Slashdot Mirror
One Man's Quest To Build True Artificial Life
Atriune writes "The creator of the renowned Creatures artificial life series is at it again. Fifteen years after the initial success of the Creatures Trilogy, Steve Grand continues his quest to go beyond simulation, and create real artificial life."
It's hard to tell if the approach is realistic, but it is certainly novel. Perhaps this will succeed in the areas the Lisp hackers of the '80s failed.
Another way to put it is that he is a dreamer with a goal. If you ever played any of the games of the Creatures series (especially Creatures3 and, astonishingly, Creatures Village, both available at GOG.com), you probably know that the norns felt quite real and sometimes behaved in bafflingly intelligent-looking ways. So, although I don't think he'll reach his ideological goal to 100%, I'm looking forward to see how Grandroid turns out.
Wlil Grandroid be written in ApeScript? I wish he'd pour his Grandroid ideas into the Noble Ape open source project.
It features a number of autonomous simulation components including a landscape simulation, biological simulation, weather simulation, sentient creature (Noble Ape) simulation and a simple intelligent-agent scripting language (ApeScript).
The Noble Apes see their surroundings, have memories (internalizations of the external world), remember their encounters (almost drowning results in a fear of water) which leads emotion simulation & even relationships, can crudely communicate and breed with each other, and even dream!
If the Norns were "alive" in Creatures -- The apes are even more so, and a great deal closer to the goal of emergent sentience (esp. than to as yet nonexistent/unpublished Grandroid code).
What the Noble Apes don't have is a detailed graphical representation of themselves based on their genetics... I hoped that perhaps SPORE would help with this, but it was closed source.
The biggest failing of software (esp. some AI/AL simulations) is the desire to attempt to realize an idea or concept instead of truly collaborating with like-minded individuals and utilizing their works / ideas to realize a greater solution.
What I find interesting is machine learning through evolution -- Start with a VM filled with random noise & a few inputs. Devise a goal & selectively "breed" the instances that are closer to reaching the goals.
IMHO, Yeah, he's quite full of himself... so much so that he'll start a new project.
Meh, I'll bet it will be a fun game, regardless.