4 ms·
Sure, "move fast and break things" can't be naively applied but that doesn't mean it can't be applied. Instead of sending up 1 super precise super accurate sup
by AdamJacobMuller 5y ago
Sure, "move fast and break things" can't be naively applied but that doesn't mean it can't be applied.
Instead of sending up 1 super precise super accurate super reliable telescope let's send up 1000 mostly POS telescopes and combine the results using AI of the 900 which actually make it to orbit and return useful images.
I'm hardly saying that's a 100% working plan, but, I think that kind of paradigm shift which is required.
- bcraven 5y agoBut how are you going to validate your scientific discoveries if all the images are coming through the pipeline: "POS telescope > AI upscale"?
- hypertele-Xii 5y agoSounds impossible to park 900 satellites on the same L2 Lagrange point. Hope you realize this telescope is to be positioned on an infinitesimally small and unstable point in empty space, on the "far side of Earth" from the Sun. Because that's the spot with the best view, apparently.
- zardo 5y agoThey don't park them right on L2, JWST will orbit L2 thousands of kilometers away from the exact point.
- HPsquared 5y agoThe AI would lack training data.