7
$\begingroup$

There have been a few attempts to transmit messages to the stars with the assumption that there may be extraterrestrial civilizations somewhere out there to receive them. These messages (such as the infamous Arecibo message) were constructed based on certain assumptions about the way a message could be structured so that it could be decoded and understood by intelligent beings on a technological par with us. The Arecibo message, for example, has been subsequently deemed flawed due to the inability of test subjects to decipher it.

In Carl Sagan's novel Contact (and its movie adaptation), there is a depiction of a message received from space, the structure of the message, and the effort to interpret it. It highlights some of the challenges of both constructing and interpreting a message where neither sender nor recipient knows anything about the other's language, culture, or level of science and technology.

The question is: if you are going to send a "message in a bottle" to the stars as a radio transmission, and you just want it to be the equivalent of "hello, we're here and we want to be friends and get to know you", how do you design and encode such a message so that a recipient on a technological par with us has a good chance at understanding it? What technical choices do you make? How do you construct it so that it can say something meaningful without requiring the understanding of any human language?

Addendum

Seems I need to add some further clarification to my question. I am not asking about choice of medium or spectrum (radio vs light, gamma pulse, whatever) but instead, given a suitable medium with which we can transmit a signal within which we have placed a constructed message, how do we arrange the information in that message so that a civilization with no prior knowledge of human language or culture could correctly interpret it? Further assuming it will be a stream of bits, how do we encode our message into bits? Given that the recipient has received our bit stream from their receiving apparatus, what steps would they have to perform to recover the intended meaning from that bit stream?

$\endgroup$
  • 1
    $\begingroup$ Too broad for me to tackle with an answer, but you might enjoy MJD's analysis of the "Cosmic Call" message: blog.plover.com/aliens/dd/intro.html $\endgroup$ – Russell Borogove Apr 5 '16 at 2:55
  • $\begingroup$ A number of people have constructed messages ostensibly from alien civilizations as puzzles, and working through them can be stimulating. puzzling.stackexchange.com/questions/17119/… $\endgroup$ – Russell Borogove Apr 5 '16 at 2:57
  • $\begingroup$ On May 16 there'll be a NASA Astrobiology lecture on the topic. We only know of a subset of biology, physics and math. "Same level" might never exist if the evolutionary paths are very different in this apparently endless combinatorics of variations. They might not have intelligence, but something much better which never happened to occur on Earth. $\endgroup$ – LocalFluff May 14 '17 at 13:06
  • 1
    $\begingroup$ @RussellBorogove Just read about the Dumas-Dutil message which, if I understand correctly, is doing three things: (1) telling the recipient things they are already expected to know, to facilitate learning of the language (alphabet, structure/syntax), (2) using that language to say something about us, and (3) asking about them in return. $\endgroup$ – Anthony X May 14 '17 at 17:49
4
$\begingroup$

I think this problem is often approached too explicitly, with the assumption that we have to explain all the basic concepts before someone can decode a complex message.

However, research into unsupervised deep learning of embedded features in images and text shows that an explicit set of concept definitions is not required to capture the meaning of the source material. Once you have machine-level representations of meaning, you can then translate to different formats, like image to text, or English to Spanish (http://arxiv.org/pdf/1309.4168.pdf).

With this in mind, deep neural networks are often trained on large corpuses of text, at minimum the contents of Wikipedia. So if you wanted teach aliens about us without requiring them to understand human language, you could transmit Wikipedia and be done with it, assuming that they have good computers.

If you only needed to say "we are here," then human language is not necessary, as shown in the Pioneer and Voyager line drawings. Perhaps someone else here can take a whack at helping aliens visualize a transmission containing a bitmap.

$\endgroup$
4
$\begingroup$

With lasers we could alter Earth's light curve to an artificial shape (maybe an Ulam spiral portion?), so that we would be visible to whatever star system we pointed our lasers at. The star system would have to be able to view our transit of the Sun. This is from a recent paper "A Cloaking Device for Transiting Planets" by David M. Kipping and Alex Teachey in the Monthly Notices of the Royal Astronomical Society. The authors propose cloaking the ingress and egress of the transit and leaving the main transit undistorted, since this is not produced by any known natural process.

It is actually cheaper to broadcast a signal this way than it is to cloak a planet, which was the main focus of the paper.

$\endgroup$
  • $\begingroup$ Cool paper, it must have been quite fun for them to write! But I don't understand how one might encode an Ulam spiral into a light curve. The spiral and its variants are 2D but the light curve is 1D. $\endgroup$ – uhoh Aug 13 at 23:02
  • 1
    $\begingroup$ @uhoh It's possible to represent 2D data in a 1D format, but in this case it would probably make more sense just to encode the series of prime numbers. Thanks for calling me out on that. $\endgroup$ – called2voyage Aug 13 at 23:16
  • $\begingroup$ How about pie? $\endgroup$ – uhoh Aug 13 at 23:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.