Jump to content

Wikipedia:Reference desk/Archives/Science/2017 October 11

From Wikipedia, the free encyclopedia
Science desk
< October 10 << Sep | October | Nov >> October 12 >
Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


October 11

[edit]

Cosmic-ray-based random-number generation?

[edit]

Our article soft error indicates that digital electronics' reliability is affected by cosmic rays, and increasingly so as the transistors grow smaller. Have any solid-state devices been built to specifically respond to cosmic-ray photons, so that noise in their interarrival times and/or landing positions (e.g. shot noise in the cosmic microwave background?) can be used as an entropy source for random-number generation (which I understand to be problematic in solid-state servers, absent the chaotic turbulence that a spinning hard drive induces or the human-induced jitter of local keyboard/mouse input)? NeonMerlin 04:41, 11 October 2017 (UTC)[reply]

It can be done, but if you are building your own it is a lot more convenient to use the americium-241 from a smoke detector as a source.
That being said, buying a bitbabbler white TRNG is a lot better and cheaper than anything you can build yourself. [ http://www.bitbabbler.org/ ]. --Guy Macon (talk) 05:55, 11 October 2017 (UTC)[reply]
For TRNG, read TRNG. --69.159.60.147 (talk) 10:28, 12 October 2017 (UTC)[reply]

Reconstructing TV

[edit]

Is it possible, in theory, to reverse-engineer a TV set and use it to reconstruct the info transmitted via the TV signal if you have zero prior knowledge of TV technology, simply from the characteristics of the signal? For example, imagine that some space aliens want to learn as much as possible about Earth from the radio signals they intercept -- would it be possible for them, given enough brainpower, to reverse-engineer a TV receiver and start pirating Star Trek? 2601:646:8E01:7E0B:0:0:0:EA04 (talk) 10:01, 11 October 2017 (UTC)[reply]

Yes. If you look at an analogue TV signal (far easier) you will see that it has a broad structure which repeats several times a second (one frame). Within this, there is a second, finer, structure that also repeats (one line) Analog television#Structure_of_a_video_signal. Assuming that a society which understands analogue radio signals has also invented raster scanning, it's not a great leap for these aliens to assume that this is what they're seeing. The synchronisation pulses are fairly obvious and it's not hard to construct a device to reconstruct the signal into an image. A similar approach was made deliberately for the SETI transmissions from Arecibo: if you broadcast a signal that has patterns within it as two prime numbers, a numerate culture is likely to recognise this as "unnatural" and to try examining them as a two dimensional pattern.
Colour is harder, because that's a sub-modulation within the overall brightness signal. But, if they can work to its higher frequency, they're likely to observe that there is some sort of signal there. Most of the obvious forms of modulation work the same on most planets, so again, they're likely to recognise what was used.
Rather harder is the problem of rendering the signal into the same sensory experience. We have trichromacy and developed TV to suit. The aliens might not, or might not have the same colour sensitivities. It would be difficult for them to match our red-green-blue experience onto their infra-red, ultra-violet and gamma radiation sensory pods. Only if we shared some "Rosetta stone" for this, such as a photograph of a rose, which they also knew from their flying saucer landings here, would they have a key as to how the encoded signals mapped to their senses. Or perhaps they'd decode an Open University program and be able to recognise a chemistry experiment involving iodine and its distinctive purple colour. Andy Dingley (talk) 10:24, 11 October 2017 (UTC)[reply]
Discounting them visiting us, in which case they could just grab a TV, I can't see a pic of a rose working as a Rosetta Stone, unless they happen to have them growing there. Try pics of nearby supernova remnants, etc. StuRat (talk) 17:52, 11 October 2017 (UTC)[reply]
I think it would be possible for them to reconstruct the color signals by adjusting the color values until bodies of water look more-or-less the same on the screen as they do on their planet (adjusted for the differences in their star's spectrum, of course), and then working from there. 2601:646:8E01:7E0B:0:0:0:EA04 (talk) 02:08, 12 October 2017 (UTC)[reply]


The question is very broad, but it breaks down into two sub-parts: (1) What technical details are needed to extract useful information about our technology? (2) How can we convey such details in a way that makes no assumption of prior knowledge about our technology, culture, language, biology, and so on?
This is a topic with a lot of history. Some important background reading is in order!
Our article on the Pioneer plaque links to Carl Sagan's 1972 publication, A Message from Earth, which discusses how an extra-terrestrial intelligence might be able to interpret our specially-crafted science and engineering diagrams. Our article on the Voyager Golden Record describes the coverplate, and links to the NASA summary of the diagrams. JPL maintains a website with more detail. Our article also links to more content and academic publications on the topic - there has been much written on whether a discoverer of these diagrams could meaningfully interpret them. The recurring theme is that we must reduce all the technical assumptions back to principles that are inherently deduced from fundamental properties of the universe. We can't assume that aliens know about steel, but we can expect them to know about magnetism; we can't expect aliens to know about transistors, but we can expect them to know about the field effect. We can't expect them to know about digitization, but we can expect them to know about entropy. We must reduce complex systems into a description that is referenced only to universal properties of energy, matter, and mathematical law.
It seems - to me - that an analog television from a few decades ago is easier to reverse-engineer than a modern digital television. An analog television has fewer parts; physical signals have simple mathematical correspondence to the information that they convey; and so on. Modern devices contain a lot more entropy and obfuscating pieces. Although it is difficult to make broadly-correct statements about the "reverse-engineering" capabilities of an alien intelligence, especially if we expand our thought-experiment to allow for technologies unimagined by Earth scientists, we can make a more restrictive claim: simple analog electronics and simple mechanical devices carry less entropy than elaborate microfabricated digital devices. Entropy, being a sort of universal constant, has some fascinating properties: on the one hand, it allows us to pack more information into a smaller volume - but perhaps a surprising consequence is that it makes our devices appear more random to an observer who has no prior understanding of our technology. High entropy entails statistical consequences: as entropy rises, it becomes exceedingly unlikely that a reverse-engineering effort can succeed, because it becomes statistically intractable to generate a correct mathematical model to deduce a device's physical inputs and outputs. This problem already exists today, with even the most simple electronic devices: the mathematics of the combinatorial explosion play out pretty unfavorably, when one must guess which voltage sequences to apply to a set of, say, 64 wires before all available energy in the entire universe runs down. This is, of course, assuming that the reverse-engineer knows what a "volt" is, and that it's something they need to apply to a "wire"!
Carried to the ultimate extreme, it is very conceivable that the "extra-terrestrial" discoverer of an ultra-modern microchip would not be able to provably distinguish it as a manufactured component created by intelligent human life. On inspection of its bulk physical properties, it looks very similar to an artifact of thermal noise or cosmogenic geology. A microchip might carry petabytes of digital data, plus a sophisticated digital circuit that (when appropriately powered and connected) can decode all that information into a useful format - and all this can fit into a package of mostly-silicon-dioxide-and-aluminum that is not much larger or more massive than any other granule of sand found floating in space as a chondritic asteroid. If an alien intelligence found it, they might not consider it likely to warrant study - not any more than the next granule of sand that impacts the alien spacecraft.
As our species progresses, and places more complexity in microscopic manifestations of high-entropy algorithms, it seems more and more implausible for anyone to reverse engineer. Even a skilled engineer with a laboratory full of equipment and detailed schematics can barely make sense of all parts of a modern VLSI device. And a human engineer knows how our stuff works, conceptually!
Broadly speaking, many SETI enthusiasts embrace the very plausible idea that this failure to recognize presents one of the more scientifically-grounded solutions to the Fermi paradox. In other words, it is conceivable that intelligent life fails to communicate with other intelligent life because advanced technology is indistinguishable from natural phenomena. We don't really expect to see an alien space-ship: it's much more likely that we'll see a dull infrared glow from a distant star. Reciprocally: we should not expect the space-aliens to ever find our televisions - let alone for them to study and reverse-engineer and operate them. In fact, we should not even expect them to receive our radio signals, let alone to interpret the information carried in them. The central half of Carl Sagan's book, Pale Blue Dot, outlines quantitatively and qualitatively why we cannot reasonably expect our radio or optical signatures to be meaningfully detected.
More likely, as much as it might be painfully nihilistic for us to admit, the only trace we shall leave on this universe is a faint metallicity signature in the solar spectrum of our sun. That spectral line would have been present in our Sun anyway, with or without the formation of planet Earth; with, or without, the development of a saline ocean suited for the evolution of the mammal; with, or without, the descent of humans toward the invention of the television and nuclear weaponry and moon rockets and fidget spinners.
If we are lucky, some intelligence elsewhere in our universe, distant from us in time and space, might evolve a way to look up at their sky and observe the color of one among zillions of little stars, and recognize that we might have existed, and created good televisions, maybe.
Nimur (talk) 16:16, 11 October 2017 (UTC)[reply]
One problem is that these aliens might have no concept of fiction, and not be able to even imagine why we would the broadcast "fake" information in Star Trek. They might interpret it as some type of military disinformation to scare off would-be attackers. StuRat (talk) 18:00, 11 October 2017 (UTC)[reply]
Galaxy Quest was not a documentary, I am sorry to let you know... --Jayron32 18:13, 11 October 2017 (UTC)[reply]
I didn't say that it was. Nonetheless, there's no basis to assume that aliens must always develop a penchant for telling fictional stories. StuRat (talk) 19:23, 11 October 2017 (UTC)[reply]
Or that they would understand the concept of "military". A good test are Dolphins/Whales and Octopuses[1]/Squid/Cuttlefish. They are highly intelligent and able to communicate with each other but we are not able to communicate with them. --Guy Macon (talk) 22:25, 11 October 2017 (UTC)[reply]
Group warfare is common in many communal species on Earth, such as army ants, where they have soldier ants, so that concept may be more common than the idea of telling fictional stories. We can communicate with many animals, such as trained dolphins and dogs, at least in a limited way. And it's quite possible that the vocalizations we don't understand as having any specific meaning really don't, they could just be an "I'm here and I claim this space" sound. If they repeat the same sound over and over, this makes it less likely to be some complex communication.
I believe there are examples of other species lying, such as a monkey species which will give a warning call to make others run for cover so it can eat their food, but this "crying wolf" isn't really the same as inventing an entire story. StuRat (talk) 22:37, 11 October 2017 (UTC)[reply]
You folks made my point exactly -- my hypothesis was that, having no previous knowledge about Earth or its inhabitants, the aliens would assume that all of what they see on our TV is mostly true (I call it -- you guessed it -- the "Galaxy Quest effect") (of course, they would probably allow for a certain amount of exaggeration for propaganda purposes, including mutually exclusive exaggeration by different warring nations, but they would not realize just how many of the stories are entirely fabricated). And that, in turn, would make them believe that the Earthling civilization is both extremely warlike (which is true) and a highly advanced spacefaring civilization (which is only partly true, because in fact we're only beginning to be a spacefaring civilization), and even that a select few Earthlings are wizards who have supernatural powers (which is baseless as far as modern science can tell), and that therefore any attempt to invade Earth would be an extremely risky endeavor and not worth it. 2601:646:8E01:7E0B:0:0:0:EA04 (talk) 02:18, 12 October 2017 (UTC)[reply]
StuRat hasn't made any point, or done anything to verify or refute any hypothesis. All he's done is ramble about what he thinks, as though the mere act of his brain producing a thought somehow makes that thought useful. As usual, he's provided no helpful way to answer the question beyond "Stu's brain thought this, so it must be true". You can always safely ignore anything StuRat says, it adds little in the way of actual reliable source material for you to read anyways. My only comment here was a subtle attempt to remind people of that. Subtlety, unfortunately, which was lost on Stu. --Jayron32 11:00, 12 October 2017 (UTC)[reply]
Actually you missed my implied point, that the aliens would be smart enough to know that the technology in our sci fi shows is impossible, so would therefore think it was some attempt at deception. Being unaware of fiction as entertainment, they may well take it as military disinformation. As for what Jayron says, he's also said he's always wrong, so I don't intend to pay any attention to it. StuRat (talk) 01:13, 14 October 2017 (UTC)[reply]
Perhaps -- unless parts of our science are wrong and this technology is somehow possible! 2601:646:8E01:7E0B:756C:F81D:F1A7:3FB4 (talk) 03:02, 14 October 2017 (UTC)[reply]
One area where they seem to always underestimate our technological advances is in hand weapons. Presumably they would have the ability to track our eyes and always hit whoever we look at when we pull the trigger, so there'd be no more of this shooting and missing back and forth for several minutes that makes for entertaining TV. Star Trek did once mention a "wide angle setting" for the phasors which could kill everyone in sight, but then rarely used it, as this would be too easy. In the end, about the only thing those futuristic hand weapons seem to have over guns is more shots before they run out, and, in some cases, smaller sizes. StuRat (talk) 03:14, 14 October 2017 (UTC)[reply]
Nimur - you remind me of the Fortress of Solitude scheme I had to store data in ice. If I recall "the most perfectly encoded signal is indistinguishable from noise" isn't actually Shannon's theorem but has something vaguely to do with it. Still, I'd think that whatever the entropy, eventually folks would get the notion to mount thrusters on a star, which is the kind of thing astronomers tend to notice. Wnt (talk) 19:11, 11 October 2017 (UTC)[reply]
I am pretty sure that the saying is "the most perfectly compressed signal is indistinguishable from noise". --Guy Macon (talk) 22:25, 11 October 2017 (UTC)[reply]
You can't mount thrusters on a star because they would melt from the heat! 2601:646:8E01:7E0B:0:0:0:EA04 (talk) 02:21, 12 October 2017 (UTC)[reply]
Unless you do it at night. ←Baseball Bugs What's up, Doc? carrots03:28, 12 October 2017 (UTC)[reply]
I'm thinking they'd make them out of magnetic fields, like prominences only with some more engineering. (Is there a field of magnetic engineering yet? I know some people make ball lightnings in microwaves and stuff, but I mean, is there something a little more elaborate?) But even known alloys come relatively close to solidity at surface of the sun temperatures; there may be room for conventional metals development. Wnt (talk) 13:01, 12 October 2017 (UTC)[reply]
If you wish to see a fictional portrayal of this, aliens reverse-engineering a TV signal is a key plot development in the 1997 film Contact, which includes scientist characters discussing the difficulties of the procedure. I seem to remember that there was discussion at the time that TV signals were not powerful enough to reach much beyond the Solar System, but I can't now find a source for that claim. Matt's talk 13:23, 14 October 2017 (UTC)[reply]