[throwaway_yy2Di]

    can it be easily ruled out?

Yes: parallax breaks it. If something's colinear with a star today (and occults it), it won't be colinear 6 months later, when the Earth's on the other side of the sun.

At 2 light years for instance, that spacecraft would trace an apparent ellipse 3.3" in diameter, while Tabby's star (1,480 ly) would be stationary (0.004" parallax). In comparison, the star's apparent disk is just 30 μas wide (0.00003").

[tzs]

If I'm doing that math right, it works out that to not be ruled out by parallax, the object would have to be almost 2 AU in diameter if it was within a few light years of Earth. That seems pretty ridiculous--now we have them not only building a megastructure, but flying it to Earth!

However, I wonder if there is a way to save the idea without requiring a ridiculously large spacecraft? Suppose instead of one ginormous ship, it is a fleet of big ships each big enough to cause irregular dimming if that ship happens to get between Earth and KIC8462852. If the fleet is spread out, it could then be that one ship or another ends up between Earth and KIC8462852 at several times during one Earth orbit around the Sun.

That would give us the irregular dimming throughout the year.

For the slow continuous dimming, perhaps that is something going on with the star itself. That could be what made the inhabitants of the system decide to build a fleet of big ships and leave their weirdly behaving and very scary star.

[e12e]

Maybe the visible size of the ship is bigger, as it's unfolding / building a solar sail to brake using our sun's radiation? (like a parachute on a drag racer). Ah, but 2au is probably still way too big?

[sitkack]

What if the parachute was a big magnetically contained gas cloud ?

[fjarlq]

Excellent point... that answers my question. Thanks!