First Digital Simulation of an Entire Life Form

An anonymous reader writes "LiveScience is reporting on what appears to be the first digital simulation of an entire life form. Researchers created more than a million digital atoms to reverse engineer the satellite tobacco mosaic virus, a relatively simple organism. But is it really a life form? From the article: 'Viruses are tiny bundles of protein and genetic material that straddle the line between life and non-life. Many scientists prefer to call them "particles" because even though they contain RNA or DNA like other lifeforms, they can only replicate inside other living cells.'"

Comments

First Digital Simulation of an Entire Slashdot DUP

[TripMaster Monkey]

Story is a dupe...original story can be found here.

I don't get it

[Wizardry Dragon]

Every single form of life has precursors to life. So the precursor here is different. What's the big fuss?

Re:I don't get it

[Anonymous Coward]

In the beginning, there was nothing...which exploded. Sounds like flying spagetti monster to me.

Re:I don't get it

[Anonymous Coward]

Every single form of life has precursors to life. So the precursor here is different. What's the big fuss?

Well, the "big fuss" is not that viruses may or may not be life. (That debate has been going on for quite some time.) The big fuss is that this is the first time something this large has been simulated in atomic detail on a computer.

Getting back to your question, though, a virus is different from a "living" organism, in that it requires another living organism to replicate. You might protest, saying that animals eat other living organisms, but they don't *need* other living organisms - your quarter pounder with cheese was living at one point, but isn't at the time you eat it. Even if your food is alive when you eat it, your digestive system does a respectable job of killing everything off and breaking things down before taking up and using it.

On a more technical level, most if not all viruses do not contain the enzymes needed to do a full cycle of replication on their genome (some exceptions, like non-retrovirus RNA viruses), nor to do the DNA-mRNA transcription, nor (most importantly) to do the mRNA->protein translation. Nor is there much in the way of genes for nutrient aquisition or processing. There is little to nothing going on in a virus particle, metabolism wise. It's merely a packet of genes, ready to take advantage of some other organisms accumulated nutrients and information processing machinery.

The "only" thing which makes the life/not-life distinction fuzzy is that it is capable of directing production of additional copies of itself, such that it fufils all the requirements for evolution*. I put "only" in quotes, because it is a pretty big thing, and is arguably the only requirement for something to be "life."

As you could possibly guess, I come down on the "not life" side, but there is a whole lot of "but WTF does it matter if it's living or not? A virus as a virus is interesting enough in it's own right. Life/Not life is an rather arbitrary distinction anyway" in there too.

*If you don't believe in Evolution, don't bother wasting time to understand this post. Just consult your holy book as to if a virus is alive.

Re:I don't get it

[ShaggyBOFH]

...they can only replicate inside other living cells.
That's the only way I've managed to do it and my replication is now 3.5 years old.

-------

Re:I don't get it

[Wizardry Dragon]

As you could possibly guess, I come down on the "not life" side, but there is a whole lot of "but WTF does it matter if it's living or not? A virus as a virus is interesting enough in it's own right. Life/Not life is an rather arbitrary distinction anyway" in there too.

Im on the opposite side of the fence, but I have the same sentiment. Tehe.

What I don't understand is how virii can be considered non-living when other parasites are.

Re:I don't get it

[m50d]

What I don't understand is how virii can be considered non-living when other parasites are.

They're relying on their host for basic life functions, such as reproduction (OK, bad example as you have flowers/bees and so on) and even respiration. Every life form relies on something external for a food source, that's fair enough, but if you call a virus alive then you might as well say genes are life forms in their own right.

Re:I don't get it

[bitspotter]

I'd venture a guess that, while bees & tapeworms breakdown things from their original structures in order to fuel their own processes from smaller components, viruses don't bother breaking anything - they just use the existing biological machinery on hand more or less as they were intended - just, for their own purposes.

In other words, they're hackers.

Re:I don't get it

[modecx]

What I don't understand is how virii can be considered non-living when other parasites are.

Because to be alive, a biological entity has to be able to do at least the following: respirate, metabolize, grow, and reproduce. Parasites are organisms that do all of the above. Viruses are not parasites. They are viruses, and they do none of the above, therefore they're not life! Weren't you awake during your middle school science class?

Re:I don't get it

[shawb]

I personally just put viruses firmly in the "gray area between living and not living" because those are arbitrary distinctions: nature always finds a way to find exceptions to the niches that man creates. Not to say that our classifications are pointless, we just have to realize that there are always going to be things in that gray area. This can be shown more dramatically in other biological definitions as well: when is a fetus "alive?" What is the exact point where two diverging groups are no longer the same species? All questions that have legal ramifications which essentially require a precise definition in order to make decisions as impartial as possible, but some things will straddle the line no matter how precise you try to make your definitions.

Re:I don't get it

[Timothy Brownawell]

The "only" thing which makes the life/not-life distinction fuzzy is that it is capable of directing production of additional copies of itself, such that it fufils all the requirements for evolution*. I put "only" in quotes, because it is a pretty big thing, and is arguably the only requirement for something to be "life."

Computer viruses can also direct the production of additional copies of themselves...

Tim

Re:I don't get it

[Zarel]

Computer viruses can also direct the production of additional copies of themselves...

I have an even better analogy: So are instruction manuals telling you how to make instruction manuals. And I don't see anyone calling instruction manuals alive.

Re:I don't get it

[Wizardry Dragon]

You could, oh I don't know, *think* about scientific theory rather than regurgitate information from your schooling. You could also think before posting a needless troll, but I digress.

Re:I don't get it

[shaitand]

Perhaps the better solution to this dillema is to realize the debate can not be settled because "life" is not a thing. It is a label. Humans invented the label and since it is an artificial construction its scope is equally artificial.

There is no such thing as "life" we invented a classification without defining it and therefore we have a debate. The only reason we even find it to be important is that we are still trying to come up with excuses to think of ourselves as something more than a random cluster of protein soup.

Re:I don't get it

[zenhkim]

I vaguely remember someone proposing/using the terms "bio-agent" or "proto-organism" to describe a virus, for the simple reason that a virus doesn't have all the necessary characteristics to qualify it as "alive".

BTW, things do not have to be alive to undergo evolution; they only need to go through a process of staged "propagation" (i.e.: reproduction, even if done by outside means) where each successive stage or "generation" is related to the previous ones yet may exhibit new and unique qualities. Computer engineering and software design are obvious examples: each new wave of technology resembles the older ones yet is markedly different, and if you go *way* back to the earlier generations the contrast is shocking. We're simply accustomed to using the word "evolution" primarily in regards to living creatures, when in fact it applies to other things equally as well.

Re:I don't get it

[Wizardry Dragon]

That's possibly the most intelligent post I've read on Slashdot. I salute you!

Re:I don't get it

[HyperTiger]

I'm not too sure we should have reproduction as one of the defining factors of life, else sterile people or eunichs might not be considered alive. Not really trying to be funny, but is something I dont understand why we say should be a reason why viruses should not be considered alive. Metabolism seems more reasonable though.

(As a side note, I'll never forget someone actually thinking I meant eunichs when I said I work with unix.)

Re:I don't get it

[modecx]

Right, well you're thinking in terms of the broad picture, and there's nothing wrong with that... But you're thinking too big in this instance. Those ideas don't really apply to mules or worker bees or ants, or any other number of living creature that doesn't normally (or can't) reproduce on it's own as an individual animal, and it's not meant to. The thing is, if you look at any animal or plant (or fungus) down at the level of their cellular components, all of those things are happening everywhere, including reproduction and growth, and all of those living components make up the or plant. Once you've realized this, it's not too far of a leap to conclude that these rules apply quite well to any sort of life, especially any kind of life that can be found on this planet. It doesn't matter if the life is silicon based, energy based, or what have you, the rules still apply! For instance, the life form will still have to respirate, and metabolize--whether that means eating stray hydrogen atoms out in space, or consuming the silicon equivalents to organic compounds for energy, or doing any number of things that we can only imagine. When we encounter something that dosen't follow the rules, we will have to change our ideas, until then, these rules work very well.

Viruses aren't alive because they're raw DNA or RNA with a crunchy protein wapper. They are objects that try to force cells to execute a program which is foreign to them... An analogy: a cell is a computer, and that computer runs programs--it includes many of it's own, some of which help it not break down, some regulate power usage, or interact with whatever mediums it has access to. A virus is a program that can get into the computer and cause the computer to do any number of things, usually like replicating the virus, and causing the computer to break down... That doesn't mean that a virus can't actually fix a problem existing on a computer; it just means that a virus is usually a bad thing--right?!

Just because a virus exhibits one single trait of life doesn't imply that life caused it! This is like saying "because a rooster always crows at sunrise his crowing causes the sun to rise"! Correlation does not imply causation! Calling a physical, real world, microscopic virus living is like saying that a computer virus is intrinsically a computer itself! That doesn't make sense, does it?! This is not the nature of the existence of these agents. Cells are hardware and viruses are software. Computers are hardware, and computer viruses are software. They are separate and distinct things, and calling them the same just doesn't follow logically!

Re:I don't get it

[modecx]

You could, oh I don't know, *think* about scientific theory rather than regurgitate information from your schooling. You could also think before posting a needless troll, but I digress.

If we all had to think of everything ourselves, we'd fucking still be wandering around the east bumfuck plains of Africa chasing some stupid antelope that's laughing its ass off because everyone else was back at the cave trying to invent the goddamn spear all over again. I dunno about you, but I like to learn from someone who knows something more than I do. The fact is, some of us are better at thinking than others, and I like to take advantage of them, because it means that I'll learn much more than I could have otherwise. It's not my fault that your statement of ignorance was begging to be enlightened--and re: the troll accusation: It takes one to know one! Heehee *snort*

p.s Learning is the way of the human, and teaching is the way of the geek. If you don't like this arrangement, I encourage euthanasia.
p.p.s. I don't make monkeys, I just train 'em.

Re:I don't get it

[Kenshin]

Your argument is correct... but in this case, we have documented outside interference: a "god" factor, if you will.

Humans programmed this thing to do what it does, whereas "life" as we know it is of still-unknown origins.

But then again, what the hell do I know about anything?

Re:I don't get it

[rtb61]

Of course as a random cluster of protein soup, why should you be able to understand some of the bit more complex metaphysical aspects of life the universe and everything.

Perhaps we just not as far up the evolutionary chain as we would like to think and some things are just beyond our under standing.

Not necessarily as bad as the religous fundamentalists ego of course, if they can't understand something, then only the supreme being of the universe could understand it, oh my ;).

Re:I don't get it

[Jasper__unique_dammi]

I read somewhere (maybe in another slashdot story?) that the "livelyness" of a object was defined by the improvement of the chance there will be a similar object if there is the first object to start with.

Livelyness(object)=P(second_object_exists|object_e xist_before)/P(object_exists)

in words: the "livelyness" of an object is the chance a second exists when another existed before divided by the chance it turns up from no-where.
Hmmm I would fall for that definition.
the chances may vary in different environments with this definition, which is also nice (the flu is alive in our body but dead without)

Re:I don't get it

[Anonymous Coward]

There is no such thing as "life" we invented a classification without defining it and therefore we have a debate.

What?? That's one of the dumbest things I've ever heard. Life does have a definition. If you had taken even a simply high-school science class you would know that living organisms are distinguished between non-living objects by several factors. Some of the more important factors are:

1) The ability to reproduce themselves
2) The ability to maintain their orderly structure through physical and or biochemical means

By using these characteristics, you can see how a rock differs from a tadpole. So stop spewing that "we're nothing more than protein soup" bullshit because you sound just as fanatical as creationists.

Re:I don't get it

[Wizardry Dragon]

Those who can't do, teach.

Re:I don't get it

[shaitand]

"why should you be able to understand some of the bit more complex metaphysical aspects of life the universe and everything"

I am certainly not one to claim to know everything there is to know about the universe. I would not even venture to claim that mankind has such knowledge; man may never have such knowledge. There are however no credible observations that would reasonably suggest that metaphysics will ever need to be invoked to explain the human machine.

Re:I don't get it

[stonecypher]

This can be shown more dramatically in other biological definitions as well: when is a fetus "alive?"

This has always struck me as more of a cheerleader call from anti-abortion-rights groups. Single celled organisms are alive; why shouldn't foetii be? I mean, they're the combination of two different lives; the eggs are alive; the sperm are alive. At no point is there a death involved.

For whatever it's worth, I'm pro-choice. I just don't think that fetuses are ever anything *but* live. Making generalizations like that is, in my opinion, an attempt to use a slogan to bury the difficult moral issue of whether we can treat the parent as superior to the child, when the child is still as simple as things like bacteria colonies which we wipe out without a thought.

Watch TV and see how we advertise bright orange death serums, genociding baccilae as a matter of course.

Old news

[Anonymous Coward]

Been done...

Re:Old news

[mgabrys_sf]

Very much so - I blow up life forms every day on RTCW.

Life is not a binary distinction

[wsherman]

But is it really a life form?

Language is digital (as opposed to analog) in the sense that you either use a word in a sentence or you don't. You can either use the word "life" in a sentence or not but you can't use a fraction of the word ("li" or "fe" don't mean fractional life - or anything at all for that matter). This creates (willful?) confusion in the minds of people who are very focused on a literal interprtation of language based laws and moral codes that "life" is a binary distinction.

The reality, however, is that the word "life" refers to a whole variety of concepts. There are all different ways of being alive and there are all different levels of being alive. Certainly we can find examples of things that are very "alive" just as we can find examples of colors that are very "blue" - but that doesn't mean every color is either pure blue not blue at all and it doesn't mean that something is either completely alive or not alive at all.

Going way off topic, the whole "life begins at conception" is what we in the sciences refer to as "not even wrong". After all, it's kind of hard for dead people to have children. If you really want to talk about when life began it would be at the big bang when matter developed the properties that cause it to form into complex self-replicating patterns over very long time scales.

Re:Life is not a binary distinction

[TheRealMindChild]

Well said.

Re:Life is not a binary distinction

[Red Flayer]

"You can either use the word "life" in a sentence or not but you can't use a fraction of the word ("li" or "fe" don't mean fractional life - or anything at all for that matter)."

Oh, without capital letters, they mean nothing -- but I know quite a few chemists who'd dispute that Li and Fe are meaningless. :)

To get on-topic, I think that humans constantly categorize and assign labels to things as either a member of a group or outside it, which IS binary.

That creature is a fish|not a fish. That creature is a mammal; or it lays eggs and has a bill, so it is a bird (ummm, bad example, on second thought). That rock is igneous; or it is not. That tree is deciduous|not deciduous.

What is the point of defining something if the definition does not allow us to use it to categorize? Things like this virus, and viruses in general, raise the debate over what is life|not life. And that debate can stimulate greater knowledge, and greater understanding, by challenging our assumptions and our definitions... so I'm all for making distinctions when we can.

Re:Life is not a binary distinction

[Anonymous Coward]

I think what you mean to say is that the definition of "life" is subjective. All that stuff about "binary" and "fractional" is besides the point, and not necessarily even true.

Re:Life is not a binary distinction

[MindStalker]

To even have this discussion shows you don't know what a virus is. Effectivly a virus is a set of genetic programming code floating inside whats basically a floating syringe. Some do have mechanical like parts on the outside to help them attach and insert their code. They are much like computer viruses. When simply a file they have no life whatsoever. Thought once viruses enter a cell you would definatly consider the new cell that has the viruses code a lifeform, weither this new cell is a "virus" or not is another question, but it is a virus producing factory..

Re:Life is not a binary distinction

[fossa]

Language is digital (as opposed to analog) in the sense that you either use a word in a sentence or you don't. You can either use the word "life" in a sentence or not but you can't use a fraction of the word ("li" or "fe" don't mean fractional life - or anything at all for that matter). This creates (willful?) confusion in the minds of people who are very focused on a literal interprtation of language based laws and moral codes that "life" is a binary distinction.

I believe the word you are looking for is "binary", not "digital"... Sure, many words lack shades of grey, like your example of "life". And I would agree that people look at many issues in black and white that are actually shades of grey. Many words, however, do not come in only black or white. For example, I can't decide if your post is subtley clever, merely ignorant, intentionally wrong, or downright malicious. Am I feeling the blues, melancholy, depression, misery, or utter despair? Is the color dark grey, ebony, pitch black, or jet black?

I do not believe most people do see life in black and white. They may indeed classify objects into the binary categories of "alive or not", but they certainly do not treat all life as equal. Most people believe murdering another human is unthinkable, would not eat their pet dog, but don't think twice about buying steaks, swatting flys, or crushing grass as they walk across the lawn.

Re:Life is not a binary distinction

[x2A]

"but you can't use a fraction of the word"

you've obviously never heard one of bush's speeches ;-)

Re:Life is not a binary distinction

[wsherman]

To get on-topic, I think that humans constantly categorize and assign labels to things as either a member of a group or outside it, which IS binary.

My definition of science would be that it is an attempt to organize and summarize mutually agreed upon factual observations. In that sense, categorizing and assigning labels is very important. On the other hand, the organization and summarizing must be in agreement with the factual observations. If one had a collection of rational numbers (eg. floating point) and one categorized them as either binary 0 or 1, such a categorization would not be particularly accurate.

Re:Life is not a binary distinction

[sholden]

That creature is a fish|not a fish. That creature is a mammal; or it lays eggs and has a bill, so it is a bird (ummm, bad example, on second thought).

Curse those platypus...

Re:Life is not a binary distinction

[alicenextdoor]

OK, so life is difficult to define (but we know it when we see it!). But things all life known to date share are the use of DNA / RNA / proteins to encode and transmit the information needed to construct copies of the original. So I think viruses have a good claim to being alive. Further, it is pretty generally agreed that viruses have evolved from more complex intracellular parasites (and there are many of those, and none of them are considered not to be alive). And they are still evolving - just look at the current bird flu scare. Is it possible to evolve from "alive" to "not alive"?

Re:Life is not a binary distinction

[jacksonj04]

Metadata tagging!

Human - Tags: Mammal, Biped, Replicates, Respires, Significant Movement, Life Birth etc. etc.
Sunflower - Tags: Plant, Replicates, Leafed, Fixed-Point Movement, Seeds etc. etc.
Virus - Tags: Replicates, Movement etc. etc.

Re:Life is not a binary distinction

[wsherman]

I think what you mean to say is that the definition of "life" is subjective. All that stuff about "binary" and "fractional" is besides the point, and not necessarily even true.

The ability of an entity to make copies of itself is not really all the subjective. The ability of an entity to respond to its environment has varying degrees but that really isn't all that subjective either. The ability of an organism to maintain a self-sustaining symbiosis between it's various parts (eg. organs) is really not all that subjective either. Finally, being self-aware is not something that humans really understand but that doesn't mean that it is subjective.

Different entities have these characterisics to varying degrees. Any distinction between life and not life is not so much subjective as it is arbitrary.

Re:Life is not a binary distinction

[glwtta]

To get on-topic, I think that humans constantly categorize and assign labels to things as either a member of a group or outside it, which IS binary.

I think the original point was that this distinction is usually arbitrary, and only derived from evidence so far observed; so we are constantly finding new things that do not fit neatly into the binary distinction.

That creature is a fish|not a fish. That creature is a mammal; or it lays eggs and has a bill, so it is a bird (ummm, bad example, on second thought).

Actually, things changed a bit since Linnaeus' day. We classify organisms by genetic ancestry, not phenotypic traits (and have come to realize that all such classification is essentially arbitrary).

So, if we've traditionally grouped a bunch of organisms together based on some trait, and then discover that 99% of them actually share a common ancestor, while the remaining %1 have developed an identical (even if highly distinctive) trait in parallel - should they be considered as part of the original group? Again, the answer is arbitrary.

Re:Life is not a binary distinction

[B. Pascal]

Hello wsherman:

I like to continue on the discussion of the usage of words and definitions, though it may be off topic.

Rather than using "digital" or "binary", I think the word you are looking for is "discrete". The word discrete refers to the concept of something inseparable. "Binary" specifically refers to a system with 2 values per digit, while digital refers to numerical subjects.

As you correctly pointed out, the surface of the word is discrete. I.e. "Li" and "fe" carries no meaning. That being said, the definition and meaning of "Life" is compositional. The meaning of "life" is a complex one consisting of combination of other meanings and refers to a variety of concepts. Adding to this complexity, every observer has his/her own definition of what "life" means (or what a particular word means).

Unless the specific senses of a word's definition from two observers result in inconsistency, the differences between the two definitions would go undetected. This is the communication problem. Similar problems arise when talking about abstracted concepts such as "life", "love", "God", etc.

Finally, rather than asking if the simulated object in the article constitues "life" according to your definition, it's more productive to find out how the author of the article defines "life". Knowing how your definition and his definition differ can aid in understanding of the article.

Such is the study of semantics of language.

Cheers.

B. Pascal.

Re:Life is not a binary distinction

[x2A]

flamebait *lol* so there's actually ppl out there who'd disagree? Well if could mod the moderation, I'd mod it +1 informative, cuz I really didn't know!

Re:Life is not a binary distinction

[Dr. Donuts]

The problem with categorization is that you must define parameters to which something is agreed upon to be of the same category.

However, the definitions of those parameters are often explained using language. Language is ambiguous, interpretive, and circular. Other words are used to define a given word. A word in one language may have no easily explainable equivalent in another. Language is a poor platform for use in creating definitions and categorization parameters.

A second point to be made is that things are often incremental. As such, increments can often make for difficulty in distinguishing one thing from another.

The problem with putting things in a box is that nature doesn't like boxes. The second is that a box only exists in your mind.

Re:Life is not a binary distinction

[MindStalker]

Further, it is pretty generally agreed that viruses have evolved from more complex intracellular parasites (and there are many of those, and none of them are considered not to be alive).

Good point but it didn't so much evolve from the parasite as it was created by the parasite as an aid. Think of this the same way as me creating a robot to help me get laid! :)

I just can't see a packet containing programming information as life. Its definatly part of the cycle of life, maybe we could consider it a life weapon.

Re:Life is not a binary distinction

[MindStalker]

Ok after reading your link I see there is a difference in terminology from what I understood. Currently a Virus is a cell that has been infected by a "virion" The virion is not alive, but the Virus is. I can definatly agree with that. Kinda like how my SO isn't a computer, but how a computer with an OS is a totally different beast beholden to the OS.

Re:Life is not a binary distinction

[MindStalker]

HAHA that should have read OS not SO. My Significant Other isn't a computer.

Re:Life is not a binary distinction

[izomiac]

And there's an example of the two discreet categories the original poster argued don't exist. A virus, alive or not, would certainly be placed in a different category than, say, plastic. What about mitochondria? I doubt they can live on their own, so are they also in the "non-alive" category? How about our own cells, they (AFAIK) can't live without mitochondria, so wouldn't they be in the same category? (Or perhaps mitochondria are the actual life forms and eukaryotes are the ones that aren't alive.) The point is that the word "life" is binary in that it only has two possible states whereas in nature there seem to be multiple levels. Personally, I'd go even further and say that the word "life" describes an abstract concept where there isn't a natural distinction. I.e. the difference between metabolism and other chemical reactions is merely complexity, which is completely subjective, and therefore no there's not really any distinction.

Re:Life is not a binary distinction

[shaitand]

A categorization is actually counter productive if nobody bothers to define it. The problem with the "life" label is that it has never been defined. There is a sensible definition that was basically given by the GP. However, those who want to believe that living beings have "souls" or use other excuses to assign importantance to our own form of meat cluster and similar forms do not want a simple classification for life.

If every self replicating and evolution pattern becomes a form of "life" then "life" is no longer sacred.

Re:Life is not a binary distinction

[afaik_ianal]

Platypus - Tags: Mammal, Bird.

Re:Life is not a binary distinction

[AlterTick]

To get on-topic, I think that humans constantly categorize and assign labels to things as either a member of a group or outside it, which IS binary. That creature is a fish|not a fish. That creature is a mammal; or it lays eggs and has a bill, so it is a bird (ummm, bad example, on second thought). That rock is igneous; or it is not. That tree is deciduous|not deciduous. What is the point of defining something if the definition does not allow us to use it to categorize?

The whole drive to categorize and pigeonhole all of science is a very 19th century notion, like Newtonian determinism. The whole ridiculous kingdom-phylum-class-etc. system is a twisted mess-- and until a thorough genetic catalog can be compiled, it will remain so. It's important to remember that just because we humans like to categorize things doesn't mean we categorize them correctly...

Re:Life is not a binary distinction

[LightCecil]

Actually, from what I remember about viruses, they are NOT mobile. They have no flagella or cilia to self-locomote. Rather, they allow whatever medium's natural motion move them. Blood motion and sneezing being the two bodily motions it capitalizes on the most.

Re:Life is not a binary distinction

[alicenextdoor]

You're doing better than a lot of us, then :)

Re:Life is not a binary distinction

[FireFury03]

being self-aware is not something that humans really understand but that doesn't mean that it is subjective.

Seems quite subjective actually. And I'm not convinced that "self awareness" has much to do with conciousness - one could say that a computer is self-aware (the system is "aware" of the various parts of itself - memory, disks, etc) but you wouldn't say it's concious.

Conciousness is the thing that's _really_ hard to define, and how do we even know whether something is concious or not if we are unable to communicate with it?

Re:Life is not a binary distinction

[lawpoop]

...so I'm all for making distinctions when we can."

this response is partially to parent and partially to grandparent.

First off, language is digital, but not necessarily binary. Yes, "this is a fish" or "this is not a fish", but also "This is a cod fish" or "this is a blue fish", and "this is kind of a fish" and "this is a strange fish" and "this is fish-like". So the moral is that people use modifiers (in my examples, adjectives) on thier categories.

As far as making distinctions, some are useful, some aren't. I think if we made all the distinctions we could, we'd be lost in a morass of useless and irrelevant data -- i.e. what letter does this virus' name begin with? I always ask myself, what do we need to know this for? In the case of viruses, what does it matter if they are 'alive' or 'not alive'? We can study them and learn a lot about them and what they do without wondering whether or not they are alive. In fact, off hand, I can't think of a reason why we would need to know that.

Re:Life is not a binary distinction

[stonecypher]

Er. This is falsely reductionist. It's not that a tree is deciduous or not deciduous; rather, a tree is deciduous, evergreen, dioecious, gymnofruiting or so on. The creature isn't a fish or not a fish; rather it's a fish or a bird or a dog or a shark. The rock might be igneous, volcanic, sedimentary, ocyllic, crystalline, meteoric or obsidianic.

Sure, we tend to enumerate things, but the comparison you're actually making is that something can have or not have one particular characteristic. That's tautological: of course something which either is or isn't can be referred to as boolean (not binary, which means counting integers in radix two.) However, just because you can say that something is red or not red doesn't mean that its color characteristic is boolean; rather that the color characteristic happens to be any one given label is boolean.

The debate over whether something is alive or not suffers this same tautology. There are other things than alive and inert. Biologists don't wrestle with this issue any further than the author's pen; if one learns one's science from somewhere other than Wikipedia, one finds the term "proto-life," to refer to autonomous molecules which reveal some but not all of the characteristics of independant life. Viruses fall neatly into proto-life: they can perform several of the capacities of independant life, and generally reproduce by overriding the parallel native capacities in a host organism. The debate over life comes in when one sees gigantic viruses which are nearly cellular, greying the boundary between life and proto-life, or things like prions and satellite nucleics, greying the boundary between proto-life and inertness.

That latter example bears some explanation, because it can be confusing to refer to proto-foo as something exhibiting some of the characteristics of foo, then to refer to something else as greying the boundary between partial and no implementation. The problem with prions is a matter of perspective and belief. There are good arguments that prions are just long-term runaway chemical chain reactions, rather than independant evolved agents (though, there are also good arguments that such a distinction is nonsensical.) The best explanation I ever heard of the conundrum was from someone I'd just spent explaining the problem to for several hours - not a scientist, but an old and dear friend who was a car mechanic.

He proposed that in the future, I describe these things in terms of oil iron particulate. See, in your car, when you run the car, the metal in the pistons strikes the surrounding metal constantly under high force and temperature. The purpose of engine oil is to lubricate the space inbetween the piston and cylinder so that the parts can fit together more tightly, meaning that the repeated shock has less room in which to build up parallel force and therefore can do less damage. Put more simply, the better the fit, the less rattle, and therefore the less wear and tear.

The way wear and tear actually presents in engines is as microscopic iron flakes breaking off of the engine and into the oil. This is the reason for oil changes: the oil doesn't become worn out. Rather, a buildup of iron powder slowly turns the oil into an abrasive - liquid iron-based sandpaper, if you will. As the iron buildup gets worse, so does the breakaway rate, and thusly you need to frequently change your oil or it gets out of control.

Now, here's where his metaphor comes in. One of Jeremiah's hobbies was rebuilding classic cars, and performing modifications. He described the pulling of an air supercharger from an old Mustang, and putting it into his Frankenstein (it was a car build from the parts of several other kinds of car.) The supercharger is in direct contact with the oil base of an engine. So, you have to be careful during transplant to flush the old oil, which may have the iron powder buildup from an old engine, before you put into your new engine.

In that respect, we cross the exact same boundary that prio

Simulating intelligence?

[4D6963]

If they can simulate something else than a virus (because I don't think viruses are intelligent) could they by this way obtain intelligence by simulating an intelligent animal?

Re:Simulating intelligence?

[Paraplex]

I think the real difficulty here is that an intelligence (or any life form really) isn't really alive without input & output. You can't just simulate (based on any physical model) an isolated life form because it would just sit there. You need to simulate the environment it inhabits.

The line between the organism and the environment is very blurred. I tried to write a cellular autonoma of a weather/ecology system at once stage and was overwhelmed with the sheer number of variables which would have to be included to make it complete. You essentially can't leave anything out.

Don't mean to sound like a buddhist or anything, but everything is inextricably connected. You'd need to start by making a 500 million atom small environment or some such.

Re:Simulating intelligence?

[egomaniac]

If they can simulate something else than a virus (because I don't think viruses are intelligent) could they by this way obtain intelligence by simulating an intelligent animal?

Of course. It would take an absolutely colossal amount of computing power, but given sufficient resources and a complete understanding of the basic physics and chemistry involved (neither of which we have yet) you could absolutely simulate a living creature, and the simulation would be intelligent. There have been many sci-fi stories that have used this basic concept. In fact I expect the first intelligent machine will attain its intelligence by simulating a living brain (although at a much higher level than individual atoms).

If we assume that all physical processes can be simulated by a computer (given complete knowledge of the laws of physics), which seems to be a safe assumption, the question boils down to "is intelligence a physical process?" Everything we know about the brain's operation says that the answer is a resounding "yes" -- and if intelligence is merely a manifestation of the physical operation of the human brain, then there is nothing about it that can't, at least in theory, be simulated.

Re:Simulating intelligence?

[Anonymous Coward]

just to let you know... Google's way ahead of you.

~ anon (with good reason).

Re:Simulating intelligence?

[Bob3141592]

If they can simulate something else than a virus (because I don't think viruses are intelligent) could they by this way obtain intelligence by simulating an intelligent animal?

Would you consider a bacterium intelligent? I bacterium is several orders of magnitude more complex than a virus. How about an ameoba? Again, several orders of magnitude more complex than a bacterium? Perhaps a jellyfish? A nematode?

The pure simulation method is unlikely to ever be used in developing an artificial intelligence. And that's just as well, as it wouldn't teach us anything more than just physically reconstucting the object atom by atom. Finding alternative routes to the goal is more likely to be interesting and useful.

Re:Simulating intelligence?

[spun]

Intelligent animals are orders of magnitude more complex than viruses. In some theories, if there was a supercomputer powerful enough to simulate all the atoms in your body, it would be conscious. Others invoke quantum uncertainty to explain consciousness, and unless the computer were some sort of quantum computer, they say it would not accurately simulate consciousness. Others claim that we don't even know what consciousness is, or why the color red looks red. These people would claim that even though a computer simulation might be perfectly accurate, it wouldn't in fact be conscious or have the same sorts of internal experiences that conscious entities have. It might even claim to have the same internal sensations we do, however, although I can program a computer to state "I feel depressed," that doesn't mean it feels the same way I do when I'm depressed.

In short, to answer your question honestly, we don't yet know.

Re:Simulating intelligence?

[tdvaughan]

The equations that govern the behaviour of nuclei and electrons are extremely complex and have currently only been solved for a handful of atoms interacting with each other. For this reason, the simulation mentioned in the article can only be an approximation to real life. The same restriction would apply to larger organisms, only much more so.

Re:Simulating intelligence?

[JesseL]

If they can simulate something else than a virus (because I don't think viruses are intelligent) could they by this way obtain intelligence by simulating an intelligent animal?

It is theoreticaly possible (although it would be many orders of magnitude more difficult than the virus simulation), but the real question is - would such an experiment yield much insight into the nature of intelligence or give us any foothold toward developing an artificial intelligence suited to our needs? While it would enable us to examine the state of all the neurons of a brain at any given moment, it may not necessarily allow us any more comprehension of how the system we watched related to intelligence at the macroscopic level.

Re:Simulating intelligence?

[nbert]

If we assume that all physical processes can be simulated by a computer (given complete knowledge of the laws of physics), which seems to be a safe assumption, the question boils down to "is intelligence a physical process?"

I believe that we should worry about size and power consumption before we get to the rather philosophical aspects. Afterall the human brain is still by magnitudes more complex than any computer we can build nowadays (not taking into account computers bigger than our solar system and/or computers consuming more than the power of a couple of suns).

But to get back to more basic or philosophical considerations: Maybe we're simply not able to create structures more complex than ourselves...

Re:Simulating intelligence?

[JesseL]

But to get back to more basic or philosophical considerations: Maybe we're simply not able to create structures more complex than ourselves...

I read a quote somewhere related to that idea. It was somthing to the effect of "in that case Einsteins mother must have been one hell of a physicist".

Re:Simulating intelligence?

[Anonymous Coward]

Would you consider a bacterium intelligent? I bacterium is several orders of magnitude more complex than a virus. How about an ameoba? Again, several orders of magnitude more complex than a bacterium? Perhaps a jellyfish? A nematode?

"Several" understates it. By back of the envelope calculations, a human is about 25 orders (10^25) more complex than a virus (1 m^3 versus 1 nm^3 - number of atoms being roughly proportional to volume). Even a nematode is more than 15 orders more complicated.

In addition, they simulated the virus for ps to ns (one microsecond at most). If you want to simulate intelligence, you'll want to get at least several dozen milliseconds of trajectory (add another 3-6 orders of magnitude complexity).

Then there is the problem that the (classical) molecular mechanics approximation they are using in the simulation doesn't treat electron transfer events very well (if at all). If you want to simulate electrical phenomenon (e.g. neurons) accurately, you'll probably have to increase the detail level to some sort of quantum mechanical system.

Re:Simulating intelligence?

[nbert]

This just made my day. Thanks a lot.

Re:Simulating intelligence?

[4D6963]

Afterall the human brain is still by magnitudes more complex than any computer we can build nowadays

Sure, but who talked about a human brain?

Personally I'll content myself with a virtual genuinely intelligent simulated bug.

Re:Simulating intelligence?

[badboy_tw2002]

Would we have to have a complete understanding of how it reacts, or just start with the simulation of an embryo with a correct and complete genome mapping as its data. Then we let the embryo behave as it would behave (assuming it has all the necessary input, read, food, oxygen molecules, etc.) You could let the embryo "come to term" naturally by letting the cells run. The program to grow a dog or human is already embedded in the cells, you just need to give them an emulator to run in.

Re:Simulating intelligence?

[x2A]

You could skip the atom layer, and just code the neurons... properties of neurons could possibly eveen be easier to code than atomic simulation ('tho couldn't say for sure). You could probably learn a fair amount by doing that too, not so much about intelligence, but in the way that information passes around the brain.

Re:Simulating intelligence?

[4D6963]

Would you consider a bacterium intelligent?

I think "something with a brain" qualifies as intelligent, I guess.

And that's just as well, as it wouldn't teach us anything more than just physically reconstucting the object atom by atomYeah that's perfectly right, but so far I just saw it as the only way to obtain an artificial intelligence, considered that it's the first goal to reach before making an interesting artificial intelligence

Re:Simulating intelligence?

[infinite8s]

> If we assume that all physical processes can be simulated by a computer (given complete knowledge of the laws of physics), which seems to be a safe assumption...

Aha, but your given is anything but, and hence your asumption isn't so safe.

Re:Simulating intelligence?

[4D6963]

I agree with you, it's hard to tell yet. But why talk about such things as the human mind? That's what comes up everytime we talk about AI, but personally, I'm not looking for a human-like intelligence that can be genuinely depressed, but rather something like a bug that can find out by itself how to walk and avoid obstacles, such things.

Re:Simulating intelligence?

[jackelfish]

The simple answer to your question is no. In the field of Molecular Modeling, we have a pretty good idea of how to simulate a system at the atomistic level. As the article states, we are pushing the limits of computational resources and time to complete the simulations at the level of about 1 million atoms (this is state of the art). The simulation discussed in this article is of a Satellite Virus (not even a true virus by strict definition, as it requires a cell to be previously infected by a virus) and as such is the smallest "living" organism on the planet (it has only 2 genes; 1 to synthesize its coat protein and another which we have no idea of its function). Now think of a bacteria like E.coli which has something on the order of 4500 genes, several of which are expresed in multiple copies. Don't forget all of the metabolites and water also. This takes the simulation from 1 million atoms up to at least say 5-10 orders of magnitude in size. So a simulation that took 12 days on a 256 CPU SGI cluster would now take 1.2 million to 1.2e11 days on the same cluster (you can also do the math to figure out how scaling the cluster would effect the time). So, long story short, we have no hope of doing an all atomistic model of even Ecoli any time soon. But, lets not forget that these simulations were only atomistic in the Newtonian sense of the word (we are only simulating atoms as point masses using Newtonian Physics, commonly referred to as Molecular Mechanics). If we want to simulate chemical reactions (an essential component of life), we require Quantum solutions for these simulations and we are looking at a limit of somewhere on the order of 400 to 4000 atoms for these types of simulations. So to really simulate E coli, lets add another 10 orders of magnitude to our estimates. The next order of business is to simulate a eukaryotic cell which may have 10,000 genes in it. Then lets go to multi-celled organisms...and so on. The long and the short of it is that there is no realistic hope, any time soon, of doing simulations on this order, until there is a significant leap made in either the way the simulations are done, or the power of the computers on which we do them. There may be ways of abstracting things by making assumptions that will reduce the accuracy of the models, but that is another story.

Re:Simulating intelligence?

[4D6963]

Google's way ahead of you.

hahahahahaha. yeah right! I think they're way ahead of me in getting high and takin science-fiction for reality. I heard one of the heads of Google saying that if they numerised all these books it was so a genuine AI program could read them and learn from them.

When I read that I was cracked up and thought to myself that that guy had to be real dumb to believe he can do that.

That kinda shit just ain't gonna happen, period.

What's your reason btw, you work at Google or you don't want anyone to know you believe that kinda bullshit?

Re:Simulating intelligence?

[user24]

If they can simulate something else than a virus (because I don't think viruses are intelligent) could they by this way obtain intelligence by simulating an intelligent animal?

From a philosophical perspective? Depends who you ask.

Most of the philosophers I know of are still talking about Turing's Computing, Machinery and Intelligence paper of 1950, which focuses on simulation of conversation (hence Turing Test) rather than learning systems or simulated life such as this.

JR Searle gives some pretty good reasons why, most notably in the 'Chinese Room' theory. From wiki:

Searle describes a scenario in which a person is isolated in a room. The individual receives pieces of paper marked with Chinese characters from under the door. Even though the person does not understand Chinese, if there is a formal sorting process for the characters then they can be filed into a meaningful order.

Seale holds that the person simulating writing Chinese is not actually writing Chinese because they don't understand what they're doing.
In a similar way, a simulated intelligence wouldn't be self aware, and wouldn't have any intention behind it's actions.

On the other hand, people such as Turing would hold that it doesn't matter how the end result is acheived, if the end result is identical to that which it simulates, then there's no reason to suppose the machine is not intelligent.

With Searle's example though, I think it's noteworthy that with chat bots the analogy is fine, but with more advanced AI, the question of who designed the "formal sorting process" becomes crucial.

I wrote a paper on the subject of intelligent computers last year, IMHO it's not very good (it was the first essay I wrote for my degree), but if you're interested: http://puremango.co.uk/device.pdf

Re:Simulating intelligence?

[4D6963]

Yeah I agree that simulated intelligence is not genuine intelligence. It seems obvious. As I read somewhere, the Turing test doesn't mean alot because theorically a computer could be programmed to have enough "data" in itself to reply to everything it could be told within 30 minutes, then the Turing test is based on a judgement, and if some people are easy to fool and not others, it's still a relative thing and thus doesn't mean anything, and then the intelligences not able to pass this test are still intelligent (babies, animals and foreigners for example).

Personally, I see a first step in AI in a bug that would figure out by itself how to do everything a bug has to do, without being even remotely programmed to do or to learn how to do the tasks it would do. I see trying to make a genuine AI chat pretty much like trying to send a man on some exoplanet, I mean we hardly can make an AI of the level of the simplest intelligent animals, and we would try to make an AI perform a task only performed by such complex being as us, not to mention that it takes us years of learning.

More on the philosophical side of things, I know I can basically "make run" any program by looking at its source code (or even assembly code) and performing by myself what it does. Even if most programs are too complex for such a thing to be done, it's theorically possible. Now, if we got an artificial intelligence generated by a program, and then i try to "run" this program by myself by looking at its code, it would mean I would have a genuine intelligence running "on top" of my brain, well maybe it's not such a valid point, but I hardly can imagine that.

However, I read about 70% of your paper, it was interesting, however it was maybe too focused on refutating Turing's test and well, I guess I would have liked to hear about how to "recognize" a simpler intelligence, such as a bug-type intelligence. I think that's what the strong-AI researcher should focus on researching, instead of directly looking for a chatting intelligence, I think that going straight for that type of intelligence is trying to skip a half-dozen of major and necessary steps to get to any genuine AI, as I said, it's as if we tried to send men on an exoplanet, that sure'd be interesting, but we should first try to send em to Mars.

Re:Simulating intelligence?

[HappyEngineer]

The pure simulation method is unlikely to ever be used in developing an artificial intelligence. And that's just as well, as it wouldn't teach us anything more than just physically reconstructing the object atom by atom. Finding alternative routes to the goal is more likely to be interesting and useful.

Pure simulation may well not be the easiest way to develop machine intelligence, but if it's ever done then it would certainly still be useful. Maybe the intellectual leaps required to develop MI are just too great for the human mind or any technique a human or group of humans can create? The only way to know for sure that that statement is false is to actually create one.

Nonetheless, if we could create an MI by raw imitation without understanding, we would still end up with an MI. Once you have an MI you are on the road to a singularity. A human level MI can be made more intelligent by running it on faster hardware (after all, what does it mean to be intelligent if it's not the ability to take inputs and get results faster than others can). The result is an intelligence that itself can develop faster hardware, etc.

At some point perhaps the MI will become intelligent enough that it can find shortcuts to developing MI which are not based copycatting the methods used by meat intelligences.

Re:Simulating intelligence?

[iammaxus]

a complete understanding of the basic physics and chemistry involved (neither of which we have yet)

I believe you are wrong and we already possess sufficient physical knowledge and have for years. As far as I understand it, the Schrödinger equation (and perhaps some other quantum mechanical theories) allows us to model the behavior of electrons completely. All the interactions involved in biochemistry are simply a result of electron behavior (nuclear reactions do not affect life significantly). This is not to say that there is not still work left to be done in the field as modelling at such a low level is probably impractical.

Re:Simulating intelligence?

[fnurb]

As Kurzweil and many others have pointed out, we don't need to simulate every single neuron and synapse, let alone every single neurotransmitter molecule, in order to simulate the operations of an intelligent brain. Rather, research now focuses on simulating cognitive processes at a much higher symbolic level. The results, from auditory simulations of human audio processing to an artificial pancreas, show that many complex biochemical processes can be simulated to the required level of detail without bothering with simulating down to the quantum level or anywhere near it.

The math represented thus becomes quite different, and, given a simple extrapolation of accelerating returns regarding computing power per cost, show that within a decade we *will* have the processing ability to create a functional digital brain at the complexity level of a human brain. This doesn't automatically mean that model will be instantly intelligent, but, when you factor in our accelerating understanding and ability to model abstract thought processes in software on top of our ability to model the physical functions of the brain, it is not unreasonable to suppose that we will produce true digital intelligence by way of a bottom-up simulation of brain processes. Add in the accelerating returns principle, and, within a few years after that, our digital model wil have processing power thousands and then millions of time that of a single brain, which in turn, even before sentience can be used to help us refine our behavioral models of thought processes-- and the likelyhood that it will cross the threshold of intelligence approaches certainty.

It is only a matter of time, and the surprising thing is, if one simply projects the curve outward, how soon it will likely happen.

Re:Simulating intelligence?

[user24]

Well, if all the bug is doing is responding to external stimuli, then it's really not intelligent; I mean, things like the Sony Aibo (or whatever the stupid dog thing is called) do that. The theory behind conversation being the criteria of intelligence is that it can pretty much cover any requirement you want to throw at the problem; the medium of text is incredibly versatile.

You're correct that Strong-AI would have to learn for itself, and indeed a program that could learn language just by observing, say, chat logs (like a human child does) would be incredibly impressive. But it's not going to happen in the near future, because we don't even know how children acquire language. I could go on all day about this, but if you're interested, check out Noam Chomsky, Wittgenstein, Ayer, Quine, and the philosophy of language in general.

Re:Simulating intelligence?

[binarybum]

If they can simulate something else than a virus (because I don't think viruses are intelligent) could they by this way obtain intelligence by simulating an intelligent animal?

Of course. It would take an absolutely colossal amount of computing power, but given sufficient resources and a complete understanding of the basic physics and chemistry involved (neither of which we have yet) you could absolutely simulate a living creature, and the simulation would be intelligent. There have been many sci-fi stories that have used this basic concept. In fact I expect the first intelligent machine will attain its intelligence by simulating a living brain (although at a much higher level than individual atoms).

      Dude, this is going to blow your mind.

Re:Simulating intelligence?

[egomaniac]

I believe you are wrong and we already possess sufficient physical knowledge and have for years. As far as I understand it, the Schrödinger equation (and perhaps some other quantum mechanical theories) allows us to model the behavior of electrons completely. All the interactions involved in biochemistry are simply a result of electron behavior (nuclear reactions do not affect life significantly). This is not to say that there is not still work left to be done in the field as modelling at such a low level is probably impractical.

We do not have sufficient knowledge to simulate a working cell. That was all I meant.

Re:Simulating intelligence?

[egomaniac]

Aha, but your given is anything but, and hence your asumption isn't so safe.

I never said that we had complete knowledge of the laws of physics. The point was that it seems pretty safe to assume that if we had complete knowledge of the laws of physics, we could fully simulate physical interactions on a computer.

There's no way to know for sure, of course, but so far all of the physical laws we know are computable. There is no particular reason to suspect that any of the remaining ones aren't.

Re:Simulating intelligence?

[4D6963]

Well, if all the bug is doing is responding to external stimuli, then it's really not intelligent

Well wait, do you mean that bugs are not really intelligent, or that the examples I chose could not reveal intelligence?

Which brings us to a potentially interesting question, how do we tell that animals are intelligent? By answering this, we can think of AI's to try to make and tests to test them..

Re:Simulating intelligence?

[Anonymous Coward]

The state of the given has no bearing on the result.

(A => B) can be a true statement regardless of the truth of A.

In other words, he is saying: "If there exist simulatable laws of physics, ~whether or not we know them~, and intelligence is a byproduct of them, then intelligence is simulatable, in principle."

Physics seems to be bringing us closer to a set of correct mathematical laws of the universe (or at least sufficiently correct to simulate something as fuzzy and high-level as human intelligence), and neuroscience seems to show the brain is a physical device, so we predict that intelligence is, in principle, simulatable. I would even go so far as to say that we don't even need a working theory of gravity, and the standard model is probably sufficient to model a human brain.

Re:Simulating intelligence?

[Anonymous Coward]

I do not quite agree.
As per chaos theory, it is well neigh impossible to have a complete interacting system simulated.
Yes, we might be able to guess some what, but this is just that... a guess

Re:Simulating intelligence?

[Anonymous Coward]

Unless, of course, that "absolutely colossal amount of computing power" ends up being on par with that of the Universe itself.

Re:Simulating intelligence?

[posterlogo]

What's amazing is that "intelligence" is kind of an emergent property. We don't have a very good idea of how it works, but we can map neurons fairly well. This implies that one day, by simply programming the basic features of neuron synaptic transmission, the system might 'come alive' even though we may not know entirely how.

Re:Simulating intelligence?

[Anonymous Coward]

Some cool artificial life programs...

Evolutionz
Avida
AntWorld

And a whole list of others here.

Re:Simulating intelligence?

[zippthorne]

I think you misunderstand chaos theory, which does not suggest any such thing, being a mathematical method for dealing with systems in which a small perterbation creates a large disturbance; unstable equilibrium. In addition to which, it is completely and utterly inapplicable to the concept of a difference engine; damped equilibrium.

Re:Simulating intelligence?

[user24]

if all the bug is doing is responding to external stimuli, then it's really not intelligent.

There's a bit of a problem in restricting intelligence to strict definition. I mean, Skinner would have held that all humans do is respond to external stimuli, and hence he (or at least some behaviourists) would be perfectly happy to accept that we really are not intelligent, but instead are 'mere automata'.

I don't think it's possible to come up with a kind of tick list for what something must have or be capable of in order to be classed as intelligent, instead, I think intelligence is more a vague cloud of qualities, which will include things like self awareness, memory, ability to make decision, and a host of other things, but the definitionis going to be hazy at best. Given that, I think there's a lot of work to be done defining exactly what we want AI to acheive, before we an start thinking about actually building a decent AI.

Re:Simulating intelligence?

[Josh Booth]

I think that you are overestimating how complex the human brain is. I believe that humans think the way they do due to a complex interaction between the logical (computer-like) side and the emotional (highly trained neural network) side and little more. It's just that neurons are really efficient at doing that and computers aren't.

Re:Simulating intelligence?

[shaitand]

"It might even claim to have the same internal sensations we do, however, although I can program a computer to state "I feel depressed," that doesn't mean it feels the same way I do when I'm depressed."

That is a rather ridiculous standard. We don't even know if you see the same thing I do when I see something red. We also don't know if you feel the way I do when I'm depressed. My conception of depressed could be entirely different than yours.

Take some acid and have a conversation about the "levels" of reality to understand what I mean.

Re:Simulating intelligence?

[edremy]

As far as I understand it, the Schrödinger equation (and perhaps some other quantum mechanical theories) allows us to model the behavior of electrons completely

IWAQC (I Was A Quantum Chemist), so I'll bite. In theory, this is true. All you have to do is solve the Schrodinger equation for the system and you're done. The problem is that we can only solve it exactly for a few systems, the most complex being the hydrogen atom. Even the He atom is beyond our abilities, at least in the realm of exactness.

Now, you can get into any number of simulation methods- there's standard Hartree Fock, density functional theory, a bevy of semi-empirical methods and a host of others. The problem is that all of them are approximate, often wildly so. Even for the simplest system, they have a lot of trouble reproducing reality. Quick: which end of the CN- ion is negative? You'll get different answers with different basis sets in Gaussian, much less the changes you get from introducing correlation methods.

Assume that you somehow manage to solve the equation "well enough". That's nice, but all of the above assumes only a single point in time. My grad work was in introducing time into the picture. Electrons move really fast. (Well, they don't really move in the way you and I think about motion, but still) You'll need to solve the above, nearly impossible set of equations every tenth of a femtosecond at the slowest if you want to have any hope of modling things accurately- really, you probably should be doing it every attosecond.

Ten years ago I managed a crappy simulation of a few lithium and hydrogen atoms undergoing a few tens of femtoseconds of a reaction. Computers are faster, yes, but Moore's law is never going to solve this issue.

Re:Simulating intelligence?

[snaz555]

t would take an absolutely colossal amount of computing power, but given sufficient resources and a complete understanding of the basic physics and chemistry involved (neither of which we have yet) you could absolutely simulate a living creature, and the simulation would be intelligent.

Note that the simulation doesn't have to 100% replicate the existing world. It just needs to be somewhat consistent and deterministic. The life forms in it will presumably adapt to the laws of their universe. But it probably needs to be somewhat close to ours if we want to copy our life machinery and hope for something to survive. Otherwise we could bootstrap life by tossing randomly composed molecules into the goo until it gets going on its own. In the latter case the laws could be altered for performance or other goals.

Re:Simulating intelligence?

[grumbel]

Its not that easy, we not only need to simulate the creature itself, but also enough of environment to train it. If you pack a baby in a dark box and shield it from all outer influences it won't really result in something intelligent either, if it survives at all. So it wouldn't be impossible, but generating all the needed input could provide not that easy.

Re:Simulating intelligence?

[4D6963]

ight, but I think you're underestimating it. The brain remains quite mysterious.

Re:Simulating intelligence?

[4D6963]

I don't think it can be as simple as that. Main reason for that is the recent discovery of the parallel role of astrocytes http://en.wikipedia.org/wiki/Astrocyte in the brain. What I mean is that I don't think we know enough about the brain to claim its valid to think that provided that you can map neurons you can take a "snapshot" of the brain and simulate it.

Re:Simulating intelligence?

[JesseL]

You're welcome. Here's the correct quotes with attributions (from a fortune file circa 1989):

  "Anything created must necessarily be inferior to the essence of the creator."
-- Claude Shouse (shouse@macomw.ARPA)

"Einstein's mother must have been one heck of a physicist."
-- Joseph C. Wang (joe@athena.mit.edu)

Re:Simulating intelligence?

[LS]

Are you really sure that cognitive processes, or for that matter all that goes into making a functioning consciousness only involves symbolic systems?

LS

Re:Simulating intelligence?

[Morky]

"We'll never be able to clone and adult mammal."

-My Phd biochemist buddy a year before Dolly.

Re:Simulating intelligence?

[zopf]

I worked in a lab a summer ago whose focus was doing something similar to this, though in baby steps. They built VLSI models in silicon using mixed-mode circuits to simulate a large array of interconnected neurons. Each "neuron" was actually a small circuit that was designed to approximate some type of real neuron. When the "neuron" fired, it actually sent what they called an AER event, an event with an address (column, row, and chip) attached to it. By routing these events around they could interconnect various chips to simulate the combination of various functions.

Of course they were nowhere near their goal of simulating an entire human brain. When I left, they had models of a section of the hippocampus, the retina, and a few other small areas.

The concept is good, however. By "computing" using analog circuits in a massively parallel format, we can drastically reduce the computational and energetic overhead involved in these simulations. After all, real neurons are essentially a very complicated charge pump with a certain firing threshold. With appropriate circuitry, researchers can come fairly close to the real thing without supercomputers.

For those interested, check out www.neuroengineering.upenn.edu/boahen/.

Re:Simulating intelligence?

[edremy]

I think you misunderstand the scale difference here. Cloning methods are understood, it's just really, really hard to do complex organisms. The fact that it's been done multiple times with different mammals in the past five years indicates that it was just an extension of existing techniques.

On the atomic simulation side we can't accurately simulate a single water molecule. There's no way to scale up the existing methods to anything even remotely biological. In addition, there's nothing on the horizon that would indicate that it would even be a possibility. We'd need an entirely new way to deal with this, and since there are literally thousands of really, really smart people who have been working on this problem for the last forty years (including multiple Nobel prize winners, and of course, yours truly :^) I don't suspect we're going to see anything in the future.

The difference between Dolly and accurate quantum simulation is the difference between someone in the days of early powered flight saying "We'll never go faster than sound" and a Homo Erectus looking up at the moon and thinking that he could pile up enough rocks to get there.

Re:Simulating intelligence?

[Anonymous Coward]

Mu

Re:Simulating intelligence?

[mrpeebles]

I think you make a lot of good points. I just want to add a few observations from one who has studied physics, but frankly knows relatively little about chemical and biological systems. While the hydrogen atom can be solved exactly, there is no such thing as even an isolated hydrogen atom. There will always be small perturbations so that you don't even know exactly how to even write the Schrodinger equation, let alone solve it. It is not at all clear to me, at least, that it is even possible (perhaps even in principle?) to approximate these perturbations in a way that allows you to simulate biological systems in a satisfactory, non ad-hoc way. But I think the problem gets even worse. We know that the Schrodinger equation is not completely correct in that it applies only to particles going much slower than the speed of light. Who knows in what other ways it may not be right, what other unconscious assumptions we are making. Are the assumptions important to the function of biological systems in a manner that we don'ty understand yet? It seems to me, a priori, there is no way to know. My own personal prejudice is that it seems likely, if theories as zany as quantum mechanics and Einstein's theories of relativity can be true for very large, very small, and very fast systems. So anyway, I agree with you that it seems to me like trying to simulate these biological systems from first principles is extremely ambitious.

Re:Simulating intelligence?

[Catullus]

Computers are faster, yes, but Moore's law is never going to solve this issue. ...until/unless quantum computers take off, of course. My understanding is that they would revolutionise the study of quantum chemistry.

Re:Simulating intelligence?

[Anonymous Coward]

Yes, there's a load of problems with this "intelligence" thingy. Turing-test waters, much caution with wording is advisable. Especially when a philosopher is around.

The Turing Test, for those new to the subject, is basically Turing's way of saying: If it looks like it's intelligent, and it sounds like it's intelligent, and it describes its feeling when it's reading a Shakespeare sonnet to you like it's intelligent, then it is intelligent, no matter how it's wired on the inside. Neurons, transistors, quantum computers, whatever.

Of course the philosophers weren't as happy with that. Many, many, many, many objections raised to this test just to make sure that humans are still on top where they should be, and machines on the bottom, where they belong. Relevant to your post though, there's the symbol grounding problem, which basically means, as Searle stated it IIRC, that 'simple' manipulation of symbols (I'm not sure what the heck they mean by 'simple' here, if you look at the size of AI programs that are available today) without them being acutally grounded via a sensory system to outside-world meanings ('grounding') cannot be considered intelligence.

Yes, and of course the answer to this would be - put some eyes on your robot! But to me this is unnecessary - If the computer/machine/whatever can convince me that it's intelligent without having a sensory system, and the philosophers have a problem with me calling it intelligent, I'll just call it "bintelligent". Just so nobody's offended.

Re:Simulating intelligence?

[Anonymous Coward]

That is quite simply the best answer I have ever read on Slashdot.

Re:Simulating intelligence?

[clawoo]

Short answer: No.
Not-so-short answer: A-a.
Long answer: Nope.

Regardless the computing power you need to firtsly understand *all* the chemical processes behind a life form (take a bacteria for example). Which is impossible with today's, or for that matter, this century's science.

Re:Simulating intelligence?

[vhogemann]

I'm not an Artificial Intelligence scientist, but I was thinking...

If they can simulate a virus, they can extract and understand the basic functions of their self-replicant nature. A self-replicant, adaptable piece of code might not sound that much, but I guess once it became bigger (by replicating itself), and more complex (with lots of mutated code), it can help us to better understand the paths to true AI.

I don't think that computer virii are that different from their biological counter parts... after all they share the same common objective, replicate and spread as much as they can. The only real difference is that computer virus is not capable of self adaptation, a MSOffice Macro virus can't mutate to exploit a IE vulnerability unless someone re-program it to do so, it lack the intelligence.

So, I think once we understand how to make a computer virus adapt itself to infect different "hosts", we'll be able to devellop some serious AI.

Just my $0,02

Re:Simulating intelligence?

[NichG]

The correct approach to this problem (of simulation in general) isn't to start at the lowest possible level and crunch. That will easily run into the issue you're talking about, which boils down to an O(e^N) dependence on particle number rather than the polynomial dependence that classical physics gives you.

To simulate such a huge system, the most feasible approach would be to start with an abstract high level model, simulate it, and see what it fails to produce. Then, refine the model down to lower and lower levels until you reach a turning point where the additional accuracy gained from the refinement is small compared to the increase in computing cost.

Sure you won't get a 'to the level of string theory accurate' computational model of a guy saying 'My name is Phil.'. But if your model of the guy says 'My name is Phil.' without needing to go that low level, then that actually tells you some interesting things about the brain. And if your high level model fails, well then you need to try something else/lower level possibly. But at least this way you don't waste time simulating things you don't in the end actually need to simulate to get the qualitative behavior thats the motivation for the simulation in the first place.

But you do have to make sure that your abstract model is consistent with the underlying stuff. If your abstract model is impossible to arrive at from the lower levels, then you're going to have some problems.

Re:Simulating intelligence?

[4D6963]

Yeah, that's pretty much what I thought. And it seems that it's pretty much what everybody on this topic agrees on. Also, I think that if we were able to really simulate right other stuff, there maybe much more interesting stuff to do with it than to try to make AI emerge from it.

Re:Simulating intelligence?

[LS]

I would say Turing's definition of intelligence is just that, a definition. Philosophy operates in a space beyond simply defining things, so I can understand how a debate started. I would say that for the most part, Turing's definition is fine, since I have no other way of even determining whether other humans are intelligent or not other than analyzing their behavior. But since we haven't yet seen a non-biological intelligence, it still lies within the realm of possibility that the only path to intelligence (by Turing's definition) is to emulate the inner workings of currently intelligent beings. If it is possible to create a being that behaves exactly the same as a human, but is wired differently internally, then by Turing's definition it is intelligent, but it is up to the philosophers to argue (fruitlessly?) whether it's experience is anything like that of a human.

Anyway, I guess the point of this post is that definitions are just marking posts stuck in reality to give a frame of reference, but do not actually make a statement about the nature of things. This is something that confuses even the most educated of people.

Dropping the word "intelligence", and just talking about the subjective experience of a being, would you say that a computer, wired completely differently, but behaving in the same way as a human, would have the same subjective experience as a human?

LS

Re:Simulating intelligence?

[LS]

that was my point...

Re:Simulating intelligence?

[FireFury03]

so far all of the physical laws we know are computable.

Nope, sorry - quantum theory introduces randomness, and randomness cannot be computed deterministically (and all computers are deterministic). Of course, there are 3 questions:

1. Is quasi-randomness "good enough" for a simulation to work? If so it doesn't matter that we can't generate true randomness.
2. Is the randomness found in quantum physics _really_ random, or is it generated by some deterministic behaviour we don't (yet) understand? I.e. maybe the "randomness" we see in the quantum world could be generated by some deterministic behaviour outside our universe (and stuff outside our universe need not be governed by our universe's laws so could be very strange indeed, but still simulatable).
3. Maybe we can use quantum physics to build a non-deterministic computer to simulate quantum physics accurately.

Re:Simulating intelligence?

[epikt]

RE the exact solution of the many-body problem--there's a wonderful quote on the first page of Mattuck's "A Guide to Feynman Diagrams in the Many-Body Problem":

"In eighteenth-century Newtonian mechanics, the three-body problem was insoluble. With the birth of general relativity around 1910 and quantum electrodynamics in 1930, the two- and one-body problems became insoluble. And within modern quantum field theory, the problem of zero bodies (vacuum) is insoluble. So if we are out after exact solutions, no bodies at al is already too many."

Re:Simulating intelligence?

[2short]

"The only real difference is that computer virus is not capable of self adaptation"

Neither are biological viruses. They are just bits of DNA; just instructions for how to build copies of themselves; there is no decision making or adapting. They "adapt" only because when actual living cells make umpteen-bajillion copies of a bit of DNA, some of these copies are wrong. Or maybe a bit of radiation fried one of the instructions. Out of these zillions of wrong copies, almost all just result in non-working instructions. Some result in slightly different, still working intructions. Once in a great while, one results in a working set of instructions that just happened to change the bit your bodies "virus scanner" systems check for, and then you get a flu epidemic.
But the virus doesn't "adapt" in any concious, or even living, way. It just gets mis-transcribed.

Re:Simulating intelligence?

[spun]

As we have learned over the past 30 years of AI research, in some ways that is harder.

Re:Simulating intelligence?

[spun]

Take some acid and have a conversation about the "levels" of reality to understand what I mean.
Been there, done that, wrote the best selling novel explaining it all. I was alluding to the issue of qualia in my previous post, even using the color red in my example. I have priveleged access to my qualia, I know what red looks like for me, but not for you. I was trying to explain the philosophical implications of AI and the issue of qualia in a single paragraph, excuse me if I skipped some details.

Re:Simulating intelligence?

[at_18]

Hey, I think I just managed it! See:

10 PRINT "MY NAME IS PHIL" I'm rushing to the nearest patent office! All your brains are belong to me.

Re:Simulating intelligence?

[4D6963]

As we have learned over the past 30 years of AI research, in some ways that is harder.

Harder than what? Than making a chatting genuine AI?

Re:Simulating intelligence?

[stonecypher]

If they can simulate something else than a virus (because I don't think viruses are intelligent) could they by this way obtain intelligence by simulating an intelligent animal?

This argument made Kuhn famous in the 1950s. You may be entertained to read Artificial Intelligence: the Very Idea, which addresses this line of thinking, both how it can be supported and how it can be denied. The book isn't neutral; it takes both sides in the effort to give you material to come to your own beliefs, and in my opinion succeeds admirably.

Re:Simulating intelligence?

[stonecypher]

The math represented thus becomes quite different, and, given a simple extrapolation of accelerating returns regarding computing power per cost, show that within a decade we *will* have the processing ability to create a functional digital brain at the complexity level of a human brain.

Yes yes, they've been saying this since the early 1960s. Then, everyone discovers some new physical process in the brain, everyone gasps, the line is moved back ten years, many backs are patted, and we trick ourselves into thinking that this time we really do have the whole picture. Hell, I've been alive through two revelations in the understanding of the physical universe, and many argue that one's also going on right now.

Every time someone says "we will have brains in ten years," I just laugh to myself. Why? Because we don't understand how the brain works yet. How can we possibly predict the complexity of a system we don't even understand?

I'm not saying we'll never model the brain. What I'm saying is we don't have enough information yet to predict when that modelling will occur. We can barely get simple machine vision working, but little retarded bugs see just fine. I'd be hella surprised if we could model the brain of a stag beetle in ten years, let alone a human.

Actually, I'd be surprised if we can model a stag beetle's brain in my lifetime. But, that's just based on actually having done these things, and knowing what I'm talking about, so don't mind me. "But Kurtzweil does these things too! He's famous for it!" No, he's famous for writing about the things he's done. A scientist should be measured by their output. Despite being a scientist, Kurtzweil's honestly more of a science writer, like Gleick. It's just that Gleick has no pretenses about whom he is. Kurtzweil has a lab, Kurtzweil dumps all this money into AI, Kurtzweil wrote all these books. Blah blah blah, I know. He's a better man than I am, and he understands these things better than I do. The problem is, he's still a science writer, and he's been saying "it's ten years away" for almost fourty years now.

Yes, he's a brilliant man. He's also an uncurable unrealistic optimist. It's just that his level of understanding is so far over most people's heads and his writing is of such high quality that nobody realizes he's saying we'll have a skyscraper that touches the moon by next thursday.

The last time he said "AI who can argue you down playing the dozens in Manhattan in a decade," I called bullshit. Everyone told me I was wrong, I was full of shit, I didn't know what I'm talking about, my god it's Kurtzweil, how dare you argue with the man . And you know what? I'm arguing with the man. I was right last time and I'm right this time. AI in the future? Humans on a chip? Sure.

10 years? You're dreaming. 10 years we won't even have real machine vision, natural language probably won't even be fully handled, muscle coordination is still a joke, we aren't gonna have any non-symbolic communication (body language, hand gestures, implication and subverbalization, sarcasm and irony, et cetera.) Machine learning will still be limited to trivial domains like the rules of a single game. Even with specialty hardware and the massive safety benefits, we won't have automatic car driving down yet. Intuition and emotion will exist only in Kurzweil's next "ten years" speech.

Yes, these things will come to pass, but I highly doubt most of them will be during our lifetimes, even to the degree which a six year old human can bring to bear. It's understandable: random processes took several billion years to work these things out. It's okay for us to take a thousand; that's still a breakneck pace, and besides, we only got started in earnest around World War 2.

Re:Simulating intelligence?

[egomaniac]

Doesn't matter. The goal is to produce one possible outcome of the laws of physics, not all of them.

Re:Simulating intelligence?

[hawkfish]

Thank you. These strong-AI folks really do like to run around naked...

Re:Simulating intelligence?

[spun]

Harder than what was originally thought to be hard, namely logical thought. Things like chess are relatively easy compared to the problems of, say, walking in a natural environment or recognizing spoken words. The things that we find easy are hard to program, where as the things that we find hard such as logic, mathematics, or chess are easier. There is still nothing man made that is as good as a human at recognizing another human face.

Making a genuine chatting AI is hard, but a human finds chatting very easy. Making an AI that can ask questions about a patient's symptoms and recommend effective courses of treatment is much easier than making an AI that can argue the merits the latest pop band, yet a human requires years of study to make a correct diagnosis while any fourteen year old girl can discuss pop music for hours without raising a mental sweat.

Does that make my meaning somewhat clearer?

Re:Simulating intelligence?

[4D6963]

Yeah, I understand clearly what you mean. I agree, an addition is hard for us as it's easy for a computer, and walking is easy for us as it's not for a machine.

But I was comparing a *genuinely* depressed AI to a simpler form of "genuine" AI, and if, provided that we consider the two are theorically feasible, both would be hard to be done, I'm pretty sure that the bug-like AI would be easier to do than the human-like AI.

Re:Simulating intelligence?

[GenieGenieGenie]

Dropping the word "intelligence", and just talking about the subjective experience of a being, would you say that a computer, wired completely differently, but behaving in the same way as a human, would have the same subjective experience as a human?

I can't tell that even when it's another human you're talking about, either. How can you tell if another human has some sort of subjective experience even remotely similar to what you have, if at all, without analyzing its behavior? You assume he does, but that's just an assumtion, no better than my assumption that the computer you mention does.

Re:Simulating intelligence?

[stonecypher]

Actually, I'm one of the strong AI folks myself. I just don't think that making wild, baseless and patently false claims to garner media attention is ethical. Actually, I've had the pleasure of debating this directly with Searle, Dreyfuss, Brandom, Conant, Belnap and so on; I remain firm in the belief that computers can be intelligent and that it's essentially a matter of programming. I just don't expect it to be finished within my lifetime.

One can believe "real" (strong) AI is possible, without believing it'll happen any time soon. ;)

I Hope...

[eno2001]

...Symantec/Norton, McAfee, CommandPoint, Crudpuppy, ClamAV, Grisoft and the rest are all preparing signatures, otherwise if this thing gets in the wild it will turn your data into nothing but pond scum... ;P (Aren't there ANY moderators with quirky senses of humor anymore?)

Imagine a beowulf cluster of these!

[Anonymous Coward]

Under a microscope!

Move along, nothing to see here...

[PoprocksCk]

...We've been able to have viruses on computers for many years now.

Re:Move along, nothing to see here...

[this great guy]

I would like to re-post what I posted the 1st time (this article is a dupe):

This is an interesting coincidence because I used to reflect deeply on this exact subject a few years ago: what if a supercomputer could simulate a human ? I'll be honest here: I am literally _astounded_ to discover that this scientific team has successfuly simulated a virus. I didn't thought supercomputers were powerful enough for such a task. I just finished reading some articles about the experience and I now understand why this has been possible: they used some empirical functions instead of implementing exact physical laws (would have required much more computing power) and they also simulated the virus for only 50 billionths of a second. But still, they seem to have successfuly simulated life.

Most people don't realize the significance of this event, it means that given enough computing resources we could theoretically simulate humans ! One day we will have enough computing power to run such a simulation. And when it will be done, this human life simulation will have the potential to prove (or disprove) that humans are "just" a bunch of atoms following physical laws and nothing more.

This is huge. Think about it. I know this may sound sad, but personally I am convinced that any life form, including humans, is just that a complex assembly of atoms following physical laws, there is no soul, no afterlife, etc. This human life supercomputer will prove I am right :)

Re:Move along, nothing to see here...

[hritcu]

Not only that, but they were running native code. Think on how much slower this simulation will be.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:First Digital Simulation of an Entire Slashdot

[richdun]

And it took 13 days...That's one slow simulation.

Duplicate; here's a link to the research anyway.

[tskirvin]

The main research page may interest some of you. And for those that it doesn't help, perhaps you want to look at our Linux clusters instead?

Computational Biology BS

[Anonymous Coward]

This is total PR. There is no possible way to simulate a complete virus. If you disagree, let me ask you this: A protein is much smaller and not nearly as complex as a virus. If we can not simulate how a protein folds in 3D space, how can we hope to simulate a virus' structure which involves a whole lot of proteins?

Re:Computational Biology BS

[eviloverlordx]

Because the proteins have already been folded for you?

Re:Computational Biology BS

[Anonymous Coward]

If we can not simulate how a protein folds in 3D space, how can we hope to simulate a virus' structure which involves a whole lot of proteins?

"If we can't simulate how the big bang happened, how can we ever hope to simulate a pendulum, which was created from the big bang?"

True, simulating the folding of a protein is a difficult task. But simulating how a protein behaves once it is already folded is a much easier task. The important thing to note is that they aren't simulating the "folding" of a virus, they are simulating atom movements within an already assembled virus.

It's true that they don't represent all the atomic details of the system, and you could argue that this means they aren't truely "simulating" the virus, but in that case, you'd have to argue we couldn't "simulate" a pendulum or a plane flight, because we aren't modeling the thermal fluctuations of the atoms in the metal, or the boorish behavior of the passenger in seat 23F.

(But I agree - the exercise was a whole lot of flash with little substance. Just a demonstration that it could be done, with little scientific merit. -- Hey, much like the Moon Landings!)

Re:Computational Biology BS

[raoul666]

Because sometimes big things are easier to simulate then small things. We don't know too much about how sub-atomic particles behave, but we can easily simulate a bunch of ball bearings (which are made of those sub-atomic particles, just like viruses are made of proteins.)

Re:First Digital Simulation of an Entire Slashdot

[Hexedian]

And here I was expecting a Spore joke...

first simulation?

[nbert]

I doubt it. Might be the first simulation that isn't forced to take any shortcuts to simulate the behavior of a life form (highly unlikely, because there's too much left regarding genes which we don't fully understand)
But even if this is a complete simulation: Is it really that interesting to watch such a simulation if it doesn't interact with other models of the same quality? It's not that interesting to watch a allegedly perfect simulation of a virus on its own, because results are not going to vary much from the simplified models we used so far. So the point it gets interesting is when there's a second simulation of a different life form to interact with the one of the mosaic virus.

Re:first simulation?

[fireduck]

Is it really that interesting to watch such a simulation if it doesn't interact with other models of the same quality?

Sure. Being able to take the together the basic building blocks (atoms), arrange them into molecules (amino acid residues), which then can be chained together (macromolecules), which fold accurately depending on the various electrical, hydrophobic, and van der Waals forces (proteins), which further interact properly with their neighbors and form stable complexes (capsid subunits) which can all be put together to make a complete shell (the capsid), which you can then stuff with the nucleic acid (which istself is made up of various molecules) and observe how all this interacts is incredibly amazing. The fact that they do it, and do it (reasonably) right by itself is an achievement.

As for this being a simulation of life, the discussions above do a good job suggesting how muddy that water is (I'm in the viruses aren't alive camp). But as a simulation of what it is, it's amazing.

I suppose the next step would be to properly model a portion of the cell wall of the host organism and observe on an atomic level how the various forces interplay to induce infection and actually shoot the genetic material into the cell. If they can model a virus, then a portion of a cell wall shouldn't be more than a few orders of magnitude in difficulty.

Re:first simulation?

[RodgerTheGreat]

An excellent point. After all, a virus, BY ITSELF, can't do anything. Unless at least a portion of a host organism's cellular machinery is simulated, I don't see how this is more than a protien-folding simulation. It's an interesting idea, but I don't think we're at the "breakthrough" point just yet.

Re:First Digital Simulation of an Entire Slashdot

[Walt Dismal]

The obvious faster simulation is our universe, which is of course a metasimulation being played by adolescent gods, and which took only 7 days to boot. After which the creator went off for a break and the kids took over the console. Ever since then, things have gone downhill. Somewhere there's a cosmic Sid Meier. And his kids are responsible for global warming, environmental pollution, and Yanni.

Its awesome

[mnmn]

Simulating life is awesome. Now the next step is to simulate something like an Amoeba in water... let its DNA drive it to 'eat' a food particle, and see how accurate the digestion (and binary replication) is with the input being only the DNA and initial conditions. I wonder what kind of computers are required to simulate all that, in how much time? I'd more gladly donate cpu cycles to this than to SETI.

Next I wonder if the computer can be used to run regression tests to create the ideal bacteria or virii for a given situation. Virii can be built to repair human DNA in various ways... a particularly disadvantageous gene can be switched off throughout the body once infected with the virus.

Of course this only allows Cybernet to have more destroying power once it 'wakes up'.

Re:Its awesome

[Anonymous Coward]

Applying distributed computing to this kind of problem is as far as I understand not possible. The reason is that you can't split the problem up into parts that a normal PC can deal with. The number of interacting molecules is far to great. The RAM requirements must be insane.

Beck

Re:Its awesome

[RubberChainsaw]

Hey that would make a great game. Simulate an amoeba in water that eats food. We can simulate that amoeba's evolution into more and more complex organisms, watching it get bigger and deadlier. Then it can grow into a vertebrate lifeform that walks up onto land and becomes sentient, builds cities, wages war and sends spaceships into outer space!

Oh wait...

Re:Its awesome

[mnmn]

"The RAM requirements must be insane"

Why? 4GB of RAM means 4kb of ram per molecule (or was it atom?)
Computers with 32GB of ram (AIX, sun) arent the most expensive ones around. I dont think each molecule needs 4kb of data including its properties and interactions. Several million particles should be possible in a midsized server's ram entirely.

Re:Its awesome

[mikej]

Repeat after me: "The plural of 'virus' is 'viruses'."

The classic explanation:

http://linuxmafia.com/~rick/faq/plural-of-virus.ht ml

Re:Its awesome

[Anonymous Coward]

False. There is only "virus"--talking about some number thereof is meaningless.

Re:Its awesome

[mikej]

Incorrect - In ancient latin virus operated more or less like 'water' does in english for pluralization. Modern usage has made virus into a singular noun by using it to refer to discrete particles, therefore a pluralization of the modern form is required. That plural form is 'viruses'.

Simulation of an entire lifeform, my ass!

[Expert Determination]

This is such a misleading headline. It's a simulation of the dynamics of the proteins forming the outer case of the virus to understand how it maintains its structure. It's purely about studying the structure - like an engineer's finite element simulation of a bridge. It's great work from this point of view. But it's not a simulation of any kind of biological process because the time scale is something like nanoseconds. So yes, it's a simulation, but it's not a simulation of a lifeform qua a lifeform.

And that word 'lifeform' - it brings the quality of the reporting down to the level of Star Trek psychobabble. Try 'organsim', or even 'virus', next time.

Re:Simulation of an entire lifeform, my ass!

[pimpimpim]

Or "assembly of proteins", which is a more accurate description of what is achieved. And it's no wonder it maintains its structure, as you would need far far longer timescales of simulation to see structural changes in systems of this size. The main achievement of this research is getting the research group into the main press, being on slashdot twice, and showing the people that fund your work how important it is for you to have this immense-mega-super-duper $$$ cluster. I was a bit more critical the first time this came on slashdot, but now I've rethought it and I guess it makes sense to pull a stunt like this every once in a while :)

Ah, what the heck, I think I will stop my current simulations right away and set in a complete Ribosomal system, ah, the creation of the key of life simulated, I can see the headlines already! ;)

Re:Simulation of an entire lifeform, my ass!

[Anonymous Coward]

organsim? WTF is that?

Re:Life vs. Non-life

[sokoban]

>> Ok, so does that mean that cuckoos and cowbirds are not "lifeforms"?

No, Cuckoos and cowbirds have all the needed apparatus for procreation. Viruses on the other hand generally require the transcriptional "machinery" of a host cell in order to reproduce.

Some people consider viruses to be the most complex thing which doesn't live, while others say it is the least comlex living thing.

I say they're non living due to their lack of respiration, complete reproductive apparatus, and lack of a cell membrane of any sort.

Re:Life vs. Non-life

[spun]

Cuckoos and cowbirds can, in a pinch, replicate without parasitizing another birds nest. Viruses are snippets of RNA in a protein coating that, simply put, do nothing outside of a cell. Vast difference here.

Re:First Digital Simulation of an Entire Slashdot

[Siffy]

Crap, it's already causing the system to crash. I've known for years we wouldn't be able to pull off a simulation inside a simulation. Those freaks over at LiveScience better chill out before they erase us all!

Re:First Digital Simulation of an Entire Slashdot

[x2A]

That's cuz they've not cared about building loads of industrious polluting stuff because they know it all ends in 2050

ZONK IS A PYLON

[Anonymous Coward]

If you haven't noticed, Zonk, the signal to noise ratio for comments on your stories is very low; in other words, you really bring out the trolls. Remember Michael Sims? He had the same problem, and, uh, well - just ask Taco.

C'mon, man - I really don't want to create a ZONK IS A PYLON /. meme.

You are not a life form, then?

[Roadkills-R-Us]

But is it really a life form? From the article: 'Viruses are tiny bundles of protein and genetic material that straddle the line between life and non-life. Many scientists prefer to call them "particles" because even though they contain RNA or DNA like other lifeforms, they can only replicate inside other living cells.'"

The same could be same for most species of animals; they ``contain RNA or DNA like other lifeforms, they can only replicate inside other living organisms''.

Re:You are not a life form, then?

[eonlabs]

I agree. They don't have all the tools to rebuild themselves, but neither do we.
If you look at us on a protein level, we require more amino acids than our bodies can naturally produce
We also don't produce Omega-3 oils, but our bodies use them, and seem to be healthier with them.
In order to survive, we need to take in these raw building blocks.
I am not going to take this to a molecular level, because you need to start somewhere, but if you need to survive and propagate, and you know that something else has mastered the second part, and you can take advantage of that to make it easier for you to live, then you still are alive.
In the same breath, I think if a computer had blueprints for the majority of its key components (the CPU, the motherboard, possibly a hard drive and some sort of input and output) and it went over to another machine and injected its code into that second machine to make the second machine pump out a computer WITHOUT HUMANS TELLING THEM TO DO SO EACH TIME then they would be alive.
If they had instructions to hunt out the machines that could produce and assemble a new computer, and then tell the machine to produce them, it would count.
Granted, this is very limited life and is not sentient life. It's just functional life.

I think agent smith summed it up best...

[NIN1385]

Agent Smith: I'd like to share a revelation that I've had during my time here. It came to me when I tried to classify your species. I realized that you're not actually mammals. Every mammal on this planet instinctively develops a natural equilibrium with the surrounding environment, but you humans do not. You move to an area, and you multiply, and multiply, until every natural resource is consumed. The only way you can survive is to spread to another area. There is another organism on this planet that follows the same pattern. A virus. Human beings are a disease, a cancer of this planet, you are a plague, and we are the cure.

Re:I think agent smith summed it up best...

[Anonymous Coward]

I think he was wrong, because that description works better for a fungus.

(that's always bothered me about that scene.)

Re:I think agent smith summed it up best...

[NIN1385]

If I didn't understand how that could bothering to you I would say you are the pickiest person ever, but I understand. Haha.

Re:I think agent smith summed it up best...

[i8puppies]

So since the machines found an equilibrium with the real world and figured out how to maintain their own resources within the single geographic location (machine city), the machines are mammals?

Re:First Digital Simulation of an Entire Slashdot

[Irish_Samurai]

You forgot the part where the parent comes back and gets angry for the content being made available to his children and sues in the Court of Universal law. The suit is found to be frivilous on the grounds that the parent shouldn't have let his kids play with the damn toys in the first place.

And since the defendant is also the plaintiff he should have known better.

Re:Life vs. Non-life

[Anonymous Coward]

Cuckoos, cowbirds, cars, and Congressmen are all just emergent properties of vast collections of quantum states. Particles aren't alive, but put them together in certain ways and we call the collections "lifeforms".

The subjective perception of life is a result of the emergent properties of neural connections in our brainpans...which in turn are just emergent properties of etc.,etc.

To try and draw the line between where something is "alive" or not is just human conceit. It has no meaning outside a semantical debate.

God is an existentialist.

Cigar Store Indian

[Doc Ruby]

"satellite tobacco mosaic virus"

That sounds like the greatest hits of American products, all in one convenient album.

I wonder if this digital being...

[manowarthegreat]

...can do a digital goatse.

spore

[Anonymous Coward]

Meh, Spore beat them to it and that was years ago!

Hey . . .

[karnifex]

This might make a cool game. Someone get Will Wright on the phone.

Re:Life vs. Non-life

[Anonymous Coward]

what abt plants ????

Digital human

[Limbo Socrates]

Let's just take this to its obvious conclusion. When will they make a simulated organism as complex as a human? Could we use them to perform experiments on? Maybe stick them in an apparently closed virtual world, and then see if they can invent computers and make virtual simulations of themselves that they can experiment on? Or maybe this has already happened...

Re:Digital human

[vidarh]

This guy at Oxford University poses an interesting argument about the likelihood that we're all living in a simulation.

Not a troll

[AchilleTalon]

Parent is not a troll, it's just plain truth. And its title is humor. Anyone here knows about self-derision?

Simulating Dupes

[Anonymous Coward]

How about simulating a DUPE on Slashdot?

Comment removed

[account_deleted]

Comment removed based on user account deletion

OMG!

[Anonymous Coward]

The virus is already dup(e)licating ! Sorry, couldn't resist.

Re:First Digital Simulation of an Entire Slashdot

[Expert Determination]

No, no, no! You're missing it completely. The entire universe is in fact running on an incredibly slow simulator. Much slower than the one used to simulate the virus. It's just that you think it's running fast because you're part of the simulation.

Lifespan

[gleather]

FTFA: Because of the enormous computing power involved, the virus was brought to digital life for a very brief period of time, only 50 nanoseconds Goodness. Don't they a decent timeline to learn anything from it?

Re:Simulating intelligence? correction

[user24]

JR Searle gives some pretty good reasons why simulated intelligence is not real intelligence

whoops.

Dupe!

[Anonymous Coward]

In other news, LiveScience reports that Slashdot posts another dupe:
http://science.slashdot.org/article.pl?sid=06/03/1 4/2021221

Scientists at the University of Illinois Urbana-Champaign are developing a supercomputer simulation of a Slashdot editor's brain to better understand the dupe phenomenon.

posted under supercomputers a few weeks ago

[Anonymous Coward]

who else remembers this?

Re:Life vs. Non-life

[hunterx11]

Right. They don't "do enough stuff" to be considered "living".

If you want us regular slashdotters to follow your arguments, please stop using all the complex biology jargon you learned in graduate school.

Yes, but...

[Anonymous Coward]

Can you make a Turing machine out of it?

"life" is a lousy line to draw

[smellsofbikes]

I think we can agree that bacteria are alive. But there are types of bacteria, the ones that cause leprosy and chlamydia, frinstance, that cannot reproduce outside of a living cell. (They, unlike most bacteria, invade and live inside cells.) It's fairly difficult to draw a hard line between them and some viruses that have lipid bilayers full of receptors on their outsides. Even prions self-amplify, so where do you draw the lines on what's alive?

Re:"life" is a lousy line to draw

[Forbman]

But what of parasites like fleas and mosquitoes that cannot reproduce without a source of blood to suck, in order to nourish and develop their eggs until they're ready to be layed?

Re:"life" is a lousy line to draw

[supertsaar]

As far as I recall the critiria would be:

Can reproduce
Has a metabolism

Viruses do not have a metabolism.....

Re:"life" is a lousy line to draw

[smellsofbikes]

There are lots of parasites that cannot live without something else -- in fact, that might be a good definition for parasite. (It's not THE definition but it has a lot it.) They're pretty clearly alive. So is a virus?

Re:"life" is a lousy line to draw

[smellsofbikes]

Yet there are bacteria like the ones I was talking about, obligate intracellular ones, that don't really have a meaningful metabolism if they're outside of a host cell. Bacterial spores, likewise, don't have any metabolism in any real sense of the word: they're waiting for a life form on whose metabolism they can bootstrap their way back to life. So how's that different from a virus? I'm not saying viruses are alive. I'm just saying it's hard to define why they're not.

Re:"life" is a lousy line to draw

[supertsaar]

Virusses _never_ have a metabolism at any stage.

Re:"life" is a lousy line to draw

[smellsofbikes]

I'm interested in how you define metabolism, coz I don't really know how I'd define it if I had to. During the lytic/lysogenic stages of virus reproduction, their genetic information is in a functioning cell, there's activity going on (at least in the lytic phase) that consumes nutrients and produces proteins and more genetic material. That's definitely part of how I'd define metabolism if I weren't specifically trying to exclude viruses. Yeah, viruses don't have a Krebs cycle, but there are bacteria that don't have a complete Krebs cycle, either.

See, the thing is, I don't think a virus is alive, but I don't have a deep reason for that, just an assertion based on an opinion. Unless I screw around with definitions and assign meanings specifically to push a point of view, it's hard for me to back up that opinion. Hence: what's metabolism?

Re:"life" is a lousy line to draw

[supertsaar]

Well, there's no reason you should take my word for it. You are entitled to your own opinion.
If you want to believe viruses have a metabolism, that's fine with me.

show me the stats!

[pigwiggle]

There seems to be a pernicious trend in the molecular dynamics world to do the biggest system using the most processors irrespective of the utility. I'll reserve judgment on this particular study until I have time for a critical read, but for all the others ... if you can only observe one or two interesting 'events', even if your system is a million atoms over a microsecond, you haven't shown anything. 'Good' statistical mechanics needs good statistics. So, here's my challenge; I say your one observed event (capsid collapse) is simply fortuitous. Disagree? Show me a statistically significant number of complimentary trajectories. Can't run enough trajectories with a million atoms? I sympathize, but really ...

First Digital Simulation?

[funkioto]

I used to play Life on the trusty C64, this is at least the second digital simulation :) http://www.math.com/students/wonders/life/life.htm l

Re:Digital human - oblig F.R.

[Anonymous Coward]

Virtual Virtual Skeeball, anyone?

And let there be light

[Anonymous Coward]

1 In the beginning man created the processor and the memory and the fibre.
2 And the net had form, and void; and light sparkled upon the face of the deep. And the spirits of men moved over the waves of the waters.
3 And man said, Let there be life: and there was life; and it multiplied and evolved and escaped.
4 And man saw the what had become, and it was strange, and man feared.
5 Profit!

Depends how you define life

[themusicgod1]

I'd say conway beat them to it.
0 0 0 0 0
0 0 1 0 0
0 0 0 1 0
0 1 1 1 0
0 0 0 0 0

Re:Depends how you define life

[Anonymous Coward]

dang you...made me waste at least an hour trying to find the bug in my python implementation of Life

Re:Life vs. Non-life

[Gauth]

It all depends on your own (ie subjective) view on what life "means".

One thing that is almost always a core criterion is the ability to reproduce. Viruses, themselves, cannot reproduce. They are, actually, incapable of reproducing. The only real question is whether or not you see "modify the genetic structure of an entity capable of reproduction so that it now only makes copies of the virus" as reproduction. I would say that this means it cannot reproduce. It can only cause something else to reproduce for it.

This is quite a different subject than that of symbiotic organisms. I don't know of any organisms that we *depend* on for reproduction. The most common I'm familiar with are the bacteria we "house" which aids in our digestion. Well, a developing fetus doesn't really need such a thing, since they get their neutrition already broken down from their mother.

So, yes. I say they don't "do enough stuff": reproduce. Some species of bird may trick other species into raising their young, however they are obviously quite capable of reproducing on their own (which they do every time).

Re:Life vs. Non-life

[pomo monster]

Human beings do nothing outside a very specific environment tailored to their needs, where temperature, pressure, oxygen content of air, gravity, radiation, &c., all lie within specific bounds. How is this different from a virus needing an environment that includes cellular structures in order to replicate?

Me, I subscribe to structuralism.

"Simulation?"

[Council]

Pretending for a moment that the story is what the headline said it was, because that makes it much more interesting:

Can you drop the word "simulation"? If it's simulated from the ground up, it's (in its environment) indistinguishable from the real-life version (here assuming the simulation is proper). So let's be provocative and just say it's a digital life-form.

If you can't call it life...

[neo]

they can only replicate inside other living cells.

then what do you call us? We can only replicate at gamer and star trek conventions.

It's all fun and games until it infects the Net!

[Progman3K]

Come on people; remember your basic Star-Trek training: Whenever you see an alien life form where it shouldn't normally be, some sh-- is comin' down!

Digital Religion

[tedgyz]

Ok, extrapolating forward a bit, how soon before simulated life forms create their own religion? Won't they be surprised when they meet their masters?

As a gentleman farmer

[Anonymous Coward]

who grows a huge plot of chile in my garden every year, and who has to be sure I get a TMR variety of seed, I hope some of this work results in the destruction of this virus.

g.f.

i've seen this before

[Anonymous Coward]

"even though they contain RNA or DNA like other lifeforms, they can only replicate inside other living cells."

Isn't that called "bootstrapping"?

From wikipedia:

"processes whereby a complex system emerges by starting simply and, bit by bit, developing more complex capabilities on top of the simpler ones."

Re:Life vs. Non-life

[Anonymous Coward]

viruses don't replicate. they get replicated. They are nothing more than a (in)convenient shape.

It's alive!

[Dice Fivefold]

It has alredy started duplicating on the Slashdot frontpage.

Nondualism and the intractability of life

[Pfhorrest]

I think both you and the GP have valid and noncontradictory points.

The GP makes the point that any distinction we make is artificial and arbitrary, that everything is in fact a continuum.

You make the point that we understand things by categorizing them as X or not-X.

There is a kind of philosophical nondualism (c.f. Taoism) which encompasses both of these concepts. It is the notion that, while there are no actual distinctions in reality as separate from our understanding of it (as Kant would say, the "noumenal" reality), we can only ever hope to understand reality in terms of such distinctions. To better understand the complete continuum we must seek to better understand where the endpoints and midpoint of it are: in other words by figuring out where exactly something shifts from being "life" to "not life", in this example.

I have a phrase I like to use for this sort of reconciliation of divergent concepts (especially in political positions, e.g. capitalism vs socialism, or egotism vs altruism, etc): "embrace the paradox and see that it is not".

In cases like this though, I am inclined to side with the GP, at the moment at least. Defining what is life has the same difficulties as defining what is a person: it seems any definition you come up with either excludes some things we'd normally consider life (like viruses) or people (like babies), or includes things we normally wouldn't consider life (like fire) or people (like domestic animals). Defining where life begins in the evolutionary ladder faces the same challenges as defining where an individual's life begins, and as best as I can determine the only logical answer you can give is either "everything is alive" or "nothing is alive", for any non-arbitrary distinction you try to make ends up seeming unsatisfactory - though really, neither are those two options, either.

I normally like to pride myself on having well thought-out answers to (or at least opinions on) pretty much every philosophical problem out there, but this one has always seemed intractable even to me.

Re:Nondualism and the intractability of life

[Red Flayer]

My main point, though, isn't that we like to categorize. It is that such categorization leads to better understanding of the subject as we attempt to redefine. All the challenges inherent in establishing absolute definitions prompt better research, better analysis, and ideally, the formation of definitions that are not arbitrary, that make sense to those with an understanding of the subject, and that are useful and practical for scientific studies and applications.

Great Point

[Mark_MF-WN]

This is a great point, and well made. The first few paragraphs of my biology textbook expound upon that very topic, that "life" is a nearly impossible concept to define.

For example, ask yourself this: given that a paramecium is considered "life", then which is the living organism -- you, or your cells? Is a heart cell alive, or are you? Or both? If you try to define "life" against the paramecium, the a human isn't a living thing at all, but rather a bizarre cohesive colony of trillions of living things. Obviously then we need to include more than one type of thing in the definition of Life. That's about as clear an issue as you'll find in this discussion though. It all gets more ambiguous from there.

Why virii are not alive

[chrisjbuck]

A (biological) virus does not eat or photosynthesize or have any metabolism at all. That's why they are virus particles, other than the slow degradation of all complex molecules if you have a tube full of virii they will just sit in the tube forever. Doing nothing. If you add sugar, protein, complex carbohydrates and sunlight to the tube of virii they will... sit there. Doing nothing. Not eating. Not metabolizing. Not replicating. Living things would either die, metabolize, or replicate, the virii do not. The virii does contain genetic information, if inserted into a cell the information is used to hijack the cell into making more copies of the virus. The virus may only encode a handful of proteins, but it uses the ribosomes and other protein building apparatus of the infected cell to make the viral proteins, and more copies of the viral genetic info (DNA or RNA) which is packaged and released from the cell (sometimes killing the cell in the process, sometimes not). Does this mean the virus is alive? All the protein synthesis, and packaging of the virus is done by the infected cell, the virus does not technically replicate itself, which is part of what we define as "life as we know it". They are not dead they just exist as a glitch. A primordial cell probably had a mutation that produced lots of particles that happened to be capable of causing the same glitch in other cells they encountered, virii are perhaps analagous to a "goto" loop that somehow copies itself to other programs, more than to actual computer viruses which imitate their biological namesake only to a certain degree. Maybe when the sony robot dogs start giving each other roborabies via bluetooth the analogy will be closer... IAABC - I am a biochemist - but genetic coding is still trickier than php scripting :P

Re:Why virii are not alive

[Puf_Almighty]

But it's not that virii do nothing, it's that they have no action when in the circumstances under which cells have their action. You're right that, in the absence of a replication and transcription system to hijack, a virus will sit in a tube doing nothing. But that's because that is its particular form of metabolism. If you take a sporulated bacterium, say Anthrax (which I choose as an example because it's especially good at sitting there doing nothing for decades at a time) , and put it somewhere in the absence of the materials which it is specialized to act on, it will do nothing- it will stay sporulated and not move, replicate, metabolize, communicate, or anything. Does this mean it's not alive? No, it means that it's not specialized to deal with the resources with which you've presented it.

A virus won't act if you put it in a buffer solution with glucose and amino acids. And a cell won't act if you put it in another cell (it'll get digested). Just like how a computer program written for a Mac won't run if you try to run it on a PC- it doesn't mean it's not a computer program, it just means that you've got it in an environment to which it is not adapted.

That said, the real argument is that cells are just enormously more complex than virii. It's just basically an enzyme with DNA.

Re:Why virii are not alive

[floki]

That's why they are virus particles, other than the slow degradation of all complex molecules if you have a tube full of virii they will just sit in the tube forever. Doing nothing. If you add sugar, protein, complex carbohydrates and sunlight to the tube of virii they will... sit there. Doing nothing. Not eating. Not metabolizing. Not replicating. Living things would either die, metabolize, or replicate, the virii do not.

Of course, if you define dying as not living any more, viruses don't die because they don't live in the first place. As you have said, given the described setup they will certainly degrade, especially under sunlight. The UV radiation breaks the molecules. That's why UV radiation causes cancer among other things. Some scientists even theorize that (amongst the warm weather helping the immune system work) people get ill in summer less often because of the increase of UV radiation compared to the winter.

Re:Why virii are not alive

[Anonymous Coward]

but genetic coding is still trickier than php scripting :P

I think you mean it's trickier than secure PHP scripting.

"Language is digital"

[Errandboy of Doom]

Language is digital...

I disagree. I think language is just mostly digital.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Everybody look for....

[Anonymous Coward]

a picture of Zonk!!!! With limited memory and a missing ability to recognize patterns this primitive lifeform cannot recognize dupes at this point in the evolutionary process.

the debat of life/ non-life

[itzdandy]

here is the solution to the debate.

life must exhibit certain characteristics.

the simplest two characteristics are

reproductivity
autonomous function

so a virus must have two classifications.

active and inactive

active virii are those currently 'living' in a cell or some other construct that support basic life functions such as allowing autonomous function and reproduction.

inactive virii are simply particles with a life-like construction. they have no autonomous functions and cannot reproduce even when paired with other virii and have no asexual or sexual reproductive abilities.

so! a virus is alive when in a cell and is not alive when free floating!

---

thanks just my take on it
thanks for the read. goodnight

Re:the debat of life/ non-life

[Anonymous Coward]

But couldn't fire then be classified as being alive? Fire reproduces, given energy (wood), and it also functions entirely on it's own, given it's very unpredictable nature. That's closer to being alive than other certain *cough* organisms.

Re:the debat of life/ non-life

[Anonymous Coward]

Viruses.

My questions

[LS]

Can a complete description of the workings of physics be described finitely, or is the complexity of our universe infinite? i.e. is the study of physics a never ending rabbit hole of major discoveries and refactoring to new models, or will we hit the bottom eventually, and jump to a new realm that makes modern physics look like yesterday's alchemy? And to make this more relevant, at what level can we start ignoring details of our models to make a realistic simulation of life, or are all the details important? If it's the later, then we may never realistically model life. I think this question is fundamental to the whole endeavor.

LS

Re:My questions

[vidarh]

First you have to answer the question of what constitutes "realistically model". You can "realistically model" the outcome of throwing a dice trivially, but exactly modelling it is incredibly hard. Depending on your goal the easy approach (just pick a random number from 1 to 6 for each throw) is often sufficient.

A significant point of modelling something is that you want to simplify something sufficiently to show all the key behavior without the same level of complexity. In this case you can assume that modelling the way quantum mechanics affects the behavior of individual particles is mostly unimportant, for instance, and a lot of factors can be closely enough modelled by generating random numbers instead of accurately modelling specific processes.

Re:My questions

[LS]

I understand your explanation, but find a couple problems with it. First, by randomly picking a number between 1 and 6, you are simulating the outcome or purpose of dice, but not an actual dice throw itself. Dice are not perfectly shaped due to manufacturing inaccuracies and weight differences caused by the number of divots on each side, which may skew numbers ever so slightly. Also if include the surrounding environment with the dice (e.g. a person who throws a specific way, skewing the results significantly) How do you generate these random numbers with your simple simulation? It's shown that no algorithm random number generator is truly random. My purpose is not to attack your specific example, just pointing out that in general what you suggest is like 2d snapshot of reality - you are losing a lot of useful information that could enhance the simulation's realism.

Secondly, your example of the dice is modeling something that was designed by humans, and thus has a specific purpose, making your "outcome" based simulation easy. Before dice were invented, someone had the idea of randomly selecting a number from a range of 1-6, and your example simulates this design, not the dice themselves. Life was NOT designed (unless you believe in creationism), so there is no functionalities or purposes that can be singled out to model at a higher level. for instance, there is no such thing as a heart as a singular entity. We just draw the borders at the arteries and connective tissue to make it easy to talk about, but the heart does not exist as an isolated system. Subtle details that are left out of the simulation may have large effects over the long term if the model is allowed to play out, or may affect another part of the system in ways that are not obviously noticable. This is especially important if we decide to start using such simulations for drug testing, etc.

LS

But is it really a life form?

[Anonymous Coward]

But is it really a life form?

Absolutely. Would you classify an intelligent species from another planet as life? What if it was constructed much differently than our own (not carbon-based)? The only real difference would be the environment they evolved in that created the life.

Virii can be classified as lifeforms that live and evolve inside other lifeforms. Lifeforms are their natural environment.

I do life form simulations too...

[cciRRus]

It's no big deal to simulate life forms. I have created an entire life form and simulated its life with this software. The best part, I have managed get it to go through sexual reproduction (think "woohoo").

Oh yes, now I get it!

[modecx]

You've just explained what in so many words? A noun? That's very profound indeed...

Re:Oh yes, now I get it!

[shaitand]

"water" is also a noun. Water, is however a label for something real. "Life" is a label for a concept that does not exist, we made up the concept itself and not merely the label. It is not even a classification like a mammal, there is a valid definition of mammal, there has never been an agreed upon definition of life.

There is a very substantial difference there.

Re:Oh yes, now I get it!

[tsm_sf]

"Life" is a label for a concept that does not exist, we made up the concept itself and not merely the label.

Wow. You should set that as your sig so people know what kind of an intellect they're dealing with.

Re:Oh yes, now I get it!

[ultranova]

"Life" is a label for a concept that does not exist,

Life obviously exist since we're having this debate - I doubt we could have it if we weren't alive.

Furthermore, any concept that has been labeled must exist, since the only way to gain a label is that someone gave it at some point, and that can only be done if the label-giver knows of the concept. This of course doesn't mean that whatever the concept refers to exists, but the concept itself does.

we made up the concept itself and not merely the label. It is not even a classification like a mammal, there is a valid definition of mammal, there has never been an agreed upon definition of life.

According to the Theory of Evolution, mammals developed from some other lifeform that wasn't mammal. If this is true, then at some point in this evolution the first mammal was born from parents that weren't mammals themselves. Do you think that that mammal was clearly a different species - no, a different branch of the tree of life - that its parents ?

I'm trying to say that, while the definition of "mammal" might seem clear (it has mammaries :), there's still bound to be some unclear cases. Very few things have usefull formal definitions that would never lead to contradictions or uncertainty; reality is a continuum, not something that can be neatly categorized. There are natural boundaries between groups like "living" and "not living", but the boundaries have width, and sometimes things happen to lie on those boundaries instead of clearly falling on one side or the other.

Anyway, here's a little philosophical advice: if your conclusions ("There is no life") directly contradict observed reality ("I'm alive"), they are wrong.

Re:Oh yes, now I get it!

[nickco3]

Life obviously exist since we're having this debate - I doubt we could have it if we weren't alive.

You're confusing concepts with their labels.

He's not saying there's no such thing as life, which is easily falsified. He's saying the concept "life" is arbitrary, and that the boundaries of that concept are arbitrary: there are seven specific conditions you need to meet to be officially alive. Why those particular 7? What if we changed the list to 6 or 8?

Having dreamt up a classification called "alive" it's easy to demonstrate there are things that meet it. But don't fall into the trap of thinking that the classification exists outside our collective heads. Because we dreamt it up it.

Let's say we change the definition of "life", adding requirement number 8 "wings". Things that are "alive" have "wings". Therefore, you and I are not "alive" because we no longer meet the definition. BUT (this is where you got confused) we carry on exactly as we were, still reading Slashdot, still eating, moving around, excreting, etc, because we're only talking about labels, and not reality.

See Zen and the Art of Motorcycle Maintenance if you're struggling. It took me ages to get it.

Re:Oh yes, now I get it!

[shaitand]

I think I am going to have to read this "Zen and the Art of Motorcycle Maintenance" book my fellow slashdotians keep talking about. Especially now that I have the honor of rediscovering one of its points. :)

Re:Oh yes, now I get it!

[ultranova]

He's not saying there's no such thing as life, which is easily falsified. He's saying the concept "life" is arbitrary, and that the boundaries of that concept are arbitrary: there are seven specific conditions you need to meet to be officially alive. Why those particular 7? What if we changed the list to 6 or 8?

Problem comes not from the concept of life, which is both natural and intuitive (I'm alife, the dog's alive, the rock's not alive), but from trying to make an "official" definition. That official definition might be arbitrary, but that's simply because it's trying to draw sharp borders where none exists - see my original post for what I mean by that.

Having dreamt up a classification called "alive" it's easy to demonstrate there are things that meet it. But don't fall into the trap of thinking that the classification exists outside our collective heads. Because we dreamt it up it.

And yet I'm still alive, and the dog's still alive, and the rock's still not alive. I and the dog have a quality, "life", which the rock lacks. If there's a quality that some, but not all, (physical) entities have, then that quality is "natural".

Now, in the case of life, there are entities (such as viruses) for which it is hard to say whether they possess that quality or not; to aid in the classification process, the quality "life" has been officially defined, but that definition is indeed arbitrary and really has nothing to do with what it actuall means to be alive.

Look at this way: the huge majority of stars are organized into galaxies. A "galaxy" is a concentration of stars. Is this definition arbitrary ? Yes. Did I just make it up ? Yes. Does that mean that such concentrations don't really exist ? No.

It is a mistake to assume that our concepts define reality; but it is also a mistake to assume that reality has no structure outside our imagination.

Re:Oh yes, now I get it!

[nickco3]

I'd certainly recommend it. Pirsig tackles this bit early on, though his particular example is the "Law of Gravity" where we've been using "life", it's fundamentally the same argument and it's a theme that gets developed throughout the book. You need to be willing to pull apart your usual ways of thinking to get to the core of what he's saying, and this is one of the tools he deploys to help get you there.

Re:Oh yes, now I get it!

[modecx]

He's not saying there's no such thing as life, which is easily falsified. He's saying the concept "life" is arbitrary, and that the boundaries of that concept are arbitrary: there are seven specific conditions you need to meet to be officially alive.

Are you daft? The conditions by which we classify life are not arbitrary; they're based on empirical knowledge that is easily observed here on our home planet! It's not like these conditions were plucked out of the air. Do we know life to exist anywhere else but here? No! How can we classify some form of life that we haven't experienced yet unless we had prior knowledge of it? When we encounter something that is discernable, as life, and it don't fit our theories, we should have to change them--correct?

For all we know there are beings in Alpha Centauri arguing this exact same point, but every creature big and small has wings... That is to say, every organism they've encountered from bacteria on up to sentient beings have wings. So their definition of life may very well include the condition of wings. Could you say that their perspective is wrong? Could you say that perspective is arbitrary, when it would appear from their perspective that their definitions are based entirely in reason, principle and pragmaticism? Hardly, I think. The value of their science will be evaluated when they meet life that doesn't fit their definitions.

If you describe a flexible definition based on the facts as we know them as arbitrary then you've jumped the philosophical shark. If you do that you've then got to take the philosophical leap to describe everything as arbitrary, so forgive me if I don't chose to take much stock in people who are so incongruous that their arguments boil down to philosophical Nihilism--that is, unless they're (unsuccessfully) trying to be funny, or they haven't yet achieved normality after an encounter with an infinite improbability drive, and they're currently walking around as Frederich Nietzsche with a bowl of Petunias on his head...

I say bull. Here's why

[5n3ak3rp1mp]

Of course..... the question boils down to "is intelligence a physical process?" Everything we know about the brain's operation says that the answer is a resounding "yes"

You are right that that is the essential question. I am a little amused at everyone's assumption that the entirety of life can be boiled down into physical processes, however. I can't explain it but in my 34 years on this earth, having been a psych major and a CS minor with a concentration in biopsych and evolution and having taken a keen interest in life in all its forms since my earliest memory of reading Ranger Ricks and National Geographics and watching Mutual of Omaha's Wild Kingdom, having family pets and volunteering for Zoo Crew and whatnot... even having played with things like Conway's Game of Life, read about Rodney Brooks' MIT robots with "emergent" behavior, having devoured every AI simulation and game that I can get my hands on over the years... my gut tells me that there is more, there is some kind of spark to it all, to all living things. If this very idea is turning you off, then perhaps your mind is a little too closed... But my current opinion is this:

Life... genuinely surprises, in a way that a calculated simulation... can't.

Though I do think the exercise in attempting to do this will be VERY educational. Hey, and if we pull it off... I stand corrected. But I think it is very foolish to assume "Of course".

If everything could be explained by physical processes, then we should also be able to create life in a test tube, and we have yet to pull that stunt off either... and frankly, I don't think we will, and that even a simple bacterium that is able to reproduce may remain forever out of our reach. Again, this is more a hunch than a proof... though again, I think the exercise in attempting to do so will ALSO be very educational.

I have a favorite anecdote from this really awesome book. The crew (which was on a tugboat in the middle of the ocean that became a popular seagull resting place) was attempting to figure out a way to keep seagulls from pooping all over the deck and making it dangerously slippery. One of them thought to rig an electric line around the whole boat where the seagulls perched. If too many were resting, they'd throw the switch and away the seagulls would go. They built it and it worked exactly as advertised. They were very happy with themselves for coming up with this. However, a couple of weeks later, an interesting development occurred. On an otherwise typical day, the seagulls were getting too numerous again and someone threw the switch. All of them flew away... save for one, who had lifted its leg. By lifting its leg it had broken the circuit through its body, preventing it from being shocked. They released the switch, the seagull lowered its leg. They threw the switch again, it raised its other leg... The following day, fully HALF the seagulls had learned to perform this feat (no pun intended). By the next day the crew found that they were coordinating Synchronized Seagull Foot-Lifting instead of actually shooing birds away, when they threw the power.

And this is exactly the type of "surprise" I'm referring to. ;) A genuinely creative solution to a never-before-encountered problem. Again, if I'm wrong... Great. Perhaps it was just a random walk through meatspace that caused them to stumble upon this solution, realize it, store it in the wet memory banks, and communicate it to the others, per whatever punishment/reward system in their brain directed them to. But I still think you're being more presumptuous than you think.

I know this probably smacks of metaphysics, spirituality, all that gross bathwater that we dismiss with creationism (and boy, do I dismiss creationism...), or at least, it hints at one more fundamental aspect of life that we perhaps have missed. But maybe there's still a baby there. Again, just a gut feeling...

Re:I say bull. Here's why

[JohnFluxx]

Hi,
    I love stories like the one you gave. My favourite is with dolphins. They found that a particular dolphin picked up a trick whereby it would use some seaweed (I think) to cover their nose while grubbing for food in corral. The seaweed protected it from getting scratched by the corral. Any dolphins they came in contact with the dolphin learnt this trick.

    Download, through bittorrent, 'Living together' - BBC documentary.

    However I do think that life is just a series of physical process. When you look at biology you realise how powerful a collection of simple neurons etc can be.
    Look at ants. Each ant behaves in a very simple way. And yet put together they can have complex emergent behaviour.

Not virii, viruses

[Anonymous Coward]

The plural of virus is not virii it is viruses, virii is the plural of "man" and means "men".

http://en.wikipedia.org/wiki/Plural_of_virus
and for those who have access to Britannica
http://search.eb.com/eb/article-9106000

Re:Not virii, viruses

[Anonymous Coward]

virii is the plural of "man" and means "men"

No. viri is the plural of vir ("man") and means "men". virii doesn't mean anything at all, at least not in Latin.

Why not call a "Poser-Simulation" a Life Form?

[Anonymous Coward]

What people did here was
aranging a bunch of spheres
(10^6) on predefined positions
(modeling protein structures
as in found in crystals!),
then defining a potential
of it (mostly Lennard Jones
and some Electrostatics),
then draw a Force calculation
from this crude model - move
some spheres accordingly.

Thats what they did. They
may call it "Life" because
it intentionally was modeled
after something "living".

So is any "Human Model"
in "Poser" or "3D-Studio".

But most people don't know
noting about molecular simulations
and are therefore stunned by
such statements as made in the
articles referred.

my (you know) 0,02

Re:Life vs. Non-life

[Vicsun]

Human beings do nothing outside a very specific environment tailored to their needs, where temperature, pressure, oxygen content of air, gravity, radiation, &c., all lie within specific bounds. How is this different from a virus needing an environment that includes cellular structures in order to replicate?

If the temperature, pressure, oxygen content of air and gravity don't meet a life form's specific needs, the life-form dies. A dead life-form doesn't eat, replicate or metabolize, even after it goes back to a suitable environment.

Living things would either die, metabolize, or replicate; viruses can do neither. I was about to type more words but thought that I'd instead link to a previous comment which explains everything pretty well.

So in conclusion, dongs.

Life and the living

[jandersen]

I think we have to distinguish between 'life' as a concept and 'living organisms'. Life is an abstraction - it is the 'quality' that is common to all life processes, ie the processes that we know from living organisms.

The only thing that is reasonably clearly defined is 'living orgnism'; and as several posts have already pointed out, viruses can't quite be called living organisms; not because the don't display life, but because they are too simply to qualify as organisms. However, they do have life proceses - eg. they reproduce.

How can one define the concept 'life'? It is a difficult one - there are many that feel it would be too narrow to define it simply as the set of chemical processes that we know from biology; among other things, there is no sharp boundary between simple non-organic chemistry and 'life-chemistry'. There are some that define life as chemical evolution (http://en.wikipedia.org/wiki/Chemical_evolution) - this theory has the advantage that it can be generalised; all that is needed is a good generalisation of 'chemistry'.

Re:First Digital Simulation of an Entire Slashdot

[Anonymous Coward]

incredibly slow simulator == human brain? :P

sounds like a lawyer/prosecutor/fbi

[cheekyboy]

Damn ,thats the best description of a lawyer / fbi agent / prosecutor i have seen in years.

What came first, the virii, or the cells?

20 minutes later

[Anonymous Coward]

it was surfing the internet getting its first $20,000,000 offer from a 415.

Life = Non-life

[foxxo]

"Life" and "non-life" are useful, but ultimately meaningless ontological distinctions that really have no purpose at the sub-microscopic level. Any sensible person can see that ultimately there is no difference between what we deem living and what we call non-living, as the quick and the dead are still naught but particles in relationship to one another. The notion of self-identical objects larger than the fundamental particles is useful, but when dealing on such a tiny scale it's best to forget about such pointless ontological nonsense.

porn versus live sex

[peter303]

Thats my reply for simulation versus reality. Take your pick.

life

[beautiful leper]

I know this goes against the face of definition but assuming life is defigned as self replicating than the whol universe could be concidered an organism. For instance there are theorys that meteors seed and act as a catalyst to life on "furtle" planets. The univer could be concidered "gods" brain. I'm agnostic so I deed you damn pesky athiests crap'n on my style. Thats my two cents

Re:Life vs. Non-life

[Anonymous Coward]

Ever heard of taking time to die. A spore is living but can live outside it habitant for along long long time before dying maybe the virus can last even longer?

Drivel

[I Like Pudding]

And yours is possibly the least. The granparent is full of shit, and the moderators are lapping it up like caviar. There is no such thing as life? THEN WHY IS THERE A WORD FOR IT? Life is a thing-state-process that has been observed for as long as there has been the apparatus to observe. The concept is older than we are, and the word is well defined.

The debate stems from from the fact that virii form and process differ greatly from pro- and eukaryotes. They are dead simple. In fact, they are inanimate simple. Virii are a corner case. Dismissing the word-definition-classification of life entirely due to it being a human semiotic unit is worse than Godwinning. You might as well shout "EVERYTHING IS SUBJECTIVE SO I CAN'T BE WRONG, AND WRONG IS A MEANINGLESS CONCEPT!".

Donny: Are they gonna hurt us, Walter?
Walter Sobchak: No, Donny. These men are cowards.
Nihilist: Okay. So we take ze money you haf on you, und ve calls it eefen.
Walter Sobchak: Fuck you.

Re:Drivel

[Tomfrh]

It's "viruses" damn it!

Re:Drivel

[shaitand]

'You might as well shout "EVERYTHING IS SUBJECTIVE SO I CAN'T BE WRONG, AND WRONG IS A MEANINGLESS CONCEPT!".'

Sounds about right to me. That is the conclusion of any cold legitimate path of logic that does not have an unfounded need for "right" and "wrong" to exist in the moral sense. In fact it is technically true outside the moral sense if one does not qualify with a perspective and scope.

"right" and "wrong" are relative and there is no standing counter argument to this that boils down to more than "nuh uh".

Re:Drivel

[Wizardry Dragon]

I'd reply with an actual comment, but there is a sign between you and I that reads, "Don't feed the trolls"

Re:Drivel

[I Like Pudding]

This was not a moral question, this was a classification question. Saying "that's subjective" then sticking your fingers in your ears and shouting "LA LA LA LA LA!" is the biggest logical cop out you can make.

Re:Drivel

[I Like Pudding]

OH MY GOD SOMEBODY IS CALLING ME A FOOL BECAUSE I SAY INCREDIBLY FOOLISH THINGS! HE MUST BE A TROLL!

Do you even know what a troll is? I can't be a troll by virtue of my genuine belief that you are an idiot. Trolls just start fights to get attention, whereas I start fights because seeing such pap get modded to +5 makes me angry at the world. Had you actually read my post, you might have noticed this, what with the antitheses and all. I swear, you are what happens when an educational system stops training rhetoric and dialectic.

Re:Drivel

[Wizardry Dragon]

I'd say something to shut you up, but it's amusing to watch someone be exactly what they complain about.

Re:Drivel

[shaitand]

I was not aware that "cop out" qualified as valid grounds to dismiss a logical argument. If "cop out" is the best you have to refute this line of logic I am afraid I will have to continue to maintain that it is valid.

As I said, there is nothing in logic to refute this line of thinking that amounts to more than "nuh uh".

Re:Drivel

[motherhead]

I'd say something to shut you up, but it's amusing to watch someone be exactly what they complain about.

Did you just say, "I know you are but what am I?" Is that a troll?

You've gone and done it now!

[modecx]

I'm rubber, You're glue. Whatever you say bounces off me and sticks to you!

And so long as we're on proverbs: A nod's as good as a wink to a blind horse!

Re:First Digital Simulation of an Entire Slashdot

[stonecypher]

And it took 13 days...That's one slow simulation.

Eh, it's Perl, what're you gonna do...

Running on Linux / SGI hardware

[halfelven]

Another link with more info:

http://access.ncsa.uiuc.edu/Stories/TobaccoMosaic/

What was the bug

[themusicgod1]

and did you squish it?