（预览）Apple 如何改变移动通信经济学、SpaceX 想如何利用频谱、航空公司与运营商、Yann LeCun 离开 Meta

Hello, and welcome to a free preview of Sharp Tech. Hello, and welcome back to another episode of Sharp Tech. I'm Andrew Sharp, and on the other line, Ben Thompson. Ben, how are you doing?

I’m not good, Andrew. I’m pretty upset, to be totally honest with you.

I’m not doing well either, just for the record. I’m feeling chastened on this end of the line. Do you want to explain why you’re not feeling good and feeling upset right now?

Look, we started out on a great page. You said you were going to take one of my recommendations for winter.

Mm-hmm.

I’m feeling good. I’m glad you’re listening to me. Such a breakthrough.

Go ahead and drop another one, yeah.

It only took 2 to 3 years, which is good for you relative to—

Uh-huh.

You had to go through the period of just completely rejecting the opposite of what I say. I try to add on to it. We’re talking about jackets here. I’m like, “Oh, by the way, my favorite company, Kizik, has winter boots that are awesome. You could just slip them on, and it’s great.”

Yep.

And the disdain and dismissal with which you—

Immediate “Not doing that.”

No, it wasn’t just that you wouldn’t do it. That’s fine. The disgust with which you expressed it hurt.

I mean, look, nothing is nearer and dearer to your heart than the Kizik brand, and honestly, I feel a little guilty—

Actually, there was a contractor here at the house this week who was slipping his shoes on and off. I’m like, “It’s fine. Just leave your shoes on. It’s not a big deal.” And he’s like, “Oh, blah, blah, blah, blah. You don’t know what happened.” And I’m like, “Well, you should get Kiziks. You’re always taking your shoes on and off.” And he’s like, “Do you work for them?”

Ben is performing guerrilla marketing all over the world on behalf of Kizik.

Well, the great thing is, you know, we talked about how the 1 comp I ever accepted was from them, and then it got lost in the mail. I never received it.

Yeah.

Which is great, because I feel—

It was the universe validating your policy.

That’s right. I feel more empowered to endorse them to everyone. So, yes, I do have another take while we’re here.

Okay.

They’re my favorite. The whole new generation of D2D startups—sell online, all that sort of thing—obviously, they’re my favorite. My other take is, do you remember all this D2C stuff really started with razors?

Uh-huh.

There was Dollar Shave Club, and then the whole thing was, “They’re all the same. 5 blades, da-da-da-da.” So—

Oh, yeah. It was a tentpole part of the millennial business generation. I bought some Harry’s razors, for sure.

Yeah, so I went to Costco when I moved back, and I’m like, “Well, I’ll try some Harry’s razors,” because they were—of course, they’ve realized that Facebook takes all their money online, so now they’re trying to sell in retailers.

Uh-huh.

Terrible. Just a terrible razor. You can use it—

Yeah.

Once. You cannot—so you know what? Good for CPG. Good for P&G. You know what I’m going to do? I’m going to go buy a Gillette razor like I used almost my entire life, and I’m going to be happy about it. So—

Totally.

I feel like, good for the old guard.

Honestly, it was an important epiphany for me. When I say it’s a central aspect of the millennial experience, early in my 20s or maybe my mid-20s, Harry’s sounded like a better razor, and I felt very superior using Harry’s razors. Then I looked down one day and I was like, “These razors just suck.” I’m going back to Gillette and going with something I can trust. I’ll pay extra money for it, and it’s a much better way to live. But they get you with the marketing, and I felt like I was using a better, cooler—

No, it feels cool and hip, yeah.

Yeah.

And the terrible thing is, because I bought it at Costco, I have all these freaking razors to work through. But that’s fine, because they’re only good for 1 shave anyway, so I’m plowing through them. Unfortunately, I’m part Asian, so I don’t have to shave that often. What are you going to do?

Yes. Well—

I’m Asian. I’m a quarter Asian. But the Asian genes that I got mean I don’t look Asian.

Breaking news on the show here, yeah.

Yeah, no, I don’t look Asian, but I also cannot grow a beard. So it takes me longer than you would think to finish off this stupid pack of razors.

Fair enough. One day I’ll break the seal on owning my first pair of Kiziks. They won’t be Kizik boots, unfortunately, though, because it snows about 3 days a year in Washington, DC.

No, that is literally the best articulation of the Kizik idea. Putting on boots is really annoying. It’s one of those things where you just don’t go outside or don’t do stuff. It’s like, “Oh, I’ve got to put my boots on,” or you put your tennis shoes on because they’re easier, then you fall on your rear end.

Snow all in your shoes.

It hurts. You hurt yourself.

Sure.

Oh—

Yeah, well, the thing is, I just don’t own any pair of boots here in DC.

Oh, but—

That’s my coastal privilege.

Well, the other problem here is your feet get cold. You wear boots for warmth. That’s the biggest case here.

Well, maybe I’m built different over here in Washington, DC. I can handle some cold feet. In any event, enough prelude. Ben, we’ll dive in and begin with a story from Bloomberg. They write:

“EchoStar agreed to sell more spectrum licenses to Elon Musk’s SpaceX for $2.6 billion as it works to unwind parts of its 5G wireless network. The deal adds to a September agreement that the satellite TV and wireless company inked to sell SpaceX about $17 billion in wireless spectrum. Between that sale and another to AT&T for $23 billion, which EchoStar brokered under pressure from federal regulators, the company has been forced to give up on becoming a 4th major US wireless carrier as originally envisioned.”

So there’s the EchoStar story, but SpaceX is the star of the story. SpaceX has now purchased nearly $20 billion worth of spectrum this year.

Is it that high? I guess it is, yeah.

You’ve written about it a few times on Stratechery. We haven’t discussed it on the podcast. What is driving these investments? Where is SpaceX trying to go?

I think it’s really interesting. I’m glad we’re going to talk about it. The 1 funny line in this story, though, I just have to call it out: “The company has been forced to give up on becoming a 4th major US wireless carrier.” No, EchoStar declined to make the investments necessary to be a 4th wireless carrier.

And so they are appropriately being forced to give up the spectrum to people who will actually use it. This is downstream from some of the AT&T Mobile stuff. I can’t even recall; I’d have to go back and refresh myself on the details. Spectrum is a scarce resource. EchoStar has been sitting on a bunch of it, and we need that spectrum.

Yeah.

So—

Out of curiosity, what would the investments have been from EchoStar in order to make that spectrum more useful?

You have to actually build towers that use the spectrum.

Okay.

I think Boost Mobile is their thing, and they’ve built a little bit, but being a carrier is not a great business. I think that’s actually an interesting place to start. You go back way back when, and—well, let’s go back just to around the iPhone days.

Mm-hmm.

Carriers wanted to own it all. The carrier everyone dreamed to be was NTT DoCoMo in Japan: fully integrated. Of course, they had spectrum and built towers. They also designed their own phones. All the coolest mobile phones came from Japan, and the absolute coolest mobile phones couldn’t be used elsewhere because they only worked on NTT DoCoMo’s network.

And you had a lot of the things that people only realized they could do on phones once the iPhone came out. You could do them in Japan in that time period. It was a real classic example of the Clay Christensen concept that, at the beginning, stuff has to be fully integrated because it’s not—

Good enough. So in the U.S., when I got my first cellphone, I got a Samsung phone—I don't even remember what it was. I think it was on AT&T, or maybe it was Sprint. I think it might have been on Sprint back then.

Oh, wow. Throwback, yeah.

I had 4 or 5 lines of text, and, being me, managed to accrue a $1,000 cellphone bill because I figured out how to get the WAP browser, and I had an RSS reader on there. This was the summer I was living in Washington, D.C., because I was interning there, and I would go sit on the lawn of the National Mall and scroll through and just read the internet for an hour. That cost a lot of money.

That sounds pretty great.

Yeah, I know.

And may actually count as touching grass, just for the record.

I was literally lying on the grass.

Yeah.

But I was doing this on 4 lines of text, black-and-white. Pretty ridiculous. Had I been in Japan, I could have had a larger screen and probably a better experience.

Mm-hmm.

And they were doing it all. Everyone wanted to do that. They wanted to own the experience. If you got a cellphone, you only got the cellphone from the carrier; you bought your cellphone from the carrier, and it was tied to that carrier. They wanted to own the whole experience, because who wants to just build cellphone towers?

Sure.

At the end of the day, one cellphone tower is like another cellphone tower. It becomes a matter of who can build more. Your profitability is a function of how much you're willing to spend on capital investments, which is a tough game to play.

Mm-hmm.

So you roll around to the iPhone coming out. The company that had won that game by and large was Verizon. They had the best coverage; they still, I think, mostly have the best coverage. Obviously, there are a gazillion commercials citing all sorts of different things, but especially if you want at least 4G speed anywhere, I think Verizon is probably still the best. Again, don't quote me—

Okay.

—but I'm such a sicko that I actually have 2 carriers just to make sure I have as much coverage as possible anywhere. Look, I haven't changed at all.

Exactly. A sicko 20 years ago, still a sicko today. Love it.

Yeah. Well, I've been a T-Mobile customer for a long time because T-Mobile is by far the best if you live internationally. First, they were the first one to have a good international plan, where by default you got a certain amount of data in basically any country you went to, and then you got SMS and texting, and all of that just worked seamlessly.

Mm-hmm.

Everyone has that now, but T-Mobile is still the only one where, with AT&T and Verizon, you have to be in the U.S. most of the time, and then you can have a good plan abroad. T-Mobile will let you live abroad, only come to the U.S. occasionally, and still maintain your plan.

Okay.

So for anyone living abroad, T-Mobile is still the best deal, the best plan. The problem is T-Mobile's coverage has just sucked. It's fine in Madison; you get out of town, and it falls apart very quickly. I still value that international aspect.

Mm-hmm.

My wife is back and forth between here and Taiwan. She has a line also. So I still have T-Mobile, but I got Charter, the cable company.

I didn't know they did cell service. Wow.

Oh, man. That's going to break the heart of Craig Moffett. Yeah, no, there was a deal a while ago where Verizon had to give them a network. So both Comcast and Charter run on Verizon.

Mm-hmm.

They run at a lower prioritization level. I wouldn't tolerate that for my primary line. I want to be top priority. I don't want my data pushed aside for anybody else. I want it to work all the time.

Yep.

In this case, it works because what I want is coverage if I'm out in the boonies and T-Mobile doesn't work, which it probably won't. They did acquire U.S. Cellular, which I'm excited about, because U.S. Cellular has great coverage in Wisconsin. Anyhow, way too many details that no one cares about.

No, a fun little digression into Ben's fetishes as far as carriers are concerned.

Oh, I mean—

But—

We talked about that. It's very important to me to be online and connected all the time.

Everywhere.

So I have 2 phone carriers, and I'm—

An absolute sicko on the other line of this podcast.

Well, even though I got fiber, I still have Starlink on the roof. So I have 2 home internet providers—and I have 2 cellular providers. It's ridiculous. Oh, actually, probably—

Just in case there's an earthquake that disrupts the fiber connection that's running into your house now.

I did maintain my carrier in Taiwan as well, so I guess I have 3 cellphone carriers. But anyhow.

All right, so the economics of carriers, though. That's where we're beginning.

Yeah, so Verizon is generally the best, but Verizon wanted to leverage being the best into owning the full customer experience. It's akin to the whole thing with GM and CarPlay.

Mm-hmm.

“We want to own the dashboard. We can sell services,” blah, blah, blah, whatever it might be.

I mean, it's fairly intuitive that if Verizon can make the best phones, that's going to be a differentiator that makes me choose Verizon service, number 1.

Well, they never wanted to make the phones. They just wanted to control them, have control, have their services on there, and push people toward their favorite partners. It's the old integrated sort of playbook. You can see why you want to do that. The problem with integration—

I mean, think about if iPhones were only available on Verizon. Everybody would have Verizon service.

Well, we're going to get there. The problem with integration is, if it lets you do something that wasn't possible, you can win. If you have some way to lock in that advantage, you can maintain your win. But over time, if you don't have those 2 things and you can't scale to the broader thing, your advantage disappears.

I talked to the CEO of Unity this week, who had one of his stops at AOL back in the day, and we were talking about this in passing. AOL in the '90s was so easy. It was the way people got online.

Getting online otherwise—it's actually funny. I realized this week that the company I got fiber from, I was actually a customer 30 years ago, because I was in the “I'm too good for AOL” camp back when they first started.

Oh, of course you were.

Yeah, so I wanted to get a real internet connection. The company was called SuperNet. I remembered this week, “Wait, I pushed my parents to get a SuperNet ISP account—

An early adopter—

—way back when.”

Out there in Madison.

Yeah.

Great.

No, so I actually haven't told SuperNet that. I have to—

But AOL was the idiot-proof solution for the rest of us 20 and 30 years ago.

And they had so much more stuff on AOL. I remember being online in the '90s and being annoyed that I didn't have access to AOL, because there was so much on there that wasn't available on the internet broadly.

Uh-huh.

But the internet grew up. You got lots of stuff on the internet, and suddenly having to go through AOL—it could access the broader internet, but it just became this burden and this tax. It was cheaper and better to just get a regular connection to the internet and then—

And a browser. Yeah.

How did you find stuff there? Oh, Google.

Mm-hmm.

Google, in many respects, is what broke AOL. The internet grew up, and there was lots of stuff, but if you didn't know how to find it, it wasn't useful to you. Then Google made it possible for you to find everything.

Yeah.

And suddenly the whole reason for AOL just fell down under its own weight. This is a story that you see happen. What makes Apple unique is that, by integrating the software and the hardware, it does something no one else can do. Operating systems are so powerful because apps are built for them. It's the key thing you interact with all the time.

People don't churn from operating systems, and because you can only get Apple's operating systems on Apple hardware, Apple can make a lot of money by selling that hardware at a premium price.

Sure.

The reason this came to bear was with the iPhone, and I'm sure there have to be business cases about it. This is one of the things I wanted to write about when I started Techre. I'm like, “I need to write about this,” because people don't really get this. It's one of the all-time examples of reconfiguring a value chain around you, which is that Apple had this built-in customer base of people who first loved the Mac, and then a larger portion of people who loved the iPod.

Mm-hmm.

Even with the iPod, they had the iTunes component and the iPod component. It was hardware and software that worked together, and they got a lot of people using iTunes who then got an iPod. Now, some people got an iPod and thus used iTunes, but it was very much working together in that regard.

Yeah.

They come out with the iPhone. The iPhone's amazing, and everyone already has the software to manage the iPhone because they have iTunes. It becomes—Apple has this thing coming out, everyone is eagerly anticipating what it's going to be, and Apple goes to Verizon and says, “We're coming out with this iPhone. You have the best network. Of course, we want to be on the best network. Let's make a deal.”

Yeah.

Verizon's like, “No,” because Apple did not want to have Verizon branding on their phone. By the way, your phone—

Mm.

—used to have the carrier's brand on it. They did not want Verizon controlling the software. That's Apple's key to success: owning and controlling that. From Verizon's perspective, it's like, “No, we're leading—”

We have the leverage. Yeah.

“We have the—” Exactly. What Apple did was sign a deal with what was then Singular. Cingular took the opportunity of this deal to rebrand AT&T.

Mm-hmm.

Cingular was what happened when AT&T was split up and then sort of reunited over time. What happened, which had never happened previously, was that people started leaving Verizon for AT&T because they wanted the iPhone.

Strictly for the phone.

That's right.

Strictly for the hardware. Yep.

That never happened before. There was never a phone—even when Motorola had the Razr—where people clamored to get it on their network. They weren't—

Mm-hmm.

—leaving their networks to get that phone. Apple did this in market after market. It was so fascinating. In every market, they usually ended up launching with the second- or third-place carrier because the leading carrier resisted. The ultimate example of this was Japan. SoftBank was in a distant third place. They said, “Fine, we'll take it.” Japan is a massive iPhone country, and NTT DoCoMo was one of the last carriers ever to carry the iPhone, precisely because NTT DoCoMo was the AOL of phone carriers.

Hmm.

They had the full experience and the better experience. They were superior, and of course they wanted to hold onto that. Apple goes in, and SoftBank goes from this distant third place to—I can't remember if they passed NTT DoCoMo. I think they at least tied them as far as—

Okay.

—share. They just obliterated NTT DoCoMo. Finally, around 2014—I think it was with the iPhone 6, because NTT DoCoMo and China Mobile came out in, I believe, the same year, around the same time—they gave in and started supporting the iPhone. Now the iPhone is just totally dominant in Japan. Everyone who stuck with NTT DoCoMo was like, “Oh, thank God, finally.”

By Apple controlling the hardware and software, they completely reconfigured the value chain for carriers, and—

Right, and turned every carrier into a commodity at the end of the day.

That's right. Everyone over time is always like, “Why doesn't Apple become a carrier?” Apple loves to integrate things. They could provide a better experience. It's like, well, number one—

Well, let them spend the money.

No, they can't provide a better experience because experience isn't a function of Apple's skill set. The experience is a function of spending capital—

Yeah.

—to build towers. That's a bad business for Apple in particular, which is making massive margins selling these things. More than that, because there was already competition in the market, what actually happened was that Cingular made a lot of money on the iPhone by selling super-high-priced plans. They also spent an astronomical amount of money because iPhones were using so much data.

Mm-hmm.

So they had to spend that money just to stay competitive, and that happened to everyone. There was this massive build-out of equipment simply because the amount of data people were using exploded.

To serve the iPhone user base.

That's right.

Yeah.

People forget—

That makes sense.

The first few years of the iPhone were like this: There would be long periods where you just couldn't use it because the towers were so saturated, particularly in big cities. It's much better now, but it's much better now without Apple spending a dime on it.

Right.

Right?

Because everyone else but Apple spent a lot of money in order to ultimately serve Apple's interests.

The whole world did that. More than that, actually, one of the reasons why Apple's earnings were so predictable through about the iPhone 6S—

Mm-hmm.

—is because what would happen was Apple, as this is becoming clear, was gaining leverage in every market, and every carrier that didn't have an iPhone was bleeding customers.

Yeah.

So now they're coming to Apple begging for the iPhone. What Apple would say was, “Okay, you can have the iPhone, and you have to sign a sales guarantee of how many iPhones you're going to sell.”

Ah, okay.

People wondered why Apple's forecasts—

They know exactly—

—were so perfect.

—how many iPhones are going to be sold. Yeah.

It's because it was contractually guaranteed how many iPhones they would sell. What was super funny, and I didn't even realize this until after the fact, is that I think it was the iPhone 6S. Suddenly, Apple kept missing earnings—

Uh-huh.

—both up and down. They sold more than they expected, then fewer than they expected. It's like, what happened? Did they become totally incompetent? No. What happened was, with the iPhone 6, they finally ran out of carriers to add.

I was gonna say.

They got NTT DoCoMo, and they got China Mobile.

Sign one of those deals.

Right. And so they could no longer lock in guaranteed sales because they no longer had total leverage in the marketplace.