(... and that's before we consider that cheap launch infrastructure means we can also address the problem by launching telescopes, which would put high-quality imagery completely unobstructed by atmosphere in the hands of far more astronomers than can touch that tech today).
As I understand it, it's not quite as simple as that when the number of satellites gets large.
> that's before we consider that cheap launch infrastructure means we can also address the problem by launching telescopes
The cost of the launch is not the largest part of the cost of having space telescopes.
I'm not sure "simple" is the term I'd use, but given the satellite ephemera and timestamps, the algorithm for knowing where the satellite is and therefore ignoring that chunk of sky when the satellite passes through is well-understood (and we can certainly mandate that every launching organization maintain correct and up-to-date ephemera).
Telescopes observe a tiny, tiny sliver of the night sky. The density of satellites necessary to interrupt them continuously would be something well past the expected launch capacity of this century alone (and would create its own navigational problems were we to achieve it).
> The cost of the launch is not the largest part of the cost of having space telescopes.
Not as they've been done, because the cost of launch was so high that it could only be justified by putting a perfect-mirrored device that could see all the way back through spacetime to the first microseconds of the cosmos in orbit. Now imagine the economics when you're putting a hundred up at that price, not one (and you can use algorithms to combine signal from them to account for lower resolution).
... imagine having, with the right math, an aperture the size of the orbital diameter of the Earth itself...
For part b) you are completely incorrect in your assumptions here. The majority of the cost and effort of telescopes is really in the instrumentation (the 'cameras' that are strapped to the back of the telescope). To get the most out of an observing facility, we need access to these to change them, fix them, and improve them. This is not possible with a space-based mission.
Long-duration observations can't be disrupted badly by a quick satellite jaunt across the sky. That's just a noise blip. Less than 0.00001% of the data collected. Isolate it and scrub it.
Similarly, wide-field observations have a lot of data to work with to figure out what the field is supposed to look like vs. what it looks like with a satellite-pass event. Isolate them and scrub them.
Satellites and deep stellar phenomena are different scales of event at different ranges and speeds. A Fourier transform should make them show up as obviously separable.
(... and again, all of this is assuming you have to scrub this data without a-priori knowledge, when the disrupting event is a well-documented passover with clear ephemera. If it is mandated that satellite companies publish the ephemera, which I think is a perfectly reasonable burden upon them, it should be even easier to separate signal and noise).
I do see your point regarding (b) though; space telescope science may require a wholly new approach relative to terrestrial science.