Doomscrolling is a black magic ritual. You lay in the same position. Turn off the lights, deeply focused on the feed. You’re shown memes, tragedies and “useful” information (it’s all slop) and before you know it you’re stuck in a trance. Your thumb moves without thought. You’re in flow but it’s artificially engineered by the dopamine loops the algo was built to trigger. You remember you have things to do but the black mirror pulls you in deeper by showing you something that stimulates you more. Before you know it, all your free time is gone. Every original idea you may have had is gone. You’re polluted with open loops and feel restless all the time. Every day you go to your altar (your bed), pull out your rosary (your ig feed) and you endlessly worship slop. Time blurs together and years have gone by yet you repeat the same pattern. Why? because you’ve never been taught how to engage with your own boredom, with your own pain, with your own discomfort. You resist it like it’s killing you, but what’s killing you is what you’ve turned into a part of your life (scrolling). It’s inverted. If you simply learned how to be courageous and face your immediate discomfort with pain and boredom. You’ll allow your natural wisdom and creativity to rise and begin living a life you truly want to live.
@Erdayastronaut This idea is so good that decision makers need to take it more seriously. @SciGuySpace posted about this https://t.co/Iz9mMZXWGF
@Erdayastronaut how would you feel about someone submitting this idea to the challenge?
@MattWhitmanTMBH This is a good take, but on another dimension, I have seen a pattern throughout history where The Tempter works to corrupt God's people/church, moving them to legislate a moral code over grace.
We see this in the Pharisees during Jesus' time and in cycles of history since then.
Science research is mostly invisible to the public. To many it's obscure & irrelevant. And returns on science investments carry longer horizons than quarterly or annual reports
Yet it's the engine room of tech-innovation. Cut funds for science & you ultimately stall your economy
View of Earth from 900 million miles away, with Saturn's rings in the image, taken by Cassini spacecraft.
That dot is us, all 8 billion of us. It's all an insanely lucky miracle.
I'm grateful for all of it & all of you. Love you all! ❤️
PS: Now, I'm off to partake in the great American Thanksgiving tradition of over-eating while getting into a heated philosophical argument with family 🤣
Every machine in a Hospital that diagnoses your body without cutting you open is based on a principle of Physics, discovered by a Physicist who had no interest in Medicine.
If you think the world doesn’t need Basic Science, or that somehow Science has failed you, think again.
Nothing has ever harmed Christianity so much as outspoken incurious Christians who are certain about almost everything yet loving toward almost no one.
@Erdayastronaut Hey! Welcome to Seattle!!
I think I know the event you might be in town for...
Either way, let me know if you'd like a local fan to buy you a drink or something.
Yesterday, I posted that large language models are non-deterministic, and people went bonkers.
People called me stupid, asked me to go back to school, and used so many other names I can't remember. I had the crème de la crème of this site come after me.
Today, I want to elaborate.
A deterministic model will always return the same answer given the same input. My argument is that the set of large language models we use today can't fulfill that contract.
Yesterday, most replies tried to school me about the temperature parameter when using an LLM. Yes, you can set this parameter to 0 to try to force a model to return the same answer, but in practice, you'll find this isn't guaranteed.
Theoretically speaking, temperature = 0 makes a model deterministic. In practice, this is no longer the case.
There are several reasons:
• Non-deterministic hardware operations can produce different outputs. Two GPU kernels can diverge slightly run to run. This may introduce slight differences in your outputs.
• Floating-point operations are tricky, and even when using greedy decoding, a model might suffer from tiny numeric drifts that shift the outputs.
• If you are using a model scaled across multiple servers, you are not always hitting the same model instance, which increases the chances of finding issues with differences in hardware.
• Your model provider will patch, retrain, and fine-tune the model you are using without telling you. What was true yesterday isn't guaranteed today. (I'm kinda cheating with this last point, because it's not related to the nature of models, but it's still a problem we have to deal with.)
So, in practice, the LLM you are calling is non-deterministic, even at temperature = 0.
OpenAI's documentation tells us that we can expect "(mostly) deterministic outputs across API calls."
Gemini's documentation says that "responses for a given prompt are mostly deterministic, but a small amount of variation is still possible."
I'm not sure about you, but "mostly deterministic" is not the same as "deterministic."
People telling everyone they want to live in a better world and then just walk down the street and don’t throw that piece of rubbish they encounter in the nearest bin
Day 1: You don't know how to use abstractions, so you write flat code.
Day 1000: Your code is fancier and full of abstractions. It's "flexible", "scalable", "observable", and a bunch of other "-able".
Day 3000: You realize over-engineering is the biggest mistake of your career. You start writing flat code again.
I’m proud of this one.
I brought a solar telescope to Florida to capture a Falcon 9 rocket launch transiting the sun. While these have been captured before, never with the details of the sun’s chromosphere, which makes this one the first!
See the video or get the print below 👇
If you say "the left needs to quit condoning violence," you need to also condemn those you agree with politically who use violent rhetoric
If you say "the right needs to quit condoning violence," you also need to condemn those you agree with politically who use violent rhetoric
Tend your own garden first.
That being said
Any Libertarian who condones the use of violence for political change I fully oppose, and they are in direct contradiction to the aims and principles of the Libertarian Party.
Here's a fact, the biggest purveyor of political violence has been the government itself, who bombs people around the world and supports others doing the same. Who violates our privacy here at home and uses mass media to distract and divide us.
The rhetoric and, more importantly, the actions of the government have been far worse in size and scope than any pundit or commentator.
This is regardless of the party in the White House, btw.
No. Fuck your sides.
They need you to believe neutrality is cowardice so you never notice the real cowardice is outsourcing your identity to a team.
Every side is just a funnel into the same machine. The same laws, the same debts, the same bureaucrats, the same wars, the same surveillance, the same permanent class of rulers playing musical chairs while you cheer for the jersey you picked.
Refusing to pick is the only way to stay whole. To keep your voice from being drafted into a chorus you don’t control.