544: Sonos SNAFU, Nvidia’s Biology Dreams, Moderna’s Wild Ride, Organizational Force Multipliers, Amazon Ads, Canada's Oil, Alcohol, and Bob Dylan

"The stakes here are enormous"

Do not force people to choose between doing what is smart and doing what helps them save face.

–Deepak Malhotra

🔊➡️ 🔇🤬😡😤 I wrote about Sonos speakers a few times (f.ex. in Edition #404 — it’s not missing, haha).

There's a story of Good Tech Gone Bad aka The Great Sonos Meltdown that I've been wanting to share but never published — I couldn't quite get it right at the time, but I think it's worth telling now that we know the ending.

It’s about how ANGRY I was at Sonos for months last year when they released a new version of their app that broke *everything*. 😤

The problems were maddening: Music would stop playing randomly, speakers would drop out during multi-room playback, and the app would lag so badly it became nearly unusable. For a while, I couldn't play music at all.

It was incredibly frustrating, and what made it worse was the confusion about what was going on — there was little information online about the issues. I couldn't tell if I was some edge case, on my own, left to fend for myself, or if this was a common issue that was top of mind for the company to fix ASAP. I scoured customer support forums and tried everything I could find, but nothing worked.

In desperation, I factory-reset everything, hoping that might fix it. It backfired and made it even worse — bizarre bugs prevented me from even setting the system back up for a few days.

I was so fed up that I considered writing off the whole thing and buying something else —I even discussed shorting the stock with a friend (looking back, it would have been profitable, though I don't do shorts). But while there are lots of companies making great audio gear (NAD, Rotel, Denon, Cambridge Audio, etc) and speakers (Axiom, Paradigm, Klipsch, KEF, Polk, etc) there's a surprising lack of direct Sonos competitors in the multi-room, small-footprint space that offer both good sound quality and good software integration.

So I was kind of stuck in audio purgatory…

But there’s something I’ve learned over the years:

• Sometimes the best thing to do is nothing. Wait for a better move to reveal itself.

It reminds me of this great scene from the show ‘Boardwalk Empire’:

I decided to do nothing and see what happened. It was frustrating having these expensive paperweights around the house (I could still use them via AirPlay, bypassing the Sonos app, but that severely limited their functionality).

Thankfully, the customer backlash was loud enough that the company had no choice but to apologize and refocus all engineering resources on fixing bugs and restoring a baseline of functionality.

They began shipping stability updates bi-weekly and things steadily improved. My Sonos are now as stable and responsive as they ever were! Hallelujah! 🙌

The fallout was significant, it cost the CEO his job, and soon after the Chief Product Officer was also fired. I don’t mind that — we need more accountability in the world.

When you mess up your company’s core products, damage the brand deeply, and are slow to admit and fix it, there should be consequences.

💚 🥃 🙏☺️ If you are a free sub, I hope you’ll decide to become a paid supporter in 2025:

---

🏦 💰 Business & Investing 💳 💴

💉 Moderna’s Wild Ride, Round Trip Edition 📈🚀📉💣

I wouldn't have predicted a couple of years ago that Moderna would underperform the S&P500 index since BEFORE the pandemic.

Or to phrase it another way: If in 2019 I had a crystal ball showing me how much revenue and profit this company was going to earn over the next three years, I don’t think I would’ve predicted that the stock would underperform the index over the next 5 years.

This is a company whose revenues grew by 24,000%+ YoY during Q1 of 2021. Soon after, they had 75% EBITDA margins 🤯

Their revenues went from about $50 million/quarter to $7 billion/quarter over a matter of months. Basically overnight, their net income grew to be *multiples* of their entire historical revenues... and what's left of that a few years later?

Now, it's almost like it never happened... You know that device in Men in Black that zaps your memory clean 🕶️

Biotech is a hard game when even the once-in-a-generation winners don't stay winners for long ¯\_(ツ)_/¯

🏗️ Organizational Force Multipliers: Bullets & Barrels, Soldiers & Generals, High vs Low Agency 🎖️👷‍♂️💭

I like this metaphor by Keith Rabois during a presentation at Stanford:

So I like this idea of barrels and ammunition. Most companies, once they get into hiring mode…just hire a lot of people, you expect that when you add more people your horsepower or your velocity of shipping things is going to increase. Turns out it doesn’t work that way. When you hire more engineers you don’t get that much more done. You actually sometimes get less done. You hire more designers, you definitely don’t get more done, you get less done in a day.

The reason why is because most great people actually are ammunition. But what you need in your company are barrels. And you can only shoot through the number of unique barrels that you have. That’s how the velocity of your company improves is adding barrels. Then you stock them with ammunition, then you can do a lot. You go from one barrel company, which is mostly how you start, to a two barrel company, suddenly you get twice as many things done in a day, per week, per quarter. If you go to three barrels, great. If you go to four barrels, awesome. Barrels are very difficult to find. But when you have them, give them lots of equity. Promote them, take them to dinner every week, because they are virtually irreplaceable. They are also very culturally specific. So a barrel at one company may not be a barrel at another company because one of the ways, the definition of a barrel is, they can take an idea from conception and take it all the way to shipping and bring people with them. And that’s a very cultural skill set.

It’s a good way to put something that I usually discuss in terms of “high-agency vs low-agency”, or “soldiers vs generals”, or maybe “heroes vs NPCs”.

But I like bullets vs. barrels because it slices the field from a slightly different angle and reveals something that the other metaphors miss.

Yes, if you’re low-agency, you should definitely try to change that and become high-agency.

But you don’t have to be a barrel. You can do amazing by being a great bullet.

Many bullets may be low-agency, but the best ones are high-agency. But the reverse isn’t true; I don’t think there are any low-agency barrels.

And interestingly, I don’t think Generals have to be barrels. Some are bullets that lead lots of soldier bullets.

Maybe I’m stretching the metaphors too far, but it’s definitely my experience that some people are just built different — they effortlessly generate lots of ideas and run with them, execute better and faster, and whatever you throw at them, you get good ideas and feedback back.

This connects to something I once wrote about, what I call ‘generative people’ — those rare individuals where if you give them an idea you’ll get more than 1x ideas back. That's probably one of the key characteristics of "barrels" 🤔

Via Ben Thompson (💚 🥃 🎩)

🛒 Amazon is Turning Its Ad Platform Into a Product for Third Parties 💵

Amazon has mastered the art of turning cost centers into profit centers — they consistently build infrastructure in a way that can later be opened up to external customers via API.

The classic examples are:

• Internal logistics → Fulfillment By Amazon (FBA)

• Computing infrastructure → AWS

• First-party e-commerce → Amazon Marketplace

• Payment processing → Amazon Pay/Buy With Prime

It looks like there’s a new one to add to the list:

Amazon Retail Ad Service enables online retailers to show product ads on their search, browse, and product pages to help their customers discover new selection and make informed purchase decisions. […]

Retailers will determine ad creative formats, where ads appear across their apps and websites, and how many ads are shown. They’ll also be able to determine what customers see after clicking on an ad, such as directing traffic to the product page, providing a ‘quick view’ of the product, or enabling the customer to add the product directly to their cart.

For retailers concerned about Amazon getting their data and using it to compete with them, Amazon claims that they keep it separate:

Amazon Retail Ad Service operates on dedicated systems with stringent access controls, ensuring retailer information is separate from Amazon Ads and other Amazon businesses.

To put this in perspective: Amazon’s advertising business is MASSIVE — in the ˜$50bn annual run-rate range, and still growing at a decent clip (19% year-over-year in Q3 2024).

In the advertising world, it's only behind Google and Meta, which is impressive considering it's much younger.

🤖🔎🧬 Nvidia’s Accelerated Biology Dreams: Partnerships with Illumina and the Mayo Clinic 🧫🦠🔬💊

If you’ve been following Nvidia for a while, as I have, it’s been clear that Jensen has long seen biology as one of the next “Big Things™️” to be digitized and accelerated.

He’s been talking about it for years, pushing his teams to develop products and services for biotech companies and academic researchers…

This latest announcement shows that they are still pushing hard:

Nvidia Corp. announced tie-ups with Illumina Inc., the Mayo Clinic and other health-care organizations as part of its push to encourage the adoption of artificial intelligence by that industry.

Illumina, a gene sequencing company, will adopt Nvidia’s generative AI platform for chemistry and biology, the world’s largest chipmaker said Monday in a statement. The Mayo Clinic is deploying Nvidia’s latest hardware and some of its software to improve digital pathology, the company said.

Friend-of-the-show Dylan Patel previously speculated that Nvidia may buy Illumina. I don’t know how likely that is — the cultures may not mesh well together — but this new partnership makes sense.

Dylan wrote at the time:

We think an acquisition of Illumina answers the following critical questions:

1. How can you develop full-stack solutions in genomics when 90% of data sequenced is captured with only the aid CPUs, gating new data production?

2. How can you develop full-stack solutions in genomics when the default flow of data from Illumina tools is straight to Amazon’s cloud, a quasi-monopoly capturing incredibly high margins, but providing very weak analytical tools?

3. How can you create intuitive workflows democratizing healthcare when data and analytical tools reside in silos?

4. Who will unify and standardize these disparate data sources and workflows in an industry inherently inept at building software?

5. How can you flip script from Eroom’s to Moore’s Law without building AI-informed solutions capable of spanning the healthcare ecosystem?

6. How can you claim to be the purveyor of accelerated compute without enabling this compute for the largest and most important dataset in humanity, the hardware of life?

(I don’t know if Dylan still thinks this — he has of course a right to change his mind)

The data aspects are particularly interesting. Here’s something that Nvidia mentioned about the Mayo Clinic deal:

we are partnering with the Mayo Clinic, the #1 hospital in the world for many, many specialties. And we are accelerating the next generation of AI-driven digital pathology.

Mayo today announced the Mayo Digital Pathology platform. This is built on an incredibly state of the art robotic digital pathology labs that has amassed a huge and unique data set of over 20 million whole-slide images that has 10 million associated patient records.

These datasets are likely fairly unique and extremely valuable for training models that can perform diagnostics and research in the space.

But the actual biology isn't the only target here — AI can transform many other aspects of healthcare and drug development that fall more into the realm of "logistics". Automating these processes could speed up progress and reduce costs (scheduling patients for clinical trials, record-keeping, filling out forms and double-checking for errors, testing more hypotheses, finding subtle interactions between molecules that humans may miss, etc).

📖🎙️ Podcast: Jensen Huang and the Nvidia Way

Speaking of Jensen, my friend David Senra (📚🎙️) just released a great episode on Tae Kim’s book ‘The Nvidia Way’, which doubles as a biography of Jensen:

• 🎧 #376 Jensen Huang: Founder of Nvidia

I encourage you to check it out. I’m almost done reading the book and I have a lot more to say on it. Stay tuned!

🇨🇦🤝🇺🇸 Canada’s Oil Production: A One-Customer Business 🛢️🛢️🛢️

I tried to find a graph that showed recent years, but this is the best one I found (it shows how basically all growth came from bitumen and synthetic).

97% of Canada’s oil exports go to the USA. There’s no real second place.

Source.

---

🧪🔬 Science & Technology 🧬 🔭

Weight Loss Through Pharmaceuticals: Let's Drop the Moral Baggage 💊 💉📉

Dr. Mike’s riff about using pharmaceuticals (like GLP-1 agonists) to lose weight is incredible.

It starts at 36min into the interview. The video above should start at that timestamp, and I’m also transcribing it here:

Chris: I think some people have an issue with the fact that they feel like it's kind of cheating, you know I had to get there the hard way...

Dr. Mike: I'd love to tackle that one.

I see where people are coming from — for the respectful part of my talk which ends now, and here begins the disrespectful part of my talk — but it's pure nonsense.

You use air conditioning. When you break your leg, you go to the hospital and you get antibiotics. You cheating mother*****! Aren't you supposed to be eating rocks and beating your head into a caribou to kill it or something like that?

As a matter of fact, the first tool industry millions of years ago that protohumans made, one of them is like the Oldowan tool — they discovered these flint rocks that are shaped in a no longer natural pattern, widowed down into something like a little club head or a little spike that you can at least like pull an animals guts apart easier or even kill something with — they were cheating! That's the first cheating they've ever done. Artificial, and it made them better at stuff.

Your ability to lose weight can be a test of your willpower or you can take the f***ing pill or the injection and with no added willpower lose all the weight you want.

And then you can test your willpower on tending to your family better, being a kinder person to others, being someone of moral authority to others by how you live your life, coaching Little League football and being on time for the kids instead of standing a f***ing Hardee's line getting your third cheeseburger for that day.

You can start a business, you can save the world, you can clean up the atmosphere — everything, we're not running out of problems.

So if we have body fat as this thing that's going to sharpen our will and we get rid of that, the set of problems remains approximately infinite still and you can sharpen your will on what's left over.

If you think you're tough improving your willpower to yourself by losing fat you're right, but you're not [doing other things] because you're draining all of your willpower on that one thing.

It's a good thing we have pills to help us with as much as possible so then we can have more bandwidth to do all the other good stuff that we want to do.

I think that’s correct.

The moral superiority element of having to do things the hard way here but not elsewhere needs to go. It’s a personal choice. I’m glad there are options for those who won’t/can’t do it the hard way and they shouldn’t be judged for that.

The stakes here are enormous: The health impact of obesity is so great that it’s sad to think that some people are reluctant about doing something that could make them live decades longer, get to know their grandkids, while feeling much better along the way, because they fear judgment and/or have been convinced it’s a “weak, cheater’s option”.

It’s purely arbitrary. As Dr. Mike points out, we use all these other things that aren’t natural and increase our convenience and efficiency — why not also have this in the toolbox for when it’s appropriate?

🏥 👧🏻 A Victory Story: Childhood Cancer Death Rates Keep Falling 👦🏻 🩻

Some charts you want to see go up 📈 but these make me very happy when they go down 📉

Thanks to scientific research, medical advances, and public health efforts, the world has made significant progress in treating childhood cancers. [...]

One reason for this progress is that scientists have learned more about the genetic causes of childhood cancers. This has helped identify children at risk earlier and develop targeted treatments with fewer side effects.

There have also been advances in immunotherapy, stem cell transplants, radiation, and surgeries used to treat different types of childhood cancers.

Consider this success story:

Survival rates for acute lymphoblastic leukemia in children have increased greatly, thanks to improved treatments and bone marrow transplants. Genetic research helped identify specific mutations which led to the development of highly effective targeted chemotherapy drugs.

Source.

🇺🇸✋🍺🍷🍸The Great American Sobering: Attitudes toward alcohol are rapidly changing in the US 🥃🍹

Source.

---

🎨 🎭 The Arts & History 👩‍🎨 🎥

🪕🎸 ‘A Complete Unknown’ — THE Film for Bob Dylan Fans

A few days ago, I saw ‘A Complete Unknown’, James Mangold's Bob Dylan biopic that covers the period from his arrival in NYC up to the legendary Newport Folk Festival concert.

Let me cut to the chase: I *really* liked it. Actually, I may have loved it. This is an "A-" or maybe an "A" film — I need to watch it again to be sure.

What makes it special?

They *really* went for it. Unlike other musician biopics that sprinkle a few songs between dramatic moments, this one puts the music front and center, which makes sense when looking at a musician’s life. I'd estimate that about 70% of the film is either music performances or songwriting scenes.

I love that, but it will divide the audience. If you don’t know Dylan and don’t know the music, it’ll probably be too much — you may have a bad time. But if you’re a fan and know the songs, it’s very satisfying!

They did a great job making you feel like you’re hanging around the Village in the early 1960s — the attention to detail is great, every car, every piece of clothing, every cultural reference feels authentic (with some small liberties for artistic license).

The cast is phenomenal: Chalamet, Norton, Fanning, Barbaro... They nail it.

I’m particularly impressed by the long takes with lots of dialogue and music where you see them play the chords, do the fancy finger-picking, and sing up close. Nowhere to hide!

Here's my plea: if you like Dylan, please go see it. Let's send a clear signal to Hollywood that we want more films like this. 💌

Want to maximize your experience?

Here's your homework: Listen to Dylan’s albums up to Highway 61 Revisited AND watch the Martin Scorsese documentary 'No Direction Home'. You'll be perfectly primed to get the most out of this gem.

Fun fact: Joan Baez is played by Phoenix from Top Gun Maverick!