Janice Kai Chen at the Washington Post on pigeons versus the internet:

At certain data volumes and distances, the pigeon is a quicker option for large swaths of rural America, where internet speeds can lag far behind the national average.

And not just rural America. As I write this from the nation’s capital, speedtest.net reports 24 Mbps up. Federal agencies should bring back pigeons for sending large files back and forth.

They are converting a modern office building into condos a few blocks down from my apartment, and by the looks of it they may as well have torn everything down and built it anew. I hope they will do that will all the brutalist federal garbage downtown, the FBI building first. Meanwhile, the late 19th-early 20th century townhouses scattered around DC have been switching seamlessly from commercial to residential and back for a hundred years now for little to no cost.

Optimization and scale: they work great, until they don’t. Just ask a salaried physician working for a conglomerate in the medical-industial complex, a large-scale operation which is being optimized to death (sadly not its own, but that of its component parts — patients and health care workers alike). All those large reptiles and mammals are extinct for a reason.

We discussed the problem of scale at the first RWRI I attended back in August 2020, the Beirut explosion still fresh in everyone’s mind. Less than a year later, a big ship blocked the Suez channel, as if to reinforce the message. I expect Nassim Taleb’s next book will have a chapter or three on the problem, even if “scale” doesn’t make it into the title.

What goes for biology, architecture, and logistics also goes for industry, and if there is one hyper-optimized massive-scale operation around, it’s Apple’s iPhone production. If and when its production chain comes toppling down, it will not be a black or a gray swan event, it will be snow-white, which is why I suspect (or, as an iPhone user, hope) they have contingencies.

And in practicing what I preach, I have slowly been transitioning away from GTD levels of hyper-productivity and into a 40,000 weeks mindset. Whether this is a sign of wisdom, experience, or just plain old age, well, who is to say? Why not all three?

The holiday season starts as soon as I get my first email from the Wikimedia Foundation, reminding me that “last year, you donated x dollars…”. And those emails work, despite some recent made-up controversies. Wikipedia is the wonder of the modern world.

There are books you read once and toss out ⊕ See also: anything by Malcolm Gladwell, Michael Lewis, or any other permanent airport book store resident. and those which keep on giving, and René Girard’s I See Satan Fall Like Lightning falls firmly in the latter category.

One concept I have come to appreciate more thanks to the book is that of the scandal. ⊕ As in: “the reaction of moral outrage and indignation about a real or perceived transgression of social norms”, not the TV show. Though I’m sure the show has its fans. Girard has a book — or rather, a collection of essays — with “scandal” in the name, but both he and Luke Burgis focus more on mimetic desire and how it can lead to conflict; the build-up of scandal is “just” a stepping stone, something natural and completely expected of humans. This may be related to their catholicism, but I am neither a theologian nor a philosopher, so I’ll refrain from speculating further.

What I won’t refrain from, however, is flipping my thinking from human desire causing scandal to the human desire for scandal. It a phenomenon not exactly like, but closely related to, schadenfreude — the pleasure in the misfortune of others. In fact, a search for Skandalfreude does return some relevant hits — one of them from Stefan Zweig, no less — so let’s use that word to describe the pleasure in being scandalized or, more broadly, the desire to be scandalized. And I suspect that, similar to schadenfreude, there is a Gaussian curve of people’s propensity for experiencing it in general and, flipping the axes, a Gaussian curve of the number of people with the propensity for it when it comes to a particular topic.

The first bell curve pits the “Karens”, ⊕ I am misusing the term Karen here, and contributing to it becoming a suitcase word. For this, I apologize. You could replace it with “the woke” in your mind’s eye — again, a misuse! — and get the same intended result. Funny how that works.scandalized by everything, opposite the phlegmatics, scandalized by nothing. On the ends of the second curve lie the haters — who think that a particular company, person, ideology, etc. is evil — and the fanboys, to whom that same entity can do no wrong.

Regarding the second, entity-based curve: the higher the profile, the fatter the tails. This we all know intuitively. When Apple causes an uproar for their “shot on the iPhone” Scary Fast event, it is the skandalfreude fat tail poking its head. Knowing how many people pour over every one of Apple’s actions just wanting to be scandalized — to the point of paying to be scandalized — it is a small miracle these storms in a teacup don’t happen more often.

So with those two curves in mind, Girard’s insight is this: mimetic desire makes only one of their ends “sticky”, the one that “likes” scandal. With a critical mass, it is no longer a bell-shaped curve at all, but leans towards the exponential. The entity to which this happens becomes a scapegoat and is stoned to death, in the true sense millennia ago, nowadays only metaphorically. This is why Apple and the rest of Big Tech should tread lightly: their tails are fat enough that even small missteps, or perceived missteps, cause controversy. A full-blown mistake or, worse yet, a true transgression, would hurl the skandalfreudeian mass ⊕ Or has that already happened?towards the deep end and make them into a scapegoat for all of the society’s ills.

The practice of doomscrolling could be viewed in this light: trawling thorough the timeline, waiting for the next opportunity to experience some skandalfreude, maybe even jump onto a stoning bandwagon or two, at no personal cost. Much of online behavior seems less bizzare when viewed through this mental model, and for that alone it is a good one to have.

On this day 85 years ago, at 8pm Eastern Time, Orson Welles performed The War of the Worlds live on radio. Things escalated quickly. I will give major nerd points to Apple for even an oblique reference to the radio drama at tonight’s event, but I am not holding my breath.

I read with great interest Sam Altman’s essay *How to Be Successful. ⊕ Five years after it came out! Better late than never. It presented many interesting nuggets of wisdom, such as:

You get truly rich by owning things that increase rapidly in value.

…

The best way to make things that increase rapidly in value is by making things people want at scale.

So, can someone please explain to me, like I am a nine-yer-old, how did Sam Altman become successful? It couldn’t possibly have been Loopt, a little-known geo-social service. Well, the very last sentence in the essay, added as a response to Hacker News comments, gave it away:

I am deeply aware of the fact that I personally would not be where I am if I weren’t born incredibly lucky.

He should have started with that! Silicon Valley people apparently like talking about “first principles”, and it seems clear to me that the first principle of being wildly successful (or “truly rich”, which Altman tellingly equates) is to be lucky; everything else is ex post narration. I want to read an essay that starts with luck: how to recognize it, how to expose yourself to it, how to benefit from it. This will inevitably become an essay about probability, so Nassim Taleb would be the perfect person to write it, and indeed he did! Only the topic is much too complex for a short essay so yes, it’s a book, and to fully appreciate it you may as well read the whole of Incerto.

And if I am contrasting Taleb with Altman — which may be an unfair comparisson to Altman as there is a bit of an age differential — here is what Sam says about hard work:

I think people who pretend you can be super successful professionally without working most of the time (for some period of your life) are doing a disservice. In fact, work stamina seems to be one of the biggest predictors of long-term success.

The parenthetical is doing a lot of work here, and the thing left unsaid is that with this ethos you can end up working really hard your whole life and end up safe from poverty, but not wildly successful… like most Americans! Here is prof. Taleb:

Solid financial success is largely the result of skills, hard work, and wisdom. But wild success (in the far tail) is more likely to be the result of reckless betting, extreme luck, & the opposite of wisdom: folly.

Indeed.

P.S. While the essay reads better than the recent Techno-optimist hocum (a low bar), did he really need 17 people to review his drafts, including… Diane von Fürstenberg? Seriously?

We are at the pediatrician’s office for an annual physical. The gowned offspring is holding my phone, reading a school-assigned book. The phone is also logged into a work meeting, which I’m listening to via AirPods. That same phone is also a hotspot for my laptop, which I’m using to write this. So there is a reason Apple is a trillion-dollar company, and I do hope they sort out their issues with China, lest we lose some of the magic.

Speaking of honeymoon periods and tech: Bing has become unusable, with slowdowns, plain unavailability, and the occasional gobbledygook. So — goodbye, Edge, it was nice while it lasted. Bard is faster, better-formatted, and available on Safari, to which I keep coming back.

As the holiday season approaches, let me note that a great use for LLMs, be it ChatGPT or Bard — and I’ve switched to Google’s lately — is for gift ideas. I now someone who likes DixIt and Mysterium, and it came up with Obscurio in seconds. It also pointed me to a good game for a tween who likes cats. For a six-year-old, too. And not a sponsored link in sight! Let’s see how long the honeymoon period lasts.

Tim Harford:

The proportion of genetics papers with autocorrect errors was estimated in 2020 to have reached 30 per cent. The Human Gene Name Consortium decided to rename the genes in question, wisely accepting that this would be easier than weaning researchers away from Excel.

At the intersection of science and technology lies the festering boil that is Microsoft Office.