• Post Reply Bookmark Topic Watch Topic
  • New Topic
programming forums Java Mobile Certification Databases Caching Books Engineering Micro Controllers OS Languages Paradigms IDEs Build Tools Frameworks Application Servers Open Source This Site Careers Other all forums
this forum made possible by our volunteer staff, including ...
Marshals:
  • Campbell Ritchie
  • Liutauras Vilda
  • Tim Cooke
  • Jeanne Boyarsky
  • Bear Bibeault
Sheriffs:
  • Knute Snortum
  • paul wheaton
  • Devaka Cooray
Saloon Keepers:
  • Tim Moores
  • Stephan van Hulst
  • Ron McLeod
  • Piet Souris
  • Ganesh Patekar
Bartenders:
  • Tim Holloway
  • Carey Brown
  • salvin francis

Genes for Machines

 
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
An Interesting idea how seriously should one take it , especially with regard to the comment about ANT and MAKE :
(Continuing the discussions on making Systems less complex).

Genes for Machines
=========================
Originally published February 23, 2003 at just.osono.com and June 19th, 2002 at just.osono.com
A Genome, from dictionary.com:
------------------------------
1. The total genetic content contained in a haploid set of chromosomes in eukaryotes, in a single chromosome in bacteria, or in the DNA or RNA of viruses.
2. An organism's genetic material.
A Machinenome, from Osono:
---------------------------
1. An encoded digital file that contains the "instructions" to build an application, operating system, or device.
2. A digital organism's genetic material.
onward....
A Genome, from dictionary.com:
1. The total genetic content contained in a haploid set of chromosomes in eukaryotes, in a single chromosome in bacteria, or in the DNA or RNA of viruses.
2. An organism's genetic material.
A Machinenome, from Osono:
1. An encoded digital file that contains the "instructions" to build an application, operating system, or device.
2. A digital organism's genetic material.
onward....
BACKGROUND
The following is from: http://www.ornl.gov/hgmis/posters/chromosome/faqs.html
What is a chromosome?
Chromosomes are tightly coiled microscopic structures made up mainly of DNA, which consists of four different building blocks called bases (A, T, C, G). The four bases are repeated millions of times to form each chromosome.
How big are human chromosomes?
Human chromosomes range in length from 50 million to 263 million bases. With few exceptions (e.g., red blood cells), each of the trillions of cells in the human body contains a complete set of chromosomes--the genome. If all the bases in the human genome were spread out 1 millimeter apart, they would extend from Memphis to Los Angeles.
How many chromosomes are in the human genome?
The nucleus of most human cells contains two sets of chromosomes, one set given by each parent. Each set has 23 single chromosomes--22 autosomes and an X or Y sex chromosome. (A normal female will have a pair of X chromosomes; a male will have an X and Y pair.)
What are the sizes of the individual human chromosomes and how many genes are estimated to be located on each?
Chromosome # Genes # of Bases
Chromosome 1 2968 279 million bases
Chromosome 2 2288 251 million bases
Chromosome 3 2032 221 million bases
Chromosome 4 1297 197 million bases
Chromosome 5 1643 198 million bases
Chromosome 6 1963 176 million bases
Chromosome 7 1443 163 million bases
Chromosome 8 1127 148 million bases
Chromosome 9 1299 140 million bases
Chromosome 10 1440 143 million bases
Chromosome 11 2093 148 million bases
Chromosome 12 1652 142 million bases
Chromosome 13 748 118 million bases
Chromosome 14 1098 107 million bases
Chromosome 15 1122 100 million bases
Chromosome 16 1098 104 million bases
Chromosome 17 1576 88 million bases
Chromosome 18 766 86 million bases
Chromosome 19 1454 72 million bases
Chromosome 20 927 66 million bases
Chromosome 21 303 45 million bases
Chromosome 22 288 48 million bases
Chromosome X 1184 163 million bases
Chromosome Y 231 51 million bases
What are genes? What role do they play in disease research?
Genes are chromosome pieces whose particular bases (e.g., ATTCGGA) determine how, when, and where the body makes each of the many thousands of different proteins required for life. Humans have an estimated 30,000 genes, with an average length of about 3,000 bases. Genes make up only around 3 percent of human DNA; the remaining DNA has important but still unknown functions that may include regulating genes and maintaining the chromosome structure.
Researchers hunt for disease-associated genes by looking for base changes found only in the DNA of affected individuals. Numerous disorders and traits mapped to particular chromosomes are displayed in this Website. Some disorders, such as cystic fibrosis (chromosome 7) and sickle cell anemia (chromosome 11) are caused by base sequence changes in a single gene. Many common diseases such as diabetes, hypertension, deafness, and cancers have more complex causes that may be a combination of sequence variations in several genes on different chromosomes, in addition to environmental factors.��
MachineGenes
All life on earth is built from the DNA sequences within its��� Genome. These sequences have evolved over the last 4 billion years on earth, and are successful because one set of chromosomes comes from a mother and one set comes from a father, and the environment introduces random mutations to genes. The result is a constantly evolving planet of organisms, all changing to suit the requirements of their environment.
If we want machines as complex as even a pepper seed, we will need to adopt a similar mechanism when constructing hardware and software systems.
The typical Human Genome consists of 3 billion bases (A, T, C, and G). In terms of digital storage, each million bases of a DNA sequence take up one megabyte of data storage, and this equates to 3 Gigabytes for the entire Genome (yes, somebody has already put their Genome on an iPOD ��� how rad is that?). But, you could start small with digital genomes. In fact, before you even start small, you need to figure out how you are going to do it.
Since computers are digital, they use a sequence of zeros and ones to describe very complex data (everything, in fact). This method has proved incredibly successful, and you probably would not want to change anything at this point (like, say, completely mimicking the genome by using base 4 instead of base 2). The point is to use the same methods and concepts, not to build airplanes with flapping wings. We are looking for the emergent behavior of gene expression and evolution, not a simulation of a biological process.
So, do you use a binary file? With a binary file, you would need a ���Rosetta Stone���, or some kind of program or agent to read it, and you would probably need different versions for different operating systems. The agent or reading program would also have to self extract itself from the digital genome before it could read it. I guess all this would be really cool, but it starts to get messy.
Do you use a text file, such as an XML file? The overhead of the ascii text format would be large, but at least you could always read it manually on just about any system. Presumably, everything will know how to handle XML in the next year or so, so you wouldn���t have to worry about it. Maybe, you combine the two files: a binary sequence wrapped in XML.
Is this like ANT or MAKE? Build systems are similar in concept, but when talking about a digital genome, we are considering a file that can describe, in one sequence, how to build an entire application from nothing, on any machine, plus change and mutate over time to optimize itself for whatever environment it finds itself (a linux server, a windows server, a PDA, a cell phone, a switch, a watch). Like a new baby, at the end you would have a fully functioning ���framework for work, adaptation, and learning��� with a bunch of ���instincts���, and a ravenous appetite for learning about its environment, for gathering resources it may need (memory, processor time, apache, storage space, etc.), and optimizing itself though evolutionary programming.
So, if you start from scratch, and use base 2, and decide to use a binary file wrapped in XML, and to include a self extracting agent; what does it actually do? What does this file create?
You create a piece of software or hardware in a default state. This default state includes the ability to sense the environment, to repair itself, and to learn and aquire knowledge over time. Sensors would be network interfaces, packets, memory registers, storage space, etc. For repair, you have the original machinenome, which is used as a blueprint in case anything gets broken- repairs are made by rebuilding the application to default state from the original encoded machinenome information. Learning is in the form of a "memory", which in a software system turns out to be a distributed storage mechanism with associative properties.
Posted by thrawn at 03:18 PM
 
(instanceof Sidekick)
Posts: 8791
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
There was a show on TV last night about a guy in England who bought a $1m super-computer (claimed roughly == 10,000 P4s) for the purpose of building mutating genetic programs that would evolve into machine intelligence. He didn't get much past "hello world" before the documentary went to print.
 
HS Thomas
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Stan, what was his name ? Lord Sainsbury , the Science Minister ?.
Found this article :

Genetic Algorithms AND AI
It would seem it's far away in the future.
[ December 03, 2003: Message edited by: HS Thomas ]
 
I suggest huckleberry pie. But the only thing on the gluten free menu is this tiny ad:
create, convert, edit or print DOC and DOCX in Java
https://products.aspose.com/words/java
  • Post Reply Bookmark Topic Watch Topic
  • New Topic
Boost this thread!