A few weeks ago someone had posted a thread regarding self-replication. There was a demo program which looked a lot like Conway's game of life, but was geared toward artificial chemistry. The goal was to work toward discovering self-replicating particles inside this simulator. I can't seem to find the thread, now, but I found the following article which is perfect for that poster. I hope others also find it useful. -Kirk Nature 437, 636 (29 September 2005) | doi: 10.1038/437636a Robotics: Self-replication from random parts Saul Griffith1, Dan Goldwater1 and Joseph M. Jacobson1 Top of pageAutonomously self-replicating machines have long caught the imagination1, 2, 3 but have yet to acquire the sophistication of biological systems, which assemble structures from disordered building blocks. Here we describe the autonomous self-replication of a reconfigurable string of parts from randomly positioned input components. Such components, if suitably miniaturized and mass-produced, could constitute self-fabricating systems whose assembly is brought about by the parts themselves. http://www.nature.com/nature/journal/v437/n7059/abs/437636a.html;jsessionid=EBD15BE3894C48B0047C3BA664133F96

Thanks. That looks a lot like something I would read... if I had access to the article. But I guess that means I'd have to pay? =(


By the way, I've also been working recently on self-replicating scripts. They are similar to BASIC. The scripts read and copy their own commands line-by-line. The line-copy routine occassionally mutates and causes changes to the arguments or the command itself, creating new functionality.

Is there anything similar (besides CoreWars or writing full-blown self-replicating viruses in assembler)?

Quote: Original post by crimson fury
Is there anything similar (besides CoreWars or writing full-blown self-replicating viruses in assembler)?

Yeah it's called Lisp. The most self-mutatable language around.

And this is my pi daily messages per-day so I must be done.

You can probably find the article in just about any library. I would be amazed if any library did not have Nature. The article itself is only 1 page, and the more interesting part is the supplementary info which you can access from the link on the right of the page I pasted in my first message. From supplementary info you can download a movie of the whole process in action. Nice.

As for mutable languages, I think there are papers that go back to von Neumann and perhaps even further. Search Google Scholar or Citeseer for all the links in the world. And, the articles in Citeseer are free pdf or ps files.

-Kirk