Sometimes I wonder if innovation in the AI space has stalled and recent progress is just a product of increased compute. Competence is increasing exponentially[1] but I guess it doesn't rule it out completely. I would postulate that a radical architecture shift is needed for the singularity though
Interesting story. I have noticed that certain parameters are often ignored not because they're unimportant, but instead because they're hard to quantify.
Another story is how the first product made by Jobs and Wozniak at Apple was a hacked phone that could be used to make international calls for free. They called the Vatican and pretended to be Henry Kissinger wanting to speak to the pope - they noticed and promptly hung up before they reached him
This reminds me of a story I read once about how Bell Labs engineers wanted to find the ideal length of a landline telephone chord. They did this by secretly shortening the chord over time on test users' telephones and waited to see how short they could make the chord before they noticed. If the plugtest is hardware-hardware testing Bell Labs engineers were doing human-hardware testing.
I tried to find a source for the story but I couldn't find one. I think I read it in The Innovators by Walter Isaacson, but I can't remember it exactly - maybe I'm misremembering it.
The HN crowd is not representative of the entire market. Most people don't care about the operating system and only want something that
1) is simple to use
2) they already know
3) they happen to already have (most people keep their phones for many years)
Also, the largest phone market in the world is the developing countries market. Cheap phones are supreme right now
I agree. I just don't agree with misinformation not being protected as free speech. Surely having an INGSOC decide what is truthful enough to be shared is detrimental to free expression and thought. Heliocentrism was also misinformation at one point.
It is basically just a proxy. I don't see how censorship could be an antidote to a "subversive political influence campaign" - if anything you're describing censorship
At first I wanted to write a comment about how cars and sex are apparently very well linked. Upon thinking it over once more, it appears that the real link to sex is privacy, which is of course obvious. Thinking over it once more, we're brought back to the real selling point of cars: total privacy. Public transport is on paper really good, but it is totally devoid of privacy - which means that it is bad in reality. In other more provocative words, public transportation is bad because you can't have sex on the bus
Yes. The reason the year of the Linux desktop has yet to arrive is because most people don't understand this joke. Linux is powerful because it is made for power users (although certain distros are changing this)
I think the obvious thing to do here is to say "Always bet on symbolics".
What separates text from images is that text is symbolic while images are visceral or feelings based. In the same way, text comes in short when it comes to the feeling you get when seeing an image. Try to put in to text what you feel when you look at Norman Rockwell's Freedom of Speech or a crappy 0.5MB picture of your daughter taken on an iPhone 3. Hard isn't it? Visual and symbolic are not isomorphic systems.
Examples of symbolic systems like text are sheet music and Feynman diagrams. You would be hard pressed if you tried to convey even 2KB of sheet music in a book
No, that is not what I mean. The efficiency of a piece of knowledge is not only a function of its intrinsic value, but also how easy it is to understand. Sure, the people who are expected to read the document are smart and this is probably the best way to do it, but even Lockheed engineers are fallible.
If anything, the enemy will be defeated before they have had the time to understand the document in case it gets leaked xD
It is often that seemingly irrelevant factors play a big role. In this case, a 141 page highly dense (and frankly boring to read) document is in its essence a liability. Engineers get bored too and it is obviously more fun to just code rather than to read a document that might aswell have been written by a lawyer.
This is also why car makers name their cars things like "Jeep Expedition" or "Ford Escape". The name doesn't change the car, but it does make it more exciting.
I think almost all multilingual people would agree that writing cordially is easier in their native language - whatever that language may be. Expressing heartfelt messages in the language you spoke when developing your identity and emotional maturity is more about just that, rather than what the language happens to be.
English is not a programming language. You're only disagreeing with my articulation here, which is irrelevant in relation to the thing of the matter - namely what I mean rather than what I type into the keyboard physically.
Preventing someone from getting value out of their work is theft - not matter how it is done. Copying a dead person's work isn't theft because a dead person can't create value, but stealing a dead person's car is still theft, because something of value is gone.
Stealing a car you were never going to buy and making an exact replica of a car you were never going to buy is two entirely different things.
Good OPSEC is surprisingly simple and boring. Essentially, it just boils down to using tor and not accidentally exposing sensitive information, which is how Ross Ulbricht got caught. (okay, it is more than that but in essence it is true)
There are probably many people in prison right now because tor is awfully slow. If you don't have the patience for tor you probably also don't have the patience for prison.
[1]https://arxiv.org/html/2503.14499v1 *Source is from March 2025 so make of it what you will.