Slashdot Mirror
"Evolved" Caches Could Speed the Net
SpaceDilbert writes "According to New Scientist, evolutionary algorithms could make many network caches twice as efficient. This article describes a study carried out by a US researcher and two German academics, who "evolved" algorithms to determine what data should be held at a cache and for how long."
According to New Scientist, evolutionary algorithms could make many network caches twice as efficient.
That's easy, just cache 4 boobs at a time instead of two
It would be cool if it didn't suck.
It would be interesting to see exactly which algorithms they are talking about here. I wouldn't be surprised if they drew some ideas from garbage collection algorithms also.
Well, I have found one flaw in the methodology:
I would think this would breed out of the caching system any affinity to locality-of-reference.One of the things I did each morning when I was running a cybercafé was "prime" the Squid cache by running a little script that did a wget -p on all of the URLs in the portal page, and a few sites that were not. And it definitely did improve performance for most users.
One of my unrealized dreams would be some sort of speculative-fetch algorithm for Squid that would basically do a breadth-first fetch on a page while the user was busy reading it.
Of course in my not-so-humble opinion, the biggest problem with any caching system is the population of websites that, through either malice or incompetence develop content that is cache-hostile, and call it "experience enhanced".
How does the Slashdot Effect happen given that no slashdotters ever RTFA?
"An important consideration is what incentives there are for caching information for other users."
:(
So the ISP's will have to upgrade their soft/hardware to make this work?
Will that be worth it ?
In my opinion the article was a bit light on details
This is the sig that says NI (again)
I wonder if they evolved logic to counter the slashdot effect. 1. Scan slashdot.org for new stories every five minute. 2. Scan new story for links. 3. Cash those pages.
From what I can tell a good majority of the traffic that my machine receives is worm traffic. Would these genetic routines be setup to disregard those as cache data? If that's the case would they be setup to just block that data?
That alone would save me quite a bit of traffic as people on my subnet hit me constantly with their infected machines.
66.41.161.120 hit my machine 57 different times (that isn't individual requests, that's total times).
For the uninitiated, elements are added to an LRU cache until it fills up; thereafter, whenever a new element is added, space is made for it by throwing away the least-recently used one. Note, least recently used, not the least recently added, i.e. the oldest, since an element that was cached long ago may be used all the time, and so be well worth its place in the cache. For example, consider the company-logo image that your browser caches when you visit a new site and that is embedded in every page on that site. However old it gets, it's likely to continue to be used while you're on the site. As soon as you move to another site, it gradually shuffles its way down the deck until it falls off the bottom - which is precisely what you want.
--
What short sigs we have -
One hundred and twenty chars!
Too short for haiku.
It already takes forever for DNS changes to propagate through every network, which can be extremely frustrating when you have a high bandwidth domain. There definitely needs some optimization on the DNS front.
GroupShares Inc.
-------
artlu.net
So now Slashdot is going to get cached for a long time and I'll never get first post again :(
Vonal Declosion
I"ve always wondered if something along the lines of cache complimented with a Bittorrent type of scheme couldn't help speed up the internet. that way bits would be mirrored all over, and a server could pull them in faster since more servers did less work each.
just something I'm thinking about today, well, that and the Kerry/Edwards pairing.
PCBRwer342$#
free ipod and free gmail!
From the article:
He[Pablo Funes] suggests networks might in the future be designed to work out who deserves the most help for themselves. "Sophisticated network behaviours might implement rules for reciprocity and trust," he says. "And conversely, for not cooperating with other others who try to abuse our resources."
The future of network security? Imagine the next computer virus outbreak: Every network in the world could recognize the virus type activity and allocate them lesser or zero resources, maybe sending them a "Virus detected, please run antivirus software or contact your IT Department" notice, and detecting outside attacks from viruses and automatically flagging them as unsafe, and not give much(or any) attention to traffic from or to that site
Stuff like this is what keeps computing interesting, I think. This technique can be used by almost every business in a situation in which optimization might be necessary. I assume that covers most tasks someone would want to accomplish. By utilizing these algorithms to explore search spaces previously thought to be too large I predict we'll experience some pretty explosive advances in the near future in areas from farming to software development. But that's kind of obvious.
Crystal Meth: Would you ingest somthing made from a poisonous gas and an explosive metal? You do it every day -- Salt!
Just have the algorithym keep all fleshtone images in the cache...
"One can even imagine each host evolving its own optimal rule."
call me old-fashioned, but I find the mindset of a host evolving a little too "new fangled" for me. is there going to be a big stir-up now on whether or not CS profs can teach this kind of thing in the classroom? maybe it's the creationist in me...
The algorithm is just a standard ga, with the alleles being numeric values, not algorithms.
Crystal Meth: Would you ingest somthing made from a poisonous gas and an explosive metal? You do it every day -- Salt!
Actually, I do genetic algorithm / genetic programming research at the university of michigan. It's unlikely that these guys are using genetic algorithms to develop a new algorithm, but are rather using an existing algorithm and *tuning* the associated parameters using a GA. Given a list of parameters, GA's work by finding the best combination of parameters. As a result, the settings could be constantly tweaked (say on a daily/hourly basis) and different servers could still have different regional settings. My only problem with the concept is that it still depends on the tuning of pre-existing algorithms... but still - the results they share (2x improvement) is encouraging.
Once again, Al Gore has his hand in the shaping of the internet.
I'm sure everyone in the Slashdot community will miss him - even if you didn't enjoy his work, there's no denying his contributions to popular culture. Truly an American icon.
1. Scan slashdot.org for new stories every five minute.
2. Scan new story for links.
3. Cash[sic] those pages.
4. PROFIT!
I didn't see any info on their implementation in the article...did I miss a link or something?
tasks(723) drafts(105) languages(484) examples(29106)
If the search space is not dominated by local maxima, basic hill climbing (go for the best neighboring value) will work. And it will be fast. If the function is differentiable, it can be orders of magnitude faster than other methods, because you can use a variant of Newton's Method.
If the search space is small, random search (just guessing) will work by exhaustively searching the space. This is obvious, but tends to be ignored in academic papers all too often.
This discussion also applies to neural nets and simulated annealing.
Now this article at least describes a problem for which a GA might actually be useful. Many such articles don't. But they haven't demonstrated that you need a bumpy hill-climbing algorithm.
This is why, despite all the hype, GAs, neural nets, and such aren't used all that much. The search space has to have the right properties. Not too small, not too big, bumpy, but not too bumpy.
The is even a link to an online Tron game where us humans can play versus his evolving algorithms. The win/loss stats for his algorithm is approaching even. Given that humans can also evolve in the Tron game play, I imagine that the algorithm will have a head start over the new influx of slashdot visitors and start to win more often than not over the next week. I never got to play though. The SQL db to mange the stats was already down then I tried.
What happens if someone caches Metallica, or the new super hot J-Lo movie? Then how is the guy who brings the director a warm towel ever going to make his money?
You guys should really think before you go out and start making technology that blatently abuses copyrights
</sarcasm>
Evolution is the change in allele frequencies over time. Pure and simple, that's all it is. Check any evolutionary biology book. The notion of giraffes sprouting gills is absurd and not even remotely what evolution is except in the creationist strawmen. What prevents a whole lot of small changes from adding up? No one has discovered a barrier or any reason that this wouldn't occur. It's like saying that water can move grains of sand, but there's no proof that a lot of water will eventually erode a beach.
The evolutionary algorithm will have a range of all possible algorithms that can be developed, so in a sense it is limited to "test various algorithms", though it would be testing all possible algorithms. Similarly biological evolution is limitted to testing various imperfect self-replicators, meaning all possible imperfect self-replicators. It is further constrained by the current state, but then that's the problem with non-biological GAs as well, the King of the Hill effect.
"Priming the cache" and "doing a breadth-first fetch on a page" are both things that create *more* network traffic on the off-chance that it might save some number of microseconds for the user. It's basically a Tragedy of the Commons situation. Everyone would be better off if no one pre-fetched links, but any given person is better off if they don't cooperate with that global model. So everyone just grabs what they can get and everyone is worse off.
Can we use this tech to fix DNS? It's a really obsolete system, IMHO.
ANY environmental adaptation of genetic code over multiple generations is evolution. It does not have to be a major change. BTW, creationists are the only ones that make a distinction between macro-evolution and micro-evolution. If fact, creationists are the only ones that I have ever heard use the terms.
Technoli
All it will do is automatically "test" various algorithm codes until it finds the one that works best for its current situation.
Are you sure? The article says:
The algorithms that reduced network traffic and improved download speeds best were then used to "breed" a new population of algorithms. Breeding involves combining different pieces of an algorithm and introducing some random mutations. The process can be repeated again and again to improve efficiency.
Looks like what is happening can be described as evolution. You said:
Breeding, OTOH, merely pushes various existing genetic codes to become dominate. e.g. A wolf might develop a thicker coat of fur in a colder climate, while his southern cousin continuously sheds. In both cases the code already exists, the most useful one for the environment merely bubbles to the surface.
As I understand it - from my reading of Climbing Mount Improbable by Dawkins - you've just described evolution. Perhaps you're using a different, far more strict definition of the word?
I know I'm going to get myself into trouble for saying this, but "breeding" is NOT evolution!
That is a nonsensical statement. Breeding is fundamental to evolution, whether it be sexual reproduction or asexual splitting. Of course, breeding itself isn't evolution, but it is the process through which evolution occurs.
You seem to assume that the genes are unitary, descrete entities - that evolution occurs when, (to use your example) a giraffe suddenly gets the correct genes to develop gills. Genetics is much more complex than this.
Actually, if you're in canada, your users should be just fine... However, I don't know how this ruling would affect caching services.
In the states, there's already several states that have laws which prevent an ISP from setting up any kind of cache. So, I don't even know what good this would do a few states, since they can't legally impliment caching services.
Score:3 Insightful? when did creationist trolls become insightful?
Nature makes no distinction between your so called macro and micro-evolution. Rather there are to separate processes at work that combine to produce evolution.
1. Survival of the fittest (what you call environmental adaptation) starts with a population with a mixture of genes and states that those best able to reproduce in the current environment will probably make up a larger percentage of the next generation.
2. Genetic crossover during reproduction and random mutation, happens to keep the gene pool mixed and introduces new adaptations. This provides the "raw material" for survival of the fittest to work on.
Large changes are unlikely in nature because 99.999999% of major random changes to an organism will result in something that is less fit than its siblings (a giraffe with gills for instance will either suffocate in the savannah, or be unable to eat if by some miracle it also finds itself in the ocean), small changes however have better chance of being an incremental improvement that will become dominant in a population through natural selection. These small changes WILL compound over time to produce major changes in physiology (for example the way whales have adapted back to living in the oceans (actually for a really good example go read Richard Dawkins description of how eyes have evolved in many different creatures as it strikes down very effectively the old creationist chestnut of "the eye is far to complex to have ever evolved in stages")).
ARRGGHHHHH why did I bother writing all this, I know you are a troll.
Most computer scientists loosely use the term "evolutionary programming" to talk about algos that have inherently unpredictable ("emergent") results rather than modelling actual evolutionary processes observed in nature, although that's also a fair part of it all. (Incidentally, I also believe the "Science" topic assigned to this story is wrong for this reason.)
The meta-algo is evolutionary programming in that the algo fianlly "developed" by the meta-algo is apparently result that isn't immediately apparent, indeed, one that perhaps unpredictable by humans.
More than mere navel gazing.
why the hell is "evolved" in quotes???
just because algorithms aren't alive doesn't mean they can't be evolved.
(notice how nice this sentance still looks without the quotes)
music - http://www.subatomicglue.com
A search space of one kilobyte would have 2^1024 cases to mutate through. Despite any effecient evaluation of each case, this method is going to take forever. Genetic algorithms rarely work in such general cases (moving through the entire bit-space).
No info about the genetic algortihm used. I though New Scientist was a science paper...
What's in a sig?
The team used a network simulator to test out different caching strategies. They created a simulation of a branch of internet network where data could be copied and stored at every major intersection. They used this simulation to test algorithms used to configure the caches.
The algorithms take known variables, such as the number of times a piece of data is requested, the number of points it has to pass through and its overall size, and work out whether it should be stored and for how long.
So they're evolving storage time and location. At least that's all it makes clear.
Crystal Meth: Would you ingest somthing made from a poisonous gas and an explosive metal? You do it every day -- Salt!
Evolution implies ... e.g. An ameba suddenly mutates into a multi-celled configuration. Or a giraffe suddenly develops gills (Emphasis added).
Rubbish. There's nothing sudden implied.
My Karma: ran over your Dogma
StrawberryFrog
If you had studies GAs, you would know that GAs do "create" new information that doesn't exist in the original solution set (i.e. any given generation of potential solutions).
GA systems both propogate "good" information, as well as generate and test new information incrementally across generations. Of course, this is done within whatever limitations you choose to impose on the solution set (i.e. if you're using a fixed N-bit string to represent entities in the solution space, then you'll search across all 2^N possible solutions in the N-bit string space).
Fundamentally, GAs are just a way of searching a solution space for good solutions. They are slow, but can tackle a more diverse set of problems easier than the traditional backpropogation-based neural network. I would consider neural networks to be more in line with your description of an 'adapting' learning algorithm.
The following analogy can only be taken so far, but you can compare backpropogation-based neural networks to be akin to hill-climbing algorithms - you start from one solution, move some limited distance towards what you think is a better solution. For a neural network with N vertices, you can think of neural network learning as hill-climbing in an N-dimensional space over whatever field used to specify edge weights.
GAs, on the other hand, do a much more disparate search over any solution space. And it is surprising how well it finds near-optimal solutions even for problems where the relation between solutions is complex and non-intiuitive (e.g. minimums for seemingly chaotic functions).
-Laxitive
Back on topic, this means that this "evolved" cache won't do anything it wasn't programmed to do. All it will do is automatically "test" various algorithm codes until it finds the one that works best for its current situation.
It's common for genetic algorithms to start out as random data. The structures of the algorithms arise over many generations. The major guiding force behind the algorithm development is how the "fitness" of the algorithms is measured, and that in no way defines the structures of the algorithms.
It really sounds like you are completely unfamiliar with genetic algorithms. If so, your final statement is pretty damn hypocritical.
Whoever modded this up should be slapped in the face.
with Critticall tool.
Original site.
Slightly OT. How does one determine that their cache is working properly? I have a cache and it doesn't appear to make that much of a difference with content, or DNS.
This should be interesting also for the Freenet Guys. Their Datastore currently uses something like LFU i think. Implementing a better algorithm could reduce Hop Times and Performance of the whole Network.
One KB would be 8^1024 cases. Genetic algorithms have been done before on 1800 bit search spaces. Genetic algorithms don't move through the entire bit-space.
Garbage collection is required to be correct, but is allowed to keep extra stuff around as memory permits. Everything that must be kept can be calculated deterministically without future knowledge, and the same holds true for everything that can be discarded. Approximation merely allows the what can be discarded calculation to progress faster. In garbage collection it is not correct to throw out something that will not be used in the future unless it is also unreachable.
For cacheing there is no requirement of correctness. Performance is improved when things that will be used in the future are kept. There is no correctness requirement for keeping things around, and indeed best performance often requires discarding things that will be used in the future. In addition, there is no way of determining what will be used in the future without knowledge of the future. The correctness requirements of a cache are related to security (don't cache sensitive information) and staleness (don't cache stuff that is too old), but even those may be relaxed rather than strict requirements.
Fundamentally the two problems are very different, and algorithms that deal are successful in the two different cases will likely be very different.
Are we impressed? A mere doubling in efficiency should be achievable by even the drunkest engineer in all but the most well studied problems.
Let's talk an order of magnitude and then you will get my attention. Caching...Hah! I want fibre to my wetwire.
+1 Insightful
This paper has a bunch of examples.
50% of all bits sent over the internet are 0s. Just cache that and we have a 50% cache hit rate. :)
Sometimes people call "microevolution" "natural selection".
Gamingmuseum.com: Give your 3D accelerator a rest.
The problem with caches is whether you can trust them. Even assuming you can keep the caches properly up-to-date, how do you prevent cache poisoning from taking place?
Evolution is the change in allele frequencies over time. Pure and simple, that's all it is. Check any evolutionary biology book.
No, it's not that simple. Evolution has been defined as changes adding up enough so that a single celled organism becomes a multi-celled organism, which develops RNA and DNA, which then grows in complexity to develop various "macro" level systems such as a cardio-pulminary system, a digestive tract, a nervous system, and a central control "brain" system. Each of these changes requires a macro change of which no predecessor exists in the current chain of organics.
The notion of giraffes sprouting gills is absurd and not even remotely what evolution is except in the creationist strawmen.
Actually, my first example was the very real issue of a single celled ameba becoming multi-celled. A giraffe sprouting gills would be an evolutionary change, but is highly unlikely. A more realistic example is a fish developing lungs. According to the current theories, most life originated in the ocean. I don't remember if the current thought is that lungs and gills developed simultaneously or if they developed in unison. Either way, the effect is much the same. A low order life form developed features not currently in its genetic makeup.
It's like saying that water can move grains of sand, but there's no proof that a lot of water will eventually erode a beach.
That's a perfectly valid assumption. How do you know that grains of sand don't get carried down to replace the ones that are eroded? Without proof that the grains are carried away without being replaced, you only have a supposition. You must have evidence to make it a workable theory, and absolute proof to call it a fact.
The evolutionary algorithm will have a range of all possible algorithms that can be developed, so in a sense it is limited to "test various algorithms", though it would be testing all possible algorithms.
Which is exactly my point. No new algorithm will be generated through the "genetic breeding". Only existing algorithms will rise to the top.
Javascript + Nintendo DSi = DSiCade
Natural selection is one of the methods (the other being sexual selection) by which all evolution (macro and micro) proceeds.
Technoli
I'd like to see an ameba gradually become multicelled. I just get this image of a far side cartoon of an ameba sitting around with a cup of coffee in the middle of meiosis, saying "I don't care if you haven't finished splitting off, I've got tickets to the ballgame so I'm dragging you along."
Evolution is *not* the "change in allele frequencies over time." Or at least that's a secondary issue, historically, conceptually, and causitively. Well, on the last, allele frequency is pretty important (but still not exclusive). But evolution as a concept is firstly a question of PHENOTYPIC change, even before it's about GENOTYPIC change.
Evolution was a well known phenomenon (e.g. from the geological record) long before the work of Mendel was known. Specifically, the rather well known Darwin had no concept of an allele when he came up with the Theory of Evolution by Natural Selection. In fact, Darwin's approach was most certainly not the only explanation of evolution floated in 19th century biology. For example, Lamarckian mechanisms had a plausibility at the time... in fact, Lamarckian mechanisms still represent an important underexamined area of evolutionary mechanisms, for historical reasons (Lysenko etc.)--see, for example, the work of Ruth Hubbard and the role of viral insertions into the genome. So the whole allele thing is only part of one (important) mechanism, only understood relatively late in the scientific study of evolution.
Moreover, even in modern terms, mutation still exists; admittedly, the high-school text book focus on point mutations of genes is vastly overplayed. But just looking at allele frequencies misses also both the role of regulatory genes, and also ignores structural changes in chromosomes. Genes and smaller sequences migrate within and between chromosomes--this has nothing to do with alleles. Stephen Gould's work on phentotypic growth regulation in _Ontogeny and Phylogeny_ is good here, as is the tome _Structure of Evolutionary Theory_.
Buy Text Processing in Python
ANY environmental adaptation of genetic code over multiple generations is evolution.
No, it's been shoe-horned into the theory. Evolution is a concept that a new, higher level trait will develop when no prior concept exists. For example, a single celled organism will become multi-celled, multi-celled with develop "cell types", and "cell types" will organize to produce higher systems (circulatory, digestion, nervous, brain, etc.). That's the theory. It's a fine that it's being worked on, and I have no issue with that. But adaption to environment is a different concept. They can theorize that it's related in some way, shape, or form, but until someone can show that new genetics suddenly "appear" through the same process, you can't force the two concepts together.
If fact, creationists are the only ones that I have ever heard use the terms.
I don't know where you ever got that idea. I'd never heard of the terms until evolutionary texts started using them. As usual, Wikipedia has the info.
Javascript + Nintendo DSi = DSiCade
The article and my own words stated the same thing. Information on the previous "generation" is used to adapt to the current environment. No "new" algorithm is "evolved", merely a new combination of existing algorithms.
As I understand it - from my reading of Climbing Mount Improbable by Dawkins - you've just described evolution. Perhaps you're using a different, far more strict definition of the word?
My statement is that "macro-evolution" is the basic concept of evolution. i.e. A single celled organism can become multi-celled, develop DNA, separate into different cell types, etc., etc., etc. "Micro-evolution" is really just a shoe-horning of environmental adaption into evolutionary theory. While it adds the concept of "genetic mutations" to the DNA, it fails to consider the fact that DNA is a trait that had to "evolve" itself. A "simple" one celled organism would have lacked the concepts necessary to process and create complex DNA.
Javascript + Nintendo DSi = DSiCade
I'd like to see an ameba gradually become multicelled. I just get this image of a far side cartoon of an ameba sitting around with a cup of coffee in the middle of meiosis, saying "I don't care if you haven't finished splitting off, I've got tickets to the ballgame so I'm dragging you along."
That's like asking to see lead gradually become gold, then still calling it lead. Now there are many singled-celled animals that form colonies that live vary closely together. I also know that in the mid-depth water colum there is at least one creature that blurs the line between multi and single-celled animals. Genetically the cells are distinct enough to be concidered separate species. However, they are specialized to the point where they can't live without each-other. I am sorry I can't recall the name of this animal/colony of animals, it was only mentioned briefly in a deep-sea documentary.
Caching information can also be used as a backup of data in case the server crashes or has its data compromised by a virus , hacker, whatever you feel like ..
so what happens if you, for instance, have a security hole in one of your "smart" servers?? or even a breach in the protocol structure (DoS)?
you could get the server to "breed" algorithms that would stop all servers by either corrupting the data in all servers that are providing the service or just DoS-ing them.
I guess i'm not yet convinced that this solution is of any good for real world network which the whole structure is based on insecure protocols.
maybe after IPV6, IPV9, who knows?
What is best in life? To crush your enemies, to see them driven before you and to hear the lamentations of their women.
Breeding is fundamental to evolution, whether it be sexual reproduction or asexual splitting.
Breeding is fundamental to life. Life is fundamental to evolution. Evolution tries to describe how organic soup could form simple organisms that become progressively more complex.
You seem to assume that the genes are unitary, descrete entities - that evolution occurs when, (to use your example) a giraffe suddenly gets the correct genes to develop gills. Genetics is much more complex than this.
Genetics is certainly more complex than that. But as for a sudden change of no gills->gills, it depends on who you talk to. Some theorists have suggested that changes did indeed happen rapidly. This would explain how the fossil record shows changes happening at a rapid pace, with very little in the way of "in-between" states. Other theorists suggest that things were more gradual, but at a level that is more difficult to notice. Either way, they agree on one point: a "simple" organism went through large degrees of change to develop biological processes that didn't exist when the the organism was first formed. Thus in some way, shape, or form, the organism was able to "create" new concepts.
The genetic algorithm we're discussing doesn't do that. It simply creates a new "mix" of algos that work best for the current environment. While there's nothing wrong with that, the algo isn't going to cause the computer to eventually develop an intelligent neural net. (Unless a human added neural net code, that is.)
Javascript + Nintendo DSi = DSiCade
They create new "information", not new "algorithms". Evolution supposes that organisms create new "biological processes", analogous to "algorithms".
Javascript + Nintendo DSi = DSiCade
I wish the mods had given you a +5. My peeve is exactly what you describe: that "evolution" is being used as an analogy rather than a realistic description of what it does. "Genetic Algorithms" is a fine name, and "breeding" is an accurate concept. But "evolving" is misleading to a very extreme degree. There's no need to describe the process as "evolution", yet CS researchers persist in doing so.
That being said, it can be a great tool for auto-adjusting software to its environment. If it works out, it may even be applicable to robotics where the concept can help the robots act more "intelligently". (e.g. Stop trying to drive over rocks and "learn" to drive around them.)
Javascript + Nintendo DSi = DSiCade
The two processes you are describing are the same thing. One simply takes a lot longer. Someone else in this thread made a great analogy about how saying that microevolution is possible while macroevolution is impossible is like saying that water can wash away individual grains of sand, but can't erode away a beach.
One of the problems with creationists is that they have a hard time conceiving of the incredibly vast amount of time that has passed which makes macroevolution possible.
I don't know where you ever got that idea.
What idea? That creationists are the only people I have heard use the terms? From the fact that they are the only ones I have heard use the terms. Read it carefully. No where did I say that they aren't scientific terms. I am well aware that they are. Creationists have latched onto them and gotten it into their heads that they are two different things.
Technoli
The evolution that is described in this article is anything but unguided. Researchers took existing carefully developed code. They selectively combined parts of this code. They studied the resulting effects to produce better code. This looks much more like Intelligent Design [discovery.org] than anything Chuck Darwin ever dreamed up.
No, if it were truly following the "intelligent design" theory, there wouldn't be any trial and error to begin with. The code would be written once, from the ground up, and it would work just as well as any code possibly could, without ever needing to selectively combine anything.
Also, your textbooks had Evolution wrongly defined. It is not random, unguided change. The mutations themselves may be random, but the environment (including predators) is what guides the changes, by causing favorable mutations to progress through the gene pool while unfavorable mutations die out quickly.
-CausticPuppy "Of all the people I know, you're certainly one of them." -Somebody I don't know
Sir, I requested intelligent discussion, but you yourself have taken to trolling. First you make an accusation ("creationist troll") which neither has any bearing on the subject, nor is in any way true. You will note that I never suggested that evolution is a "made up" process. That is outside the realm of this discussion. All I claimed is that the "Macro/Micro" division is an as-of-yet-unsupportable position that causes a great deal of confusion. The algorithm described in the article will not "evolve" in the "macro-evolutionary" sense, but will instead adapt in the "micro-evolutionary" sense.
1. Survival of the fittest (what you call environmental adaptation) starts with a population with a mixture of genes and states that those best able to reproduce in the current environment will probably make up a larger percentage of the next generation.
Exactly. It allows the organism to best meet its current environment. However, it cannot be the vehicle for the creation of new, nonexistent biological processes. The mechanism for that must be much different as organisms must have evolved from a "simple" organism, to a "complex" one.
2. Genetic crossover during reproduction and random mutation, happens to keep the gene pool mixed and introduces new adaptations. This provides the "raw material" for survival of the fittest to work on.
Random mutations only gain you so much, and in all cases are also harmful in the same way they are helpful. Sickle-cell anemia is an example of a mutation that is used to adapt to the environment (resistance to malaria), but has negative side effects on the gene pool (causes early death in nearly all who contract it). The process that took a "simple" organism to a "complex" organism must therefore be far more sophisticated.
ARRGGHHHHH why did I bother writing all this, I know you are a troll.
Insult number 2. Why can't people discuss this topic without throwing around insults? I'm not sitting here trying to convince people that evolution is an incorrect theory. I'm merely complaining about the confusion caused by mixing the ability of an organism to adapt using its existing genome as opposed to the concept of a more complex organism forming from a simpler one. I suppose I knew what I was getting into when I posted.
Javascript + Nintendo DSi = DSiCade
For a simple method to solving complex problems. There's also Fuzzy logic.
Why?....Why?....Why?....AHHHH...
The two processes you are describing are the same thing. One simply takes a lot longer.
No, they are theorized to be related. Macro level changes require a mechanism more sophisticated than that of "micro-evolution". "Micro-evolution" requires an existing genome from which to pull, and only allows for a small degree of genetic mutation. As I replied to another poster, Sickle-cell anemia is a perfect example of a mutation that "helps" by resisting malaria, but it has the side effect of an early death for the organism.
In fact, SCA makes the perfect point that genetics is a zero-sum game. To gain certain advantages, some must be given up. This always confused me when I was working in the livestock genetics field. All the charts of the cows' features always appeared to sum to the exact center. i.e. The perfect cow was a perfectly average cow. Well, I spoke with my colleagues and learned I was correct. All we were doing was breeding for some "desirable" features (often fads) at the expense of other features of the cow.
One of the problems with creationists is that they have a hard time conceiving of the incredibly vast amount of time that has passed which makes macroevolution possible.
No, I have no problem with the amount of time. I have a problem that "micro-evolution" is not a sufficient process for "macro-evolution". Theorists divide the two because they are not provably related. From Wikipedia:
There are two proposed mechanisms for macroevolution. The first way is through the extrapolation of microevolutionary processes. Tiny microevolutions, over sufficient time, add up and accumulate in isolated populations and eventually result in new species. The second way in which "macroevolution" is believed to occur is through sudden and rapid changes. This theory, punctuated equilibrium, put forth by Stephen Jay Gould, is based on the fact that there are critical genes (such as the homeobox) in all living organisms, and a small change in them could cause drastic changes in the organism, resulting in a new species quite rapidly.
[snip]
The additional information needed for these structures did not arise from the mutation, of course, but existed elsewhere in the animal's DNA and was replicated at the novel location.
No provable link yet exists between micro and macro evolution. Until such a link is proven, the term "micro-evolution" continues to be misleading.
What idea? That creationists are the only people I have heard use the terms? From the fact that they are the only ones I have heard use the terms. Read it carefully. No where did I say that they aren't scientific terms. I am well aware that they are. Creationists have latched onto them and gotten it into their heads that they are two different things.
Then I believe, sir, you are trolling. If they are scientific concepts, then claiming that creationists are the only ones who vocalize them is an underhanded way of insulting anyone who discusses the topic.
Javascript + Nintendo DSi = DSiCade
No, it's not that simple. Evolution has been defined as changes adding up enough so that a single celled organism becomes a multi-celled organism, which develops RNA and DNA, which then grows in complexity to develop various "macro" level systems such as a cardio-pulminary system, a digestive tract, a nervous system, and a central control "brain" system. Each of these changes requires a macro change of which no predecessor exists in the current chain of organics.
That is one of the facts that it explains, not the definition. You're doing the equivalent of saying that a ball falling is the definition of gravity. Evolution is the collection of mechanisms by which these occur, which under the modern synthesis is by various alleles changing over time in a population.
Actually, my first example was the very real issue of a single celled ameba becoming multi-celled.
And oddly, stages between single cell and multicell still exist. The Volvox are single cell creatures that only function in colonies. A colony is a sphere that can grow a daughter colony inside it, then give birth to a new colony. Yet all of the cells involved are generic, not specialized into tissues. Sponge and jellyfish are on the other end of the gap. They both show tissue level specialization without organs.
A giraffe sprouting gills would be an evolutionary change, but is highly unlikely.
A giraffe sprouting gills would be a challenge to modern evolutionary theory, not an example. Modern evolution relies on the existing material being gradually changed or coopted.
A more realistic example is a fish developing lungs. According to the current theories, most life originated in the ocean. I don't remember if the current thought is that lungs and gills developed simultaneously or if they developed in unison. Either way, the effect is much the same. A low order life form developed features not currently in its genetic makeup.
Such as a mud skippers and lung fish? They both occupy niches and use mechanism similar to the first air-breathers using modified gills and air bladders, respectively. Not only are the transition species possible, but they're descendants still exist.
Which is exactly my point. No new algorithm will be generated through the "genetic breeding". Only existing algorithms will rise to the top.
That's interesting. I can write a program that contains the Axioms and rules from ZF Set theory, then randomly applies them to create new theorems. But that would be pointless since only theorems that existed in the orginal data and algorithms "will rise to the top". The whole of mathematics is pointless since all theorems are tautologies from the axioms and rules of manipulation.
"In fact, evolution can be precisely defined as any change in the frequency of alleles within a gene pool from one generation to the next."
Yes, you *can* define evolution that way (to be precise, one group of biologists, the population geneticists, do define evolution that way). However, that's an entirely different statement than saying that's the *only* valid definition of evolution, which seems to be what you're implying.
As a molecular evolutionist, I find alleles to be too high level -- I deal with the evolution of sequences themselves. And a zoologist would probably claim that alleles are too low level as zoologists mostly deal in phenotypes.
Natural selection is one of the methods (the other being sexual selection) by which all evolution (macro and micro) proceeds.
Mutation and genetic drift. Also, the term natural selection as Darwin used it included sexual selection (besides ecological selection). The term is used to distinguish these kinds of naturally occuring selection from intentional breeding, which is called artificial selection.
I love C++
That is one of the facts that it explains, not the definition. You're doing the equivalent of saying that a ball falling is the definition of gravity. Evolution is the collection of mechanisms by which these occur, which under the modern synthesis is by various alleles changing over time in a population.
My argument is that "micro-evolution" has not been shown to be the mechanism by which "macro-evolution" can occur. The results of environmental adaption and genetic mutations are well known, but no one has yet shown how a new biological process can derive from one that currently does not exist inside the genome. For example, if humans lack the genome for an exoskeleton, and exoskeletons are "nature's selection" for space survivability, then the "micro-evolutionary theory" says that humans who live in space should eventually experience genetic mutations from which exoskeleton code will appear. The only mechanism that has been provably demonstrated is that existing codes can be reorganized to produce oddities such as legs growing from eye sockets, or extra thoraxes. (Source) Thus evolution is still described as a process, but the mechanism is still a topic of debate.
And oddly, stages between single cell and multicell still exist. The Volvox are single cell creatures that only function in colonies. A colony is a sphere that can grow a daughter colony inside it, then give birth to a new colony. Yet all of the cells involved are generic, not specialized into tissues. Sponge and jellyfish are on the other end of the gap. They both show tissue level specialization without organs.
[snip]
Such as a mud skippers and lung fish? They both occupy niches and use mechanism similar to the first air-breathers using modified gills and air bladders, respectively.
I'm not arguing the concept of evolution. I'm merely arguing that "micro-evolution" is not a proven method for such macro change. Until evidence exists to show that these sorts of biological changes can occur via genetic adaption and mutations, the name "micro-evolution" will continue to confuse things. In terms of "Genetic Algorithms", things have become particularly confused. The system will never develop a "new" algorithm, it will simply decided which combinations best meet the current environment. Nothing "evolves" through the process, merely "adapts".
But that would be pointless since only theorems that existed in the orginal data and algorithms "will rise to the top".
I hardly stated that GA's were "pointless". The idea is fine. However, it doesn't produce "evolution". A GA "caching" algorithm, for example, will never be able to "evolve" a P2P network to leverage existing caches like itself. It can't do that because the code doesn't exist.
Similarly, no code existed in nature for "micro-evolution" to "mutate" lungs, electrical impulse handling, circulatory systems, etc. Either there are properties to "micro-evolution" that have not yet been revealed, or some other (possibly related) mechanism must be responsible.
Javascript + Nintendo DSi = DSiCade
For example, if humans lack the genome for an exoskeleton, and exoskeletons are "nature's selection" for space survivability, then the "micro-evolutionary theory" says that humans who live in space should eventually experience genetic mutations from which exoskeleton code will appear.
"All too often creationists spend their time arguing with a straw-man caricature of evolution."
Well put.
I love C++
My argument is that "micro-evolution" has not been shown to be the mechanism by which "macro-evolution" can occur.
Define micro-evolution and macro-evolution and a set of criteria that a person could use to determine which of the two an adaptive event falls into.
For example, if humans lack the genome for an exoskeleton, and exoskeletons are "nature's selection" for space survivability, then the "micro-evolutionary theory" says that humans who live in space should eventually experience genetic mutations from which exoskeleton code will appear.
Selection is far more likely to make the species go extinct altogether, if I'm reading you correctly. Extinction by far the more common result of selection than success. Since the creation of nylon in the 30s, species of bacteria have evolved to eat it. If suddenly the entire planet had to survive off of eating nylon, you wouldn't see most species evolving that ability, you'd see most species going extinct. Evolutionary success usually occurs as a result of something moving from a niche that it already successfully occupies into an unoccupied niche.
. For example, if humans lack the genome for an exoskeleton, and exoskeletons are "nature's selection" for space survivability, then the "micro-evolutionary theory" says that humans who live in space should eventually experience genetic mutations from which exoskeleton code will appear.
[...]
I hardly stated that GA's were "pointless". The idea is fine. However, it doesn't produce "evolution". A GA "caching" algorithm, for example, will never be able to "evolve" a P2P network to leverage existing caches like itself. It can't do that because the code doesn't exist.
I don't believe you understand the word "evolution". Evolution is not a huge jump that rapidly redefines something, it's an almost unnoticable change over time. Evolution can be understood as a search through a problem space. The caching algorithm sets up a problem space of "all possible caching algorithms". A P2P server would be completely outside of the problem space and thus irrelavent. However, an explicit view of the problem space is far larger than the algorithm and seeding data. The GA is just a way to walk through a subset of the problem space in a way that lets you ignore large areas of useless algorithms. You certainly could create a GA capable of creating a caching algorithm or a P2P network, you'd just have to have a huge population size and millions of generations to do it. It wouldn't be worth the time.
Doesn't Freenet already do this within its own network?
Another related item is the ARC algorithm from IBM, which is an adaptive cache for block buffers.
Algorithms are just a certain kind of information.
Consider a GA system that operated on bitstrings that are considered inputs to a universal turing machine. Then, the information created is an alrogithm.
True, when designing a GA system to do that, you might have problems in creating test cases for evaluating solution fitness, but that doesn't detract from the point.
There has been research done on using GAs on lisp-style expressions, using subexpression substitution instead of crossovers, to generate useful algorithms. GAs aren't just for parameter optimization, although they are particularly suited to parameter optimization problems.
-Laxitive
Nothing "evolves" through the process, merely "adapts".
Nope. This is one and the same thing. E.g. there is plenty of molecular evidence that the blood clotting proteins forked off of pancreatic digestive enzymes by means of gene duplication and mutation. In embryonic development, some muscle cells are chemically activated to turn into bone cells. Archaeopteryx is still a tell-tale example of a gradual transition between bipedal dinosaurs and birds.
Similarly, no code existed in nature for "micro-evolution" to "mutate" lungs, electrical impulse handling, circulatory systems, etc. Either there are properties to "micro-evolution" that have not yet been revealed, or some other (possibly related) mechanism must be responsible.
The first fossil traces of lungs consist of a pair of small pharyngeal sacs in Actinopterygii. They may have populated oxygen-depleted pools and supplemented their oxygen supply by "gulping" air like some extant fish species do. Any slight increase in pharyngeal skin surface area and volume would have been of value to them, even if it just consisted of two tiny sacs at first. Tissue sensitive to electricity is found even in the lowliest of critters. Vertebrate circulatory systems serve to speed up diffusion of oxygen and nutrients, and likely started out as a simple linear transport in chordates (the link between invertebrates and vertebrates).
I love C++
Easy: consider an amoeba sitting in a soup with millions of its cousins. Now consider an amoeba starting to huddle closer and closer to these cousins. Now it starts to cuddle with its children. Finally consider an amoeba that makes identical copies to cuddle up to. Cousins, children, self. Gradual evolution from a lot of related single cellular organisms to a single multi-cellular one.
\ This was on data where the distance the document had travelled was taken into account. So, given some available input data they evolved a priority function (using GP) that servers as the center for a caching strategy. This with apparent great success.
I'm afraid this doesn't seem like a particularly appropriate use of genetic algorithims. In fact the entire piece seems to suggest a poor research project (someone who did genetic algorithms needed a thesis) which got picked up on a slow news day.
First of all I would point out that genetic algorithms are most appropriate for pure optimization problems with minimal mandatory constraints. While loading speed is important for internet caching it must be optimized inside certain constraints. A good cache system would have guarantees about how frequently web pages were refreshed/verified. For instance one might evolve a very efficent cache algorithm which only refreshes slashdot once per day.
Of course one could tack refresh guarantees on top of your evolved algorithm but the more constraints you place on top of the algorithm the less likely it is that this algorithm is more efficent than a more traditional solution. Most likely one would like to make the refresh minimum dependent on the frequency the website updates and placing a complex condition like this on top of a genetic algorithm evolved simply to maximize cache hits is likely to destroy any speed benefit it might provide.
To be fair one could amalgamate all these concerns into the evolution process. Instead of just measuring cache hits/download speed one could also check how frequently the cache page matches the current page and optimize some combination of these scores. However, this doesn't appear to be what was done in the article. Moreover, for web page cacheing some absolute gaurantee of freshness is desierable. The winning genetic algorithm might have very good performance on average but almost never update one person's report on the stock market.
In general it seems like a dangerous idea to use genetic algorithms in situations open to abuse. Handwritten code can be analyzed and it's behavior in exceptional situations relatively easily determined (as opposed to a genetic algorithm). With a genetic algorithm not only is it not clear how it will behave in exceptional circumstances but we aren't even sure what kind of situations are most likely to break the algorithm.
A web cache should robustly handle attempts to misdirect or corrupt it. If someone forges a great deal of web page requests we don't want the cache to dump truly popular pages to load the falsely requested pages. A flaw like this could quite possibly be used to mount a DNS attack using the large bandwidth of the web cache. More worrying is the possibility that a clever hacker might be able to trick the cache into replacing a valid website with a page of his choosing. Handwriting the code allows one to deal with all these potential problems and also ensures that the cache files/records are in a manageable form so sysadmins can manually flush some pages out of the cache or otherwise address errors.
Finally, it is not clear that a genetic caching algorithm would continue to work as the internet changes. If you sort through a wide swath of algorithms it is quite possible that certain features/tricks will end up being hardcoded, for instance the algorithm might refuse to cace anything starting with google.com which will then become inefficent if the internet changes. It would be good to remember the story about the army using a neural net to diffrentiate russian and american tanks and discovering it had just learned to distinguish the time of day as the american tanks had all been photographed at noon.
In short it seems that genetic algorithms have major drawbacks as cache algorithms. Since we have barely broken the surface of handwritten cache systems the effort hardly seems warranted. Before moving to genetic algorithms there are tons of good ideas which should be explored first. For instance making browsers cache aware (so the online cache can cooperate with the browsers cache like I suppose earthlink does), predictively requesting pages linked from downloaded pages, generating user profiles to better predict which pages wil
If you liked this thought maybe you would find my blog nice too:
In IE, when you hit 'back' or 'forward', it takes time to load the page!! Why?? If you tell it to work offline, then everything is instantaneous. There's no reason that it should check to see if a page has been updated after you click back if you were just there 5 seconds ago. Changing this would speed up my websurfing experience a lot more than this caching algorithm.
-grankk
Of course with systems like internet caches, where the cache is not a part of system it caches, it needs to have other measures as well. Not just obeying the expire headers on the things it caches but also trying to see when it has changed even if the expire info are missing or wrong.
And then perhaps trying to predict the update rate of the page and then prefetch pages high in the LRU tables to have fresh versions ready for the user.
So does that make you a creationist? I just read you using those terms.
(-1, not very Funny)
To me its building in more capablity to track what and where you go in the internet. Most ISP's already have these servers. I don't see how they really have helped much. Plus being another layer of crap tends to make it also another layer that won't be managed securely and lend it self to the next compromised system.
BYte69
Why? Just because you give an example of where this is true (Sickle-cell anaemia), doesn't mean it is generally true. Why can't a random mutation be beneficial without bad side-effects? A random mutation can surely be harmful without good side-effects. Why not the reverse? It isn't a zero-sum game. Sure, the kind of mutation that is solely beneficial may be exceedingly rare, but that isn't proof that it can't happen.
Also, this isn't even necessary. So long as the good outweighs the bad, the mutation will be selected for. This is exactly what happens with SCA. The carriers are resistant to malaria. If this helps them (or siblings with only one copy of the gene) to survive to child-rearing age, then it will be selected for.
Why? Just because you give an example of where this is true (Sickle-cell anaemia), doesn't mean it is generally true.
Correct.
Why can't a random mutation be beneficial without bad side-effects?
Because no positive mutation has ever been shown. Every mutation ever observed results in a delterous effect on the organism. SCA is the best example anyone has ever given of a "positive" mutation.
It isn't a zero-sum game.
Funny you should mention that. I made that exact point earlier:
In fact, SCA makes the perfect point that genetics is a zero-sum game. To gain certain advantages, some must be given up. This always confused me when I was working in the livestock genetics field. All the charts of the cows' features always appeared to sum to the exact center. i.e. The perfect cow was a perfectly average cow. Well, I spoke with my colleagues and learned I was correct. All we were doing was breeding for some "desirable" features (often fads) at the expense of other features of the cow.
Granted, I am not an evolutionary geneticist, but all evidence I've seen tends to suggest that genetics *is* a zero-sum game. You simply can't give to Peter without taking from Paul.
Sure, the kind of mutation that is solely beneficial may be exceedingly rare, but that isn't proof that it can't happen.
Proving the negative in this case is uselss as the only way to prove the negative is to prove the positive. No mutation has ever been shown to be helpful to an organism without introducing serious negative effects. If you know of a mutation that is only positive, then I would make a most captive audience to your presentation.
So long as the good outweighs the bad, the mutation will be selected for. This is exactly what happens with SCA. The carriers are resistant to malaria. If this helps them (or siblings with only one copy of the gene) to survive to child-rearing age, then it will be selected for.
Creating a 1 in 4 chance of a very early death is not a "good" result. It can help in the short term, but only if a sufficient genetic pool is maintained to later breed out the negative condition. In this example, the African population is huge, but 1 in 13 African Americans still carry this mutation. And how useful is the mutation if it must later be bred out of the genetic stock?
Javascript + Nintendo DSi = DSiCade
I don't believe you understand the word "evolution". Evolution is not a huge jump that rapidly redefines something, it's an almost unnoticable change over time.
No, the current theories state that it is a unnoticable change over a long period of time. On a higher level, evolution simply states that major changes occurred over a long period of time. It also states that those changes occurred in an "organic vacuum". e.g. Lungs developed where no lungs had ever existed before.
Evolution can be understood as a search through a problem space. The caching algorithm sets up a problem space of "all possible caching algorithms". A P2P server would be completely outside of the problem space and thus irrelavent.
If that was really the case, then higher organisms should never have developed. The problem space for the "simple" organism was to survive in its wet environment. I believe you named colonies of amebas as an example of "vestigal organisms" at one point. Without arguing that point, what forced these successful organisms to then develop different cell types? How did different cell types improve their success rate? From there, how did developing complex DNA improve their success rate? From there, how did specialized macro-organs improve their success rate?
In evolutionary theory, the "simple" life form was striving to become a better adapted life form. It obviously exceeded its problem space by a wide margin, or you and I would not be having this discussion today. Thus if GAs are really indicitive of "real" evolution, then why can't they eventually evolve into "higher algorithms"? A P2P network could very well improve the GA's success rate for its problem domain. It can start by accidentally reusing its network connections, then accidentally making contact with other cache hosts with the same mutation, then it can evolve a protocol by accidentally learning each communication packet through a torrent of random spam packets, etc.
While that sounds somewhat silly, that's not my point. My point is that evolution states exactly that sort of behavior over a long enough period of time. If a GA is a good emulation of micro-evolution, and macro-evolution is based on micro-evolution, then shouldn't we be able to compute a set of mutations that take the program out of its pre-programmed algorithms? Shouldn't it be able to leverage parameters that we didn't program it to leverage?
Javascript + Nintendo DSi = DSiCade
That wouldn't happen because it is outside the bounds of the system. Say, for the sake of argument, that it is impossible for carbon-based life forms to have telepathy. There is now no way for any carbon-based life form to evolve telepathy, yet it [the organism] is still evolving. Telepathy is outside the bounds of the system. Same with a GA system. It is evolving in the same way (very minor changes from one organism to the next that add up to be a significant change). It can come up with genetic material that was never introduced into the system (that's what the mutation operator does), it can combine genetic material in ways never before considered (crossover), and each of these possible solutions is checked to see how good it is. The better it is, the greater the odds of it's survival. I have seen GAs find solutions in areas that people hadn't even considered, just by starting from a few random points and slowly moving to where things are a bit better.
There are no pre-programmed algorithms (unless the programmers seeded it with a few known-good ones as a start). In either case, it is going to be moving away from it's starting situation as soon as it starts evolving. A bit here, and a bit there will change, and look at it a few hundred generations later, and what it has is totally different to what it started with. There is no macro- or micro-evolution, there is only evolution. The more time you look at it over, the bigger the changes. If you start with 1, and add 1 you have 2, which is close to one. Take it 100 time steps later and you have 102, which is very different from 1 or 2, but is the result of many small changes.
Do a google search on "single-celled colony". There's your intermediate stage.
My Karma: ran over your Dogma
StrawberryFrog
I think that this is largely because we have been evolving for so long that we are at the point where if a single mutation was going to have a totally beneficial effect, it would already be in the population. However, take someone with colour blindness: a disadvantagius condition (slightly so, anyway). Now, if he has a child, and in the process of reproduction the colour blindness causing gene mutated so that the child wasn't affected, wouldn't that be a beneficial mutation? (ignoring the whole diploid thing for simplicitys sake). I suspect the reason we don't see that happen is because it is much less obvious. Nowadays we are at the point where mutations come with tradeoffs, because if they didn't, then they would have already happened.
Will you make up your mind, first you agree with the grandparent poster that one particular example doesn't prove the general case, and then you make the claim that the SCA example does suggest the general case is zero sum.
Come on man, make up your mind! which side of the fence are you on?
It is worth noting that there is a natural bias towards observing negative mutations, not only are they probably far more common but also they stand out against the background, i.e. when people start dieing, or are born with gills, or no legs then people are going to notice pretty damn quick and start saying "ohhhh, mutation==bad!" however a beneficial mutation (for example a resistance to the common cold) is much more likely to go unnoticed at first, as it is a lot harder to notice or prove that something hasn't happened during a persons lifetime. Chances are by the time the mutation is noticed it will be wide spread enough in the population that it will be impossible to identify as a single mutation to a specific person sometime in the past, and it will thus look as though it is simply part of the general mix of genes in the population.
I apologise if you took offence at being called a troll earlier, I meant it in the original sense of someone who deliberately posts inflammatory content in the hopes of sparking a debate. (anti-evolution posts certainly qualify on Slashdot)
As for calling you a creationist.... well are you are aren't you, you claim to not believe in evolution, so how do you think we got here?
I just took the morning off to register at university for my masters in computer science. My research topic is a distributed caching proxy network. Obviously since i haven't started researching yet, i don't have all the details, but the basic idea, is to have a network of caching servers, each serving a group of users, but interconnected, so that if a user of a particular caching server requests a file that isn't in that server's cache, but is in the cache of another server on the network, then that server gets the file from its neighbouring cache instead of the origin server. Eventually, each of the caching servers end up with a copy of the file and the remote site is only hit once. There are obviously issues that I'll have to deal with, but thats the basic idea... kinda like squid crossed with bittorrent :)
That logic just doesn't make sense. If I type a random bunch of characters, I *could* produce the works of Shakespeare. If I change a random gene, it *could* be beneficial. How can you possibly argue that a random change can't be beneficial ever? A random change can produce anything, including stuff that is good.
Granted, I am not an evolutionary geneticist, but all evidence I've seen tends to suggest that genetics *is* a zero-sum game. You simply can't give to Peter without taking from Paul.
Again, why? One example from agriculture doesn't prove anything. All it proves is that we don't know how to do it properly. That *may* be because it isn't possible but not necessarily.
Mutations that are positive are also far less obvious. If your child is slightly stronger than normal, would you notice? If they have a crippling disability, on the other hand, you sure will.
Creating a 1 in 4 chance of a very early death is not a "good" result.
Ah, but i believe 2 in 4 are protected from malaria? Evolution has selected for the sickle-cell property because it provides an overall benefit for the group. African Americans still have the mutation but it is fading out (hence the lower prevalence). Evolution takes time, especially now that modern medical practices mean that many more sufferers survive.
Will you make up your mind, first you agree with the grandparent poster that one particular example doesn't prove the general case, and then you make the claim that the SCA example does suggest the general case is zero sum.
I have made up my mind. The parent of my post made a statement that was factually correct: SCA does not "prove" that positive mutations can't happen. I then go on to show that everything from SCA to genetic manipulation has so far shown us that genetics is a zero sum game. It's perfectly possible that it's not, but we haven't yet proved it either way. All we have is a mountain of data that suggests genetics is a zero-sum game.
Come on man, make up your mind! which side of the fence are you on?
How is that scientific? If you always approach a problem from the angle that "X must be right", then you'll never make any new discoveries. If something can be observed then proven to be true, then we must adjust our world-view and move on.
It is worth noting that there is a natural bias towards observing negative mutations, not only are they probably far more common but also they stand out against the background, i.e. when people start dieing, or are born with gills, or no legs then people are going to notice pretty damn quick and start saying "ohhhh, mutation==bad!"
This is quite true.
however a beneficial mutation (for example a resistance to the common cold) is much more likely to go unnoticed at first,
The difficulty here is that until we can isolate and identify "positive" mutations, they are nothing but speculation. It's possible that it's just a matter of time before such mutations are discovered, but we can't say they exist until we can demonstrate them. It's a bit like Tachyons in physics. Theory said they should exist, therefore they existed, right? (If it's on Star Trek, it must be true!) Turns out that Tachyons are now considered to have been a "glitch" in string theory that has now been worked out. No time-traveling phantom particle, sorry.
as it is a lot harder to notice or prove that something hasn't happened during a persons lifetime.
It's very difficult to prove a lot of things in science. That's what we have laboratories for. By direct manipulation of creatures with short life spans, scientists can observe how genetics work and test theories on how a positive mutation might be induced. With more and more computing power at their disposal, it should only be a short time before we're able to begin very fast computer modeling of entire genomes.
The first surprise has already come in the form of shorter gene sequences than predicted. It wouldn't surprise me if biologists find themselves completely rewriting the books on genetics in the near future. If there's anything that physics has taught me, it's that the Universe is not one to easily give up its secrets.
I apologise if you took offence at being called a troll earlier, I meant it in the original sense of someone who deliberately posts inflammatory content in the hopes of sparking a debate. (anti-evolution posts certainly qualify on Slashdot)
Apology accepted, but with one caveat. I never claimed that evolution was false. It's not very nice to stomp all over the theories of others just because you don't like it. My specific complaint was the outright confusion caused by the "micro-evolution" renaming of environmental adaption. I don't like it, because no link has been demonstrated. This gives people the mistaken impression that they can write software using genetic adaption type algorithms, and actually produce software that "evolves" in the macro-evolutionary sense. The truth is that the software doesn't go anywhere. It merely "adapts" to the situation at hand until the most powerful solution can be found.
As for calling you a creationist.... well are you are aren't you, you claim to not believe in evolution, so how do you think we got here?
Ac
Javascript + Nintendo DSi = DSiCade
No new algorithm will be generated through the "genetic breeding". Only existing algorithms will rise to the top.
For the last time, please educate yourself about genetic algorithms. Take a textbook example:
Let's evolve an algorithm to follow a trail of markers on a 2D grid.
The population is made up of randomly generated finite state machines, whose input is whether there is a marker in front of it and output is whether to move forward or turn left or turn right.
The fitness of each finite state machine is measured by how many markers it passes over during the course of 200 moves. Each marker only counts once.
During each new generation, the usual crossover and mutations occur.
Now, we start off with randomly generated algorithms. We end up with new algorithms that actually follow the markers. At no point did the programmer input an algorithm to follow the trail. How can you argue that no new algorithms are created with genetic algorithms?
That logic just doesn't make sense. If I type a random bunch of characters, I *could* produce the works of Shakespeare.
Yes, you could. By attempting random combinations, you could achieve the proper result in about 3 times the amount of time it will take for the universe to reach its heat death. You can test the theory with this simulation. A quick search on Google also presents this math for the traditional Million Monkeys problem.
If I change a random gene, it *could* be beneficial.
It could be beneficial. Not because we know that it could be beneficial, but because we don't know enough about genetics to positively state that it absolutely would not be beneficial. This is similar to saying that it could be possible to time travel and violate causality. It could be true, but we don't know enough to make the statement one way or another.
How can you possibly argue that a random change can't be beneficial ever?
Actually, I argued that existing evidence suggests that random changes are never positive. While a heap of data doesn't qualify as proof, there is current zero evidence of an absolutely beneficial mutation. Until such a mutation can be isolated and identified, evolution by mutation is pure speculation. If evidence actually existed, then it could graduate to a theory.
Mutations that are positive are also far less obvious. If your child is slightly stronger than normal, would you notice?
Most people do notice. In fact my first child was born rather strong. It took four people to hold him when he was getting his shots. The question is, was it a mutation, or was it existing genetics? By sequencing samplings of the population, we are starting to isolate enough genes that we may soon be able to answer that. We've already identified genes for color, strength, obesity, etc. With any luck, we'll soon be able to know if positive mutations are a real phenomenon, or just wishful thinking.
If they have a crippling disability, on the other hand, you sure will.
Of course. "Bad" mutations are almost always obvious. The problem is that if positive mutations exist, then negative mutations must outweigh them by an overwhelming margin. We can induce mutations, but they always turn out either negative or settle as dormant genes.
Again, why? One example from agriculture doesn't prove anything. All it proves is that we don't know how to do it properly. That *may* be because it isn't possible but not necessarily.
It's hardly "one example". My colleagues of which I speak are PHDs in the field of genetics. They were not only educated in the field, but they dealt with genetics on a daily basis. All genetics data to date suggests that genetic manipulation is a zero sum game. That's why Eugenics can't work. You'll never get a race of "super-men", you'll only create a race of people who have traits you desire. How shallow are you willing to make your "super-men"? If you breed for intelligence, then strength may suffer. If you breed for strength, then intelligence may suffer. If you breed for height, then your people will have less lifting power. If you breed for shortness, then leverage will suffer. So on and so forth.
Ah, but i believe 2 in 4 are protected from malaria? Evolution has selected for the sickle-cell property because it provides an overall benefit for the group. African Americans still have the mutation but it is fading out (hence the lower prevalence). Evolution takes time, especially now that modern medical practices mean that many more sufferers survive.
Yet again, if this is a "good" mutation, why does it need to be bread out of the gene pool? "Micro-evolution" speculates that positive mutations crucial to the development of an organism must have occurred and remained. You didn't acquire lungs just so they could be bread out in six generations down the line, did you? No! The mutation must have become a standard part of the genome.
Javascript + Nintendo DSi = DSiCade
BTW, you may be interested in knowing that the Monkey Simulator has reached an all time record of 17 letters of "Troilus and Cressida"!
After 9,151,670 billion billion monkey-years, that is.
Javascript + Nintendo DSi = DSiCade
Browsers don't always bother checking DNS expiration times and rechecking the address when they time out. This used to be less of an issue on IE because Windows used to bluescreen several times a day (:-) but these days it can be a problem.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks