Never underestimate the bandwidth of a truckload of hard disks.

Latency hardly matters when the majority of the lag is the 100 light year leg: you first spread out your whole message to the surface of the filter with a high bandwidth distribution medium. Then each section of the filter switches in sync, encoding the message without relying on a real-time input from a central location light seconds away

Is it necessary to use the star one is orbiting around?

Why not use an array of smaller polarizers that can be modulated when they block the path of a bright but less adjacent star’s light to the recipient?

what you are saying could also be: other civilizations would do the same as we do. send small unmanned drones into space, and not just signals. they are much slower, but allow much more information and better reaction time, if the onboard logic does that. and they are not mistaken for something else, as in this case.

Can't you delay the signal sent to the center-parts of the polarizer relative to the edge? I think the panel would appear to switch all at once to somebody far away. Seems like this would increase the bandwidth.

Problem is, a star is curved, with the limb being over two light-seconds farther away from an observer than the center. Sure, you can switch the whole surface of the star as fast as you like, with bandwidth limited by how accurately you can synchronize the switching components. An advanced civilization might be pretty good with attosecond-level stuff, but the signal gets spread out over seconds regardless of the synchronization superscience you apply.

You can phase the switches and "aim" the star's signal at a point in the sky. Lower-frequency modulation can get signals to observers not on the direct line, at a lower bandwidth, and you can probably sweep these. Worst case you can blink the whole affair slowly; a star with a five second light curve is going to be pretty interesting.

It would take 2.25 seconds for a clock signal to propagate throughout the polarizer, but, after that, all distinct elements could sync up for whoever is perpendicular to the transmitter and blast away at whatever bandwidth the signal transitions allow.

You don't need to place the polariser right next to the sun; you can use a smaller disc to modulate it's output in one direction.

Like the way the earth is large enough to block the sun on the entire moon during a lunar eclipse.

If your shield is smaller than the star's diameter, you have a penumbra effect. Rather than stellar occlusion, at a sufficient distance you have a stellar transit.

Building a static support for your polariser might also prove challenging.

Ah, yes of course. Though if your disc is large enough, such a transit should still be detectable.

You're trying to detect the modulation, not merely the transit.

Your noise floor is now vastly greater.

They could be basically the same - during each transit you could either let the light through or block it.

Now you've decreased bandwidth to orbital period.

I missed the part about some minimum speed, but it's still flexible in regards to how many levels or frames per transit you use. The point is really to be able to communicate at all - without using a star's light it's basically impossible after a certain distance.

There is that "something is better than nothing" element.

I've also been thinking of whether or not some sort of stellar laser or maser might be possible. Modulating a narrow, tightly focused beam.