Recently, there has been some media coverage of a recent sample of fast radio bursts (FRBs) that are an unusual regular integer spacing apart, raising the possibility of intelligent life.

The original paper:

We have noted a potential discrete spacing in DM of FRBs. Identified steps are integer multiples of 187.5cm−3


In case this would hold, an extragalactic origin would seem unlikely, as high (random) DMs would be added by intergalactic dust. A more likely option could be a galactic source producing quantized chirped signals, but this seems most surprising. If both of these options could be excluded, only an artificial source (human or non-human) must be considered, particularly since most bursts have been observed in only one location (Parkes radio telescope). A re-assessment of man-made phenomena, such as perytons (Burke-Spolaor et al. 2011), would then be required. Failing some observational bias, the suggestive correlation with terrestrial time standards seems to nearly clinch the case for human association of these peculiar phenomena.

Usefully sceptical coverage by Gizmodo:

No matter how you slice it, eleven data points is a small sample set to draw any meaningful conclusions from. A handful of deviant observations could cause the entire pattern to unravel. And that’s exactly what seems to be happening. As Nadia Drake reports for National Geographic, newer observations, not included in the latest scientific report or other popular media articles, don’t fit:

“There are five fast radio bursts to be reported,” says Michael Kramer of Germany’s Max Planck Institute for Radioastronomy. “They do not fit the pattern.”

Instead of aliens, unexpected astrophysics, or even Earthly interference, the mysterious mathematical pattern is probably an artifact produced by a small sample size, Ransom says. When working with a limited amount of data – say, a population of 11 fast radio bursts – it’s easy to draw lines that connect the dots. Often, however, those lines disappear when more dots are added.

“My prediction is that this pattern will be washed out quite quickly once more fast radio bursts are found,” says West Virginia University’s Duncan Lorimer, who reported the first burst in 2007. “It’s a good example of how apparently significant results can be found in sparse data sets.”

Who here knows more about this?


New Comment
5 comments, sorted by Click to highlight new comments since: Today at 4:05 AM

That reminds me .. of one of the problems with SETI.

If you look at how our current civilization uses its large powerful radio broadcast telescopes, they are used mainly for radar imaging - where a powerful one shot pulse is used to image some distant object (such as an asteroid or moon). Naturally there is little point to repeating an imaging pulse.

SETI is designed to search for information carrying signals that are either permanent or repeat reliably, and thus would fail to detect most of the radar energy output for a civilization like our own. Ping-like radar signals are detected all the time by SETI, but they are not the long repeating signals that SETI is looking for.

THIS. It is so incredibly frustrating. The sorts of signals we are likely to see are exactly the ones that are unlikely to be confirmed.

I have the skeleton of a post in my head about the fermi paradox and these sorts of intermediate solutions - what we can observe, and have and have not, what we have excluded and what we have not.

Cool - sounds like a fun post.

At this point SETI can rule out some hypothesis classes - namely those involving colonization of the galaxy by bright hot civilizations. (energy limited stellar based civs)

However, from a pure future engineering perspective, the cold dark model for future civilizations seems more likely on first principles (temperature constrained civs). A galaxy populated by cold dark ET civs would more or less look like what we see now. Perhaps some future clever technique could detect cold dark ETs, but at the moment it seems difficult short of sending out a ton of probes and waiting centuries - probably faster to just create superintelligence on earth and use that to figure it out.

I've written a little about the cold dark model and the 'fermi paradox' - here.

It may just be a coincidence, but I notice this was published at the end of March. Historically there have been a lot of spurious results published on or immediately before April first.

Historically there have been a lot of spurious results published on or immediately before April first.

Is there anything I can read on this phenomenon? I've never heard of it before but it'd be pretty important if real.

New to LessWrong?