7 ms·
The problem with these papers is there is no compilers to perfectly test the logic except peer review so no one knows if its horse shit or not until someone com
by deytempo 8y ago
The problem with these papers is there is no compilers to perfectly test the logic except peer review so no one knows if its horse shit or not until someone comes along and comes up with some way to prove it wrong, so I would not be so quick to say “if that’s an obvious contradiction, the author would have caught it”
- phs318u 8y agoPhysics is highly mathematical and as such, can indeed be "tested" (i.e. proven) to be consistent with mathematical laws. This is fairly straightforward (for suitably mathematically literate people). The real issues are: 1. can the theory make testable predictions that can be verified by observation? e.g. string theory has had issues with this. 2. what does <wierd-mathematical-construct> "mean"? e.g. negative mass, spooky action at a distance, spacetime singularity. In practice, only 1 really matters. However, 2 tends to bug a lot of people (consider the arguments around what the quantum world really represents). I'm one for whom 2 is important. "But what does it really mean?" Back to your original point, looking at the paper [0], it doesn't seem too ridiculously hard to validate the math for someone familiar with this space (see what I did there?), unlike me who has forgotten everything I was taught [1]. [0] https://arxiv.org/pdf/1712.07962.pdf https://arxiv.org/pdf/1712.07962.pdf [1] I have a physics degree but that was 33 years ago and I've been doing nothing but IT since.
- v_lisivka 8y ago1. OpenGL makes good predictions. Why not to switch to OpenGL terms when talking about physics?
- Retra 8y agoBecause physicists would find it totally unsuitable for their requirements?
- v_lisivka 8y agoWhy match is better for that? To compute prediction, you will need to use computer and compiler, so it better to talk directly in these terms because of (1) (see above).
- tejon 8y agoYou should go back to 1910 and tell Einstein he can't start for another 50 years.
- v_lisivka 8y agoEinstein used the best tool he had at hands - math. Now we have mach better tools: computer models, knowledge databases, AI. Math is good because math formulas are very compact, so they are easy to play with, but computer models are much better, because computer can visualize them directly, calculate values correctly, and verify result automatically. Think about Computer Aided Science.
- phs318u 8y agoI'm confused - are you trolling us? You do realise that computer science is based on math right? All the calculations that that OpenGL does are mathematical calculations. Even databases are based on relational algebra. AI - linear algebra, probability, multivariate calculus, optimization etc. There would be no computer science without math.
- v_lisivka 8y agoNo, I'm not trolling. English language just have no word for that thing. How would you name something in middle of OpenGL, Kerbal, Wikipedia, Recursive Text, Alpha Go, etc? Something where you can setup your experiment, start it emulation, and then zoom from macro level to quantum level, with description of every effect at each level, with formulas at hand, with links to experiments, papers, reviews, confirmations? Something that can actually answer your questions, teach you actual physics. Something where you can plug your own theory to see is it fits real world better that someone else theory. I have no idea how to name it. Universal Programmable Augmented Disсoverable Science Model?