In my experience, most sufficiently advanced use of PHP involves rendering the HTML. You can do it either way.

And most sufficiently complex PHP uses PHP as a full-fledged language, complete with classes, etc., and to the extent templates are used, the code in them consists of tiny partial-line-long fragments of code that call some method in some other file that is basically just code.

In other words, very fragmented code isn't caused by PHP; it's caused by the way that the developer chose to use PHP.

I still want a USB-A port on my MacBook Pro. The number of times I don't have the right cable/adapter with me is still way too high.

But no, I meant phones, where Lightning was dumped on iPad in 2018, but didn't go away on the iPhone for another half decade.

Yes. At the very latest, they should have changed when they changed the iPad over. Having a different charger for iPhone than for every other Apple device was overtly user-hostile. It meant carrying around extra crap for zero reason other than because Apple extracted four bucks worth of MFi revenue off of every licensed iPhone cable sold, and couldn't extract revenue from every USB-C cable sold.

And given how thin Lightning cables had to be because of the miniature connector and the inability to use a bigger connector because of case manufacturers building the holes too small to accommodate them, and how often the cables failed because of the too-thin wires, I'd imagine that was quite a nice revenue stream. The number of Lightning cables I went through was nuts.

USB Mini was in 2000. Micro was in 2007. Both were before the first iPhone. Don't get me wrong, micro-USB sucked. But we'd all have been better off if they had gone with mini-USB and stuck with it. We'd all have been better off if Apple had stuck with the Dock connector another two years and pushed the USB Forum (of which they are a member) to make a reversible connector happen, then switched to USB-C. We'd all have been better off if they had switched to Lighting, scared the rest of the industry into making USB-C happen, and then dumped Lighting for USB-C in 2018, starting with the iPad and then quickly progressing through the entire product line, rather than changing the iPad, but keeping the iPhone on Lightning until 2023.

Instead, Apple waited until the EU forced them to do it. By the time they finally switched, I was about to give up and move to Android, because the experience was so much better being able to share power infrastructure with my Mac instead of carrying around special sh*t-quality, low-data-speed cables just for the iPhone. Don't get me started on the fact that Apple never put USB 3 silicon into any iPhone, so they maxed out at a pathetic 480 Mbps data rate. It made photography and videography on iPhone absolutely horrible, because you couldn't get data off of them in a timely manner.

Everything about the way Apple handled Lighting on iPhones was botched just about as horribly as you could possibly botch it. Consumers put up with it, but that didn't make it suck any less.

Apple has sold two different versions of their display on Mac laptops for years — a glossy screen and a matte screen. An OLED screen option wouldn't be unthinkable. They'd just have to price it appropriately based on availability.

But you're right that doing it across the entire product line would be a lot of hardware. Even doing it for the Pro line would be a lot, largely because that's more than half of Apple's laptop sales — a sad testament to how badly Apple screwed up when they dropped the standard Macbook in favor of the thermally throttled Air. Contrast that with when the first Macbooks came out and the Pro laptops were maybe a percent or two. This doesn't leave Apple with the wiggle room to have truly high-end-only features except by making them BTO/CTO options.

It's still noise compared with the 1.1 billion OLED panels expected to be sold this year. Whether manufacturing capacity drove Apple's availability or Apple's demand helped drive capacity is unclear. Probably both.

As far as I can tell, this is an Android phone minus the phone. It is the iPod Touch of the Android world (which Apple discontinued years ago for lack of any real-world demand beyond use as a developer test device).

Why are they building this?

Oh. For "audiophiles" (in the "I don't understand how anything works, but making the back of the device out of some special kind of metal sounds like it could make the sound better, so I'll spend thousands of dollars on it" sense of the word). Now I understand.

Therefore, don't look. Yeah, this is indeed a fairly likely explanation.

To be honest, I don't think this is them ignoring privacy. I think this is them having gotten so complacent about privacy that nobody there even understands how to protect privacy anymore.

I mean the first rule of privacy on the Internet is "DNS isn't private." This is such a basic mistake that nobody with an ounce of privacy sense should make it. The URL's hostname should contain only the organization name or individual username and maybe some unique identifier if they really need that. But 99% of titles aren't secret in and of themselves, so I'll give them a pass on this one for now, even though it is still a horrible bug for the remaining 1%.

And the second rule of privacy is "Knowledge of a resource is not authorization." The fact that a resource that has not been shared can be accessed by its URL, and that apparently no attempt has been made to ensure that only people who are authorized to access that URL can view its contents means that they effectively have no access control at all. This would be an epic fail even if the first security bug weren't there.

This makes me suspect that some intern vibe-coded this thing as a prototype with no human design review at all, and that their team shipped it without actually taking the time to understand how it worked, relying too much on AI to do the review, and that the AI had inadequate training on how to implement privacy and security. Either that or the design docs did not adequately cover how the authentication worked and the reviewers did not bother to ask, which is an even bigger concern.

Either way, Meta needs to seriously clean up their act. This should have never shipped without a proper security review, and the fact that problems this severe were not caught is clear evidence that a proper security review did not happen. Identifying the underlying reasons for that failure should be a top priority, and the answer is never "Fire person X for not doing their job," because something this bad is never the fault of one person or even a small number of people. It's always systemic. It's always a lack of adequate controls, adequate numbers of eyes on design, adequate review, etc. And until they address the systemic problems that led to this outcome, I won't trust anything that Meta does going forwards as far as I can throw it.

So my advice to Meta would be that they do a very detailed internal postmortem, identifying all of the different places where this could have been caught and why it was not caught, identifying mitigation strategies to ensure that similar things don't happen in the future, etc., then implement all of those mitigation strategies, and finally create a second, publicly sanitized version of that doc and publish it externally as a "Here's what we're doing to make sure this never happens again" document. That's what they need to do to restore the faith of the folks who have witnessed this disaster.

Anything short of that level of public documentation showing that they truly understand how badly they screwed up and have genuinely taken steps to ensure privacy and security in future products would be inadequate, given the level of corporate ineptitude required to reach this point.

That's why it is using it to look over drafts. You take the AI output and then go through its suggested edits or concerns. Essentially it functions as a very fast, half-way decent proof-reader. For example, my spouse looks over a lot of drafts of things I write, but if I give her a 15 page paper, there's good work on the first 8 or 9 pages, probably a joke on page 10 about my notation, one or two comments on the next page, and then a few comments on the bibliography. Claude is not as good a proof-reader as she is, but Claude doesn't poop out after a few pages. And Claude is much faster. So reliability goes up without having to rely on the AI being reliable itself.

And sometimes it catches subtle things. For example, not too long ago, I gave a paper draft to Claude, and it noted in the bibliography that the journal volume number seemed unlikely given the year; it turned out that the volume was correct, but I had actually written the wrong journal down in the entry. Another time, Claude noted that one entry in a bibliography was from a "borderline predatory" (its words but it wasn't wrong) journal, and since I wasn't using it for much other than as part of a list of "hey, here's prior literature" and the paper in question didn't prove anything interesting, I discussed just dropping it from the citation list. Claude then without me asking, pulled up preprint I already had on the arXiv and noted it saw that paper cited there and suggested we remove the paper there also. I didn't do so, but this is the sort of thing that it can suggest. One still has the human make the decisions, but the input itself is useful. And for the most part, verification by a human for most issues is much faster than finding the issues one self.

Not really. It turns out that specialized models do not do as well as large general models. This is essentially a variant of the Bitter Lesson https://en.wikipedia.org/wiki/Bitter_lesson. In this case, there's other reason for it also. There's a lot of reasoning science and math discussions in non-specialized training data. For example, one paper I wrote with two coauthors started as a Reddit discussion. But there's also just a lot of subtle interaction that specialized models don't necessarily get. Math is easier to understand if one has a good command of physics. Engineering is better if one has math and physics. Etc. And sometimes a technique gets directly pulled over. For example, in the large OpenAI paper drop, one of the graph theory problems relied on a trick involving a weighing on the graph that involved complex numbers designed to cancel, where the trick came from a physics context that did something similar. A math specialized model would have been unlikely to find and import that trick.

Right, and the line here isn't clear. One take on NFTs is that they are nominally a tech but functioned purely as a commodity bubble. Or that NFT is a tech but it was used to make commodities which had a bubble. And the line between tech and commodity was blurry historically. When Spain entered the "platinum age," that was based in part on how to make nearly pure platinum was a tech, but the platinum itself was a commodity.

Since per a token output seems to be getting better also for open-weight models, this suggests this is more than just a business decision. And some of it is specific changes; we know that larger models with more training data for example will often produce higher quality output. But also aside from scaling, there have been architectural improvements. For example, one insight that occurred was that having lower precision weights which allowed more weights was often a positive tradeoff. That lead to changes in both the software and hardware ends, as the GPUs were modified as a result. Similar, there's been changes in what functions are used for the feedforward function.

False. The DoD banned imports of new, foreign-made consumer wireless routers due to claims of national security, supply chain, and cybersecurity risks.

There's no evidence that crappy consumer-grade routers with firmware written by contractors to hit a specific set of bullet points on a spec sheet are going to be any more secure merely because the low-quality firmware was installed on the devices at a factory in the United States. The national security risk is a total farce. And the supply chain problem is still a supply chain problem even if it is assembled in the United States, because few (if any) of the components inside it are made in the United States, nor do U.S. manufacturers exist for most of those parts.

So what you actually have is a lunatic executive branch leader whose tariffs got thrown out by the Supreme Court who then weaponized the DoD and the FCC to raise the price of many electronics goods, despite no evidence that doing so will actually create any short-term or long-term improvements in the near-total lack of U.S. component manufacturing (which for silicon parts is itself dependent on foreign silicon wafer manufacturing, because U.S. manufacturers consume far more silicon wafers than U.S. manufacturing can produce, though ironically, most of the sand that goes into them was mined in the U.S.).

At this point, disgruntled employees are almost certainly orders of magnitude more honest than the current administration which has been repeatedly been caught lying, distorting facts and straight up making shit up and calling it alternative facts.

Trump has done its best to sow doubt, anything that disagrees with them by default has much higher odds of being true than literally anything they could say.

At this point if trump administration said the skyvis blue, sun is yellow when viewed from earth and grass is green - I would instantly assume all of those things were false. Thats the state of fucked up this prick has created.

Fuck you and anyone else so stupid as to believe him. You deserve the ass rape he gives you.

This is actually surprisingly rare. Most of the time we've berated Apple, it has been because they've dragged their heels on basic stuff like USB-C. I see no reason that Apple dragging their heels on OLED displays will be any different.

I've been complaining about Apple displays not being bright enough to use outside since my first active-matrix Mac laptop. Ironically, the passive matrix black-and-white Mac displays from the early 1990s actually could be used outside easily. You changed the contrast to a mode where it behaved like a reflective LCD panel. So from my perspective, Apple is finally getting back to the level of usability that I had with my PowerBook 145 more than thirty years ago.

In fact, I can't think of any examples of what you're describing, and certainly none where users would not have been better off with Apple introducing the feature sooner and then making it better in a subsequent generation. After all, having a feature, even if it is not refined, is usually better than not having it at all.

In general, when Apple is late, they're just late. They're either doing it because they have some financial advantage to delaying it (Made for iPhone revenue from the Lightning connector), because they didn't have the engineering resources to integrate it for a particular release cycle, or because they didn't think the components available were good enough (e.g. early folding displays).

But Apple has been shipping OLED panels on phones since 2017. Almost a decade later, laptops are finally catching up. Maybe they thought the early ones wouldn't last as long as people kept their laptops, maybe the cost for larger panels was high enough to discourage laptop purchases, or maybe they just dragged their heels because they knew they wouldn't have anything interesting in the feature pipeline and wanted people to have a reason to replace their laptops. Either way, they're fully a decade behind in display technology, and it's about time they finally caught up.

Looking forward to the touchscreen as well, at least in principle, though in practice, I doubt it will get a lot of use. :-)

The business case for LLMs is a bit broader than that. They don't need to replace that many workers if they make enough workers more efficient or more reliable (such as simply having LLMs look over drafts). There's also at this point, clear usage cases for scientific and mathematical research but that's using the very high end models, and admittedly is a pretty small market, not enough to justify massive expenditure at this level. The tulip example is interesting; what you are touching on their is the distinction between a "tech bubble" (like railroads or internet) and a "commodities bubble" (tulips, housing, Beanie Babies). Note that the degree to which the tulip bubble was real is questionable. See https://theconversation.com/tulip-mania-the-classic-story-of-a-dutch-financial-bubble-is-mostly-wrong-91413, and the line between a tech bubble and a commodity bubble itself can be blurry (see for example Bitcoin and NFTs). One of the difficulties here is that the LLMs keep improving, and the cost per token is going down, and the amount one can do per a token is going up. But that's happening in part due to massive investment in both infrastructure and research. How sustainable any of that is is unclear.

There are at least four problems with this. First, linear algebra is from a math standpoint insanely powerful. For example, all of representation theory https://en.wikipedia.org/wiki/Representation_theory is using linear algebra to understand groups. Much of the study of modular forms https://en.wikipedia.org/wiki/Modular_form involves understanding the behavior of modular forms over SL_2(Z) which has heavy linear algebra elements to it. People who see linear algebra as somehow a weak thing are often people who took a single semester of it and didn't do much else with it.

Second, the idea that these AI systems are just linear algebra isn't accurate anyhow. While there's a lot of linear involved, a major part of how transforms like LLMs work is the intervening non-linear layers, which use ReLu https://en.wikipedia.org/wiki/Rectified_linear_unit, Softplus, or a similar function. So, summarizing this as just linear algebra isn't accurate. It is linear algebra with some spice.

Third, even if this were somehow just linear algebra, that shouldn't therefore be a reason to think that there's any limits to what is going on. At the end of the day, the empirical success is what matter, and systems which can do things like prove the quasi-Riemann Hypothesis or solve other major math problems are pretty impressive regardless of what is under the hood. So, if one does think this is just linear algebra, that's not a reason to dismiss the systems but a reason to increase your estimate of what can be done with mere linear algebra.

Fourth, slowing down connections like this may make crashes *less likely.* In particular, the main way we could have a serious bubble here is over-investment. Delays including power grid delays force bottlenecks on the investment level, and make that sort of problem less likely.

We may be in a bubble here. Certainly some elements look very bubble-like. In the late 1990s, during the dot-com boom, one had people making perfectly normal businesses but just tacking on "on the internet" to it and it would get millions of dollars of funding automatically. See for example pets.com . And we're seeing some of that now; we've had a whole bunch of businesses see their stock price shootup just because they mention AI. So we may be in a bubble, and it may crash now, or might crash never or decades from now. (Arguably the US had an 80 year automobile bubble that didn't crash until the late 2000s which then forced the auto bailout.) But the other thing to keep in mind is that a crash does not mean the tech goes away. In 1873, the US had a devastating collapse of the railroad bubble, and yet railroads continued to grow, with the amount of track not peaking until around 1910 https://transportgeography.org/contents/chapter5/rail-transportation-pipelines/rail-track-mileage-united-states/. Similarly, although the internet bubble burst, by 2015, one had more internet use than even the most internet hyping person in 1998 would have likely dreamed of. A bubble bursting does not mean the tech itself goes away.

The set of problems that only a human can solve is getting rapidly smaller. The set of puzzles that a human can solve, that an AI cannot, and which take a small amount of time for a human and aren't highly inconvenient may already be empty.

No, it isn't. To be fraud, the lie has to be material. Would him saying that he fought in the military have disqualified him? No. Because being in the military of a communist country doesn't automatically make you communist. It might have caused a deeper investigation, and that investigation might have concluded that he was, but given that the regime had already fallen and any records would probably have been nonexistent at that point, that's a pretty unlikely outcome — doubly so decades later.

Second, the lie has to be deliberate. In U.S. law, there's not a presumption that an error is deliberate; the government has to have some evidence to support such a claim. And given that he has been dead for a long time, there's no way to get him to admit that it was deliberate, so in the absence of some sort of documentation, this one isn't provable anymore, either.

Third, even if you could prove both of the first two, the fraud was still committed by her father, not her, which makes it moot after the fact.

Continuing to spew this same nonsense over and over doesn't make it true. First, he was a citizen. Until such time as you get a court to agree that his citizenship should not have been granted (years after his death), all of this is purely academic.

Second, even if you did so, although removal of citizenship is considered retroactive to the time of the original grant of citizenship, you still run into the presumption of regularity when trying to extend that retroactive removal to a retroactive nullification of the citizenship of an additional person who relied on a presumption of the regularity of the prior immigration decision. That just won't fly, because as far as was known to her at the time, he was a citizen, and as far as was known to the immigration people processing the paperwork at the time, he was a citizen. And none of that changes even if his citizenship were removed retroactively, which means *she* did not commit fraud.

Further, his citizenship was likely not material to her being approved for citizenship. She didn't become a citizen because he was a citizen. She became a citizen *when* he became a citizen. Had she applied independently, even if we take the most outlandish interpretation of the law and assume that his military service would be disqualifying, it would not be disqualifying *for* *her*. And at that point, the courts would be effectively punishing her for not waiting a few extra months and filing a form independent of her parents. No court is going to do that, both because doing so would be a gross miscarriage of justice and because the specific form that she filed is almost certainly not material to whether a young adult would qualify for citizenship.

Plus there's that pesky little problem of you not having any actual evidence to support your absurd theory in the first place, which means that while you might get a lot of idiots on your side in the court of public opinion, you'd be laughed out of any actual courtroom.

Bzzt. Reading comprehension fail.

I didn't say you have to be high up in the military of a totalitarian regime to be a member of the party. I said you aren't guaranteed to be a member unless you're a military officer who took control of the country in a coup or someone who is very close to that person. That's what "does not constitute" means.

In a lot of foreign countries, military service isn't optional. And that was actually true in Somalia for many years, though I suppose that was after her father served. Regardless, there are many reasons for fighting in a military, including financial reasons, as an alternative to jail time, etc., which is why you cannot automatically assume that a member of a country's military supports their government's actions or is a member of their ruling party.

Moreover, that law was written with a very specific purpose in mind, and its purpose was not ensuring that no male over 18 from that country could ever become a U.S. citizen. Its purpose was preventing those who posed a high risk of spying on our country from becoming entrenched in our society and manipulating it from the inside. Those who did not support the ideology or were fleeing it presumptively are the polar opposite of that group. So claiming that they are party members is borderline nonsensical in the absence of actual hard evidence... which you do not have, or else you would have provided it.

Again, your claim has absolutely no legal merit, because it relies on a fundamentally absurd interpretation of the law in question.

If that claim is false, then I'm the President of the United States.

Her dad applied and was granted citizenship. Therefore that claim is not false. And even if his citizenship was obtained under false pretense, he still had it at the time when she applied, so that claim was not false at the time that she applied. Absent any evidence of her knowing that he lied to become a citizen, your argument has zero merit from a legal perspective.

You see, in the United States (and any other country whose law is based on English Common Law), there is an automatic presumption that government actions have been handled properly, and that you can rely on that presumption for all other purposes. The legal term is "presumption of regularity". Reliance on that presumption prima facie cannot be fraudulent in the absence of direct knowledge to the contrary.

So unless you can prove beyond a reasonable doubt that she *knew* her father got citizenship through fraud, once again, your argument has no legal merit whatsoever.

Please learn a bit about how U.S. law actually works before spouting such utter nonsense.