Home
Was
Darwin
Wrong?
Home | Intro | About | Feedback | Prev | Next

Similarities and Dissimilarities of Computer Viruses
and Biological Viruses.

Gert Korthof, 2 Feb 2006 (updated: 28 May 2015)

Overlapping but non-identical sets of units of evolution and units of life
Overlapping but non-identical sets of units of evolution and units of life.
(figure slightly modified from The Principles of Life).
Prions and Alife (Artificial Life) could be added to viruses and memes
because they are evolving but not alive.


#COMPUTER VIRUSESBIOLOGICAL VIRUSES
s i m i l a r i t i e s
1infection of specific targets (.exe or .com files)infection of specific targets (host cells)
2attach to .exe or .com filesintegrate in DNA
3spread to other computersspread to other hosts
4parasitism: copied by hostcopied by host cell
5one virus per fileno re-infection of same cell
6initially infected file is functional initially infected cell is functional
7user does not immediately notice infectionhost organism does not immediately notice infection
8software can be made immune to infectionnot every cell is infected
9specificity for Operating Systemhost specificity
10species (kinds) of virusesspecies (families) of viruses
11degrees of harmfulnessdifferent degrees of virulence
12difference in susceptibility of computersdifference in susceptibility of individuals and species
13antivirus software on computerimmune system
14percentage of computers protected by anti-virus softwarepercentage of individuals in population immune to virus (vaccinated)
15PC's came first, viruses laterhost organism evovled prior to infecting virus
16contain information, have length expressed in b(ytes)contain information, have length expressed in b(ases)
17source code causes behaviour of virusgenotype causes phenotype including behaviour
18virus has small size relative to host softwaresmall genome relative to host genome
19not living according to Ganti definitionnot living according to Ganti definition
p o t e n t i a l    s i m i l a r i t i e s
1mutating virus virus mutates
2activation of virus depends on dateseasonal activity of virus
3software version dependent actionage dependent action of virus
4virus infects new host softwareinfection of new host species
5anti-virus software introducedevolution of immune system
6anti-virus stoftware comes at a priceimmune system is costly for the organism
7arms race virus and anti-virus softwarearms race virus and immunesystem (vaccines development)
8spread via Trojan horsespread via vector
9hidden presence of viruslatency; initial symptom free period
10polymorphic virus polymorphic virus
11virus disables virus scannervirus attacks immune system
12Darwinian evolution of mutating virussesDarwinian evolution of mutating virusses
13detected by virus signature detected by virus signature
d i s s i m i l a r i t i e s
1created by humans created by biological evolution
2source code known to author of the virussequence of new virus not known
3no 2D or 3D formalways 3D form / structure
4virtual (digital) material; based on molecules
5no auto-immunityauto-immune diseases
6useful viruses do not existsome viruses have usefull effects for the host


Comparing natural evolution and evolutionary algorithms

From: 'From evolutionary computation to the evolution of things', Agoston E. Eiben, Jim Smith Nature, 521, 476–482 (28 May 2015) doi:10.1038/nature14544


PropertyNatural evolutionEvolutionary algorithms
FitnessObserved quantity: a posteriori effect of selection and reproduction ('in the eye of the observer').Predefined a priori quantity that drives selection and reproduction.
SelectionComplex multifactor force based on environmental conditions, other individuals of the same species and those of other species (predators). Viability is tested continually; reproducibility is tested at discrete times.Randomized operator with selection probabilities based on given fitness values. Survivor selection and parent selection both happen at discrete times.
Genotype–phenotype mappingHighly complex biochemical and developmental process influenced by the environment.Typically a simple mathematical transformation or parameterized procedure. A few systems use generative and developmental genotype–phenotype maps.
VariationOffspring are created from one (asexual reproduction) or two parents (sexual reproduction). Horizontal gene transfer can accumulate genes from more individuals.Unconstrained vertical gene transfer. Offspring may be generated from any number of parents: one, two or many.
ExecutionParallel, decentralized execution; birth and death events are not synchronized.Typically centralized with synchronized birth and death.
PopulationSpatial embedding implies structured populations. Population size varies according to the relative number of birth and death events. Populations can and do go extinct.Typically unstructured and panmictic (all individuals are potential partners). Population size is usually kept constant by synchronizing time and number of birth and death events.


Further Reading


 

Valid HTML 4.01 Transitional
Valid CSS!
guestbook (moderated) home: wasdarwinwrong.com wasdarwinwrong.com/korthof78.htm
Copyright ©G. Korthof 2006 First published: 2 Feb 2006 Updated: 28 May 2015 F.R: 17 Feb 2016