Joined: Sep. 2002
|...one thing that the story mentions that I hadn't really seen mentioned elsewhere (including the paper IIRC, so I'm a bit suspicious) is that the building up of the functions sometimes "burned its bridges", i.e. eventually no trace of some intermediates can be found in the final program. Would this be an instance of scaffolding? It would be cool if so...|
I'll have to trudge back through the paper again, and probably through a lineage or two or three, to see if that was the case. I know that the EETimes article had several serious errors in its description of the research, so much so that I emailed Johnson with them, and he asked if they were serious enough to warrant a Letter to the Editor with the corrections. I said they were and did so - actually, I told him to use my original email as the correction letter. This is the email I sent:
| Dear Mr. Johnson,|
It was good to see your Avida story in EETimes. However, there are a couple of misconceptions in it that distort the actual research.
First, the story says "The organism's metabolism consists in the endless execution of the sequence of instructions. Energy from the environment, or "food," is modeled as single-instruction processors (SIPs) that are fed to the CPU. The number of SIPs that a CPU receives is proportional to the length of its tape. Thus, as the CPU becomes more complex in terms of the length of its instruction tape, it is able to get more food from the environment, giving more-complex organisms a competitive advantage."
"SIPs" are "Single Instruction Processing" units, not "processors." A SIP is a quantum - unit - of 'energy' that allows processing one instruction. It is not additional processors. And providing SIPs in proportion to length does not give longer organisms a competitive advantage. It actually neutralizes genome length as a selective variable.
Second, the story says "SIPs introduce new instructions to the CPU, allowing it to grow as well as to reproduce." Nope. SIPs have nothing at all to do with introducing new instructions. That was the role of mutations.
Third, the story says "The researchers performed evolutionary runs starting with individuals that could replicate themselves but could not perform any logic operations except the simple NAND."
In fact, the initial organisms (Ancestors) could perform NO logic operations, not even NAND. NAND was in the instruction set available to be inserted by mutations, but was not in the initial organisms. Further, the primitive NAND instruction could not by itself perform NAND in the context of a critter's program: It had to be appropriately embedded in a context of other instructions that gave it access to registers and I/O. By itself, NAND in a critter's genome just sits there.
"There are only two ways we know of to make extremely complicated things, one is by engineering, and the other is evolution. And of the two, evolution will make the more complex." - Danny Hillis.