Comcast is not a rapidly growing company, though.

This isn’t hypothetical. SpaceX is increasing Starlink revenue by like 50% per year. And their current Starlink constellation, about 10,000 satellites, has been launched entirely by Falcon 9. They’ve been waiting to launch much larger satellites on starship (in fact they had versions of these ready for several years now, and recently did suborbital tests of some of them on recent starship flights). Starship is about 5-10 times the capacity of Falcon 9, is fully reusable, & has larger diameter allowing much larger satellites. They asked for approval for roughly 40,000 of these larger satellites (~3 times the size of current ones, each about 10 times the bandwidth… and half of the 10,000 are even older designs), and they may eventually do about 100,000 of them & further increase the size and reduce latency (by operating at lower altitude). It’s not an exaggeration to say SpaceX intends to increase bandwidth by at least 100x, maybe a lot more. They intend to use a lot of this extra capacity to expand into mobile coverage as well. They are leveraging their platform for incredibly important national defense capabilities as well, and they operate as their own backhaul using on-board laser links. They can service anywhere in the world that will let them, including lucrative sectors like aviation. I do think it makes sense to value SpaceX as a rapidly growing business, not as a dividend-giving, plateaued ISP like Comcast.

This is all before even mentioning the idea of orbital datacenters.

You're trying to justify SpaceX's valuation using Starlink, but it's clear from the S-1 that it makes up only for ~5% of SpaceX estimated TAM.