https://www.wired.com/1997/12/...

I had to plead in writing but they'd run out of ink, so I pleaded "not quill cup".

Back in 2002 I was working with image processing, and I came up with a video compression idea: treat the video as a 3D image and apply 3D Fourier transform, then drop the weaker components as done with similar 1D/2D schemes. This would provide a natural balancing between temporal and spatial resolution, depending on the scene. I wasn't much of a programmer back then, but I later realized this would probably be too heavy for practical use. I also learned that the same idea had been put into development around the same time as Ogg Tarkin, but it didn't get very far either.

As a bonus, the Fourier/wavelet format could be decoded in arbitrary resolutions in space or time. This idea is actually used in some music playing libraries, producing 24-bit output from a format intended for 16-bit quality.

It dit, right before it discovered the Jevons paradox.

If you travel back a couple of centuries, your 6 jigowatts will only be worth about 8 million horses.

Cow tittie milk should be labeled "cow tittie milk" to remind people where the product comes from. It's natural for mammals to drink the tittie milk of their own species when they're young, but drinking tittie milk (a) when you're grown up and (b) from another species seems downright perverse. Likewise, people could use a good reminder how the meat they eat is produced.

What do you mean "open-source"? If CUDA were open-source, you could run it on AMD GPUs or any other reasonable hardware. Instead, Nvidia provides libraries that only run on their own hardware.

In Finland we've just started the first phone-free school year. Apparently, some parents are getting doctor's orders to allow their child to keep their phone, for situations such as anxiety attacks (article in Finnish). It's a miracle how such kids would have survived before mobile phones.

34 was my age when I quit my day job, thanks to Linux and the floodgates of open source software it opened for me.

https://www.youtube.com/watch?...

I think a long-term rental is called "leash" for this reason.

I have a hard time believing that a particular encryption will remain unbreakable, quantum computers or not. At the moment, we have Shor's algorithm for factoring numbers on QCs, so we should avoid relying on the hardness of factorization. How can we be sure that there won't be new algorithms in the future that break the current "post-quantum" encryption?

During my advanced math studies, I only took a rather introductory course on encryption, including stuff like Galois fields and elliptic curves. I recall my professor saying that none of the current encryption methods (besides something like the one-time pad) are proven to be safe; we just don't know any efficient methods of breaking them at the moment.

It's easy to keep making new highs when the value of USD itself is in a free fall. Look at USD vs. EUR charts, for example.

Why not? It's originally a quotient of two numbers multiplied by 100, just like any other percentage.

"Originally, IQ was a score obtained by dividing a person's mental age score, obtained by administering an intelligence test, by the person's chronological age, both expressed in terms of years and months. The resulting fraction (quotient) was multiplied by 100 to obtain the IQ score." https://en.wikipedia.org/wiki/...

Thank you for putting this so clearly. I said basically the same thing in the first post, and I was soon modded as flamebait.

You paid for a closed-source service where you're at the mercy of the publisher's whims. What did you expect?