Re: my post today on the still-vaporware Tesla Roadster and Benedict Evans’s maxim that “Elon Musk is a bullshitter who delivers”, a reader sent me this YouTube short of Charlie Munger’s answer to a question at Berkshire Hathaway’s 2020 shareholder meeting regarding Tesla’s volatility and Musk’s self-promotion. Munger said:

My thoughts are two. I would never buy it, and I would never sell it short.

[Audience laughs.]

I have a third comment. There was a man out in Los Angeles for years named Howard Ahmanson. He once said something that I’ve taken to my heart: “Never underestimate the man who overestimates himself.” I think Elon Musk is peculiar and he may overestimate himself, but he may not be wrong all the time.

Apple Developer News:

As outlined in our App Review Guidelines, pornographic material has long been prohibited on the App Store — this includes apps that generate nudity or sexually explicit imagery, such as “nudification” apps that create sexually explicit images of real individuals without their consent. Apps that host user-generated content are required to filter objectionable material, give users a way to report offensive content, and let users block abusive accounts. Apps containing hidden or undocumented features are also prohibited on the App Store, including apps that misrepresent their purpose during review and only reveal harmful capabilities — such as “nudification” — after approval, a tactic commonly known as bait-and-switch.

Their post includes the specific changes to the guidelines. Let’s hope the guideline changes are accompanied by better enforcement.

Of course, it’s worth noting that the best known perpetrator of AI-generated non-consensual nudification is X, which (checks notes) is where Apple executives do all their social media posting, and which, yes, I argue once again marks the spot in the zeitgeist. The zeitgeist is not clean.

Joe Rossignol, reporting for MacRumors back on 9 March 2015:

Apple CEO Tim Cook on Monday at the “Spring Forward” media event in San Francisco announced that Apple TV will now cost $69, discounted from its regular price of $99. The new price for the set-top box arrives on the heels of the announcement of HBO Now for Apple TV. The new price appears to apply to the third-generation Apple TV, as the company did not announce a new set-top box on stage at the Yerba Buena Center this afternoon.

Basic Apple Guy posted a short clip of Tim Cook, on stage, announcing the price reduction. Cook said:

And it’s got a very attractive entry level price of $99. But today, we would like to make it even more accessible for more people so they can take advantage of this great new content. So we are lowering the price today to $69.

The audience greeted this with whoops and cheers. (This, despite it being a smaller media event — Apple always packs the front of the house with employee ringers. Whoops and cheers are one of the things Apple is missing out on with their post-COVID pre-filmed keynotes.)

Apple continued selling the 3rd-gen Apple TV (not to be confused with “Apple TV 4K”, which restarted the generation counting in 2017 alongside the introduction of the iPhone X) for $69 until October 2016.

Inflation adjusted, a $69 Apple TV box today would cost a little under $99. Even if we only consider the $99 price in 2015 dollars, that’s about $137 in today’s dollars. Round that up to $150 and it would still be half the price of the only Apple TV hardware Apple now sells: this week’s new 4th-gen Apple TV 4K.

The new Siri Remote, with precision Find My support (and a speaker), is compatible with all prior generations of Apple TV 4K. So if you just want the new remote you can just buy the new remote. But it costs $59, which is almost as much as an entire Apple TV (with remote, of course) did in 2015.

Mark Zuckerberg, on X:

Muse Spark Open — weights releasing today — runs great on new Mac Studios thanks to work with @johnternus and Apple.

John Ternus, replying:

Congrats to the team! Amazing to run a model of Muse Spark Open’s caliber and scale this efficiently, right from your desk. Mac Studio keeps pushing the boundaries of on-device AI!

You will recall that the M5 Ultra Mac Studio with 512 GB RAM is set to ship by “the end of October”, and we’re midway through the month. You will also recall, from that same announcement, that multiple Macs can be clustered together via Thunderbolt — four Mac Studios can be connected to each other.

For Ternus, I could do without the second exclamation mark. For Zuckerberg, it’s fascinating that he had to post this on X, despite owning Threads, because no one at Apple uses Threads.

Zuckerberg has been posting somewhat regularly to X for months now, particularly regarding AI. Alex Heath asked him about it on his Sources podcast a month ago. From that one-minute clip:

Threads is doing great. I think it’s actually ... either bigger than X at this point, or it’s like very soon about to be. But, I mean, look, there are different communities in the different places. There’s a lot of AI folks on X, and I think part of what you try to do with social media is just communicate where people are, right?

Lastly, there’s this “1984” spoof posted by Alexandr Wang, Meta’s chief of AI and mulletry, which I find both amusing and painful.

From the NetNewsWire blog:

The latest version of NetNewsWire for iOS supports iPhone Duo, which is very much a novel iPhone with toolbars running up the side and all kinds of behavior.

How we did it: we didn’t do any work at all. We built the app using Xcode 27.1 and the default iOS SDK that comes with it, and that’s it.

Here’s the thing, though: we were already using standard controls — toolbars, navigation controllers, split views, etc. — and we were paying attention to size classes and such (as opposed to just looking at device type).

Develop your app the right way and, depending on the app, it might literally be no work at all to produce a great version for the Duo. Develop your app the wrong way and it might take months. NetNewsWire is one of my most-used apps. I keep it in my home screen dock, alongside Messages, Notes, and Safari. It would drive me nuts to use a Duo without a version of NetNewsWire that adapts to the Duo’s novel display sizes.

Jonathan Gitlin, writing for Ars Technica:

Nine years after first trying to resurrect the Roadster, Tesla is apparently trying again. CEO Elon Musk might have relocated its HQ to Texas, but yet again he has returned to California’s showbiz industry to hawk his wares; last time it was a Hollywood back lot for the remote-controlled Cybercab demo, last night it was decades-old stunt technology to make a show car do tricks in the air. [...]

Musk claims that the Roadster will go into production by 2028. But the CEO has a long and glorious history of overpromising and underdelivering; in 2017 he told attendees at the previous launch that “We are making it now.” Using the company’s referral program, some people were in line to receive two free cars. But those free cars never showed up, and neither did the ones that people put down $50,000 deposits for; last year Sam Altman discovered how hard it can be to get that money back.

I like a good stunt show as much as the next guy, and this looked like a cool stunt show. But people are losing their minds if they think this stunt show makes it any more or less likely that Tesla will deliver a Roadster as promised. This Reuters story makes it sound like they think Tesla is going to sell cars that do this stuff. Last night’s stunt show has no more bearing on actual production cars Tesla may or may not ship than George Lucas’s Star Wars movie had on Ronald Reagan’s “Star Wars” missile defense system.

The Tesla Roadster is vaporware. But Elon Musk is very good at distracting from unshipped vaporware by delivering on stunts. Two years ago it was robots that were totally faked. Last night’s car stunt involved real things that actually happened but only in the way that David Copperfield made the Statue of Liberty “disappear”.

Over at Jalopnik, Amber DaSilva has another good take, which concludes:

Tesla has long made up specs for its cars, made wild claims that weren’t substantiated by users, and now it’s making more big impressive claims. The “cars” used for the demos were likely nothing more than shells outfitted with exactly enough equipment to make the tests work. The 0.81 second zero-to-60 number wasn’t even shown with an on-screen timer during the event, let alone independently verified. Tesla has yet to prove anything about the new Roadster works in the real world, or could even exist within the realm of physics that those of us not named Musk occupy.

Roadster preorders have reopened, with the company once again taking $50,000 checks to reserve a car that’ll theoretically be made two years from now. It hasn’t yet said what will happen to the people who placed a reservation for the car Tesla announced nine years ago, which looks and functions completely differently to the one that bears its name today. And — arguably most concerning — we don’t know, if Tesla rakes in more money off the Roadster name, just how much will go to some malicious end.

And on Threads, Benedict Evans reiterates his longstanding take on what makes Elon Musk so confounding to so many people:

A long time ago, I described Elon Musk as a bullshitter who delivers. This is the perfect expression of that. This car was first shown a decade ago, with another stunt, and it never shipped. Now there’s another stunt, and another huge wave of PR and admiration.

But meanwhile, there’s the scaled car company, and the self-landing rockets!

Henry Ford was a brilliant industrialist, a terrible manager and a racist piece of shit. Several things can be true at once.

If you think Musk is either completely legitimate and trustworthy or an utter fraud, you are wrong. He has led his companies to extraordinary accomplishments, and he has made promises based on complete bullshit. He is a shameless bullshitter who sometimes delivers. The sometimes is implicit in Evans’s take, because if he always delivered, there’d be no bullshit. But I think it’s worth including the sometimes to make the point. And some of the things he’s delivered are so amazing that they sure sounded like they might be bullshit when they were announced. “Space internet” sounded like total bullshit when Musk announced it in 2015, but it started rolling out in 2020, and today you can just buy it at Best Buy for $350.

Tesla makes some tremendous vehicles. Their self-driving technology is the best in the consumer market. (Not as good as Waymo’s, but you can’t buy a Waymo.) So maybe they will ship a great Roadster in 2028. I’ll pay attention if and when they do. But I wouldn’t bet a nickel that they will. And I wouldn’t bet that they won’t, either.