158: Fin-FANG Redux, Constellation Software Organizational Design, Obsidian, Bezos to Space (Elon Next?), Computational Biology, Patrick Quinn, and Mare of Easttown

"if this your first edition ever and you don’t understand what’s going on, I’m deeply sorry"

Avoid society's proxies:
- Knowledge is not intelligence.
- Lust is not love.
- Money is not wealth.
- Status is not character.
Society delivers false prizes.
Learn to reward yourself.

—Stoic Empire

(another phrasing of Buffett’s inner scorecard)

📝 Obsidian released their mobile app! (it had been in beta for a while)

I instantly paid for sync service ($48 for a year — that’s a lifetime 50% off for “early birds”, not sure when they’ll raise the price, but hurry if you’re thinking of getting it). I could’ve used iCloud sync, but I wanted something more reliable and instant, and to show appreciation for the great work with some $.

I’ve been playing with the app for a few days...

Wow!

The app is basically the desktop app, somehow shrunk to mobile. I almost can't believe it, but they kept pretty much all functionality and plugins working on a tiny screen! This is a huge design challenge, but this tiny team is some of the best software and UX people I’ve ever seen. They just keep shipping great features and fixing bugs every weeks or two.

Synching is almost instant and very solid so far.

This is important though: I don’t think it’s an app that can or should be learned on mobile first.

If you want to get into Obsidian, watch this series of intro/how-to videos, and then start with the desktop version. You’ll have a much easier time that way, IMO.

I also want to correct something I wrote in edition #128 about Obsidian not being an Electron app:

Having spent a lot of time in Obsidian lately, it’s reminding me just how nice a native app running locally can be (and I mean real native apps, not Electron apps that are basically a Chromium rendering engine wrapper running Node.js JavaScript code, masquerading as a native app — ever wondered why chat apps like Slack or Discord will use a gigabyte+ of RAM over time and feel less responsive than a complex AAA 3D game?).

I was wrong!

Obsidian *is* an Electron app, it’s just a really really good one that feels native and fast, unlike the vast majority of slow and bloated popular Electron apps. Mea culpa.

Bonus for my Obsidianers, er, my Obsidious People:

I’ve been having fun with new themes to replace my beloved “Gruvbox”, and I think you may like them. Try the dark modes of “Everforest” (that’s the image above) and “Dracula” (especially neat on mobile, IMO).

🟧 I feel a bit like a 5-year-old even writing this, but I think I just found out at 39 years of age that my favorite color may be orange. Not something I’d put everywhere, but I dig it as a highlight, sprinkled here and there.

I wear black t-shirts pretty much every day and never really felt like I had a favorite color, but I recently noticed a pattern:

The highlight color I picked for this newsletter is orange, on Twitter I set the “display options” so that my buttons, links, and DM bubbles are orange, my phone lockscreen background is hugely orange, years ago I bought some folding knives with handles that are that bright “safety orange” because I liked how they looked (f.ex. I have one of these), I always really liked the design of the Overcast icon, my main Obsidian theme (Gruvbox) has orange highlights, etc.

I guess I’m slow, because it hadn’t really registered until recently that, hey, I like orange. ¯\_(ツ)_/¯

🙈 Feeling a bit like Dr. Moreau after trying to make this Mothy McFly thing work in Pixelmator Pro:

Ladies and gentlemen, Mothy McFly!

I may have nightmares… (and if this your first edition ever and you don’t understand what’s going on, I’m deeply sorry)

💚 🥃 Mothy McFly would LOVE for you to become a supporter by clicking the button below.

And if you don’t, he may come visit you at night (likely to get stuck on a lightbulb, though)…

---

A Word from our Sponsor: In Practise

Do you invest? In Practise is a research service for fundamental investors to learn about high-quality companies.

In Practise conducts hundreds of high-quality interviews every year with CEO’s from the best companies globally available on mobile and desktop. All for the price of only $40 per month. 

⭐️ Sign up here. ⭐️

Two great recent interviews that you’ll love: Spotify and Universal Music and Topicus (the Constellation Software spin-out)

---

Investing & Business

Fin-FANG Part Deux

This is a follow-up to the ‘Non FANG Fin-FANG’ stuff I wrote about in edition #157.

Some interesting discussion of the idea on Twitter, including some suggestions on what other businesses could be in the club, if we were ever trying to launch an ETF or create a meme out of it the way FANG and FAAMNG or whatever turned out.

The first problem with V/MA/SPGI/MCO is not enough vowels. Hard to make a good acronym.

A few people recommended FICO (33.43% 10-year CAGR), which makes sense, though doesn’t help with the meme-ification. Same for MSCI (32.20% 10-year CAGR).

A potential vowel could come from Intercontinental Exchange (ICE), but I’m not sure how much a single vowel would help, and that one doesn’t quite have the 10-year CAGR of the others (a mere 17.8%)…

FIMMVS? VIMMFS? SIMMFV? MIMFVS..? ugh

Some commented that there’s no certainty that these companies will do as well going forward, etc.

I was purely looking back, not giving investment advice or saying anything about the future.

Something else that a few people commented on is that fact that a lot of the return from V/MA/SPGI/MCO came from multiple expansion. I’ve got a few things to say about this.

Multiples are a bunch of things aggregated together, and it’s very hard to tell much from them just from a 1st degree glance. Many people seem to consider stock performance that came from multiple-expansion to be “fake” or “cheating” somehow, but there’s can be plenty of company-specific and macro-level reasons for it to make sense (or not).

Interest rates are obviously a big player, and the market as a whole had a rerating over the period (also coming out of the GFC), so while the Fin-FANG may not have had as high an absolute CAGR under a lower multiple environment, the rest of the market would likely also have done worse too, and so the diff of alpha between the two would likely have remained pretty wide.

At other times, I think multiples expand because the market gains confidence and expertise (collectively) in whatever thing that company is doing.

For example, part of Constellation Software’s multiple expansion over time is probably the market understanding better that VMS is a really good place to be, and confidence increased over time in the ability of CSI to execute its playbbook; you now have 15 years of data supporting it (even if the company is bigger and there are other downsides), while in the earlier years, you were looking at only a few years of a microcap and had a lot less info about how good VMS was, so it all felt still a lot more uncertain and probably “deserved” lower multiples (on top of interest rates and other factors).

‘Constellation Software & Org Design’

Friend-of-the-show and supporter (💚 🥃) Shomik Ghosh has a great post that looks at Constellation Software’s organization structure, and how that gives it various competitive advantages:

I believe CSU’s competitive moat is driven by org design.

As you can see from the chart above, CSU is a complex business. However, the structure of the org is very intentional. CSU has honed the craft of what decisions need to be centralized and what can be decentralized. They have structured the org to maintain flexibility and agility in their core competency, acquiring companies. This allows them to continue doing acquisitions that I believe number over 100 companies a year! [...]

The interesting thing about the Operating Groups (OG) is that they all operate independently and have their own CEOs. CSU will share best practices, hurdle rates, and can help on planning, but each OG can have their own culture. Some decide to centralize decision making more, some decide to stick with their competencies more and not expand into other verticals, etc. [...]

CSU can acquire 100 companies that may total the same value as 1 PE acquisition and do that across industries, geographies, and scale at an increasingly accelerated pace as CSU’s crm of target companies grows. Now ask Vista if they will purchase a golf club software business for $10M. See what they say. This is CSU’s competitive advantage and why Leonard spends so much time studying other businesses, focusing on incentives, and tweaking the org design.

Another aspect that makes CSI’s (not a typo: CSU is the ticker, CSI is the acronym for the name) structure an advantage is how it makes the small low-growth VMSes more valuable by giving them reinvestment opportunities.

ie. A standalone VMS may have great economics, but it likely can’t re-invest most of its cashflow into similarly high-ROIC opportunities. So it’s basically a cash cow. But inside CSI, these cashflows can be redeployed at high ROICs, making them a lot more valuable.

Interview: Patrick Quinn

“Patrick is an unapologetic generalist. He deliberately keeps himself open to every good opportunity, which has resulted in 29% net annualized return since his fund's 2011 inception.”

I enjoyed this one. Didn’t really take notes while listening, so I don’t have much context to provide, but it was a solid discussion about the challenges and opportunities facing flexible, long-term investors:

• Patrick Quinn: Upside Volatility

I remember liking the part where he talk about the importance of “experiencing the investment at every level”, and not outsourcing too much of it to others, because it prevents you from building experience and pattern recognition that may surface later on as “gut feeling”…

h/t Uncovering Value

---

Science & Technology

Well, he did it — Jeff Bezos went to space! (briefly) 🚀

You can watch the video here (should start at the right timestamp for launch).

So Branson went to 86.1 kilometers of altitude and Bezos went to… *drumroll please*

107 kilometres (66 miles)

So clearly the next logical step of all this is for Elon Musk to go up to 250 km and post a meme about how he can “barely see where the others went up to from up here” or something like that ¯\_(ツ)_/¯

This is just an image I found online, it isn’t my account. I just checked and I have 11 million credits on it (the more computation you do, the more you have).

Interview: Dr. David Baker on Computational Protein Design, and Digital Biology

Baker is the head of the Baker lab at the University of Washington, where a lot of really cool stuff has been happening for a while.

Here’s the quick version of the story on why I’ve been tracking Baker for about 15 years:

I have an uncle who’s a bit out there, and around 20 years ago he showed me this little program he ran on his computer to look for aliens. It was called SETI@home and basically used a distributed-computing model to look through vast amounts of radio-telescope data to try to find unusual signals.

I was never into that Mulder & Scully stuff, but I thought the idea of running scientific workloads on my PC at home was really cool, so I participated for a few years.

Over time, other scientific projects started using the same infrastructure (now called BOINC, for Berkeley Open Infrastructure for Network Computing) to allow anyone to volunteer their idle CPU cycles to help work on projects that are simply too large and computationally-intensive for a single lab. A big one is Folding@Home (not part of BOINC).

So around 2005-2006, I discovered Rosetta@home, which is Baker’s computational protein design project.

The functions of proteins depend on their 3D shapes, and their 3D shapes are encoded in their sequences of amino acids (that’s what DNA is).

If we could predict the shape from just the sequence, and we could do it in reverse (design the shape we need and then build it with the right sequence), this unlock a tremendous amount of really useful stuff that we could do (drugs, cures for diseases, synthetic biology, catalysts to create biofuels or degrade toxic waste or whatever).

So most winters since then, I’ve had all my CPUs crunch for Rosetta@home (the waste heat just goes to heat my home, displacing some of what the furnace would produce, so the cost is minimal).

Wow, this short intro is super long, so I’ll just punch the eject button here and leave you with this very interesting interview with Dr. Baker, a man that I wouldn’t be surprised to see getting a Nobel prize someday, or maybe one of his 11 companies will become huge and change the world:

• Dr. David Baker: Director of the Institute for Protein Design, University of Washington (there’s a link to the transcript in there)

---

The Arts & History

Mare of Easttown (HBO, 2021)

Watched the first episode of ‘Mare of Easttown”.

Very promising mini-series so far.

Don’t know where it’s going, but very high quality on all levels (acting, writing, cinematography, etc).

It has that very “HBO” feel to it, and I wish I better knew all the elements that come together to make me feel that way. Just the look… There must be a playbook somewhere that they share with cinematographers and directors. I don’t mean the obvious stuff like the nudity and swearing and slower pacing and the way there’s little hand-holding to help you keep up when things get confusing (lots of secondary characters). I mean the more intangible things, like how people are dressed, how things are filmed, etc.

The physical setting reminds me of ‘Prisoners’ (also set in PA).

Winslet is badass.

Update: Saw episode 2.

Really good too, just can’t say much negative about it.

The stuff with kids breaks my heart, though. 😥

I do like how realistically messy everyone’s families are. Every secondary character feels like they have a whole life separate from the show.

Looking forward to more.