Skip to 5:41 to hear it actually played. Although I assume you'd need very good speakers and I'm not sure if the YT audio compression is good for very low tones.
I can't speak for YouTube's specific compression, but since there's less information in low frequency sounds they should compress better than high frequency sounds.
Anecdata: the bass in low bitrate online drum and bass radio in the 2000s always came through cleaner than the mids and treble.
Where "sort of" depends on your personal answer to "does spending 15 minutes recording yourself making various noises with it, then editing the clips together count as "playing"?".
Not sure what the technical terms are but there's 2 phases -
1. Placing the queens, this is a pretty basic DFS over all possible queen positions. You can visualise it by going across the board row by row and placing the queen in a random available column, then moving to the next row and placing a queen in a random available column. So for an 8x8 board with 8 queens, the 1st row has 8 possible columns, the second row has 5 possible columns, the third has 4, and so on until all the queens have been placed or there are no available cells to place the next queen on.
2. Generating the colours - This is a bit more complicated but it is essentially a flood fill with each queen as a source block. I keep an adjacency list for each colour containing the uncoloured cells it borders. At each iteration I pick a random queen/colour and a random cell from the adjacency list, then fill that cell and update the adjacency list (and remove the cell from other adjacency lists).
I'm planning on updating my colour generation algorithm to support 2 things: 1. Generating boards with unique solutions only (not sure how I'll do this without finding all the solutions). And 2. including known shapes in the generation process so there is a random chance colours takes on specific shapes (e.g. +) and the other colours will generate around them.
I don't think it's particularly relevant that the suspect is a software dev. Many people in the US are software devs nowadays. It has become a completely mainstream career option.
That’s just basic engagement strategy. Speculation and “the other tribe did it, those assholes!” Go hand in hand and will boost engagement on the site which is going to make _somebody_ money they otherwise wouldn’t have made…
Yeah, was glad to see some Reddit posts that when folks like Ben Shapiro and Matt Walsh tried to say "This is why the crazy left is OK with CEOs being shot" that they were bombarded with comments saying "It's not just the left".
On that point, Zeynep Tufekci once again shows to me how she has the most insightful analysis (I became a big fan of her during Covid, I thought she had the most measured takes). I totally agreed with what she wrote in https://www.nytimes.com/2024/12/06/opinion/united-health-car...
It's pretty obvious that's bullshit. I was pretty shocked at how broad-based the lack of sympathy for the CEO was.
Ben Shapiro and Matt Walsh got destroyed in their comments when they tried to make this into a "look how evil the left is" issue, e.g. https://www.youtube.com/watch?v=bimRF1K3Bzw , and note his YouTube comments are usually very supportive. The contrast in the audience reaction to https://www.youtube.com/watch?v=Agu2gRzUSDY , his commentary on the Daniel Penny case, is stark.
Yes, my point is that right-wing talking heads are trying to make this into a "right vs left" issue, and they are getting destroyed for that characterization by their right-wing fan base. When you say "I can't determine who is downvoting/commenting on the video", that's specifically why I added the Walsh's video of the Daniel Perry case. Or heck, look at any of his videos, which normally always upvote his main ideas and tilt heavily right. If you think it's just a left-wing brigade flooding his comments, why don't you think they'd do the same on any of his Daniel Perry videos?
It could easily be a brigade by the left-wingers, because he is explicitly blaming them.
Daniel Perry video doesn't blame anyone. And most people, even on the left (though definitely not everyone), seem to have a normal reaction to the outcome.
I am surprised that more people don't know this type of brigading to influence public opinion. Left-wing brigades are orchestrated on private Discord servers. Smaller right-wing brigades are orchestrated on private Telegram channels.
The left-wing brigaders pretending to be right-wing CEO-killing-supporters could have gotten away with it if they had limited themselves to populist MAGA spheres.
However the Ben Shapiro fanbase are boring middle of the road pro-capitalist/neocons who would never support killing CEOs. Pretending to be his fanbase was their fatal mistake.
looks like he followed/retweeted rogan, trevor noah, ezra klein, elon, AOC, rfk jr, peter thiel, bernie on twitter. very concerned with falling birthrates, anti-capitalist but pro musk, supports AI, dislikes overwatch but is a pokemon enjoyer. eats mcdonalds but is super ripped, gave positive reviews to ted kaczynski's manifesto as well as jd vance's hillbilly elegy
Lotta folks like that online, STRONG feelings but it’s not at all clear if their opinions are thought through… lots of sentiment and mismatched identities.
The problem with tribal (political) identities is that to hold independent and opposing tribal ideologies (ie pro-choice but also pro-gun) is isolating from all groups. Thus, many people end up choosing to belong at the expense of the inconsistencies in the respective group they have decided to be a part of.
I'd say that to hold anti-capitalist ideals while liking Musk is reconciled through varying lenses. Musk can be idolized for his impact on the environment and moving various initiatives forward such as self-driving. One could argue that Musk is capitalist, but it could also be argued that Musk would drive these "good" initiatives even in a non-capitalist structure, thus the respect.
I'm not supporting musk, but merely illustrating how a certain level of rationalization can take place, counter to your insinuation that "opinions may not be thought through." It's the same thinking that also gives way when understanding the psyche behind various voters in this recent election.
Seems like Github should train one LLM per license type. There'd be a GPL LLM that's trained on GPL code, an MIT LLM trained on MIT licensed code, etc. Then Copilot users could select the LLM for a license appropriate for the work they're doing.
