, it's the kind of word that as you say, you're still thinking and trying to process what you actually think about the thing, so it's a nice way to stall while the progress bar hasn't reached 100% yet..

💚 🥃 The water’s nice, jump in:

---

Investing & Business

…there are days when years happen…

Giverny Capital's Francois Rochon on the folly and futility of trying to time the market:

"In 2020, the S&P 500 generated a return of 18%. However, if you had missed the five best trading days of the year, the return of the index would have been closer to -18%."

I suspect 2020 was a fairly extreme year when it came to that, but the effect is probably still pretty strong on average.

h/t William Green

Business Pixels

When you’re looking at a TV screen or a computer monitor, you’re looking at a bunch of dots. They’re blinking quickly, showing various colors (I won’t get into sub-pixel color rendering here, but that’s cool too), etc.

What people usually care about is not the dots themselves, but what a large number of them taken together, and in context, shows.

Content rather than the container.

I feel like it’s the same when you’re studying a business.

You have a bunch of dots — some financials numbers, some facts about management, some judgements about products and competitors, some trends and hunches and personal impressions of the integrity and intelligence of various players, things are weighted as more or less likely, and these weights get updated as you learn more — but then you can’t get stuck too much on any single dot, because it only means something when you step back a bit and look at the whole picture.

I think lots of investment errors are caused by people who get hypnotized by a few pixels and forget that, while these are very shiny and a nice shade of their favorite color, on screen is still a movie headed for a bad ending.

Nvidia Follow-Up on GeForce Now

OG reader and Extra-Deluxe (💚💚💚💚💚 🥃) supporter M. R. made a good point that I overlooked in edition #136 in the section on Nvidia’s game-streaming service:

GeFore Now also creates a positive feedback loop for their DCG segment. Effectively they put themselves in the shoes of one of their customers leading to faster iteration, tighter feedback loops, and rapid product development and/or GTM tweaks to accelerate adoption of DCG products/services in the DC for various use cases - namely, remote access of VMs + Live Streaming.

That’s a good insight.

A lot of the best tech companies are large users of their own tech, and in the same way that Nvidia uses large Nvidia-based supercomputers to design their products (hardware and software — they train their large AI models on them, for example).

GeForce Now (game-streaming) makes them a customer of a different type and probably allows them to learn a lot and improve their data-center offerings.

This reminds me of a recent Snowflake transcript where they mentioned that they run the company internally on the same software as customers use, including the security stuff (ie. if it’s good enough for customers it’s good enough for us).

There are countless other examples, like Amazon running on AWS and Microsoft on Azure and Windows/Office/etc.

I do wonder how Nvidia is implementing GeForce Now, though.

Are they literally just packing a bunch of vanilla consumer gaming GPU cards in a data-center, or are they virtualizing GPUs using their beefier enterprise GPUs (A100, etc), or did they make some customer stuff specifically tailored to this use-case to keep costs as low as possible?

Diderot Effect… For Companies?

James Clear reminded me of the Diderot Effect (had heard of it years ago, but hadn’t thought about it recently):

The effect was first described in Diderot's essay "Regrets on Parting with My Old Dressing Gown". Here he tells how the gift of a beautiful scarlet dressing gown leads to unexpected results, eventually plunging him into debt.

Initially pleased with the gift, Diderot came to rue his new garment.

Compared to his elegant new dressing gown, the rest of his possessions began to seem tawdry and he became dissatisfied that they did not live up to the elegance and style of his new possession. He replaced his old straw chair, for example, with an armchair covered in Moroccan leather; his old desk was replaced with an expensive new writing table; his formerly beloved prints were replaced with more costly prints, and so on.

"I was absolute master of my old dressing gown", Diderot writes, "but I have become a slave to my new one … Beware of the contamination of sudden wealth. The poor man may take his ease without thinking of appearances, but the rich man is always under a strain".

How great is that?

It makes me wonder how this applies to larger organizations, like companies and governments.

Is this why it’s so hard for companies to stay frugal as they achieve success? Is this why Bezos insisted on the door-desks for a long time and why the 3G guys talk about cutting costs like fingernails (just gotta keep cutting over and over as they grow back).

As soon as you start getting some nice stuff, it’s very hard not to get more and more, until you’ve totally changed your ways..

Can you have a garage startup with just one of the Google perks, or once you introduce gourmet food or whatever, all the other crap now seems terrible and everybody’s pushing to an upgrade to the whole package even if they were quite satisfied with the old way until the new luxury entered the picture..?

What do you think?

Interview: Dennis Lynch, Counterpoint Global

Enjoyed this podcast interview of Dennis Lynch by Patrick O’Shaughnessy:

• Dennis Lynch - Delivering Alpha in Adapting Markets (transcript too)

I like his systems-thinking and ability to look at things from first-principles (I kinda hate that this phrase has now been over-used and is turning into a cliché like everything else — outsiders, flywheels, compounders, etc — because it’s a really useful concept if you can use it, and identify the real thing in others — I’ll just keep using it and not care what others think).

Experts are really useful when the world is stationary, when it's not dynamic and companies fit very closely into the buckets that they follow, but when it becomes non-stationary, actually expertise can become a negative.

This is a good point. It’s a subtle one, because you can’t take it as a binary thing, you have to take it how it was meant (“can become”, not “becomes negative”).

But it does help show why certain big things are missed by experts, either because they became too backward-looking, or because it falls enough outside of the realm of expertise of existing silos and the old guard doesn’t see it as part of what they need to be looking at.

Looks like Cloudflare (but that’s not a bad thing)

Trung Phan 🇨🇦 @TrungTPhan

SAAS, explained

---

Science & Technology

Anthropic Principle for UFOs

I’m seeing this topic pop up everywhere lately, a deep zeitgeist injection…

Why is UFO “evidence” always blurry and inconclusive?

In recent decades, the quality and quantity of cameras out in the world has improved on an exponential curve. The military of many countries has spy satellites that encircle the globe, as well as drones and AWACS and other platforms that we probably don’t even know about, and these things have optics that can probably resolve your eye color when you look up at the sky.

So why is “evidence” of UFOs still as blurry as in the 1970s?

Because if it *wasn’t* blurry and ambiguous, we’d see that it’s nothing special.

It’s like the anthropic principle, but for UFO.

Anything that is clear can be explained as nothing special, so all that remains is fuzzy, unexplained stuff — here the word “unexplained” has undertones of the extra-terrestrial to many, but I just mean it in a very literal way; there’s lots of stuff that we can’t explain because we don’t have good data, not because it is inherently mysterious.

If we had a lot more high-quality data on it, we could explain it — the failing is in the map, not the territory.

Also, extraordinary claims require extraordinary evidence.

The footage isn’t much, and the eyewitness testimony is basically worthless for such a hurdle/burden-of-proof.

Anyone who has looked into the value of witness testimony — even expert one — for fairly mundane cases knows to discount it heavily, but when it comes to things that seem so unlikely, it’s basically worth zero (and the track record on it isn’t very good, right?

How many millions of people have claimed all kinds of stuff over the decades, and how much of it has panned out at all over time? Not the best base rate there..).

(For more on this witness stuff, check out the Netflix documentary mini-series ‘The Innocence Files’)

Anyway, I thought I’d share this post by Erik Hoel on the topic:

the evidence is so bad it’s literally laughable. Every single video is so easily explained it is a blow to the dignity of the human race that journalists at major news organizations are willing to run with this without, you know, the briefest sort of investigation. This is tinfoil-hats levels of evidence. Everything in the videos are just dots and blips on thermal or night-vision cameras from incredibly far away, and could easily be caused by simple illusions or natural phenomena or just like, distant planes. If your two possibilities are a) distant plane and b) alien visitation, your priors should make that decision pretty simple. [...]

The video taken from a Navy ship shows a triangular or “pyramidal” UFO blinking in the sky. It’s a video that has already been soundly and completely debunked by professional skeptic Mick West. He goes through three facts:

a) the light source only appears as a glowing flying pyramid (I can’t even write this with a straight face) because the video comes from a night-vision camera that’s out-of-focus. Light sources, such as planes in the sky, appear triangular if the triangular aperture is not all the way up. It’s a well-known effect.

b) the "pyramid” blinks at the exact frequency of the lights on a Boeing 737.

c) the Navy ship was under a popular flight path for passenger planes like Boeing 737s.

Quod erat demonstratum: it’s an out-of-focus Boeing 737, not pyramidal alien spacecraft. [...]

What about the verbal reports from a couple pilots? They should be ignored. Seriously. If we’re going by verbal reports millions across the globe have been “visited” for decades. Doctors, lawyers, women and men of standing and prestige—not just dozens of accounts, but likely closer to millions. Assuredly plenty of ex-military. You should give the same credence to reports of Bigfoot—i.e., none. In the context of these sorts of things humans are fallible, illusion-prone, bipedal morons running from place to place like chickens with their heads cut off. If reports are even worthy of consideration there are way better ones than “pilots saw kind of a big white thing from an unknown distance for a brief period of time and couldn’t figure out what it was and then later we got this crappy video that was easily debunked.” Yet the media now glowingly cover the verbal report accompanying the debunked video. A flying huge tic tac!

The truth is that if we include eyewitness reports the evidence for aliens is supremely overwhelming and has been for a century. If we don’t include eyewitness reports the evidence for aliens is supremely underwhelming. It’s pretty clear what to do in that circumstance. There’s a reason it always comes down to “okay the video isn’t convincing and has all sorts of problems but what about the accompanying verbal report?” The pilots were confused about a big ballon or a bubble or a cloud or swamp gas or a mirage or a weather balloon or a bird from an odd angle, or whatever, it literally doesn’t matter, verbal reports are useless for these issues.

Erik also has a follow-up post here.

Australia Coal Miners Reaction to Tesla Acceleration

h/t my friend JPV 👋

AMD & Tesla (Ryzen & RDNA 2 Inside™)

Speaking of Tesla, AMD recently unveiled some new GPUs and made some announcement about a use case:

we're very proud to be at the heart of the PlayStation 5 and Xbox Series X and S with RDNA 2 and Ryzen CPU technology. Millions of these popular new game consoles have shipped around the world. But we're really just getting started with RDNA 2. So you might be surprised to learn the next place you'll find RDNA 2 graphics.

It's actually on the road, in the electric vehicle market with the new Tesla Model S and Model X. So we've actually -- we have an embedded AMD Ryzen APU powering the infotainment system in both cars as well as a discrete RDNA 2-based GPU that kicks in when running AAA games, providing up to 10 teraflops of compute power.

---

The Arts & History

That’s not a photo he’s holding, it’s a drawing.

Dylan Eakin: If you don’t invest yourself fully into what you do, why would others invest themselves into it?

I stumbled upon a great visual artist on Tiktok (wait, is Tiktok the heir to StumbleUpon? Old school internet nerds will remember how great that site/browser extension was for a while, it was kind of the big traffic drivers to many sites before Digg took over, and then Reddit and social media took over — but I digress)..

Here are a couple videos showing his work, and the evolution of his work over the years (from what looks like regular drawings to the hyper-detailed stuff he now does):

• Portrait of my dad, 150 hours

• Every self-portrait since 2016

But what’s most interesting is this video below, but to get to what I find interesting, you have to watch to the end, because he starts out with some perfectly ok tips, but then drops the veil and talks about what actually changed how he saw things:

• What happened

What he says is powerful, and not just about art.

Why should others care about what you do, why should they invest themselves in it, if *you* can’t fully invest in it?

If you’re holding back because you’re waiting for positive feedback, or because you feel like it’ll be less crushing to fail if you haven’t gone all in, or any of these other defense mechanisms that we have…

It’s also about being well-calibrated. Not everybody wants to be great, some people just want to do a good job and then focus on other things. That’s fine.

I think the problem is when you have jealousy or angst about not being as good as the greats, but you don’t put in the kind of effort/commitment that it takes to be great.

That’s the kind of mismatch that will only close if you either give it a real try, or realize that you’re not playing that game.

Of course, it’s not because you invest all you’ve got in something that you’ll be great at it, but it is necessary (if not sufficient) to even have a chance.

David Foster Wallace

I was telling my cousin about the artist above, and mentioned that he reminded me of David Foster Wallace (probably just how he looks and talks, maybe his approach to his craft).

It reminded of the film ‘End of the Tour’ (2015) with Jason Segel & Jesse Eisenberg.

I saw it around when it came out, and I have fond memories of it. I think this is making me want to re-watch it.

I also enjoy trivia like this, how film-makers get the right feel for a scene:

Segel: “When we meet in the movie, when he shows up at my house, that was the first time we acted together. I think you can tell we're sniffing each other out.”

Those of you not familiar with DFW, you may want to listen to his talk ‘This is Water’.

1. I know I’m paraphrasing someone here, but I can’t remember where I heard that. It may be from James Clear, but I’m not sure ¯\_(ツ)_/¯

Update: Thanks to Luis S. (💚🥃) for pointing out that this was actually from Peter Attia’s interview with Layne Norton!