Why compete with other researchers for telescope time when you can just use ALL the time!
It's honestly a perfectly sensible setup. SETI doesn't particularly care where in the sky they're looking, they just want as much data as can be acquired. Piggybacking off of everyone else's observations is just as good as pointing the antennas at random sections of the sky.
The COSMIC system is a cluster of FPGA, CPU, and GPU hardware, which at all times receives a copy of digitized data streams from each of the VLA’s antennas. The data received by COSMIC necessarily reflects antenna-pointing, LO-tuning and digitization settings commanded by VLA observers, but otherwise operates completely independently. In particular, COSMIC is unaffected by any of the interferometric phasing, channelization, or correlation processes of the VLA’s facility instrument: WIDAR.
COSMIC comprises three main components:
A fiber optic splitter, which duplicates digitized data streams from the VLA antennas and delivers these to COSMIC digital instrumentation.
A small FPGA Cluster, which receives VLA antenna digitized data streams, channelizes these to 1 MHz frequency resolution, and transmits these data streams over multiple industry-standard 100 Gb/s Ethernet links.
A CPU/GPU Cluster which receives 1 MHz resolution data over Ethernet and performs a variety of digital signal processing operations, ultimately outputting SETI signals of interest (often referred to as "hits") into a database for further analysis.
And the CPU/GPU cluster in particular is extremely interesting:
The COSMIC CPU/GPU pipeline implements a beamformer, high-resolution channelizer, doppler-drift search, and calibration correlator. The configuration parameters of these algorithms may be varied in order to best leverage the array’s behaviour as dictated by the Primary Observer.
The COSMIC CPU/GPU processing pipeline is still under development, but will have the following capabilities:
Buffering of ∼ 5 minutes voltage data for all antennas in the VLA;
Beamforming of up to 32 coherent tied-array beams;
Incoherent beamforming;
Configurable frequency resolution between 1 MHz and 1 Hz;
Inter-antenna correlation (used for gain calibration observations) with millisecond time resolution; and
Doppler drift searching for narrow band signals (in antenna or beam data) with linear drift rates between -50 Hz/s and +50 Hz/s.
In normal operation, the COSMIC system will search for candidate signals with non-zero linear drift rates, and will permanently archive small data snippets (< 100 kHz bandwidth and < 5 minutes duration) around these events.
Things should be pretty silent if we are in a dark forest. Unless the top of the food chain predators sometimes make noise waiting for naive preys that answer.
The dark forest has always seemed like too much of a just so story to me. I doubt that beings that would actually act in the way predicted would reach that level of advancement in the first place. I expect the real explanation for lack of observed extraterrestrial life is more mundane.
It makes very little sense, and has been thoroughly debunked in the literature for decades before Three Body Problem rehashed the idea.
One of the main critiques is the vast cosmic waste. It makes far more sense for the hypothesized psychopath top-dog civilization to hoard all negentropy (grab the resources) then let no other civilization emerge. Such an entity would be (1) plainly visible, and (2) have prevented us from emerging anyway.
Second major problem: why don’t we see a diversity of outcomes in unreachable galaxies? They are ALL dead.
> It makes far more sense for the hypothesized psychopath top-dog civilization to hoard all negentropy (grab the resources) then let no other civilization emerge.
Does that mean "taking" instead of "destroying" a planet? If so, doesn't it assume that FTL travel for macroscopic objects / life forms is possible?
> Second major problem: why don’t we see a diversity of outcomes in unreachable galaxies? They are ALL dead.
Can we distinguish dead from silent?
> debunked in the literature for decades
Can you recommend a book or articles about this topic? I'm ready to be convinced but I would want to see more than a HN comment before (with all its limitations)
> Does that mean "taking" instead of "destroying" a planet?
I'm not sure I understand the distinction? Or if I am understanding it correctly, I'm not sure it's a meaningful distinction. Either way I think the answer is neither: I wasn't really talking about planets. In the calculus of galactic resource management, planets in general, and especially those like Earth are an ignorable rounding error.
Primarily I'm talking about stars. 99% of the matter of the galaxy is contained within stars, and their nuclear fusion engines are transforming vast amounts of nuclear potential energy into waste heat, for no purpose other than raising the cosmic background temperature by a few microkelvin.
Planetary bodies and other non-stellar raw materials represent <1% of the accessible (non-blackhole) resources of the galaxy, and planets like Earth with solid surfaces that could support the evolution of life represent less than 1% of that, and it's locked up behind annoyingly high gravity fields. So really, we're the energy-poor, resource-poor, inaccessible interstellar backwaters that nobody would be interested if not for the presence of life.
Suppose you are a space-faring civilization hundreds or thousands of years beyond our own capabilities. Where would you get your material resources? By mining the low-gravity planetesimals, then minor and major moons, then planetary bodies. At some point you would start extracting resources from the core of Jovian gas giants and even stars themselves, a process called "star lifting."
You'd get your energy by enclosing the stars you encounter with swarms of orbiting solar panels, called a Dyson sphere. In the very far term, you'd lift matter from the star until its fusion engine shut down, causing it to become a brown dwarf. You then run the extracted light elements through much more efficient fusion power plants to directly capture as much energy as possible, and filter out heavier metals for construction purposes.
Etc. etc. This isn't just me decreeing this is how things would be done. This is a physics-based argument. Right now the ENTIRE output of the sun, minus that 0.00000005% which happens to hit the Earth, is utterly wasted. It does nothing except heat up empty space by an immeasurable amount. If nothing else, a Dyson swarm with existing technology would be 500 billion times more efficient use of solar power to support life than what the biosphere is currently doing. Enough to support quintillions of human beings in our solar system alone! And since there's ~100 billion stars in the Milky Way, a galactic-scale civilization powered just by 20th century human technology could leverage 500 billion * 100 billion = 50 sextillion times more energy than the theoretically most powerful "hiding" planetary civilization could access.
> Can we distinguish dead from silent?
Yes.
Now here's a critical point: there is intrinsically no way in which a civilization can achieve this without leaving the night sky looking different than we observe it. You can't build a Dyson sphere, and then camouflage it to look like a normal star to the rest of the galaxy. Not without re-emitting all the same energy that you were trying to capture, defeating the whole point and then some.
So there aren't galactic-scale civilizations out there. Not in our galaxy, nor in any other galaxy that we could see (I'll return to this important last point in a moment).
For the moment, I hope this answers your first question about "taking" resources. A grabby civilization would start building Dyson spheres (or something better), and capture the output of stars, strip mine gas giants for resources, etc. This will be visible, but because it is an intrinsically exponential process, technosignatures will go from 0% -> 100% over a very short time period, e.g. a few hundred years or even a few decades.
Which gets us to the next part of your comment...
> doesn't it assume that FTL travel for macroscopic objects / life forms is possible?
I'm not sure what "it" dereference to here, but the above quite possibly does imply that an effective wolf/interdictor/inhibitor/predatory civilization would have to either (1) be everywhere at once, and/or (2) have faster-than-light capability.
What would it mean for a hiding civilization to encounter a Kardashev level 2 (stellar) or Kardashev level 3 (galactic) civilization? Imagine the fully equipped USS Gerald R. Ford carrier group vs a Sentinel Island canoe, and you probably have a roughly accurate approximation of the scales involved. So like defending against an ICBM, it is vastly easier to do during the boost/breakout phase. After that, you've likely already lost. If you can't squash a civilization within a few decades of detecting stellar engineering, then it's already game over for the would-be interdictors. By the time they arrive they will have ascended to a level far beyond the interdictor's own puny capabilities.
So either:
1. The wolves are everywhere. They detect you almost immediate, and destroy you just as quickly, or
2. The wolves have faster-than-light capabilities, so that roaming observers who detect you can call upon the combined resources of the galaxy (or entire universe?) to wipe you out,
The FTL option we can justifiably rule out for reasons that are too long to get into in this already long post. Suffice to say any physicist would agree: the light speed limit isn't some arbitrary thing but rather baked into the very fabric of the cosmos itself. Nothing can go faster than light, fullstop.
So either there are out there indetectable predatory civilization(s) out that simultaneously live in every nook and corner of the galaxy, while at the same time consuming effectively no resources (energy or matter) or otherwise altering the trajectory of galactic evolution in a way that would be to their benefit. They must exist everywhere because if they miss a spot in their eternal vigilance, that life could in very short order develop into an unstoppable Kardashev level 2+ interstellar civilization. Oh, and the exist everywhere, in every galaxy, in every direction that we look, because nowhere in the observable universe do we see stars replaced with the dim infrared glow of Dyson swarms.
Or, if you're a believer in Occam's razor:
3. There aren't any wolves out there.
> Can you recommend a book or articles about this topic? I'm ready to be convinced but I would want to see more than a HN comment before (with all its limitations)
Sort-of. It was old hat when I started learning about it 25 years ago, so it's kinda diffused into the general culture of hard science-based predictions of the distant future, which is how I learned about it. But if you want primary sources, I'd start with the SETI institute and their various reports over the years, and writings published by the British Interplanetary Society. Some early writing of Carl Sagan is good, although his ideas are antiquated and need updating. Max Tengmark has written quite a bit about this more recently.
And the dark forest is not enough to hidden the sunlight if it is bright enough. So that concept can only apply to early space civilizations, assuming that space civilizations eventually use more and more energy and leave enough trail to be detected anyway.
We are still an early space civilization. We don't know if more advanced ones must have an energy trail as we don't have their science yet. We don't know even if they must expand neither, nor live in planetary surfaces or a lot of things that culture assumes from other civilizations.
You are entirely correct, hence my assumption. I think non-expansive civilizations are out of question as it's far more likely that we are never able to detect them even if they exist. The energy trail is more debatable, but that necessarily needs a violation of the second law of thermodynamics, so it still sounds like a reasonable assumption to make.
So glad to see SETI alive. Next step, building a proof-of-work token to bring back SETI@Home and leverage thousands more GPUs :']