Sure do :) I can provide more if you want, but start there, as it's a good read. Indeed, blind people are much better at understanding the consequences of colours than they are at knowing what colours things are..

You do realize that what you're describing is all of about ten lines of Javascript with the right libraries (audioTrackList property, subtitle library, currentTime property), right?

The congenitally blind have never seen colours. Yet in practice, they're practically as efficient at answering questions about and reasoning about colours as the sighted.

One may raise questions about qualia, but the older I get, the weaker the qualia argument gets. I'd argue that I have qualia about abstracts, like "justice". I have a visceral feeling when I see justice and injustice, and experience it; it's highly associative for me. Have I ever touched, heard, smelled, seen, or tasted an object called "justice"? Of course not. But the concept of justice is so connected in my mind to other things that it's very "real", very tangible. If I think about "the colour red", is what I'm experiencing just a wave of associative connection to all the red things I've seen, some of which have strong emotional attachments to them?

What's the qualia of hearing a single guitar string? Could thinking about "a guitar string" shortly after my first experience with a guitar string, when I don't have a good associative memory of it, sounding count as qualia? What about when I've heard guitars play many times and now have a solid memory of guitar sounds, and I then think about the sound of a guitar string? What if it's not just a guitar string, but a riff, or a whole song? Do I have qualia associated with *the whole song*? The first time? Or once I know it by heart?

Qualia seems like a flexible thing to me, merely a connection to associative memory. And sorry, I seem to have gotten offtopic in writing this. But to loop back: you don't have to have experienced something to have strong associations with it. Blind people don't learn of colours through seeing them. While there certainly is much to life experiences that we don't write much about (if at all) online, and so one who learned purely from the internet might have a weaker understanding of those things, by and large, our life experiences and the thought traces behind them very much are online. From billions and billions of people, over decades.

Language does not exist in a vacuum. It is a result of the thought processes that create it. To create language, particularly about complex topics, you have to be able to recreate the logic, or at least *a* logic, that underlies those topics. You cannot build a LLM from a Markov model. If you could store one state transition probability per unit of Planck space, a different one at every unit of Planck time, across the entire universe, throughout the entire history of the universe, you could only represent the state transition probabilities for the first half of the first sentence of A Tale of Two Cities.

For LLMs to function, they have to "think", for some definition of thinking. You can debate over terminology, or how closely it matches our thinking, but what it's not doing is some sort of "the most recent states were X, so let's look up some statistical probability Y". Statistics doesn't even enter the system until the final softmax, and even then, only because you have to go from a high dimensional (latent) space down to a low-dimensional (linguistic) space, so you have to "round" your position to nearby tokens, and there's often many tokens nearby. It turns out that you get the best results if you add some noise into your roundings (indeed, biological neural networks are *extremely* noisy as well)

As for this article, it's just silly. It's a rant based on a single cherry picked contrarian paper from 2024, and he doesn't even represent it right. The paper's core premise is that intelligence is not lingistic - and we've known that for a long time. But LLMs don't operate on language. They operate on a latent space, and are entirely indifferent as to what modality feeds into and out from that latent space. The author takes the paper's further argument that LLMs do not operate in the same way as a human brain, and hallucinates that to "LLMs can't think". He goes from "not the same" to "literally nothing at all". Also, the end of the article isn't about science at all, it's an argument Riley makes from the work of two philosophers, and is a massive fallacy that not only misunderstands LLMs, but the brain as well (*you* are a next-everything prediction engine; to claim that being a predictive engine means you can't invent is to claim that humans cannot invent). And furthermore, that's Riley's own synthesis, not even a claim by his cited philosophers.

For anyone who cares about the (single, cherry-picked, old) Fedorenko paper, the argument is: language contains an "imprint" of reasoning, but not the full reasoning process, that it's a lower-dimensional space than the reasoning itself (nothing controversial there with regards to modern science). Fedorenko argues that this implies that the models don't build up a deeper structure of the underlying logic but only the surface logic, which is a far weaker argument. If the text leads "The odds of a national of Ghana conducting a terrorist attack in Ireland over the next 20 years are approximately...." and it is to continue with a percentage, that's not "surface logic" that the model needs to be able to perform well at the task. It's not just "what's the most likely word to come after 'approximately'". Fedorenko then extrapolates his reasoning to conclude that there will be a "cliff of novelty". But this isn't actually supported by the data; novelty metrics continue to rise, with no sign of his suppossed "cliff". Fedorenko argues notes that in many tasks, the surface logic between the model and a human will be identical and indistinguishable - but he expects that to generally fail with deeper tasks of greater complexity. He thinks that LLMs need to change architecture and combine "language models" with a "reasoning model" (ignoring that the language models *are* reasoning - heck, even under his own argument - and that LLMs have crushed the performance of formal symbolic reasoning engines, whose rigidity makes them too inflexible to deal with the real world)

But again, Riley doesn't just take Fedorenko at face value, but he runs even further with it. Fedorenko argues that you can actually get quite far just by modeling language. Riley by contrast argues - or should I say, next-word predicts with his human brain - that because LLMs are just predicting tokens, they are a "Large Language Mistake" and the bubble will burst. The latter does not follow from the former. Fedorenko's argument is actually that LLMs can substitute for humans in many things - just not everything.

Greater Fool scheme finds Greatest Possible Fools (goverments) to keep pumping money into their zero-sum game.

Great opportunity for open source web services. :-)

There's no good reason to use it. Just encode your video for random-access streaming, set up Apache or nginx with a URL that you make sure isn't indexed, require a client cert on the directory if you really want to be careful, port forward to it from a port on your router, set up dynamic DNS, and use a web browser. No arbitrary restrictions, just your content on your terms.

None of it is unique to the device — I have backups of everything — but I still want the photos and videos on the new phone. :-)

Transferring a terabyte of data from one phone to another takes the better part of an hour even at real-world USB 3 speeds.

Counterpoint: My phone history includes:

• iPhone (original), 5 years
• iPhone 5, 3 years
• iPhone 6s, 8 years
• iPhone 15 Pro, 2 years so far

Assuming I keep the 15 Pro for 3 years (the prior minimum), that's 4.75 years average. I also buy the device with the largest capacity, and always wish it were bigger. Unfortunately, Apple doesn't increase capacity quickly enough for upgrading to help with that.

Upgrading is expensive and it is a pain in the a**. So there are multiple reasons to keep a device until it dies. Some people who buy flagship phones have the same concerns.

The main difference is that flagship phones typically get security updates for five to seven years. Low-end phones are often previous generation hardware that is still for sale, and may get security updates for as little as one year from the date of purchase. So unless you're willing to put your entire life at risk by using a phone that has gaping security holes, low-end phones are often false economy, purchased by people who see the price tag and are too broke to afford a better one, who then end up paying for replacement after replacement at a higher rate because they can't afford a phone that will actually last five or six years.

So I would expect low-end phones to get junked every couple of years, and for high-end phones to get junked when support is dropped, assuming that the owners know that the phone is no longer supported, and the rest of them just end up in a giant botnet, and they replace their phones because they're bogged down with malware a few months to a year after they go out of support.

iPhone users keep their phones longer than Android users, on average. 61% of iPhone users have owned their phone for more than 2 years, versus just 43% of Android users.

So patch availability does appear to have a significant impact on how long people keep their devices.

Why would he feel inadequate when, according to a trustworthy source, he's a better boxer than Mike Tyson, fitter than LeBron James, hotter than Tom Brady, one of the top minds in history with a near-Olympian physique, the world's best runway model, better at resurrection than Jesus, the world's best bottom (ahem) (cough) and the ultimate throat goat?

Sooner or later, we'll end up at the point where trying to maintain the ways of the past is a fruitless fight. Teachers' jobs are no longer going to be "to teach" - that that's inevitably getting taken over by AI (for economic reasons, but also because it's a one-on-one interaction with the student, with them having no fear of asking questions, and that at least at a pre-university level, it probably knows the material a lot better than the average teacher, who these days is often an ignorant gym coach or whatnot). Their jobs will be *to evaluate frequently* (how well does the student know things when they don't have access to AI tools?). The future of teachers - nostalgia aside - is as daily exam administrators, to make sure that students are actually doing their studies. Even if said exams were written by and will be graded by AI.

You're right about YouTube. I was thinking HEVC was one of their delivery formats, but apparently not.

Netflix definitely did use HEVC for delivery of some of its high-end content at one time. Whether they still do or not, I have no idea.

Either way, the fact that people are running into error messages suggests that there is some actual customer impact.

Except that nobody who buys one of their machines is going to think, "I could pay an extra $100 and my machine would work better." They're going to think, "This piece of s**t can't even do things that my cell phone from eight years ago can do. Why did I buy this, and why should I ever buy anything from this manufacturer in the future?"

This level of penny-wise, pound-foolish behavior is a sure way to permanently lose customers.

...or decoding...

You mean like... anyone who wants to watch Netflix or YouTube or any other streaming service's high-res content?