Understanding through Discussion


Welcome! You are not logged in. [ Login ]
EvC Forum active members: 66 (9049 total)
64 online now:
Tangle (1 member, 63 visitors)
Newest Member: Wes johnson
Upcoming Birthdays: DrJones*
Post Volume: Total: 887,622 Year: 5,268/14,102 Month: 189/677 Week: 48/26 Day: 0/11 Hour: 0/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Design without a designer?
IrishRockhound
Member (Idle past 3458 days)
Posts: 569
From: Ireland
Joined: 05-19-2003


Message 1 of 14 (58727)
09-30-2003 9:55 AM


I posted this on christiansunite.com a while ago, and I thought I'd toss it in here just for comparison...

Computer programs have shown that it is possible for apparently designed objects to appear that were not explicitly designed by the programmer. This is seen very clearly in the "Langton's Ant" program.

Here's the link: http://fvdp.homestead.com/files/ant_index.html

The universe in which the ant operates, and the movement of the ant itself, follow incredibly simple rules. These rules are set by the programmer (-> intelligently designed?). After many thousands of steps, however, the movement of the ant begins to construct a highway that was NOT set or included in any way by the programmer; it is a result of the ant randomly discovering a successfully self-replicating pattern.

In my opinion, we can consider our universe to be a very large program with very complex rules that has organised itself in a similar manner to Langton's Ant because it has been running for so many billion years. Perhaps God is just another programmer? In any case, I also think that this is good proof that abiogenesis can occur, and that evolution is not quite so unbelievable as some people imagine it to be. It's possible that it's all a result of the universe becoming more and more organised.

Before anyone decides to start flaming me, remember I said it's just my opinion ok? And I'd like to know what everyone else thinks.

The Rock Hound

{A note from Adminnemooseus - Also available is the topic Design With No Designer. It was started by MrHambre, currently has 47 messages, and was last posted to on 9/24/03}

[This message has been edited by Adminnemooseus, 09-30-2003]


Replies to this message:
 Message 2 by Rei, posted 09-30-2003 1:33 PM IrishRockhound has not yet responded
 Message 3 by Warren, posted 09-30-2003 3:16 PM IrishRockhound has responded
 Message 6 by MrHambre, posted 10-01-2003 1:39 PM IrishRockhound has not yet responded

  
Rei
Member (Idle past 6035 days)
Posts: 1546
From: Iowa City, IA
Joined: 09-03-2003


Message 2 of 14 (58773)
09-30-2003 1:33 PM
Reply to: Message 1 by IrishRockhound
09-30-2003 9:55 AM


I played with this algorithm a lot when I was younger, and had a lot of variants. I even made a psychodelic DOS screen saver based on it (it had 256 levels which the ants could alter, not 2; it then did smooth palette rotation on the results). It is an excellent example of complexity coming from a simple set of rules (far simpler than the set of rules our universe operates in). Not quite as enthralling as the Mandelbrot set, but the complexity that it gets does is really neat

It inspired a completely different fractal "art generator" by me which I called "roaches", that had reproducing ants that had mutable rulesets that ate away the screen (which was steadily refreshed). Of course, that was back when I was just programming in Qbasic (7th grade?).

------------------
"Illuminant light,
illuminate me."


This message is a reply to:
 Message 1 by IrishRockhound, posted 09-30-2003 9:55 AM IrishRockhound has not yet responded

  
Warren
Inactive Member


Message 3 of 14 (58794)
09-30-2003 3:16 PM
Reply to: Message 1 by IrishRockhound
09-30-2003 9:55 AM


I am not a programmer of genetic algorithms, but from what I’ve seen of them, they are tuned to seek adaptive change that is most certainly non-random with regard to the reason(s) for adaptation. You can of course claim that this is a precise modeling of natural evolution in biological life forms. Which would then tend to falsify the Darwinian assertion that adaptive mutations are random with regard to fitness.

This message is a reply to:
 Message 1 by IrishRockhound, posted 09-30-2003 9:55 AM IrishRockhound has responded

Replies to this message:
 Message 4 by Rei, posted 09-30-2003 3:26 PM Warren has not yet responded
 Message 5 by Quetzal, posted 10-01-2003 2:41 AM Warren has not yet responded
 Message 9 by IrishRockhound, posted 10-02-2003 11:59 AM Warren has not yet responded

  
Rei
Member (Idle past 6035 days)
Posts: 1546
From: Iowa City, IA
Joined: 09-03-2003


Message 4 of 14 (58796)
09-30-2003 3:26 PM
Reply to: Message 3 by Warren
09-30-2003 3:16 PM


Warren, what have you seen of genetic algorithms? What leads to your assertion that they are "tuned to seek adaptive change that is most certainly non-random with regard to the reasons for adaptation"? Give an example.

BTW, this is in no way a genetic algorithm that was discussed here. It is simply a case of complexity from simplicity.

------------------
"Illuminant light,
illuminate me."


This message is a reply to:
 Message 3 by Warren, posted 09-30-2003 3:16 PM Warren has not yet responded

  
Quetzal
Member (Idle past 4894 days)
Posts: 3228
Joined: 01-09-2002


Message 5 of 14 (58915)
10-01-2003 2:41 AM
Reply to: Message 3 by Warren
09-30-2003 3:16 PM


Warren,

You appear to be erecting a strawman. One which, btw, has been addressed in numerous threads already. Mutations are random with regards to fitness. It is only by considering the phenotypical effects (if any) on the organism in the context of its environment that a given mutation can be, post facto, determined to be "adaptive". In other words, you're wrong again.

Besides, that's not what is being discussed on this thread - these are software demonstrations of complexity from simplicity, not mutation and selection.


This message is a reply to:
 Message 3 by Warren, posted 09-30-2003 3:16 PM Warren has not yet responded

  
MrHambre
Member (Idle past 415 days)
Posts: 1495
From: Framingham, MA, USA
Joined: 06-23-2003


Message 6 of 14 (58976)
10-01-2003 1:39 PM
Reply to: Message 1 by IrishRockhound
09-30-2003 9:55 AM


Even Conway's Game of Life is a simple example of three dumb rules generating 'design'. The floaters and other strange characters that appear are not the planned outcome of intelligent design, only the algorithm working itself out over and over.

Daniel Dennett has pointed out that Darwin's genius lay in his discovery of certain sets of biological algorithms. The repetition of these processes create the phenomena we see today. If someone doesn't accept that apparent design can be created by non-goal-oriented processes, they're denying what they can see with their own eyes. Not that anyone here would do that.

------------------
I would not let the chickens cross the antidote road because I was already hospitlized for trying to say this!-Brad McFall


This message is a reply to:
 Message 1 by IrishRockhound, posted 09-30-2003 9:55 AM IrishRockhound has not yet responded

Replies to this message:
 Message 7 by Rei, posted 10-01-2003 2:06 PM MrHambre has not yet responded

  
Rei
Member (Idle past 6035 days)
Posts: 1546
From: Iowa City, IA
Joined: 09-03-2003


Message 7 of 14 (58977)
10-01-2003 2:06 PM
Reply to: Message 6 by MrHambre
10-01-2003 1:39 PM


I think one of the simplest and most beautiful examples of this is the Mandelbrot set. Such a simple algorithm. You take a graph of the complex plane near the center, having your x-axis be the real component and your y-axis be the imaginary component. You pick a point on the imaginary plane - it doesn't matter where, it'll get you pretty, complex results no matter where you pick. Then, for each complex point you want the color of, you do the following algorithm:

Given a point to plot Z, and the initial point you picked C, run:

Z = Z^2 + C

This algorithm will either converge on zero or on infinity. If you set a threshold for where to decide that it is going off to infinity, then you can look at how fast it is converging to infinity by how many steps it takes to get to that threshold.

The mandelbrot set is the plot of the number of steps it takes to get past that threshold. That's it. For all it's beautiful complexity, it is such a simple algorithm.

Simple rules in themselves create simple results. But iterative simple rules can produce amazing complexity.


This message is a reply to:
 Message 6 by MrHambre, posted 10-01-2003 1:39 PM MrHambre has not yet responded

  
Rrhain
Member (Idle past 895 days)
Posts: 6349
From: San Diego, CA, USA
Joined: 05-03-2003


Message 8 of 14 (59023)
10-01-2003 11:15 PM


I'm reminded of an assignment given in a computer science class:

Write a program that attempts to find a knight's tour of a chessboard. For bonus points, determine if the tour is cylical.

That was about it. For those who don't know, a knight's tour is where you take a knight and, using only the moves allowable by a knight, try to traverse the entire chessboard, landing on each square only once. It's "cyclical" if the last square is a knight's move from the first square.

How to go about this? Well, you first need to model the chessboard. That's simple...an 8x8 array. But then we notice that not all moves are possible on all squares. From the corner squares, there are only two possible moves.

What I noticed was that if you had moved to a corner, you actually only had one possible move left since the other one was the square you just came from which you could no longer land on. So instead of allowing all possible moves from the corner, you should restrict each space to have only legal moves available. But along those lines, I got to thinking if there was a "favorable" direction to move in general. Might certain types of moves be more likely to arrive at a knight's tour than others? Perhaps there was a way to restrict the possible moves for each square....

And after a little bit of playing around with a real knight and chessboard, I found a cyclical knight's tour. And thus, I programmed each square with only two possible moves. Thus, no matter what square you started on, you were guaranteed to get a cyclical knight's tour.

But the thing is, it was always the same tour. The only difference between one square and the next would be whether it started going in one direction or the other.

Since the assignment didn't say how to accomplish the finding of the tour, my program received an A, but it sorta missed the point. The program wasn't really finding the tour.

Some other people actually gave their algorithms a memory. That is, the knight would just start moving around the board, keeping track of the moves, so that if the knight got stuck and yet still not have found a tour, it could backtrack until it found a new direction it hadn't gone and try again.

Now, this isn't quite enough because, depending on how the algorithm searched for moves, you could get the same tour for any given starting point. That is, starting square A would give you a different tour from square B, but you always get the same tour when you start from A. This would happen if each square listed the 8 possible moves in the same order and it always chose the same direction to move first, second, etc.

There is, of course, a way to fix this: Randomize the direction of the next move. Keep track of which moves resulted in failure, but when you get back to a square with an option, don't "rotate one move clockwise" but rather "choose a random direction not tried."

A program of that final type could give you different tours, even when starting on the same square.

So do we see how simple rules can come up with complex patterns? How the claim of "the computer was programmed with the answer" is silly? Yeah, my version was me programming the computer with the answer, but that isn't the only way to do it.

------------------
Rrhain
WWJD? JWRTFM!


  
IrishRockhound
Member (Idle past 3458 days)
Posts: 569
From: Ireland
Joined: 05-19-2003


Message 9 of 14 (59107)
10-02-2003 11:59 AM
Reply to: Message 3 by Warren
09-30-2003 3:16 PM


Sorry I didn't get back sooner... Warren, your assertation is meaningless here. The ant hasn't been 'tuned' to seek anything. All it has is a simple set of rules for movement on a simple plane. My point here is that we see an ordered structure - the highway - arising from apparent disorder in a simple system, that was NOT, repeat NOT, designed by the programmer. I would assert that this is resonable proof that abiogenesis and evolution are possible, if we consider the universe to be a very complex program running with very complex rules.

Leave out the 'mutation' crap here. It's not relevent.

The Rock Hound


This message is a reply to:
 Message 3 by Warren, posted 09-30-2003 3:16 PM Warren has not yet responded

  
IrishRockhound
Member (Idle past 3458 days)
Posts: 569
From: Ireland
Joined: 05-19-2003


Message 10 of 14 (59770)
10-06-2003 2:03 PM


Just wanted to bump this topic - I really want to discuss how many programming errors God made...

The Rock Hound


  
Mike Doran
Inactive Member


Message 11 of 14 (59947)
10-07-2003 2:12 PM


Feedback loop systems are used in computer programming as well as to understand the functioning of a biological system. Rules involved here that leads to some of the paradoxes express IMHO are climate regulation feedback related. That is what I am discussing in the living earth thread. I have more comments in that thread I will post in a minute.

Replies to this message:
 Message 12 by IrishRockhound, posted 10-07-2003 2:27 PM Mike Doran has not yet responded

  
IrishRockhound
Member (Idle past 3458 days)
Posts: 569
From: Ireland
Joined: 05-19-2003


Message 12 of 14 (59951)
10-07-2003 2:27 PM
Reply to: Message 11 by Mike Doran
10-07-2003 2:12 PM


That's a bit off topic - I'm simply proposing that the similarity between the Langton's Ant program and the universe may be evidence for abiogenesis and evolution. I've read your thread on the Living Earth and I have to say that most of it is completely beyond my area of expertise (I'm a geologist).

Feel free to stick around here and post your opinion.

The Rock Hound


This message is a reply to:
 Message 11 by Mike Doran, posted 10-07-2003 2:12 PM Mike Doran has not yet responded

Replies to this message:
 Message 13 by NosyNed, posted 10-07-2003 6:45 PM IrishRockhound has responded

  
NosyNed
Member
Posts: 8965
From: Canada
Joined: 04-04-2003


Message 13 of 14 (59991)
10-07-2003 6:45 PM
Reply to: Message 12 by IrishRockhound
10-07-2003 2:27 PM


evidence for

I don't see any of this as really evidence for anything (not directly anyway). What it is though, is strong evidence against any claims that complexity can not arise from simple processes. That evolutionary mechanisms can not supply the "complexity" (whatever that is) needed is refuted.

So far I don't see anything but handwaving in refutation of this.


This message is a reply to:
 Message 12 by IrishRockhound, posted 10-07-2003 2:27 PM IrishRockhound has responded

Replies to this message:
 Message 14 by IrishRockhound, posted 10-08-2003 8:54 AM NosyNed has not yet responded

  
IrishRockhound
Member (Idle past 3458 days)
Posts: 569
From: Ireland
Joined: 05-19-2003


Message 14 of 14 (60066)
10-08-2003 8:54 AM
Reply to: Message 13 by NosyNed
10-07-2003 6:45 PM


Well, I think maybe I phrased that wrong. No, it's not evidence for abiogenesis and evolution - rather that it is my opinion that it makes a strong case for the possibility of such things happening (which is exactly what you said ).

I wonder why no creationists want in on this? I would have thought that a few would be here by now, telling me that I'm mad and chattering about irreducible complexity or something...

The Rock Hound


This message is a reply to:
 Message 13 by NosyNed, posted 10-07-2003 6:45 PM NosyNed has not yet responded

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2018 by EvC Forum, All Rights Reserved

™ Version 4.0 Beta
Innovative software from Qwixotic © 2021