InterstellarNet began in musings about a far-future, star-spanning, human civilization. I presumed the pesky constraints of relativity: no FTL travel.
One thing led to another … soon I was pondering a comm network that functioned across the light-years. And, we homo saps being a tad competitiveabout interstellar cyber attacks.
I’m a computer guy. I started writing functional requirements for the network. (As Douglas Adams advised: Don’t panic! The specs do not appear in any InterstellarNet story.)
Human explorers will depart from our solar system with familiar languages, predefined protocols, and ready-made comm gear. NASA is already hard at work on an interplanetary Internet. It was more fun to imagine how an InterstellarNet might spontaneously ariseand thriveamong species that initially lack any common language, experience, or technology.
At this point, I should mention that I’m not only a computer guy. Besides computer science, there’s physics and an MBA in my shadowed past. From that MBA program comes background in economicsalso known as the dismal science.
Sometimes I play with physics, as in Moonstruck, originally an Analog serial. As often I play with computer science, as in my cyber-fiction anthology Creative Destruction (some of whose stories, including the title story, first appeared in Analog).
InterstellarNet lets me play with all three sciences.
I’m hardly the first SF writer to wonder how distant species might establish communications. Nor am I the first to suggest that the universality of physical laws would provide common ground. (H. Beam Piper’s “Omnilingual” did the latter way back in 1957in Analog’s precursor, Astounding.)
The InterstellarNet series opens in “Dangling Conversations.” Three different sciences suggested their own challenges. On the physics side: What, exactly, do math and science let us and ET say to each other? On the computer side: How, exactly, would a meaningful message be encoded between absolute strangers? On the economics side: Why would humans persist long enough to establish a dialogue with aliens light-years distant (or they with us)?
From its inception in “Dangling Conversations,” InterstellarNet was a commercial entity. Limited to primitive communications, that early commerce involved barter.
A generation later, the series moves on to “Creative Destruction.” (That’s the famous phrase of a largely obscure economist, Joseph Schumpeter. It refers to the often brutal efficiency with which markets reallocate capital from mature technologies to emerging ones. In an example from after Schumpeter’s time, think: mainframes, then minicomputers, then PCs, and now, perhaps Internet-centric applications.)
So … imagine we’ve learned to converse effectively (albeit slooowly) with our neighbors in other solar systems. They know things we don’t. They have wondrous capabilities like mature nanotechnology. To what lengths might some people go to obtain that technology, even after government bans its import because nanotech would be too disruptive or dangerous? Once everyone knows how to barter with the four-eyed, many-tentacled neighbors, who can stop the unscrupulous from violating such bans?
Time passes. Technologies converge as an increasingly rich trading language and the increasingly robust comm infrastructure accelerate everyone’s progress. Years-long Q&A becomes tiresome.
By “Hostile Takeover,” most InterstellarNet species have swapped trade representatives: artificially intelligent agents. This, to a computer guy, is getting really interesting. What’s to prevent us from stealing the intellectual property of aliens’ trade agents? What’s to stop us from hiding malicious softwarewe can all imagine things much nastier than mere virusesin the agent software we transmit to our neighbors?
Of course whatever plots we can hatchour neighbors can, too. Uh-oh.
As I said, I wrote functional specs. Herewith, a few of Lerner’s Laws for Artificially Intelligent Trade Agents:
1. Agents run only inside mutually agreed upon containment: the sandbox. The sandbox protects:
a. The secrets of the agent from the locals.
b. The local infosphere from the agent.
2. Sandbox code is fully disclosed and fully agreed upon across the interstellar community. (ETsone more argument for open source software!)
3. Access to/from the interior of a sandbox is only by messages.
4. An agent, its software entirely proprietary to its patron species, is transmitted encrypted across interstellar space.
a. It unwraps itself inside a sandbox provided by the host species.
b. It self-destructs, its secrets undisclosed, if the purported sandbox deviates in any way from expectations.
5. Trade waresintellectual propertytravel encrypted between solar systems, and are unwrapped in secrecy by the sequestered AI agent. Goods are sold (or not) and bought (or not) as the agent negotiates within its authorized-from-home parameters.
6. Agents buy and sell information using the host species’ banking system. Credits not spent locally may be transmitted, securely encrypted, between solar systems.
There’s (much) more to it, of course. Information must be exchanged with agents without breaking quarantine. Sandboxes must be extensible without any loss of privacy or security, as AI agents amass more and more knowledge about their host species. Agents need ways to back up their archives locally without risk of data theft. Distant species need ways to install an updated agent while retaining the obsolete agent’s memories. (In “Strange Bedfellows,” an InterstellarNet story that appeared in Artemis magazine, a crisis is precipitated when the Alpha Centauri trade rep to Earth objects to being replaced with a newer version.)
It’s a long list.
Isaac Asimov’s classic robot stories (many from the hallowed pages of Astounding) revolve around loopholes and ambiguities in the Three Laws of Robotics. Many InterstellarNet stories likewise explore ambiguities and loopholes in agent containment. In “Hostile Takeover,” our neighbors six light-years away at Barnard’s Star have some very clever ideas in that regard ….
“A New Order of Things” is the latest (and most ambitious to date) InterstellarNet tale. The good news is: Interstellar travel has finally become feasible. The starships are STLI’m not about to render InterstellarNet obsolete. The better news, if only for the reader, is: Things are not at all what they seem. Aliens, AIs, alien AIs … all with their own agendas … all in our solar system.
I hope to explore InterstellarNet for a long time to come.
(2010 update) Five InterstellarNet stories, many of which first appeared in Analog, are now expanded and novelized as InterstellarNet: Origins.
"One of the most original, believable, thoroughly thought-out, and utterly fascinating visions ever of what interstellar contact might really be like."
— Stanley Schmidt, Editor of Analog magazine
For more information, see this page on Edward M. Lerner's website.
Copyright © 2006 by Edward M. Lerner