Things My Daughter Has Recently Said

Being a partial explanation for why I have not posted much recently.

Sandor:  What color is the hat?

Baby:  Purple.

Sandor:  Well, actually, it’s green.

Baby: …No, purple!


As my wife and I are engaged in a normal conversation in the car.

Baby:  No talking!


As we are eating at a sushi restaurant.

Baby:  More dirt!

Sandor:  What?

Baby, gesturing:  More dirt!

Sandor:  You mean tobiko?

Baby:  Dirt!


The Vote was Not Rigged

If you’re reading this post, you probably know a decent amount about computers, math, and (less certainly) statistics, voting machines, and how the electoral college system works.

Suppose that you were on the team of J. Corrupt Fakeypants, a presidential candidate in 2020, and Mr. Fakeypants has asked you to lead the effort to rig the vote in his favor.

Fakeypants has a number of priorities:

1. He does not want to go to jail. So he wants you to do this in a way that does not point a giant neon sign at him.

2. He wants to be President. So he wants you to do a good job of rigging the election.

How do you do it? I think it’s pretty hard.

First of all, Fakeypants needs to be doing pretty well. Nobody is going to believe that the polls missed a 20% swing in the election, and if the results seem that way, there will be lots of people investigating. Fakeypants will probably go to jail.

Second of all, if Fakeypants actually wins, your efforts run the risk of sending him to jail in a scenario in which, you know, he actually wins.

Third of all, every vote you tamper with increases the risk that your vote tampering is discovered, so you want to tamper with as few votes as possible, but…

Fourth of all, if the margin is ultra-close, you will trigger recounts that increase the possibility that your tampering is discovered.

We will stipulate that installing malware on any given voting machine is fairly easy (though each time you do it, you run a very small risk of being caught). And each type of voting machine does require a custom-created piece of malware (and each hacker you bring into your project increases the risks of leaks to your project).

Bringing a well-resourced state actor into play may help you overcome the scale problems of your project, but well-resourced state actors are also the targets of cyber-intrusion from the US security apparatus, and you have a hard time gauging how likely it is that if you ask your buddy V. Putin to help out, the NSA will know immediately.

The voting machines are for the most part isolated from the internet on the actual voting day and during the counting process. You also have no polling that you trust to be better than the overall aggregate public polling as filtered through good aggregators.

What is the method you use to maximize your chance of winning the election and minimize your chance of being caught, with the constraints above? I think that it’s a very difficult project, probably unfeasibly difficult.

Unconvinced on Trump’s Traditional Racism

I think that Scott Alexander pretty much nails the issue of Trump’s supposed white supremacy here.  A few excerpts that I agree with:

And if you believe he’s lying, fine. Yet I notice that people accusing Trump of racism use the word “openly” like a tic. He’s never just “racist” or “white supremacist”. He’s always “openly racist” and “openly white supremacist”. Trump is openly racist, openly racist, openly racist, openly racist, openly racist, openly racist, openly racist. Trump is running on pure white supremacy, has thrown off the last pretense that his campaign is not about bigotry, has the slogan Make American Openly White Supremacist Again, is an openly white supremacist nominee, etc, etc, etc. And I’ve seen a few dozen articles like this where people say that “the bright side of a Trump victory is that finally America admitted its racism out in the open so nobody can pretend it’s not there anymore.”


1. Is Trump getting a lot of his support from white supremacist organizations?

No, because there are not enough organized white supremacists to make up “a lot” of anyone’s support.


Stop centering criticism of Donald Trump around this sort of stuff, and switch to literally anything else. Here is an incompetent thin-skinned ignorant boorish fraudulent omnihypocritical demagogue with no idea how to run a country, whose philosophy of governance basically boils down to “I’m going to win and not lose, details to be filled in later”, and all you can do is repeat, again and again, how he seems popular among weird Internet teenagers who post frog memes. In the middle of an emotionally incontinent reality TV show host getting his hand on the nuclear button, your chief complaint is that in the middle of a few dozen denunciations of the KKK, he once delayed denouncing the KKK for an entire 24 hours before going back to denouncing it again.

Trump is clearly not “openly white supremacist” in the way that tons of people were 50 years ago and a very small minority of people still are.

I would believe that Trump is notably “more racist than most 70 year old white guys” in a newfangled 21st Century where racism means being more willing than is rational to believe that Arabs are all one step from being terrorists or Chinese people have an insidious plan to steal America’s economy or Mexicans all long to come to the United States, immediately stop working, start having babies, and just consume social services.

And the above beliefs are wrong and bad and stupid.  But they aren’t dressing up in white sheets and lynching black people, and conflating them is wrong.  When you try to make it look like Trump is a classic early-mid 20th Century white nationalist, you undermine your ability to address the tangle of anxieties that lots of people clearly have about the modern world.

Scientific Environmentalists are Junior Coalition Members

David Roberts at Vox has a long article on Initiative 732 in Washington state, which I found through this article at Marginal Revolution.

The Vox article is worth reading in the entirety, but the summary version is that Washington’s Initiative 732 is a carbon tax proposal that seems very well-crafted to actually reduce carbon emissions, and to do so in a way that could be a template for other states or federal level regulations.  But the initiative is receiving no support from either environmentalist organizations in the state nor from broader “left” organizations.

This is, in my view, something of a piece with why the environmental movement (and/or the left) is not going to embrace nuclear power or geoengineering.  There is a subset of the environmental movement that I’ll call “scientific environmentalists,” which have these traits:

  1. They genuinely believe that there is currently one or more environmental crises going on that could have catastrophic consequences in the course of the next few decades.
  2. They want to take actions to prevent those catastrophic consequences at all costs.
  3. They believe that the traditional tools of western rationality can tell us how to avoid those catastrophic consequences.

Now, the thing to understand is that this is a group who look pretty good on television, but they are none the less just a part of a coalition  — both within the environmental movement and in the broader left.  And that means that they basically don’t get to put any of their ideas into practice.

The environmental movement is a coalition of the scientific environmentalists along with people who I’ll call “spiritual environmentalists,” who are broadly people who regard either industry, modernity, or western rationality as suspicious and unnatural.  Spiritual environmentalists are not on board with nuclear power or geoengineering.  They aren’t necessarily opposed to things like the carbon tax of Initiative 732, but it also doesn’t excite them.  It doesn’t play to their interests, which aren’t about technocratic solutions to mitigate climate damage, but about making a visible shift away from industry, modernity, and western rationality.

The environmental movement tends to put scientific environmentalists in a more visible position than the spiritual environmentalists.  The left is committed to telling the right that the left is smarter, better educated, and more rational than the right, and the scientific environmentalists play into that narrative.  Meanwhile, the left is still trying to distance itself from the perception that they’re hippies, and the spiritual environmentalists have a worryingly hippie-ish vibe.  But despite that visible nature, my guess is that the majority of the environmental movement is spiritual environmentalists, not scientific ones.

Even moreso than that, environmentalists per se are junior members of the left coalition.  Again, they may get a pretty large amount of visibility.  Environmentalists (especially scientific environmentalists) look pretty good on TV, compared to say labor union leaders.  Especially if you’re playing to centrists.  But they don’t have any money, they don’t control large organizations that get out voters, and they don’t hold onto people’s loyalty the way that ethnic or cultural blocs do.

The way that life works when you’re a junior coalition member is that you bring your assets to the table to support initiatives created by the senior members of the coalition, and they in turn craft their initiatives to have incidental benefits to your cause, and they also pay lip-service to your cause.  You don’t lead policy as a junior coalition member.

So the disconnect is that scientific environmentalists have a prominent public facing role in the left coalition, despite being junior members of a junior member of the left coalition.  And they get lip-service, and because of what scientific environmentalism is, that lip-service means saying things like “if we don’t act now, billions of human lives may be taken by dramatic climate change.”

But don’t take that to mean that scientific environmentalism is powerful, or able to lead policy.  If they’re to get their agenda advanced, they need to find ways to attach it to the interests of more powerful coalition members.  Or find a new coalition, but honestly I don’t see any place more hospitable to them.  Maybe if Trump breaks the Republican party completely, one of its successors will be a more natural place for scientific environmentalism, but the current Republican coalition is just obviously not going to advance their agenda.

No Successor to Lithium-Ion Batteries has Arrived

Robert Cringely wrote, in March 2014,

[A]lternative battery chemistries intended to dramatically increase the range and decrease the cost of electric cars. I don’t know which of these will ultimately dominate but I am sure one will which is why I can be so confident in predicting a black swan. With dozens of groups working on the problem and an eventual market worth probably $1 trillion I have no doubt there will be a solution within the next couple of years.

He describes them as in the general range of “Twice the range, one tenth the recharging time and safer, too.”  It’s been more than two years since he made that prediction.  Alternative battery chemistries remain lab-bound and none of them are noticeably closer to being industrialized today than they were two years ago.

Predictions are hard.

No Return to Apple’s Glory in Cars

Recently, Neil Cybart got a lot of attention on Hacker News for an article claiming that Apple would “pivot” into cars.  This is a very, very stupid claim.

The reason it’s very, very stupid is that Apple isn’t going to “pivot” into anything.  Pivoting has a meaning, and that meaning is not “introduce a new product.”  It’s changing your focus.  Apple is not some small agile startup with nothing to lose and everything to win.  It’s the most valuable company in the world, and if the iPhone and iPad lines are past their peaks, they are none the less incredibly valuable properties that will continue to produce gigantic profits for Apple for many years to come, and obviously Apple, while interested in finding another big product or two, will continue to focus the majority of its organizational weight behind the iPhone until the iPhone is doing much, much worse than it is now.

Okay, you say.  But you’re just poking at the sensationalist headline.  The meat of Cybart’s claim is that Apple will make a big bet in cars and has a chance to resurge from its current doldrums on the back of a big play in cars.

That claim is not very, very stupid.  But it’s still wrong.

The car industry simply isn’t big enough to compensate for a (let’s say for the sake of argument) rapidly weakening smartphone market.  Not even if driverless cars are coming.  Not even if the ownership model for cars changes.  In fact, especially not in that case.

First, to be clear, let’s look at the Apple of 2015.  Apple of 2015 had revenue of $234 billion, with a gross margin of nearly 40% (!!!) and a net margin of about 20%, meaning it racked up somewhere in the vicinity of $50 billion in profit over the course of the year.  Holy shit, guys.

Toyota is the most valuable auto company in the world, with a market cap of $150 billion.  So just to be super clear, Toyota’s market cap is less than 2015 Apple’s revenue, and right around 3x of 2015 Apple’s profit.  It’s market cap is around 1/5th of 2015 Apple’s.  Its gross margin hovers around 20%, or half of 2015 Apple’s.  In fact, even today, the somewhat diminished Apple’s market cap still exceeds Toyota’s + Ford’s + Honda’s + GM’s.

So what does that tell us?  It tells us that even if Apple is wildly successful in the car business, it would have to fundamentally change the car business to support a valuation that comes close to the height of the smartphone wave.

Okay, you say.  But maybe Apple can make such a change.


I think it goes without saying that you can’t overturn the car industry with a car that looks cool or has a really nice entertainment system or extremely clever rear-view mirrors.  No amount of design that Apple brings to the table is going to change the auto industry from a capital-intensive, low-margin, hyper-fragmented industry to a extremely high-margin, unified one.

So what else might happen to cars soon?  Obviously, cars might go electric (or hydrogen-powered, I suppose) and they might become driverless.

The size of opportunity of cars going electric is more or less the value of fuel savings.  A 20mpg car that drives 15,000 miles per year uses about 750 gallons of gas.  At $4 per gallon, that’s $3,000.  So far, even a pretty compromised electric car costs at least $6,000 more than a comparable gas-powered one — there’s no way we’re getting from there to Apple convincing everyone to plunk down an extra 20% margin to support a smartphone-like business.  The market for people who’ll pay lots of extra money for the dubious environmental benefits of an electric car just isn’t that big: if it’s overall a big waste of money to avoid gas, people aren’t going to avoid gas.  If it’s not a big waste of money, you can’t support Apple 2015-style profit margins, and that means you can’t support Apple 2015-style valuations.

A move to full-on driverless cars is potentially a much bigger revolution.

Before we go into that, though, let’s take a realistic view of Apple.  Apple is not an AI titan.  Probably the biggest AI play that Apple has ever made was Siri.  And Siri was an acquisition.  And Siri has been, I think it’s fair to say, pretty disappointing.  Even if Apple has somehow built a world-class AI team with relatively their (characteristic) secrecy, you can’t bring a self-driving car to market without testing it on real roads in real driving conditions, and to do that lawfully, you need to notify the government, and Apple hasn’t done that (I also don’t believe that Tim Cook’s Apple just decided to cowboy it without the necessary permits).  You have to believe a real series of improbabilities to imagine that Apple has best-in-class driverless cars up its sleeve.

But let’s say that against all odds, Apple comes to market with driverless car technology that’s much better than all the known players in the market, in a similar way to how iPhones were so much better than any other smartphone in 2008.

Even then, how does Apple get to the high margin, high volume business that would support a circa 2015 valuation?  Cars are expensive.  I think it’s reasonable to say that people just can’t pay afford to pay a massive premium on the cars they already have, even if those cars are really, really, really cool.

Going to an Uber rides-on-demand model is even worse.  The entire point of driverless cars + rides on demand is that it increases the utilization of each vehicle and lowers the cost that each consumer pays for cars by allowing them to share the costs of each vehicle with some number of other consumers.  Even if Apple swallows the entire auto industry (and just think of the capital costs associated with trying to produce tens of millions of vehicles — these things are way more expensive to manufacture than iPhones), the entire auto industry does not currently add up to the market cap of Apple, and the entire point of rides-on-demand + driverless cars is that the total market for vehicles would shrink!  If it wouldn’t — if the market would grow instead — that means that ipso facto, consumers are spending more on cars in the rides-on-demand world than they were on their cars today.  At which point why wouldn’t they just buy the cars?

In conclusion:

Might Apple be planning a foray into the automobile world?  Sure, it might.

Could Apple imagine that an Apple Car would become the main pillar of its business?  I very much don’t think so, but maybe they know something I don’t.

Would an Apple-as-principally-an-auto-company future be good news for Apple?  Not compared to the heyday of smartphones it wouldn’t.  If Apple were genuinely planning to in even the broadest and most loosest of terms pivot into automobile manufacturing, that suggests that they know some really unbelievably dire news about their personal electronics business.

Much more likely, any foray into the auto business would be a side venture for Apple, one where they’d stick to a small luxury market where they can have good margins at the cost of being a relatively small minority of their overall business.