The obvious answer is that it is aliens having a rave. Duh!
Incomplete Dyson sphere or ring. Whether it hasn’t been finished or has been partially destroyed is the real question.
Wouldn’t an advanced alien race count as an astronomically interesting outlier? Just saying.
Although it certainly is just one of some unknown large number of other possibilities and as such, unlikely.
Bumping this; a bit of news – whatever’s causing it isn’t comets.
Nice nod to Bowie in the article’s subheads there. 
I’m wondering if this could be caused by aging of the photographic plates? I assume that they looked to make sure that other stars didn’t also dim.
It’s actually a binary system with a (relatively) low-mass captured neutron star. The gradual dimming is caused by the secondary siphoning off material from the photosphere, while the erratic changes in intensity reflect a combination of induced flares and partial to full occlusion by the secondary and its accretion disk. There may also be one or more other bodies with orbits perturbed by the presence of the secondary.
I actually don’t know jack about this, I just think it sounds cool.
No; a system with an actively accreting neutron star would be very bright in x-rays, and I presume this is not the case.
A tangentially related question: in one article I was reading about it, the author doubted it could be an alien civilization because covering 19% of a star in just a century seemed very implausible to him. This led me to wonder just how fast a technological civilization can expand. (I’ll set aside self-reproducing machines for the moment and consider the whole civilization as a self-reproducing machine.)
As the nearest analogy we have experience with, the Baby Boom years are an example of how fast a technological civilization can expand in good times. Measured in population and gross domestic product, how much did the United States grow during the quarter-century from 1945 to 1970? I suspect the USA matched or exceeded 19% growth per century.
Can we really say what is or isn’t plausible for an advanced civilization?
Yes, this is almost certainly some natural phenomenon we just don’t understand yet, but I don’t get the argument that it’s not plausible for a civilization to do it when we’re talking about advanced technologies we know nothing about.
True, but Phil’s article didn’t say anything about x-rays, even though he mentioned lack of infrared emissions ruling out dust. I’d think it’s something they would have looked at, though. How much would the orientation of the disk affect the amount of x-rays we’d expect to pick up? If the secondary’s orbit is such that the plane of the disk never fully faces us, might the x-ray emissions that reach us be low enough not to stand out?
I think it would depend on how they’re producing whatever is causing the occlusion. If they had a huge stockpile of raw materials in place (maybe a trojan point full of asteroids) and were producing enormous thin-film mirrors, it would be more plausible than building structural panels.
In between those extremes, they could be inflating asteroids to make bubble habitats or something. That would be an application for the mirrors.
If we want to get really sci-fi, there might not be any megastructures. They could be manipulating the star itself somehow.
This was discovered by the Kepler satellite, long after the era of photographic plates. Kepler is in fact designed specifically to look for dips in the light output of stars, and is calibrated for that purpose.
As I understand it, this effect for this star shows up only in the brightness. The star’s specrum doesn’t change at all. Most variable stars will show some sort of spectral effect, such as a Doppler shift from the surface expanding and contracting, or from it orbiting another star, or a shift in the blackbody emission from a temperature variation. Without any spectral signature, this is almost certainly due to something obstructing part of the star; it’s just a question of what.
My personal guess is that it’s a bunch of hot Jupiters in chaotic orbits. It’s very rare to find anything in astronomy with chaotic orbits, since those tend to eject or eat objects until what’s left isn’t chaotic any more… but as mentioned, we’re working with a very large sample size here, so finding one rare circumstance isn’t too surprising. Since the star itself appears to be of a reasonably mature age, this presumably means that something seriously disrupted the orbits of its planets, in the recent past.
That’s more or less what I was thinking when I mentioned a captured object: something has recently (on the scale of such things) messed with the system somehow.
I wasn’t talking about what was discovered by Kepler. I was talking about the latest news, which was posted by iiandyiiii, about what was found in the old photographic plates.
I discovered the answer to my own question earlier: During the peak of the Baby Boom years, the US GDP nearly quintupled in twenty-five years, from $228 billion in 1945 to $1075 billion in 1970. And that was the average; growth exceed 6% annually in seven of those years. In the same roughly one-generation period, the US population increased from 140 million to 205 million. Population growth exceeded 1.60% annually from 1947 (end of WW2) to 1961 (introduction of birth-control pills?), with a peak of 2% in 1950.
So let’s say that a technological civilization at least as sophisticated as ours could maintain a 1.5% growth rate annually, and that in this case that was the growth rate of energy capture by space colonies. That would work out to over 4x increase per century.
Interesting question: do the photographic plates register a linear or exponential decrease in brightness; or can they tell?
That’s the kind of thing I’m wondering about. What can they tell and how can they tell it?
How uniform were the exposures, the processing, and the chemistry of the plates over all of that time? Do the images fade at all with age? The earliest are from the late 1800s, so they aren’t exactly high tech.
My guess is that they’re measuring the brightness relative to the rest of the stars on a given plate. At least I hope that’s what they’re doing. That would at least give a relatively accurate measurement of changes in one star over time.
In regards to astronomy and photographic plates.
Photographic plates were used in professional astronomy from almost the day film was invented till at least the early 80’s. Much good and precise science was done with said plates.
Those old plates are a treasure trove of information.
Astronomers spent decades developing techniques to get good info from photographic plates.
Plates, even very aged, are plenty good enough to “measure” a star’s brightness in the past, particularly something as large as a 20% variation.
As a semi educated WAG, I’d say with good technique, something more along the lines of 1% would be easily doable.
Long story short that IS how they do it.
I got it, and it made this old electrician laugh.
Maybe all of the other stars have gotten 19% brighter. 
A Kickstarter project has successfully funded 1 year of access to the Las Cumbres Observatory Global Telescope Network to observe the star and try to catch and record and dimming incidents.
https://www.kickstarter.com/projects/608159144/the-most-mysterious-star-in-the-galaxy/description
So maybe we’ll have some answers.