Instead of sending flat messages into space, why not send an explorable environment?
This idea is in Extraterrestrial Languages, Daniel Oberhaus’ excellent history of attempts to talk to aliens (read last year).
e.g. there’s the famous Arecibo message, transmitted in the direction of Messier 13 in 1974 (the message will arrive in 22,000 years, by which time M13 may have moved out of the way). The message is a pixel grid, 73 by 23, which shows atomic numbers and a pictogram of a person.
Here’s a list of other interstellar messages, and they’re the same more-or-less: data with enough clues to say “hey try and decode me” with some fundamental information communicated as simply as possible. Who knows what alien intelligences might be like.
BUT:
Paul Fitzpatrick’s insight was that if you can send a message, you can send mathematical equations. And if you can send equations, you can send the rules of a programming language. And then you can send executable code. And then…
The idea behind Cosmic OS is that by beginning with simple math, it is possible to construct a programming language that can simulate an interactive virtual environment for an extraterrestrial intelligence. Such a rich environment would in principle allow the extraterrestrial to manipulate the program to get a better idea of the social and behavioral properties of the Earthlings who sent the message.
Here’s CosmicOS on GitHub. The code is open; it’s an ongoing project. (Cosmic OS hasn’t yet been sent into space.)
There’s a demo too. You can see the message, and run the code. There are a large number of statements, building up to abstract objects of “things” and “rooms” and “robos” (things that can move) and a few others.
Until eventually…
“New York” and “Boston” are connected, north and south, with an “autobus” that moves between them.
I mean, it’s basic.
But it shows the power of Fitzpatrick’s idea.
Instead of a description, which is what previous messages have been, an interactive environment - even a simple one - shows ontology, behaviour, and context. It allows the alien to build their own understanding of our world because they get to experience it, well, not exactly directly, but almost. It’s such a better way to transmit knowledge and understanding.
If we can transmit immersive environments to aliens, why not to each other?
Instead of sending a Powerpoint deck, why not a self-contained wiki? A packaged hypertext.
Instead of preparing a Google Doc, why not build a miniature explorable world? Not VR in photorealistic 3D, but a virtual reality of (mainly) text.
I would like to email a “file” which is a playable, navigable space of words, pictures, and embedded bots to have conversations with, at the end of which the recipient understands my ideas just as much as if I had used bulleted lists and diagrams. Their comments should come back to me as in-game questions that I can answer with environmental embellishments. This “world document” should be as easy to author, and as endlessly flexible, as a spreadsheet.
Instead of sending flat messages into space, why not send an explorable environment?
This idea is in Extraterrestrial Languages, Daniel Oberhaus’ excellent history of attempts to talk to aliens (read last year).
e.g. there’s the famous Arecibo message, transmitted in the direction of Messier 13 in 1974 (the message will arrive in 22,000 years, by which time M13 may have moved out of the way). The message is a pixel grid, 73 by 23, which shows atomic numbers and a pictogram of a person.
Here’s a list of other interstellar messages, and they’re the same more-or-less: data with enough clues to say “hey try and decode me” with some fundamental information communicated as simply as possible. Who knows what alien intelligences might be like.
BUT:
Paul Fitzpatrick’s insight was that if you can send a message, you can send mathematical equations. And if you can send equations, you can send the rules of a programming language. And then you can send executable code. And then…
Here’s CosmicOS on GitHub. The code is open; it’s an ongoing project. (Cosmic OS hasn’t yet been sent into space.)
There’s a demo too. You can see the message, and run the code. There are a large number of statements, building up to abstract objects of “things” and “rooms” and “robos” (things that can move) and a few others.
Until eventually…
“New York” and “Boston” are connected, north and south, with an “autobus” that moves between them.
I mean, it’s basic.
But it shows the power of Fitzpatrick’s idea.
Instead of a description, which is what previous messages have been, an interactive environment - even a simple one - shows ontology, behaviour, and context. It allows the alien to build their own understanding of our world because they get to experience it, well, not exactly directly, but almost. It’s such a better way to transmit knowledge and understanding.
If we can transmit immersive environments to aliens, why not to each other?
Instead of sending a Powerpoint deck, why not a self-contained wiki? A packaged hypertext.
Instead of preparing a Google Doc, why not build a miniature explorable world? Not VR in photorealistic 3D, but a virtual reality of (mainly) text.
I would like to email a “file” which is a playable, navigable space of words, pictures, and embedded bots to have conversations with, at the end of which the recipient understands my ideas just as much as if I had used bulleted lists and diagrams. Their comments should come back to me as in-game questions that I can answer with environmental embellishments. This “world document” should be as easy to author, and as endlessly flexible, as a spreadsheet.