3 ms·
The article is a pain to read: it's a word salad with just a few bits of information. I can see how such articles will be generated by GPT with a prompt like "v
by frongpik 6y ago
The article is a pain to read: it's a word salad with just a few bits of information. I can see how such articles will be generated by GPT with a prompt like "voyager crossed the heliosphere boundary and found weirdly turbulent space there". I feel like that most journalists are glorified GPT models: they expand a short prompt into a word salad so the news agency could stuff it with ads.
The really interesting question is "what is [solar] magnetism?" Wikipedia gives a thought terminating answer: magnetism is a vector field, as if vector field a real thing that can press back the sea of turbulence like the sun does.
- p1esk 6y agoIt's a force vector field. What's not real about it?
- frongpik 6y agoWhat is it made of?
- p1esk 6y agoI have no idea. Doesn’t make it any less real though.
- frongpik 6y agoThis is what I meant by a thought-terminating answer. Magnetic field is not a "vector field" because a "vector field" is a math abstraction, while magnetic field is real. Just like water is not a diff equation with boundary conditions, no matter how accurate that diff equation is.
- m_mueller 6y agoI think that is the wrong question to ask. A vector field is just a mathematical model to describe a natural phenomenon, as does curved spacetime for gravity. Those models have limitations under certain Conditions. For magnetism there are afaik quantum mechanical models available for all energy states while for gravity those seem to only work at lower energies. At the end, your question is about the nature of truth - what is behind all this. I don't think anyone can answer that so far in a scientific way, that is, in a falsifiable and not yet disproven manner.
- frongpik 6y agoThose quantum models are just more obscure and more accurate abstractions. Otherwise we wouldn't have copenhagen and other interpretations. The fact that abstractions are accurate doesn't make them more real. Example: navier stokes equations are accurate, but they aren't water; moreover, by adding complexity to the diff equations, they won't become more like water. This is what I really don't like about the modern physics: it's become an elaborate fortress of math models shielding us from the reality, and whenever we discover a missing wall in the fortress (dark matter), we just patch it with another clever math abstraction.
- m_mueller 6y agoBut that is all science is and has ever been - trying to understand the world, by measuring it and creating models that should match measurements as much as possible. What you consciously experience as reality is also nothing else - a model about the world that your brain has built up, fed with inputs about the current state.
- doliveira 6y agoWhat do you think that is (or can be) the reality quantum mechanics is shielding you from?
- thekyle 6y agoOn the bright side GPT like models can also be used to produce summaries.
- m_mueller 6y agoThen you just need an algorithm to consume the summaries and we can finally just have machines feed other machines...
- lmilcin 6y agoThat is simple. Journalism becomes another super-optimized activity. "Journalists" today typically want to expend as little effort to get as much results the reader be damned. The results are defined by your output in the views you can generate per month and this means you generally want to write a lot of articles each to be well received... by Google algorithm. You probably also want people to spend a little bit reading the article and hopefully scroll further than the first page.