> And it's not possible to simply argue to use older designs to last longer in a new mission, because you would give up multiple orders of magnitude of data col

> And it's not possible to simply argue to use older designs to last longer in a new mission, because you would give up multiple orders of magnitude of data collection. Why not both? The heavy parts of the probe, then or now, were presumably not the electronics themselves so much as the physical structure, the power system, etc etc. If you were designing a probe to last fifty, a hundred years, why not include series of fallbacks of the electronics to simpler and more basic systems, so that you could still get something out of it, after the primary state-of-the-art electronics fail?