I'm sorry, I but I don't buy that argument at all. Jon Anderson put together Olias of Sunhillow in 1976, at home, playing every instrument himself on a multitrack tape machine. He had an engineer to run as tape-op, but he could arguably have done that too with a remote control. Jean Michel Jarre did the same, also 50 years ago. Michael Garrison as well in 1979, and that's ignoring Mike Oldfield, and Vangelis who also had their own multitrack studios.

You had decent drum machines and sequencing in the 1980s and that made solo stuff even easier and cheaper. Multitrack machines were still a bit pricey but dropping rapidly, especially when ADAT arrived. By 1999 you could do the whole thing on a computer, sequence the backing tracks and drum machines, digitally recording the vocals in an early DAW and by the mid-2000s the costs had come down immensely. Virtual instruments are still a huge market.

Point is, you have been able to program and record your own singles completely solo without using an LLM for over 30 years, and for the last 20 years the cost of doing so has been inconsequential as you can do everything on a single laptop. People do this as a hobby. I have been doing this as a hobby for over two decades. Recording and mixing the songs is the fun part, and the challenge makes it rewarding - especially if you're doing the DAWless thing or working entirely on tape just because you can. Asking an LLM to do everything for you? To my mind, it takes away the parts that make it worthwhile.

24 hours is for critical infrastructure, "Control centers" as the summary puts it. For cell towers, they only mandated 4 hours which should be a lot easier.

SMPTE timecode was (and sometimes still is) used to synchronize two playback machines. AFAIK the original intention was to help lock film or video editing machines to an external tape recorder, e.g. you'd record the final master for a movie score on 1/4" audio tape with two tracks and digital timecode data in the middle. If the film or video also has SMPTE timecode data you can use this to determine the time difference between them and keep them in sync by adjusting the capstan motor.

If you had two 24-track tape machines in a recording studio, you use the same principle to get 46 tracks by striping timecode data on track 24 on both machines, and again, a synchronizer unit can ensure they run together. War of the Worlds was done this way and was probably one of the first. I think the world record was 5 24-track machines slaved together, which must have made mixing an absolute nightmare.

Later you could get a box that would decode an SMPTE timecode into MIDI timecode. That way, before using a DAW was feasible, you could compose a song on a computer, using it to control a bunch of synthesizers, and overdub them to tape. Using an SMPTE timecode you'll ensure that the computer can chase the tape and the timing will be perfect each time even if the tape speed drifts slightly, which it does.

More recently, modern DAWs can actually decode SMPTE in software, when you're importing a project from tape, or doing a hybrid tape/DAW project.

I use SMPTE for both sequencer timing and to lock two tape machines together. Obviously I could do this with a lot less effort on a single laptop, but I get a kick out of doing things the old way.

Author's rights lose meaning and rationale when the "author" is not an individual or a small group but a corporation.

Research Machines built weird PCs in the 80s based around the 186. It was almost PC compatible but had to have its own special versions of DOS and Windows 2. Unfortunately Research Machines had some kind of lock on the UK education sector, so schools that weren't still using BBC Micros or Acorn Archimedes machines, had these strange 90% IBM-compatible abominations. (Just looked it up, it was called the "RM Nimbus")

So what. When you're screwed you might as well try unorthodox things on the off-chance something might work, because hello, we don't actually understand human biochemistry perfectly. And what exactly did he have to lose, a few more or fewer weeks with life with apparently lethal metastatic cancer?

If you offer bets on things that aren't definitely and objectively classifiable, you aren't foolish, you're a scammer. There's a difference.

There's an added point here, similar to what Rory Sutherland went on about with the, according to him, unfortunate preference for rational (or rather conventionally logical) approaches in business. As of now, zero people have gotten fired for suggesting "Our sales are down, we ought to advertise more!" , even when it didn't work. Now, try to think what would happen to your job if, instead, you went "Our sales are down, we ought to advertise less!"

A while back someone posted a (now-deleted) comment on the Fediverse about how they were able to construct a single unicode glyph of arbitrary length, e.g. a single character that could require 2MB of data to store.

The salmon code appears to be doing something similar - if you look at the source code in a hex editor, you'll see that the four spaces before "is very yummy" are actually a huge stream of F3,A0,81,93 and so on, where '93' is a varying number that may contain the actual payload (a recipe for cooking salmon). I don't know enough about unicode to tell precisely how it's doing that, but essentially it's some strange unicode trick.

However, I do not understand the significance of the 'grill' defines, since the code doesn't appear to be calling it in a way I'm familiar with, but it is also doing some unicode shenanigans with that too, and the code won't build if the 'grill' defines are commented out.

but I'm not touching the ad-serving vehicle that is Chrome and its derivatives with a ten foot pole. And, let's face it, this is the only reason for anti-firefox propaganda these days - they really need to whittle it down before forcing manifest v3 through, or users will mass-switch.

Since a polite way of putting it isn't cutting it, let's try again. What the "x maintainer" in questiponn posted on the linux kernel mailing list doesn't matter because nobody reasonable gives two shits, more so in a completely different project. So, you coming with it like it's a revelation is at best ridiculous.

I've never liked the distribution - I've used SuSE, Slackware, Mint and finally Debian... but there's a difference between "I'm not a fan" and "this unapologetically sucks". The internet is full of people having issues with Wayland.

...so? In what way did that impact his X.org work? Like, when I go to a restaurant I care that the cook knows how to cook, not who he voted for.