We’ve been tracking the growing judicial revolt against the Supreme Court’s shadow docket nonsense, from individual district judges getting snarky in footnotes to anonymous judges speaking to reporters. But what happened Thursday at the Fourth Circuit Court of Appeals crosses into entirely new territory: a full en banc panel of federal judges openly criticizing the Supreme Court’s approach during a live oral argument session.

This isn’t normal. Federal judges don’t usually air their grievances about the Supreme Court in open court. The fact that an entire appeals court panel—including respected conservative judges—turned their oral argument into what Politico called “a remarkable, 80-minute venting session” tells you everything about how broken the system has become.

The immediate catalyst was trying to figure out what to do with a case about DOGE’s access to Social Security data after the Supreme Court issued one of its trademark unexplained emergency orders. But the real issue was much bigger: how are lower courts supposed to function when the highest court in the land operates like it’s playing Calvinball?

“They’re leaving the circuit courts, the district courts out in limbo,” said Judge James Wynn… “We’re out here flailing. … I’m not criticizing the justices. They’re using a vehicle that’s there, but they are telling us nothing. They could easily just give us direction and we would follow it.”

Judge Wynn didn’t stop there:

“They cannot get amnesia in the future because they didn’t write an opinion on it. Write an opinion,” Wynn said. “We need to understand why you did it. We judges would just love to hear your reasoning as to why you rule that way. It makes our job easier. We will follow the law. We will follow the Supreme Court, but we’d like to know what it is we are following.”

I’ve been writing about the law for almost three decades. I’ve never seen anything like this. Ever. Not even in the same zip code as this. These are judges crying out for help under a completely lawless Supreme Court.

And, no, this wasn’t just liberal judges complaining. Judge J. Harvie Wilkinson III—a Reagan appointee and one of the most respected conservative jurists in the country—was right there with them:

“The Supreme Court’s action must mean something,” said Judge J. Harvie Wilkinson III, a Reagan appointee. “It doesn’t do these things just for the kicks of it.”

Even Wilkinson can’t figure out what the hell the Supreme Court is doing. When you’ve lost Harvie Wilkinson—a judge so conservative and institutionally minded that he’s basically judicial royalty—you’ve completely broken the system.

The specific case that triggered this judicial revolt involves the Supreme Court’s typical shadow docket bullshit. In June, the Court overruled the Fourth Circuit’s decision and lifted an injunction against DOGE’s use of Social Security data. But they did so in the most bizarre and troubling way. After sending the case back to the Fourth Circuit for more review, it said that even if the Fourth Circuit rules that DOGE is breaking the law, the stay will remain in place.

By an apparent 6-3 vote, the justices went further, saying that no matter what the appeals court decided, the injunction would remain on hold until the case returned to the Supreme Court. Yet, the high court’s majority offered no substantive rationale for the lower court to parse.

So the Supreme Court basically said: “We’re overturning you, and also whatever you decide doesn’t matter anyway, but we’re not going to tell you why.” This left the entire Fourth Circuit panel wondering what the fuck they’re even supposed to do.

That left many of the 15 4th Circuit judges on hand for Thursday’s unusual en banc arguments puzzling at their role. One even suggested the appeals court should simply issue a one-line opinion saying the injunction is lifted and kick the case back to the Supreme Court to resolve.

Some judges thought they should just give up entirely and punt the case back to SCOTUS since SCOTUS has already said whatever they decide here doesn’t actually matter. Others insisted they had a constitutional duty to actually do their jobs:

“It sounds like some of my colleagues think that there’s no work to be done, that we’re done because the Supreme Court has told us what the answer is,” said Judge Albert Diaz, an Obama appointee.

Judge Robert King said punting on the case would be a mistake.

“We each have a commission and we have a robe and we have an oath to abide by,” said King, a Clinton appointee.

This perfectly captures the impossible position the Supreme Court has created. Lower court judges literally don’t know if they’re supposed to do their jobs or just rubber-stamp whatever vibes they think they’re getting from the shadow docket.

The whole mess stems from a series of recent Supreme Court shadow docket rulings (without much explanation) basically telling lower courts they have to follow SCOTUS shadow docket rulings (also without much explanation) as binding precedent. But as we’ve written about extensively, these aren’t reasoned legal decisions—they’re often unexplained orders issued with minimal briefing, no oral arguments, and little to no explanation of any reasoning.

This has created a situation where experienced federal judges—people who’ve spent decades interpreting legal precedent (often longer than the Justices themselves)—literally can’t figure out what the Supreme Court wants them to do.

What we’re witnessing is the breakdown of the federal judiciary as a functioning institution. When Reagan and Obama appointees are united in open revolt, and Harvie Wilkinson can’t figure out what the Supreme Court wants, the system has collapsed.

The three liberal Justices have been warning about this in dissent after dissent, while the conservative majority just keeps issuing more unexplained orders and then getting pissy when lower courts can’t read their minds. This isn’t jurisprudence. It’s government by judicial decree, where constitutional law operates on vibes and the only consistent principle is “give Trump whatever he wants.”

When federal judges with decades of experience are reduced to public pleading for basic guidance during oral arguments, we’ve crossed into judicial authoritarianism. The Supreme Court has effectively told the entire federal judiciary: “Follow our orders, but we won’t explain what they mean, and if you guess wrong, we’ll scold you for defying us.”

That’s not how precedent works. That’s not how courts work. That’s not the rule of law. It’s just nine people in robes demanding deference to their unexplained whims.

As I was covering the iPhone announcement earlier this week, there was a moment in Apple's presentation that made me double-take. Describing the new iPhone Air, the company said: "What might appear to be a single camera is actually our new, powerful, 48 megapixel fusion camera system, which works like multiple advanced cameras in one." It continued on to mention the phone's "2x telephoto," later calling it an "optical quality lens."

Reader: the iPhone Air has one camera. (And it's not new, by the way; it's the same main camera from the standard iPhone 16 and 17). The camera app may present you with a 2x option, but it will be using the exact same optics. It "might appear to be a single camera" because it literally just is.

The Air's "2x telephoto" is really just a crop. Image: Apple

The company pulls the same trick for the regular iPhone 17, calling its main lens "two cameras in one," and takes it even further with the iPhone 17 Pro. That phone has three cameras – a 13mm equiv. "0.5x" ultra-wide, 24mm equiv. "1x" main and 100mm equiv. "4x" telephoto which is an impressive amount to fit in such a small device. Yet Apple claims the phone is capable of "up to 8x optical-quality zoom," and that carrying it is "like having 8 pro lenses in your pocket."*

The iPhone 17 Pro's eight camera modes (produced by its three cameras). Image: Apple

It isn't, and these additional options are not "optical quality," an essentially meaningless phrase meant to evoke the idea of a lens capable of actually zooming in. In reality, it's a crop. Your phone is punching in on the pixels in the center of the sensor, only using a quarter of its resolution to capture the scene.

The phone may be processing it differently than it would a straight digital zoom, but at the heart of it, that is all these "optical zoom" modes are. It's also worth noting that, because of the sensor's Quad Bayer design, that center crop won't necessarily have the detail you might expect from a 12MP image taken on a standard Bayer sensor.

The "optical zoom" phrasing is all over Apple's website, including in the tool that lets you compare between models, which doesn't distinguish between the actual lenses and the crops. Also, while I'm griping, this comparison tool may be one of the most obnoxious pieces of web design I've come across in a while.

This isn't a new trick by any means; Apple has been pretending that cropping is the same as optics for a while now, but that doesn't really make it better. People are still getting confused by it, likely in part because the tech press routinely parrots the "optical quality" phrasing without explaining what's actually happening under the hood.

While Apple's presentation was particularly egregious, it's far from the only phone manufacturer participating in this marketing sleight of hand. Describing the Pixel 10 Pro's capabilities, Google writes that the phone has "optical quality at 0.5x, 1x, 2x, 5x, 10x," despite it physically only having three cameras (the 2x and 10x modes are center crops).

Another example of "optical quality."

Samsung similarly boasts that the S25 has a 50MP wide-angle camera with "2x optical quality zoom," though it at least includes a footnote saying "Optical quality zoom is enabled by the Adaptive Pixel sensor. 3x distance is optical zoom. 2x distance is optical quality zoom." That could tip off attentive readers that there are some liberties being taken, but telling the truth in the footnotes isn't the same as being honest.

While these companies are (generally) careful to modify "optical" with "quality," I'd argue this is still misleading, though I'm sure they all have some convoluted reasoning as to why they call the crop modes that**. While consumers probably aren't buying phones thinking they have more cameras than they actually do, Apple & co's. marketing may trick them into thinking they're not giving up by pressing the button to punch in.

You are giving up something for that extra reach, no matter how hard manufacturers try to make you believe otherwise

But, to put it plainly: they are. All the computational tricks in the world won't make an image taken using a quarter of the sensor the same quality as one taken with the entirety of that same sensor. You are giving up something for that extra reach, no matter how hard the phone companies try to make you believe otherwise.

In all honesty, I don't expect the phone companies to stop using this kind of language, especially if thin phones with fewer cameras continue to be en vogue. What I can hope is that photography enthusiasts and the tech press will stop regurgitating their misleading labels, and instead start educating people on how the different modes actually work.

* - Apple's math: three real, physical lenses (ultra-wide, wide, telephoto), plus two main camera crops to emulate a 28 or 35mm focal length, the "2x" center crop of the main camera and "8x" center crop of the telephoto camera and the ultra-wide's macro mode equals eight options.

** - In its presentation, Apple says the crop modes have their own "dedicated image pipelines." This is almost certainly technically correct, the most annoying kind of correct

Earlier this week, Ars spent some time driving the new Nissan Leaf. We have to wait until Friday to tell you how that car drives, but among the changes from the previous generation are door handles that retract flush with the bodywork, for the front doors at least. Car designers love them for not ruining the lines of the door with the necessities of real life, but is the benefit from drag reduction worth the safety risk?

That question is in even sharper relief this morning. Bloomberg's Dana Hull has a deeply reported article that looks at the problem of Tesla's door handles, which fail when the cars lose power.

The electric vehicle manufacturer chose not to use conventional door locks in its cars, preferring to use IP-based electronic controls. While the front seat occupants have always had a physical latch that can open the door, it took some years for the automaker to add emergency releases for the rear doors, and even now that it has, many rear-seat Tesla passengers will be unaware of where to find or how to operate the emergency release.

A power failure also affects first responders' ability to rescue occupants, and Hull's article details a number of tragic fatal crashes where the occupants of a crashed Tesla were unable to escape the smoke and flames of their burning cars.

China to the rescue?

In fact, the styling feature might be on borrowed time. It seems that Chinese authorities have been concerned about retractable door handles for some time now and are reportedly close to banning them from 2027. Flush-fit door handles fail far more often during side impacts than regular handles, delaying egress or rescue time after a crash. During heavy rain, flush-fit door handles have short-circuited, trapping people in their cars. Chinese consumers have even reported an increase in finger injuries as they get trapped or pinched.

That's plenty of safety risk, but what about the benefit to vehicle efficiency? As it turns out, it doesn't actually help that much. Adding flush door handles cuts the drag coefficient (Cd) by around 0.01. You really need to know a car's frontal area as well as its Cd, but this equates to perhaps a little more than a mile of EPA range, perhaps two under Europe's Worldwide Harmonised Light vehicles Test Procedure.

If automakers were that serious about drag reduction, we'd see many more EVs riding on smaller wheels. The rotation of the wheels and tires is one of the greatest contributors to drag, yet the stylists' love of huge wheels means most EVs you'll find on the front lot of a dealership will struggle to match their official efficiency numbers (not to mention suffering from a worse ride).

China's importance to the global EV market means that, if it follows through on this ban, we can expect to see many fewer cars arrive with flush door handles in the future.

Read full article

Comments

Google is one of the most ardent proponents of generative AI technology, as evidenced by the recent launch of the Pixel 10 series. The phones were announced with more than 20 new AI experiences, according to Google. However, one of them is already being pulled from the company's phones. If you go looking for your Daily Hub, you may be disappointed. Not that disappointed, though, as it has been pulled because it didn't do very much.

Many of Google's new AI features only make themselves known in specific circumstances, for example when Magic Cue finds an opportunity to suggest an address or calendar appointment based on your screen context. The Daily Hub, on the other hand, asserted itself multiple times throughout the day. It appeared at the top of the Google Discover feed, as well as in the At a Glance widget right at the top of the home screen.

Just a few weeks after release, Google has pulled the Daily Hub preview from Pixel 10 devices. You will no longer see it in Google Discover nor in the home screen widget. After being spotted by 9to5Google, the company has issued a statement explaining its plans.

"To ensure the best possible experience on Pixel, we’re temporarily pausing the public preview of Daily Hub for users. Our teams are actively working to enhance its performance and refine the personalized experience. We look forward to reintroducing an improved Daily Hub when it’s ready," a Google spokesperson said.

It's not surprising that Google has hit pause on Daily Hub. Google's approach to mobile AI relies on the Tensor processor's capable NPU. Pixel phones process your personal data on-device rather than sending it to the cloud, which is how Daily Hub is supposed to glean insights into your life. The problem is that it doesn't do that very well.

As we mentioned in our review of the Pixel 10 series, Daily Hub's premise is similar to Samsung's Now Brief. They both pull in data from your apps and run it all through on-device AI models to create a digest of your life that is refreshed multiple times per day. Samsung has continued promoting Now Brief as its chief AI innovation despite the fact that it rarely offers more than the weather and calendar reminders. Google's Daily Hub was in a similar place, but it may have been even less useful.

In our testing, Daily Hub rarely showed anything beyond the weather, suggested videos, and AI search prompts. When it did integrate calendar data, it seemed unable to differentiate between the user's own calendar and data from shared calendars. This largely useless report was pushed to the At a Glance widget multiple times per day, making it more of a nuisance than helpful.

Smartphones are overflowing with personal data, which is one of the reasons they are such attractive targets for hackers. Despite this, both Samsung and Google have found little success generating useful daily insights with on-device AI models. We wait with bated breath to see if Google can make Daily Hub worth using when it returns.

Read full article

Comments