Evolution of complexity/information

quote:

---

It is this distinction that took me away from conventional ideas about knowledge/information management to look at how information and knowledge is handled by the human brain. It then becomes apparent that knowledge is something quite different to information and is a unique experience for every different person.

---

It sounds to me like you guys are having a semantic difference.As I understand it, the Humanities distinguish between KNOWLEDGE and INFORMATION, in which information is the raw input, and knowledge the result of interpretation and processing.Information science makes exactly the same distinction except between DATA and INFORMATION.The commonality of the term information in two different senses trips people up a lot, it seems to me.

happy_atheist wrote:

quote:

---

It seems to me that there is a big difference between computer code and genetic code. There is a huge level of abstractness with computer code, which makes it highly independent of the operating environment. The computer code in no way dictates or creates the operating environment.

---

I'm afraid that is not as true as it appears at first glance. Take for example variable definition; the code has to carry out discrete actions to establish the appropriate circumstances for other actions. If a programme is invoked in an empty memory register there is literally nothing there to work with - everything HAS to be created by the code there and then.

quote:

---

Computer code does not create motherboards or anything else that would be constituted hardware. Computers do not get created simply from computer code.

---

False I'm afraid - we already have code that evolves subsequent code.

quote:

---

Scientific American, Programming with Primordial Ooze; October 1996; by Gibbs; 2 page(s)

​

Computer programmers ascended the economic food chain by inventing clever algorithms to make manufacturing and service laborers redundant. But some programmers may one day find themselves automated out of a job. In university labs, scientists are teaching computers how to write their own programs. Borrowing from the principles of natural selection, the researchers have built artificial ecosystems that, for a few problems at least, can evolve solutions better than any yet devised by humans. Someday such systems may even be able to design new kinds of computers.

---

I couldn't find the most recent article on these evolving structures; this one specifically dates back to '96, so maybe the one I'm thinking of was in another journal.

quote:

---

The code itself is entirely physical,

---

All computer code is physical, comprised of electromagnetic energy in particular states.[qupte] Also, since the processes involved in DNA is nothing more than a chemical reaction it is much more dynamic than computer code and a lot less arbitrary[/quote]Why do you see an inherent difference between a purely chemical reaction and a purely electromagnetic one? Why should chemicals be "more dynamic" and "less arbitrary" than electrons?We should not confuise the primitive sophistication of our technology with fundamental constraints. Evolution had billions of years to figure this out - we have had less than 80 of serious computing.

quote:

---

Firstly, computer code and DNA are quite different. Computer code is unable to directly act on on a natural environment while DNA, being a chemical, is able to directly interact with nature.

---

False - code pushes electrons; are they not part of the natural world?

quote:

---

Secondly, we do have random sequences of RNA that are useful. RNA nucleotides are allowed to randomly bind to one another, forming a random sequence. If RNA/DNA are like computer code, then we shouldn't see anything useful as a result. However, we do see chemical reactions occuring, some of which could eventually lead to a replicating reaction such as production of new RNA nucleotides and the manipulation of other RNA strands.

---

Data corruption can produce functional bugs by in a similar way. Agin I think you are falsely comparing levels of sophistication and then construing these as fundamental. There is no particualr reason to claim that random code cannot work.

Yes I think Percy has nailed the difference in perception.I fully agree that the relationship between the code and platform in computing is not the same as the relationship bewteen DNA and the body it 'designs'.But it is highly likely that we will, quite soon, with the developement of eveolving code, use it to develop both the theoretical parameters of future platforms, and possibly the specifications of the machines that will produce those platforms. This will be like a computer designing a womb which will give birth to another computer.And looking at it from that perspective the sheer awsomeness of biology becomes apparent.

quote:

---

DNA doesn't build organisms .... it builds proteins.

---

and, machine code doesn't draw pixels on a screen. Machine code puts command on a processor stack, and pops them off the stack, in a certain order. And those commands only direct the shuffling of bits which in turn shuffle electrons which eventually cause the cathode ray tube to fire yet another electron at a phosphor dot painted on the glass screen you actually look at.

quote:

---

Chemical interactions in vast nested, cyclic, complex systems
result in organisms.

---

Rather like nested procedures, eh?

quote:

---

All analgies to communication systems are very poor in this context.

---

Not on the basis you gave. Please note I have not asserted an exact identity between computer code and DNA, but it seems to me they are very very similar.

quote:

---

Neat how computer code can physically affect the environment (RAM, I presume, I'm computer literate in a limited way). Thanks for pointing that out.

---

Well, yes RAM but also all other forms of memory. That is, all memory appears as vast numbers of electric switches that can be on or off. But there is nothing essential about this mechanical structure except technical utility; in principle, you could build a functional computer out of a big enough network of canals and locks. Yes, out of mud and water. Because the key is not what the material substrate of the device is - but how it is organised. We only use silicon and electricity because a usable machine built out of canals would probably need more surface area than the planet has and would execute operations in decades not nanoseconds.

quote:

---

I think computer code is not independent of the computer. It can certainly be formulated outside the computer, but it can work ONLY in the computer. But, I may have misunderstood happy on this point.

---

I think thats perfectly valid to describe the present state of programming. But as above, the computer itself is just a mechanical medium on which the code executes. There is no reason in principle that it could not execute on a substrate that was essentially biological, even if this is beyond our means at present.

quote:

---

...just seems to support ID all the more, to me. Just a thought there.

---

Well, I sort of see where you are coming from, but it comes down to assumed premises. Because of course to me, if a human is "teaching" a computer, then this is really just one computer is teaching another.What I am getting at is that IMO we have enough knowledge now to draw direct analogies between information processing as we know it and some of the complex things that happen in biology. In 1950 what DNA did had little parallel in any scientific realm and was nearly magical. Now we can look at how a very small representation, organised in the right way, can create a cascade of cause and effect that quite independantly create complex outcomes. Having developed "instructional coding" outselves, we can make sense of DNA much more easily than we ever did before. Thus to me, it actually undermines ID, because one more element of the magic of biology is now understood. No divine instructor or blueprint is required after all.