68: Heico Q3, VC Dogfooding, Internet-as-a-Scenius, Accelerators & Co-Processors, In-Group vs Out-Group Signalling, and Hallgrímskirkja in Reykjavík (say that fast)

"entries in the matrix only update when you send them a fax"

Our whole society has moved towards a philosophy that children are not allowed to fail. As a result we give people easy "successes" at every opportunity, and don't let people experience the fact that they can fail.

The result is people whose self-image is like a balloon - really puffed up but likely to deflate suddenly on facing a real challenge.

btily, Hacker News comment

⊕ Just the idea of thinking in vectors (direction, velocity) rather than in points, does a lot of heavy lifting in life, IMO.

Of course this helps in finance, but I'm thinking of other things too, like relationships, health, life goals, building skills, etc.

It’s common for people to look at where they are (point A), and where they want to go (point B). The main perceived relationship between those points is the distance between them, and it can be discouraging when you’re still far from where you want to be.

But if you can think about the situation in a slightly more dynamic way, it’s both more motivating (a more important ingredient than it might seem), and ways to make progress become more obvious.

So instead of just the two points, you’d also try to determine your direction and velocity (not precisely like in physics, but still more precisely than if you’re not thinking about it). Are you moving in the direction of your goal, or are you moving away from it? Rapidly, or slowly? Add in acceleration as a second derivative, and you’re really talking.

Once you have that kind of mental image of what is going on, there’s more ways to attack it. Can I improve my direction? By learning more, maybe I’ll find out that I thought I was going the right way, but there’s a better way to go? After all, if you’re running really fast to get to Canada (from the US) but you’re actually going South, a change of direction will help you a lot more than just grinding harder.

Once you’re going in about the right direction, there’s always ways to get there faster and more efficiently. What are the things that could help you go faster, and what are the things that are slowing you down, and which of those do you want to keep (not everything that slows you down is bad), and which can you drop? Even if you don’t need a 180 life-change, sometimes a 5 degree course-correction still helps, especially if you have far to go.

And even if you’re not moving fast in the right direction now — it can feel that way when you’re a young adult just getting started on big life goals — you can look at the rate of acceleration. It can feel good to notice that while the current pace is slow, it’s getting faster, so you can better see that progress won’t be linear.

This is all a bit more abstract than I’d like, but hopefully it’s helpful to some of you. I don’t want to make this already long intro longer by adding concrete examples…

Real time follow-up from Twitter:

Investing & Business

Heico’s Track Record since 1990

Considering the impact of cash dividends, prior stock splits and stock dividends, one share of HEICO worth $8.38 in 1990 has become worth on a combined basis approximately $4,794, representing an increase of approximately 572 times the 1990 value and a compound annual growth rate of approximately 23% as of December 18, 2020. (Source)

h/t Rad Kapital

Q3 just came out (yes, Fiscal Q4.. ugh).

I don’t like how they clearly changed the press release format. Look guys, we’re smart enough to read, we know there’s a pandemic going on. It’s fine to provide extra context and highlight good news that may be missed otherwise, but don’t remove the calculated percentages that are usually there.

Some highlights:

Fiscal year net sales: -32%

"Cash flow provided by operating activities totaled $110.2 million, or 177% of net income"

FSG sales fiscal FY: -25.4%

FSG operating margin: 15.5% vs 19.5% YoY

ETG sales: +5%

ETG operating margin: 29.6% vs 29.4% YoY

From the call:

As demand for air travel slowly recovers, we remain very confident in our ability to offer cost savings solutions... we expect to increase our market share.. have even a stronger presence within the commercial aviation market

net debt-to-EBITDA ratio: 0.71x… this is after making 6 acquisitions during the year.

we're also fairly conservative when we look at them in terms of inventory reserves and in terms of not pressing the pricing envelope we want to make sure that we've got a very good business for generations to come

Early in the crisis, there was a destocking phenomenon. I don't really see that anymore. I think the airline now very much living hand to mouth.

our strategy with the parts business is to take a minority market share. We only go for a 30% market share. So as long as we're able to pick up that market share by terms would make sense for our customers as well as for ourselves, we keep that market share around that level because we want to make sure that our competitors also have a very good business strategy for their market share

when I speak with our sales folks, they explained to me that frankly, fear, uncertainty, and doubt were our competitors -- OEM competitors try to push these reasons why not to buy our product, [but it] really doesn't hold up

We could have generated higher operating margins, but we felt it was really important to take care of our people.... other companies say their people are the most important, but then they don't act that way. And we really tried to act that way.

Looking at the filings, I was surprised to see that the optometrist owns more HEI than Laurans, even after selling a bunch. (if you know this company, you know who the optometrist is)

Federal COVID Bill Includes…

$1.9 billion for "rip and replace" efforts to remove Huawei and ZTE equipment from U.S. networks. (Source)

‘entries in the matrix only update when you send them a fax’

Good piece on logistics by Byrne Hobart. I mostly just wanted to highlight this bit of great writing:

In one sense, the logistics system is one big linear algebra problem. Unfortunately, it's hard to solve because some of the entries in the matrix only update when you send them a fax and they fax you back something they copied off a literal clipboard.

Eating Your Own Dog Food, VC Edition

VC Logan Bartlett writes:

Michael Grimes drove for Uber for years to win the IPO for Morgan Stanley. Will be interesting to see the lengths investors go to invest in OnlyFans.

On OnlyFans, I wrote a bit about their 75% EBITDA margins in edition #64.

Interview: Brett Winton, Head of Research at Ark Investment

I know they’ve been in the spotlight a lot in recent times, and have killed it performance-wise, to the great dismay of the wider TSLAQ crowd, but I haven’t followed Ark closely.

This very thoughtful interview of Brett Winton, their head of research, by Brandon Beylo certainly made it sound like they are thinking about things clearly (which doesn’t mean that luck is involved, but it always is, so holding that against them is just hypocrisy):

I think Brett has a good framework for long-term growth/tech investing, with an explicit/systematic look at the optionality that certain companies have, even if it’s not yet showing in the filings; trying to be probabilistic about what could happen and how likely each scenario is, rather than making binary calls, and focusing on a 5-year horizon (which a lot of people pay lip service to, but few actually do) and following the exponential curves where they lead rather than chopping them off because it looks less “embarrassing” to be predicting something that is so far from the conservative consensus.

Not saying it’s a secret sauce that nobody else has or that it’ll always work in any market or that things aren’t a bit crazy right now, but if you’re going to be investing in these big world-changing technological trends, I think this kind of approach is one of the saner ones.

I also enjoyed his last answer, both because I think he’s right (even if you can meet some genius from the past, you probably won’t get anything new out of them), and because it shows he’s really thinking about the question and not just saying what he thinks is the expected answer.

Science & Technology

Is the Internet a Mature-State Scenius?

Packy’s latest post sent me on a tangent looking up Brian Eno’s (Brain Eno!) concept of “scenius”:

Brian Eno, musician, producer, and inventor of the term “scenius”, describes scenius as similar to genius except embedded in a scene rather than in genes.

The geography of scenius has several important factors:

  • Firstly, there is mutual appreciation, which is like motivational peer pressure.

  • Secondly, there is a rapid exchange of tools and techniques, in which as soon as something is invented, it’s widely shared among everyone within the scenius as everyone within the scenius is united by a common language.

  • Thirdly, there are the network effects of success, which means whenever there is a success, it’s celebrated by everyone within the scenius.

  • Fourthly, within the scenius there is a local tolerance for the novelties, which means that renegade, maverick, unusual, and revolutionary ideas are protected from tampering by a buffer zone. Scenius, in other words, is a flourishing space for nonconformity.

    [...] Since serendipity is the nature of scenius, all of the members of the scenius are flexible and have the total freedom to create what they wish. The scenius also has barriers to entry which separate the wannabes from those serious about participating in the scenius and barriers to exit which engender a sense of community among the members of the scenius. Since sceniuses are typically nondescript, you wouldn’t really notice them if you were an outsider to the scenius but to the scenius members they are very real. There is also always a sense of excitement among members of the scenius which makes the scenius members very playful. (Source)

This would help explain why so many geniuses and innovation came out from certain clusters in certain eras, like Edinburgh or Vienna in the 18th century, etc.

Austin Kleon wrote in Show Your Work:

What I love about the idea of scenius is that it makes room in the story of creativity for the rest of us: the people who don’t consider ourselves geniuses. Being a valuable part of a scenius is not necessarily about how smart or talented you are, but about what you have to contribute — the ideas you share, the quality of the connections you make, and the conversations you start. If we forget about genius and think more about how we can nurture and contribute to a scenius, we can adjust our own expectations and the expectations of the worlds we want to accept us. We can stop asking what others can do for us, and start asking what we can do for others.

The idea makes a lot of intuitive sense to me, and I think the Internet is no doubt turbo-charging whatever emerging phenomenon took place in the physical world.

If the combinatorial possibilities in Vienna were N, imagine how many times N they are now on Twitter or on niche forums about topic X?

Now anyone who’s really interested in contributing to a group of like-minded people can do so from wherever they happen to have been born, and the geniuses can now learn from more people and more sources and have a much more robust support-system to elevate their genius more than if they had been born in 1826.

The question is, are these sceniuses (scenii?) limited in size by some kind of dunbar-equivalent number, or can they keep growing and growing and become ever more productive over time?

Are niche groups focusing on certain topics acting as entities in the large soup of the internet, and can groups support each other into a kind of meta-scenius that helps make humanity even more productive/innovative?

F.ex. Maybe some software scenius can help some biotech scenius solve some of its rate-limiting issues, which then help those genius make more progress — of course that just sounds like what tool-making does generally, as the tools from one field are used in another, but I wonder if there can be some kind of more direct transmission mechanism that wouldn’t be possible without the internet. For example, some specialists from one scene may never hear about what’s going on deep in another scene in the old world, because of geography and communication costs, but today, they can dip in and out, learn about the big problems in various fields, and maybe get inspired to find a solution to big problems that aren’t as easily solvable by people who don’t have the same skill stack.

I dunno, but it’s cool to think about, and I’ll certainly be keeping an eye out for the signs of it…

In Silicon: Why the Focus on Accelerators and Co-Processors?

Another good piece by Erik Engheim:

Here’s some highlights about one of the aspects discussed that I find particularly interesting, and relevant to the wider industry:

Back in 2010 at UC Berkley the Parallel Computing Laboratory saw the development towards heavier use of coprocessors. They saw how the end of Moore’s Law meant that you could no longer easily squeeze more performance out of general purpose CPU cores. You needed specialized hardware: Coprocessors.

Let us reflect momentarily on why that is. We know that the clock frequency cannot easily be increased. We are stuck on close to 3–5 GHz. Go higher and Watt consumption and heat generation goes through the roof.

However we are able to add a lot more transistors. We simply cannot make the transistors work faster. Thus we need to do more work in parallel. One way to do that is by adding lots of general purpose cores. [...]

Transistor Budget: CPU Cores or Coprocessors?

You can keep playing that game and eventually you have 128 general cores like the Ampere Altra Max ARM processor. But is that really the best use of our silicon? For servers in the cloud that is great. One can probably keep all those 128 cores busy with various client requests. However a desktop system may not be able to effectively use more than 8-cores on common desktop workloads. Thus if you go to say 32 cores, you are wasting silicon on lots of cores which will sit idle most of the time.

Instead of spending all that silicon on more CPU cores, perhaps we can add more coprocessors instead?

This is the heterogenous computing paradigm that Apple’s M1 embraced, with two kinds of CPU cores (high-performance and low-power, with the great codenames of FireStorm and IceStorm), lots of GPU cores, lots of machine-learning accelerator cores, encryption accelerators that run AES really fast, image and video signal processing units (for fast and low-power compression and decompression of h.264 and HEVC video, for example), etc.

Erik then goes on to talk about how RISC-V is particularly well designed for this world of accelerators/coprocessors.

ARM Will Be The New x86

Thus ironically we may see a future where Macs and PCs are powered by ARM processors. But where all the custom hardware around them, all their coprocessors will be dominated by RISC-V. As coprocessor get more popular more silicon in your System-on-a-Chip (SoC) may be running RISC-V than ARM. [...]

I though the future would be about ARM or RISC-V. Instead it will likely be ARM and RISC-V. [...]

General purpose ARM processors will be at the center with an army of RISC-V powered coprocessors accelerating every possible task from graphics, encryption, video encoding, machine learning, signal processing to processing network packages.

Great stuff.

You can read the rest here (around 7/10 on the geek-O-meter, so not for everyone).

In-Group vs Out-Group Signalling

Interesting piece by Zeynep Tufekci in the reaction to her co-authored piece (with a highly credentialed immunologist) about trying “a single-dose and/or delayed booster trial for the mRNA vaccines” for low-risk groups, to see if maybe this could be a way to free up a lot more doses for high-risk groups (front-liner healthcare workers, older age groups).

On social media, where there was a wide range of responses, but a more noticeable “this has no scientific merit/how dare you as an outsider” reaction compared to when I reached out to experts within the field.

One of the reasons for this discrepancy is that, on social media, the dynamic for in-group status assertions and status competition is strong—and getting stronger for this topic as the pandemic rages on. I’m no stranger to this dynamic, as it is something that’s very common among social movements (something I’ve studied at length) and, well, pretty much any human group. Who’s in and who’s out of the group is a key force for group-based species like ours and thus “stay-in-your-lane” assertions and wagon-circling against criticisms from perceived outsiders is forceful in any human group or profession. As the pandemic progresses, and as the field feels more and more under attack (and much of it quite unfair and terrible) the dynamic strengthens, often to the detriment of the field.

However, on the other hand, of course, expertise and domain knowledge matters and there’s no doubt that the assumption that one can waltz into a topic people have put in a lifetime of work and hand-wave some solution they all missed is often not just wrong, and it can be dangerous or deadly in a pandemic.

On the other, other hand, experts are not immune from groupthink and conflating in-group signaling with protecting domain knowledge from ignorance. Outsider challenges are not always without merit, even if they are often misguided, ignorant and sometimes even downright dangerous.

She tend goes on to say that when a top scientist in the field came out in support of the idea (a stronger version of the idea than she pushed for, actually), there didn’t seem to be that same backlash.

Implied is that the backlash wasn’t so much that the idea had “no scientific merit”, but that who was associated with it wasn’t high-status enough in the field to even be permitted to mention it, despite co-authoring with someone with all the credentials — just having her name associated with it was enough to have the doors slam shut and the idea be poisoned for many.

A lot more, including more context here.

What I find interesting here isn’t so much the specifics of this instance, but the more general pattern, which is repeated again and again and you see everywhere once you know to look for it.

COVID19 Vaccine Tracker

You’ve probably already seen it, but if not, this is neat:


The Arts & History

This is the Hallgrímskirkja church in Reykjavík, Iceland (say that sentence three times fast). You can see another good view of the facade here.

At 74.5 metres (244 ft) high, it is the largest church in Iceland and among the tallest structures in the country.[1] The church is named after the Icelandic poet and clergyman Hallgrímur Pétursson (1614–1674), author of the Passion Hymns. [...]

It took 41 years to build the church: construction started in 1945 and ended in 1986, but the landmark tower was completed long before the whole church was finished (Source)

More details on here. Via Reddit.

I still think the Good Will Hunting approach to education is my favorite…