Talk:Artificial life/Archive 1

This page is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

DO NOT EDIT OR POST REPLIES TO THIS PAGE. THIS PAGE IS AN ARCHIVE.

This archive page covers approximately the dates upto November 2006, before Numsgil's major refactoring work.

Post replies to the main talk page, copying or summarizing the section you are replying to if necessary.

Please add new archivals to Talk:Artificial life/Archive02. (See Wikipedia:How to archive a talk page.) Thank you.

Start of talk page

In the first sentence, "human-made analogs"; shouldn't it be "analogues". I'm not too sure though.

There is nothing (I know of) that says that artificial (read intelligent) life cannot be a result of a neural net, artificial or not, either simulated in programmes or physically laid out. I did not add the neural network reference, but I don't see why you are dismissing it out of hand. Could you explain your position a bit more fully? - Reigh

Ehr. I dont agree with that! Life does not have to be intelligent - intelligence does not have to be incorporated into life. An neural network does not create children, instead it augments itself. Neural networks dosnt have any need for any incoming energy (food), and could therefor not be considerd to be alive in the TRADITIONAL meaning. A neural network is a mind without a body - that is no life, it is just an mind. what exactly is life? (i dont belive in life) - smaffy

First of all the fact that something does not have children does not mean that the element is not alive. Also from historical reasons, Neural Nets ARE part of the Artificial Life research. Neural Nets articles are a tool used by people in Artificial Life, and often a medium of research. Articles on Neural Nets, are part of an Artificial Life conference. So, I put it back. Pietro

The phrase particularly well defined doesn't make sense to me in:

"The field is particularly well defined by the tools it uses"

Could we just say

"The field uses" ...
"The field is defined by the tools it uses", or
"The field is associated with tools it uses"

- Jeff 23:09 Nov 4, 2002 (UTC)

Yep, it does look weird, I have rewritten the section as "the field is characterised the extensive use of computer programs..." and also the surrounding sections quite a bit. --Lexor 11:22, 27 Sep 2003 (UTC)

Quick request... can someone knowledgeable disambiguate the "reversible" link in "Toffoli later provided a key proof that CAs were reversible" under the 1970s-1980s section? It is not obvious to me whether it means reversible like a physical process or like a mathematical function, so it's not trivial for a random person to be able to disambiguate it themselves from the link. Thanks. -NereusRen 22:31, 10 May 2006 (UTC)

I think "reversible like a computation" is the best bet here. Michael Ralston 23:15, 10 May 2006 (UTC)

reverted text

I reverted this text out of the article. I'll leave it to someone familiar with the topic to decide how/whether to incorporate this info into the article in a proper format. - Kbh3rd 05:02, 23 Jan 2005 (UTC)

(My first attempt at editing... May be the moderators can fix the links or edit it if this is not relevant) There have been very important developments in the biological and genetic aspects in this field in the last 5 years. I will list these as a time line with the links in text format. Saturday, 18 December, 2004, 07:43 GMT Scientists create an artificial cell that can hold DNA to become artificial life. Some Researchers at Rockefeller University in the US have made the first tentative steps towards creating a form of artificial life. Read more in this article. http://news.bbc.co.uk/1/hi/sci/tech/4104483.stm
Thursday, 13 November, 2003, 19:04 GMT Scientists use DNA to make virus. The naturally occurring virus does not infect people US scientists have produced a wholly artificial virus using a method they claim could lead to new lifeforms. http://news.bbc.co.uk/1/hi/sci/tech/3268259.stm
Thursday, 11 July, 2002, 23:28 GMT 00:28 UK First synthetic virus created Scientists have assembled the first synthetic virus. The US researchers built the infectious agent from scratch using the genome sequence for polio. http://news.bbc.co.uk/1/hi/sci/tech/2122619.stm
Wednesday, 21 February, 2001, 04:15 GMT Synthetic virus nearing reality Scientists must overcome hurdles with DNA before synthetic lifeforms become a reality Scientists will have the technology to create a wholly artificial virus within the next five years, a major conference in the US has been told. http://news.bbc.co.uk/1/hi/in_depth/sci_tech/2001/san_francisco/1181710.stm
Thursday, 21 November, 2002, 12:29 GMT Genome man to create new life Scientists in the United States are to press ahead with plans to create a new lifeform in the laboratory. Dr Craig Venter - the man behind the privately funded human genome sequence - and Dr Hamilton Smith - a Nobel-Prize-winning geneticist - want to create a man-made microbe with the minimum number of genes needed to sustain life. http://news.bbc.co.uk/1/hi/sci/tech/2499119.stm

Updating

A lot has happened in A-Life since the 80's...

Indeed - it might be an idea to add some information about Steve Grand, his Artificial Life projects (Lucy, the Creatures series of games for PC) and his books on the subject. ~Liamlala of the Creatures Wiki

Although I should point out, Steve Grand's work is not universally accepted within the scientific world.

I notice that both John Von Neumann and Stephen Wolfram get fancy epithets. I doubt we need to inform the reader that Von Neuman is a "math and computer prodigy" or that Wolfram is a "brilliant and controversial scientist". They probably should be removed. - 69.105.95.207 16:43, 29 May 2005 (UTC)

Langton

There really should be more mention of Christopher Langton's contributions. He has, to a certain extent, been under-credited.

Possible copyright violation?

The History and contributions section is duplicated at: http://www.artilifes.com/importantcontributors.html This section was contributed by User:CatherineMunro. Does anyone know if she copied it from there, or if they copied it from here? MakeRocketGoNow 02:12, Jun 25, 2005 (UTC)

Well, I certainly didn't copy from elsewhere; I was making drastically summarized sketches of personages from the Levy book Artificial Life as I read. (Really need to go back and add some more, I was about halfway through the book when I had to stop.) The entire text of the website appears to be copied from this article, one section per page. It's okay for them to copy, but they don't credit Wikipedia anywhere, and don't mention the GFDL; we should drop a mention of them at Wikipedia:Mirrors and forks so someone can contact them about proper licensing. — Catherine\^talk 02:31, 25 Jun 2005 (UTC)

Error: Free University of Brussels termite simulation?

I've been researching the termite simulation that was done by the Free University of Brussels, and according to everything that I've found, the simulation consisted of creating differential equations to model the the construction of termite mounds, not "They wrote a script describing the behavior of termites, then modified the environment and watched the way that the simulated, script-driven insects reacted" as stated in the article.

The paper I'm referring to is "A model for the emergence of pillars, walls and royal chambers in termite nests" located at http://www.journals.royalsoc.ac.uk/media/4H8ACMVWRN4RW1L4FATN/Contributions/Y/T/P/U/YTPU13GT3GUX75TC.pdf

In the discussion section at the end of the paper, under "Limitations of the model", it states:

2. Given its formulation in partial differential equations, the model cannot accurately describe individual behaviour.

This would suggest that there is an error in the Wikipedia entry. However, given that this is an entry which has been copied verbatim to and from several online encyclopedias, I'll put this up for discussion before trying to change anything.

Group behaviour self-optimisation system creates soccer players specializations

Could anyone please write an article about application of step-by-step group behaviour self-optimization (neuron) algorithms to the football (soccer) game emulation?

I read somewhere that an institute carried out such research/experiment (in UK, USA, or in Russia), by creating a field, two gates, virtual ball and two groups of agents (two teams), and the basic rules of goal scoring and accounting (there was no arbiter agent). It was quite a while ago, so I can not remember whether I explain the experiment precisely.

At the time of firts turn, all players/agents had no specialisations and did not "know" how to behave themselves. All of them were given multiple chaotiс movement impulses until a ball was somehow crossing one of the gate lines. Winning team of agents was "learning" their experience, adjusting their behaviour pattern (neuron chain reaction), while the losing team discarded their actions in the game as unsuccessful.

After many, many turns of such learning and "evolution", not only virtual teams of agents became experienced in winning matches and scoaring goals. Both teams' agents were naturally gaining specialisations, that were created consciously by human for the current version of football (soccer): 1) Goalkeeper, 2) Defense players, 3) Midfielders, 4) Offense players.

Since I can not find through search systems original source/insitution of this research, I would like to ask people from the community for help. (By DenisRS, 7th August, 2005.)

(Of course, I did not mean any robotic football competitions and championships; I meant strictly virtual computer programmed self-organization system experiment.)

Term coinage

Computer scientist Christopher Langton coined the term in the late 1980s

I dispute that. The term was long in use. I found it being used in a list of 'predictions about the future' made by Glenn T. Seaborg et.al. in the 1960s. bogdan | ^Talk 18:30, 12 October 2005 (UTC)

Seperation and rewriting

It seems to me that there are two areas that are trying to be conveyed by the same article. The main article describes the history of what I would term software AL. Things like cellular autonoma, etc.

Then the "see also" links to what I would call engineering AL. This would involve physically creating a new life form, robotics, etc.

I think the article should be split and expanded along those two paths.

I don't think software AL should follow a chronological organization. Perhaps start with a more standard outline of the different techniques in the field. Cellular autonoma, theoretical biology, agent based models, etc., moving into the chronological history much later.

There really aren't any encyclopedic articles on artificial life on the internet that I think are of high quality, a concerted effort of pulling from different sites to provide a high quality introduction I think would be a worthwhile goal.

--Numsgil 02:37, 14 February 2006 (UTC)

I too am not particularly happy with the present layout. But before splitting it I would suggest to migrate the information about each main researcher in its own page. Like this it seems like a summary of a book on artificial life. Also many of those researchers have made contributions in other fields, so it would make sense to give them their stand alone page, and just link there, if necessary. We could just add a list of scientist per discipline, in case. Once this is done, the history will feel much lighter, and we can consider (if, and in case) how to split.--Pietrosperoni 18:32, 15 February 2006 (UTC)

I don't know that I like the idea of a list, at least not just a list, as I think that's part of what the current problem is with the current page (too many lists). A list of 20 names isn't particularly lucid. Perhaps a list of names with a short blurb or paragraph about each's contribution. --Numsgil 19:26, 15 February 2006 (UTC)

Of course not just a list. And it does not even need to be included in the page, could be a separate page or better a category page (if anyone knows how to set us something like this).

About the division, there are a number of ways to split this article. Personally I think the most natural one would be between strong alife, and weak alife. But then strong alife means anything that tries to recreate life. As such some CA would be there, some wet alife would be there too. On the other hand robotics would be not so much there, since robots are generally understood as just trying to implement one aspect of life (like motion, or sensor/decision) and not the most basic feature, like: self healing; metabolism; evolution...--Pietrosperoni 18:32, 15 February 2006 (UTC)

Strong and weak Alife as far as I understand is more a behind-the-scenes motivation than an actual difference between methods. CA could quite easily be either strong or weak alife simply by how the scientist/hobbyist views it, so I don't think it's a practical method for splitting the article. I think making the page into something similar to the biology or physics pages would be optimal. --Numsgil 19:26, 15 February 2006 (UTC)

I am really weary about a division in biology and physics, or in any way that too close mirrors other divisions in science. AL has often been a point for scientist of different disciplines to meet, and this should be represented in the way in which the page is presented. Natural sub parts (but not apartition) are in general seen as: wet, weak, strong, robotics, ac (artificial chemistries, who already have their page), ca (who have their page), agent based models, tierra models, language models. Some of those mirror science division: wet-biology and chemistry; language models-linguistic; robotics-engineeristic and other. The whole field have an incredible amount of cross pollination between subfields. Any division would make the topic more superficial. Also consider how there are many pages on internet way more complex or longer than this.--Pietrosperoni 05:29, 16 February 2006 (UTC)

I'm not sure, but I don't think you're understanding what I'm meaning (assuming you're replying to me). I mean that the layout of this page should be similar to the layout of the biology and physics pages, ie: a prepatory article designed to introduce lay people to more specific sections of the category. --Numsgil 05:40, 16 February 2006 (UTC)

Aha! Yes, now I understand, yes I do agree totally that we should strive for that! Also there is a cool quote form (I think) von Neuman who sais that he thinks nothing can be intelligent if it is not alive (or similar). The quote helped move the interest from AI to AL. I see if I can dig it.--Pietrosperoni 05:56, 16 February 2006 (UTC)

Addendum: please read the section above on possible copyright violation, in this page. I was quite right about the sources and format of the historical section. Should we just take it off?--Pietrosperoni 05:34, 16 February 2006 (UTC)

Doesn't Catherine above say that she herself wrote the article, and that the other site was probably copying from her? --Numsgil 05:40, 16 February 2006 (UTC)

she does. Said that copyright covers also summaries [[1]], yet this is still not the reason why I was wondering if we should take it off. More like the sensation that the whole part did not resonate at all with what you want from the general page of a field. If you look at other fields they do have the history, but in a separate page. And still this history in a scientist by scientist way seems too detailed, and missing the general breath we should give to this article. An historical part made up of what were the challenges, and the general aims seems more appropriate. As I said, this seems taken from a journalistic book about the subject. Not from an encyclopedia.--Pietrosperoni 05:56, 16 February 2006 (UTC)

Opinion of Biologists

It is not correct to say that the field of Artificial Life has "has not generally received much attention from biologists." The fact is that Eörs Szathmáry, a close associate (and seeming heir apparent of Maynard-Smith) is one such biologist who regularly attends academic conferences related to the field.

The lament of Maynard-Smith,

“One reason why we find it so hard to understand the development of form may be that we do not make machines that develop: often we understand biological phenomena only when we have invented machines with similar properties … and we do not make ‘embryo’ machines”.

is, in fact, a current hot topic of discussion. Consider the subjects of interest in the upcoming Workshop on Machine Self-Replication (website not yet available), a component of the Tenth International Conference on the Simulation and Synthesis of Living Systems. William R. Buckley 19:44, 20 February 2006 (UTC)

Definition in intro

I think the recent additions to the intro are too specific and give an inaccurate summary of the field: "...evolving software that is more alive than a virus. Theoretically later it will become intelligent life [Strong AI]."

Much a-life does not require "evolving software" — swarm intelligence, cellular automata, etc. ("evolution" is a loaded word in this context and should be avoided if the actual mechanism or analog of evolution is not intended)
Is the (biological) virus threshold of "more alive" widely accepted?
Without a reference, the "Theoretically later..." sentence seems like speculation.

--Ds13 19:24, 26 July 2006 (UTC)

going to agree with Ds13 --Numsgil 20:35, 27 July 2006 (UTC)

While there may be a few exceptions, most life is made of existing life. If patterns replicate in cellular automata, and they dont adapt, they're only a virus. If they do adapt, then some of them are better than others and those are kept while others are discarded, and thats evolution. Of course they dont have to be discarded, but thats just chaos, not life.

"Theoretically later" is supposed to be speculation because we have no facts about the future, but its obvious thats the direction its going, to merge with Strong AI.

About the 2 Strong AI links. All Strong AI is Strong ALife, but not all Strong ALife is Strong AI. Maybe I shouldnt have added the second link. But is there a closer synonym?

BenRayfield 18:23, 31 July 2006 (UTC)

The difinition of life

There is a great theoretical book which adresses the definition of life. If anyone is interested, it's called "The principles of life" by Tibor Ganti.

Mirjan Švagelj (mirjans_at_email_dot_si)

November 2006's New Page

I've rewritten the main page entirely. The most possibly controversial change is the total removal of the list of notable figures (some of which aren't that notable anyway). Please tell me of any major problems with my change. In a week or so I'll be deleting or moving passages from this talk page that no longer apply. --Numsgil 04:41, 5 November 2006 (UTC)

Well done, and much needed. I personally think the open questions in artficial life should remain even in the new versionas a paragraph in itself. Also, it would be interesting to make a philogenic tree of models and programs. For example Tierra being the predecessor of Avida. etc. I moved the philosophy above, so that we can refer to it in the rest of the page. And I connected to the weak alife in the optimization problem part--141.35.12.219 17:17, 27 November 2006 (UTC)

Most of the open questions seemed to relate to wet alife, which I really think should go in the wet alife article, especially since that article is so underdeveloped. I'm thinking this "main" alife page should strictly cover computer models, since that tends to be the most common "alife". Mixing the two in the same article really makes things messy. Other than perhaps some base philosophy, they differ by just about everything possible.

I think a phylogenic tree of the different programs would be very interesting, though I'm not sure all the simulators in the digital simulator page could be directly added to it. At the very least a limited diagram picture would help add some more graphics to the page.--Numsgil 20:32, 27 November 2006 (UTC)

Removal of link to article by Subhash Kak

I removed the link to http://www.acm.org/ubiquity/views/v6i42_kak.html because I felt it falls in the realm of original research, and doesn't seem to be more than casual link dropping. I'm certainly for discussion of criticism about Artificial Life in the article, but I'd prefer it to be a generalization of many people's viewpoints, and I'd like it to go in its own section. --Numsgil 04:35, 17 December 2006 (UTC)

What "artificial life" is not

We have been at this for some time. Various positions put forth, and some concensus developed, it is probably a good time to boil this discussion down to a series of bullet lists, one for each author. This way, we can remove the verbiage, and produce an article that is mutually satisfying to us all. William R. Buckley 20:47, 28 December 2006 (UTC)

"Artificial life" is not the study of a phenomenon, it is the phenomenon itself. "Life" is not the "study of living things", the name for that is "biology". Similarly, the "study of synthesized / manufuactured / artifically created life" would be "artificial biology", not "artificial life". Paul Beardsell 00:28, 18 December 2006 (UTC)

Not quite, Paul. Biology is the study of natural life. It is not the study of living things. If the living thing does not originate within the natural world, it is not the subject of biology. Biology is now, and will forever be, the more subject-specific field of study. William R. Buckley 07:43, 25 December 2006 (UTC)

I looked at the definition attributed to Chris Langton, on the zooland website. I agree with Chris that nothing in principle prevents biologists from studying living systems other than the carbon based systems found naturally on Earth. I do not agree with him that biology is about all living systems. Biology has always been dedicated to the naturally occurring living systems we find on Earth. Indeed, in the next sentence, his reasoning respecting the limits thereput to theoretical biology proves the claim, and justifies creation of a new field. It is this new field, artificial life, that accepts living systems of any form and foundation.

Though in principle biology could subsume artificial life, it is not likely to be that way. The terms carry baggage in opposition, and this will reinforce the current usage of terms. Were some life form to visit us from deep space, we would apply the sugar/lipid/amino acid/nucleic acid test to determine if it is biological. If not, then a new category would be created, and a new field named.

Biology is about carbon based life. Artificial life is about all other life. Biology knows well the effects of contingency upon language. In time, the colloquial use of terms feeds back into academia, and there too this current usage of the terms biology and artificial life will be reinforced. Hundreds of years from now, these terms will remain distinct. William R. Buckley 08:28, 25 December 2006 (UTC)

In the article, again and again, we say things like "artificial life is the study of simulations". No! It might be a "pointless life" to do such a thing, or an "exciting life", but it is not "artificial life"! Fixed in article. Paul Beardsell 00:28, 18 December 2006 (UTC)

The term "artificial life" applies to a very specific field that intersects biology and computer science. The article specifically refers to that field of research and artistic expression that is catalogued in the Artificial Life scientific journal and the periodic gathering of interested participants for the ALife conferences. Artificial Life is the study of life and life's evolution, as a subset of the larger Biology using methods and materials involved with computer science. Artificial life is often said to be a subset of Bio Informatics.

Actually, Numsgil, you are wrong on this point. As Christopher G. Langton stated at the first alife conference, which I attended, alife is the study of life as we know it within the LARGER context of life as it might be. You got that point backwards. Alife is not a subset of biology. Indeed, biology is a (proper) subset of alife. William R. Buckley 19:42, 18 December 2006 (UTC)

Another take on this seems appropriate. Identification of the proper relationship between fields of study is not always an easy task. I will not accept that alife is a subset of biology. However, it may be best to understand the relationship between both fields as having an intersection that is empty. I am torn, it would seem, between biology as a proper subset of alife, and the two being non-intersecting proper subsets of a larger whole, say the field of living systems. This may really be the notion that Langton supports. He used the phrase *life as it might be* and for me, the word *might* suggests inclusion of all cases. On the other hand, the practical reality is that alife models derive from non-biological sources. Oh, there may be a few cases where the subject matter examined is biological via alife methods. It is hard to tease these two fields apart, and alife has the larger scope, it admits more methods and more media, than does biology. At least the fields are on par, and reasonably alife contains biology.

I should comment on the word *might* versus the word *could* as used in the immediate discussion. For me, in the context given hereabout, both words refer to the notion of possibility; they are synonymous. As I recall, Langton used the work *might* at ALife I. Elsewhere, I see the phrase to include the word *could* instead of *might* such as may be used in published papers. Bottom line: life-as-it-might-be means the same thing as life-as-it-could-be. William R. Buckley 22:31, 19 December 2006 (UTC)

Yes, I see your point. Originally my point was that Artificial Life is generally defined in any university setting or in grants as a field within Bio Informatics. In reality, it's classification is difficult to define. In my mind, ALife exists in the void between proper Biology, Chemistry, and Computer Science. Biology does not neatly fit within the confines of the current ALife field any more than the current ALife field dits neatly within the confines of Biology. This might actually be a large part of why funding for ALife is so scarce compared with Artificial Intelligence (not that AI gets all that much money anyway). AI seems to fit neatly within Computer Science (not that it doesn't draw from other fields). --Numsgil 04:32, 19 December 2006 (UTC)

One more point. Alife includes the computational and the physical models. Wet alife is alife. It is not an estranged subfield, a bastard child, or the like. Alife as a field does not discriminate between the models. Both are equally part of the field. Hence, the construction of the opening paragraph of the article is not sufficiently balanced. It draws a distinction between model types that simply is not reflected within the field, and would be rejected by field researchers. This distiction needs to be removed. William R. Buckley 20:20, 18 December 2006 (UTC)

I agree, as I've reread the opening paragraph several times this last week this has stuck out at me more and more. But we can't ignore the fact that ALife in silico is almost always just refered to as "ALife" unless its being compared with Wet Alife, while the reverse isn't nearly as true. The opening paragraph could be worded better, but the article segregation of the two still seems proper. Other than sharing some basic philosophy, the two are about as different as two things can be. The methodology, materials, skills, etc. do not overlap neatly. Cellular Autonoma do not exist as a useful construct within Wet Alife anymore than physical proteins (as opposed to the idea of proteins) do for in silico Alife. The opening paragraph probably needs to start as describing ALife as a whole and the clearly differentiate between Wet Alife and in silico Alife, and then proceed to state that since the term Alife is also used to describe ALife studies in silico specifically, the article will explore that particular sub field. --Numsgil 04:32, 19 December 2006 (UTC)

You are specifically thinking of artificial life being an adjective followed by a noun. Meaning life that is created artificially. This article is about a compound word "artificial life" being used to describe a very specific scientific and artistic field that deals with evolution of agents. In your context, metaphysics is the rule. This was largely the problem with the old article, and the problem I'm specifically fighting in the current page. This isn't the article to discuss the implications of androids, bio engineering, etc. etc. It is only being used to describe a very specific field of research. --Numsgil 01:47, 18 December 2006 (UTC)

Numsgil, I argue that the article is about both the field of research, and the items researched. Thus, alife studies alife; the field studies its subject, both having the same name. I should point out that in reworking the opening paragraph, nowhere does it state that the subject of the field, and the field itself, are named identically. Indeed, there is no need to explicitly state so. Rather, this name usage can be easily transfered to the reader by connotation, through carefully worded exposition.

The change you made thereafter does help to simplify wording, a benefit to pedestrian readers. I believe my change further improves readability.

I agree, thanks ;) --Numsgil 09:32, 27 December 2006 (UTC)

What is needed is a page called 'Artificial life examples' instead of a new name like 'synthetic life' as the former gives more consistent naming and context. Efforts to build an hierarchy of articles to cover the topic of alife is a laudable goal. Consistent naming can facilitate that goal. William R. Buckley 07:15, 25 December 2006 (UTC)

I think there's a conceptual problem in calling what alife studies "artificial life". Issues of confusion aside, a single artificial life form, say an AI that passes the turing test, while maybe called a life form, isn't really part of what alife studies. That's the realm of AI. Likewise, there are some things which are studied in alife but probably wouldn't be considered alive even under the most lenient definition. Conway's game of life for instance, or genetic algorithms. What I call "synthetic" life, and the agents studied in alife, do have some overlap, however, neither is contained entirely within the other. Add to this the fact that calling what alife studies "artificial life" or even "synthetic life" denies the legitimacy of the question between weak and strong philosophies. That is, you seem to automatically consider it alive. I don't know what wet alife would call what it studies, but I think the term "digital organism" or even just agent, when applied to alife in silico, is a far more appropriate label (and far more common) than anything with the word "life" in it. --Numsgil 11:41, 28 December 2006 (UTC)

It is perhaps easy for you to see that possession of the property of being alive does not imply possession of the property of being intelligent; for example, fish are stupid. Less easy to see is that possession the property of intelligence does not imply possession of the property of life. AI is not the provence of AL. I would expect that little discussion of AI occurs in AL settings. Indeed, at ALife X, I did not observe a single talk respecting an issue of intelligence within an artificial system. Agreed is that most of the models are lacking. The key element they miss is embodiment. They are therefore models of life, not life itself. I think Bob Ulanowicz expressed his concerns with alife models using the word *entailment* though I do not recall exactly. Simply put, the processes of life are not employed in AI models, and so an AI is not an alife. I agree with the strong position, and suggest that if you do not so agree, then you are at minimum on the precipice of either bigotry (biocentrism) or vitalism (there is something extra-physical about life - something not available from the matter of the universe).

Exactly, AI is not the provence of AL. But there might be a time in the future when AI reaches the goal of something that could be considered alive, depending a great deal on your definition of life. There is an overlap between what would be considered valid objects of study in Alife and what lay people might call "artificial" life forms, but it is far from a perfect overlap. An article that discusses the idea of artificially created life and an article that discusses valid objects of study within the ALife field need to be seperate. They are not the same.

A weak alife position may be the minority within the Alife field, but it is hardly the minority in science or even the general public. The present textbook definition of life taught in any bio classroom entirely prevents anything digital from being alive. We don't understand the processes of life well enough to make any claims about its reproducibility. There may well be some "vitality" that is not reproducable, and entirely prevents us from creating anything that is properly alive (despite how it would consternate countless people of science). My personal philosophy is that strong alife is possible, but most or all present simulators have not yet reached the necessary complexity to call their agents "alive" (primarily a lack of computing power is to blame). So I'm primarily playing devil's advocate on this point. The mass scientific concensus as I understand it is that there does not exist anything that could properly be deemed "synthetic life", and there have been more than one eminant biologist to posit that life in silico is strictly impossible. Robert Rosen for instance. An article on synthetic life needs to explain that. An article on artificial life examples does not. --Numsgil 11:41, 28 December 2006 (UTC)

Rober Rosen was mistaken. Nothing more. William R. Buckley 19:55, 28 December 2006 (UTC)

Probably, but that's not my point. Obviously there are notable biologists who would take a weak alife position (if that!). Therefore, we should not assume a strong alife position in any articles, and respect both positions. Which would require the seperation of articles discussing synthetic life (a theoretical entity) and examples of the subject of study of alife researchers (a real and present entity). --Numsgil 09:40, 29 December 2006 (UTC)

Jay, first, you are not giving enough time to the construction of your point. For instance, the second sentence above is unclear. You probably mean, "Obviously there are notable biologists who would not take either the strong on weak positions respecting alife." Do remember, there are notable biologists who do take the strong position, Tom Ray being but one example. I have personal contact with a number of well known and respected biologists, who have privately expressed their acceptance of the fundamental percept of alife, that life is a process and not a material phenomenon. William R. Buckley 22:44, 29 December 2006 (UTC)

You should carefully consider the difference between a completely computational model and one including physical expression. Robots that are able to self-replicate will experience the effects of natural selection. I hold a philosophical position that computation allows any system to circumvent the effects of natural selection but, that does not mean that such systems must avail themselves of such opportunity. Indeed, living systems, up to human kind, have simply not had access to the means to effect such wresting from chance, and the impending change to technology, such that germ-line therapy is functional, will allow humans to wrest their genetic destiny from the clutches of natural selection. Robots are a part of the physical universe, and so can be expected to exhibit the processes inherent thereto (i.e. the physical universe). William R. Buckley 22:44, 29 December 2006 (UTC)

The point about what is taught in a biology classroom belies the problem with mixing biology and alife. The biological definition of life is, ahem, biocentric. Start with a bias, and you get biased results. And, it is not true to say that no digital form has been judged as corresponding to a biological form - the computer virus, and this by Eugene Spafford. To the extent that a biological virus is a lifeform, Spafford accepts computational viruses as a lifeform, this acceptance coming during the ALife II gathering. William R. Buckley 19:55, 28 December 2006 (UTC)

No doubt there is a bias in the present definition of life. But acknowledging it as such does not mean it will suddenly go away. As an encyclopedic entry, any articles which broach this subject should do so flatly and without bias, which means even presenting points of view that are wrong if they are strongly supported or held, especially when evidence one way or another is not available. As for viruses, I would point out that it is not generally accepted that viruses are alive. This is still a point of heated debate. An article on synthetic life would be well behooved to do as the virus page has done and demonstrate that this is a subject of debate. An article on artificial life objects of study does not need to do this, because it is entirely immaterial if the object of study by alife researchers is alive or not. Again, the example of genetic algorithms. Not alive, but still valid research. Clearly an article seperation between synthetic life and any articles dealing with alife research is needed. --Numsgil 09:40, 29 December 2006 (UTC)

You give here another non-sequitur. Sure, acknowledging bias fails to eliminate it, and why should I expect otherwise. The important part of your statement is the flat presentation; for this, I am all in favor. I will strongly battle for language such as that which I use in describing the field subject of this article. I will also strongly back you for organisational efforts. There are reasons to break the article into many, and several have been discussed. There is also need for citable publications, so as to justify each and every sentence of the article. The term synthetic life is probably a well used designation. The problem with using same to build articles is that the term will have very specific implications associated with its use in the field, and same is not likely to be expressed through the open editing process of Wikipedia. Further, it allows the term to be defined within a general article, and so facilitate comparison and contrast of organism categories. William R. Buckley 22:44, 29 December 2006 (UTC)

Whether viruses are accepted as life or not is not the point of my argument. Rather, it is that some express their satisfaction that whatever is the phenomena that biological viruses represent, it is exactly the same thing with computational viruses. In this one case, researchers of various fields have concurred with the view that these two things are examples of the same phenomenon. I will review the virus page and consider the other point you make, regarding article construction. William R. Buckley 22:44, 29 December 2006 (UTC)

Do understand also that some in the field are specifically interested in whether a particular example represents life, or only a model of life. For instance, my goal is the construction of life, not models of life. I seek to know exactly those necessary components such that a living system can be initiated into life, thereafter to perpetuate its processes in the face of its environment, a computational one. Wet alifers also seek the same kind of result but, for the physical universe. William R. Buckley 22:44, 29 December 2006 (UTC)

You may also find interest in the words of Richard Dawkins, who at ALife I opened his talk with the statement, "There are some embryologies which are pregnant with possibility, and then there are some that are not." The last statement he uttered during his talk is, "Of all the systems that I have seen at this artificial life workshop, Core War is the least cheat model." So, when you look at the models that are offered, understand that he who is perhaps the most eminent of Darwinists says that some computational models are promising.

I am indeed interested, but the way you present those quotes it sounds to me as if Dawkins was lamenting that the majority of simulators in existance are cheating. Core War is the ancient predecessor to most alife simulators, and Dawkins saying that that was the best model would seem to point to his belief that alife research has followed faulty reasoning ("cheating") and is headed down the wrong course. A commonly held position among many biologists, as far as I can tell. This would also explain his shying away from alife research in recent years. But I'm reading the quotes out of context, so perhaps I'm reading them wrong? --Numsgil 09:40, 29 December 2006 (UTC)

That is just wishful thinking on your part, I think. Dawkins was rendering praise, because the element of natural selection is built into Core War. He spent the interim period between the previously given quotes expressing the degree to which he was committed to a Darwinian view. He was specifically pointing out that one model shown him had an image consistent with his image of biological life. He was trying to guide non-biologists, showing them what he as a biologist found interesting, promising, and analogous. Your subsequent sentence adds to the position I give, saying that the model shown in Core War was the notable exception to the cheating of other models. William R. Buckley 22:44, 29 December 2006 (UTC)

Don't be so sure that the problem with most alife models is computational throughput. William R. Buckley 19:55, 28 December 2006 (UTC)

I think that computational limits are creating limits on what can be done with simulators. I don't think that the majority of simulators in existance, given infinite computational resources, would ever achieve anything as remarkably intricate and complex as even a single bacterium. There seems to be a complexity ceiling, which has to do with the simplistic representations of life that simulators have to use because of computational limits. No computers in existance could handle the simulation of even a single real bacterium. Protein folding alone is a not a nail but a whole railroad tie in the coffin. In another, say, 50 or 100 years when parallel computing or quantum computing allows us to broach these sorts of issues, maybe we'll start to see some truly complex digital organisms. Or maybe not. But the present computing power is insufficient to create something that could be considered alive, IMO. --Numsgil 09:40, 29 December 2006 (UTC)

Throughput is always an issue. No user wants to have to wait until they press the enter key in order to get an answer in return. Delay in computation will ever be a source of annoyance. William R. Buckley 22:44, 29 December 2006 (UTC)

Not everybody is interested in simulation. Infinite computational resources applied to a single bacterium, seems like overkill. Simulation of the physico-chemical processes, atom by atom, state by state, given some acceptable quantisation, is likely supportable upon less than infinite computational resources. You will kill your argument with hyperbole. As for protein folding, I think it is an NP problem, and that P does not equal NP. Indeed, you can formulate the protein folding problem as the corollary to the traveling salesman problem. In TSP, you know the distances between cities but not the shortest path. In protein folding, you know the shortest path (along the backbone) but not the distances between chiral carbons. Steric hinderence between residues ensures that the shortest path between the chiral carbons of sterically interacting residues is about 1.4 (as I recall) times the separation between adjacent chiral carbons. Hence, the shortest path must be along the backbone. I think from this model you can easily see that PFP is at least as hard as TSP. William R. Buckley 22:44, 29 December 2006 (UTC)

I mean not vitality, but a vital force, such as that offered in the 1800s, for the difference between life and non-life. I claim there is no vital force, life is nothing more than epicomplexity upon epicomplexity, and the substrate is immaterial.

Finally, you do not have the basis upon which to claim the position of the public vis-á-vis the acceptance of the strong versus the weak position, much less the position of established researchers (i.e. practicing scientists, regardless of field). You assert without proof. William R. Buckley 19:55, 28 December 2006 (UTC)

I know what you mean, I'm using it the same way (using vitality as the nominative form where you use vital force, both having an adjective form: vitality). Your claim is unfounded and unproven, so any presentation of it in an article on wikipedia needs to come from that direction. I would agree with your position, but there are many who would not.

It is not proven but I think well founded, if only by anecdote. Balanced presentation is agreed. William R. Buckley 22:44, 29 December 2006 (UTC)

If I may, I would relate anecdotal evidence. In my Humanities class not so long ago (a year ago perhaps) we were reading Frankenstein. The discussion moved towards wether creating living substance from non living matter was possible at all. To my considerable surprise, at least 5 or 6 kids out of my 12 to 18 person class (not sure how full it was that day ;)) argued that it was not. They didn't state in so many words, but they were claiming basically what you are claiming against: that there is an inherant vitality in living organisms that is unreproducable by man. I don't agree with that position, but clearly the set of all people who hold this position is non zero. And if my humanities class is a fair sample of the general public (probably it's not, since all students were between 20 and 25), the proportion of people who believe that synthetic life is impossible is non trivial. My discussions with and readings from biologists point to a general biological disinterest to disdain for alife. What biologists embrace ALife seem to do so from the weak alife side, meaning that it's a valid research platform for study of natural selection, and not for creating life. I do not have proof for this assertion, but could you provide proof of the opposite? Isn't this your general impression as well? In the end, this is probably non trivial enough to present both points of view, again in the synthetic life article. This issue has no bearing on artificial life research's validity (even the weak alife position says that we can understand life through these models), though, and so probably shouldn't be discussed in alife articles except from the point of view of weak vs. strong philosophy. --Numsgil 09:40, 29 December 2006 (UTC)

It is not my general impression that biologists hold disdain for alife research. The degree of acceptance of premises varies, some holding that such research cannot create life, and some holding that such research can create life. I hold the latter position. This does not mean that all alife models engage in the creation of life, and indeed I would argue that precious few current models create life. The point is one of possibility, not of reality. Again, constructing a balanced and thorough article is the goal, and we probably have covered enough of the problem to have an overview of what to do. So, each should take his turn and make changes. As we find need, more discussion ensues. William R. Buckley 22:44, 29 December 2006 (UTC)

So, the offer is extended. Write a few paragraphs describing the structure you find desirable, and I will give a reply. A few iterations, and inputs of others like Paul, and concensus will give the best result. William R. Buckley 22:44, 29 December 2006 (UTC)

Wet alife calls the subject of its study alife. Wet alife researchers are concerned to know if other molecules can be used as the store for genetic information, and if other kinds of molecules can be used for catalytic control, etc. So, perhaps a different long-chain sugar can be used to carry notations encoded in other than pyrimidines.

At the moment, the wet alife article is incredibly sparse. I personally lack anything more than a cursory knowledge of wet alife. Most likely because I come to Alife from a hobbyist's perspective, and it's hard to perform wet alife experiments in your house ;) While artificial life in the abstract studies life in all its forms, artificial life the present and real scientific field has studied primarily computational models of life with a minority of research involving "wet" alife, or that's my understanding anyway. Many things that could be called wet alife probably aren't (Biologists aren't fond of ALife in my experience). --Numsgil 11:41, 28 December 2006 (UTC)

Your understanding regarding the distribution of efforts, physical and computational, seems correct. Very little work is done in wet alife ("walife" ?). ALife X enjoyed a presentation by Norman Packard, who is doing wet alife. Maybe one other person had something similar to discuss. So, the (vast) majority of alife work is computational. The robotic component is ascending.

It occurs to me that should we find extraterrestrial life, and should it be based upon the same chemistry as ourselves, that would be a deep philosophic and scientific problem. Well, if it were just one case, maybe not so much of a problem. But, if we were to find multiple examples, and in each case determine the same basic sugar/lipid/amino acid/nucleic acid composition, then we would again need to visit the notions of predetermination of life within the known universe. While we might expect to find life elsewhere in the universe, we should perhaps expect it to be different from life on Earth.

I would expect life found on planets with similar composition and radiant heat to be remarkably similar to us in biochemistry. Carbon, Hydrogen, and Oxygen make up the majority of the (lighter) surface mass of the earth's crust. It makes sense for life to be made out of that. Silicon is the exception, but carbon makes a clearly superior building block for complex molecules over silicon, again in our environment (primarily because it's lighter I would say). I would expect a planet with more radiant heat, lower radiant heat, significantly stronger atmospheric pressure, or strongly different composition to have significantly different biochemistry. Obviously this is all theory, but I would wager a significant sum that life based on our present biochemistry is the optimal choice for existance in the narrow range of conditions that occur on Earth (or, using Gaia theory, occurred on Earth when Earth was young). Extremophiles represent outliers in this discussion, as they are hardly the norm. Life existing in the corona of the sun (should such a thing be possible) would probably be very different from what we're used to, perhaps based on matter in a plasmic state. --Numsgil 11:41, 28 December 2006 (UTC)

You might be getting yourself into trouble here. The notion that in similar circumstances life should be similar belies notions of vitalism. That the universe is designed for the emergence of life is a stretch that few scientists are prepared to make, though exceptions do exist. William R. Buckley 20:47, 28 December 2006 (UTC)

I'm not saying that the universe is predestined to make life in our image. I'm saying that our image is the (probable) optimal selection for the narrow range of conditions found on Earth. I would not begin to guess the commonality of those conditions. Rather, I would say that where (if) they arise, the corresponding life would have similar biochemistry to us. I would be surprised if macromolecules like DNA, RNA, etc. are common. I would not be suprised if alien life used proteins (meaning folded chains of amino acids) which shared many of the same amino acids we use. A good read, if you haven't already, is alternative biochemistry. --Numsgil 09:40, 29 December 2006 (UTC)

Predestined? What about predisposed, however benignly, however without intention. If what we find is left-handed, DNA, and RNA (even though modified slightly), and so forth, we would be required to reexamine same as a form of evidence for God, so profound would be the experience. I should not be surprised if the narrow range of conditions here on Earth would elsewhere yield similar results but, not to the exclusion of alternatives. Sensitivity to initial conditions, and the subsequent tendency to freeze contingency within the phenotype should be expected to deliver substantial differences between examples of life in the physical universe at all levels of expression, microscopic to macroscopic. William R. Buckley 23:11, 29 December 2006 (UTC)

I agree. As you climb up the possibility chain, pure randomness gets "frozen" in the phenotypes. I would be quite surprised if it turned out that any alien life at all was digestable, for instance. Haha, imagine the strange mirror world where all isomers are reversed. ;)

As for "digital organism" I am fine with that for the digital cases. Wet alife does not produce such an organism, so the digital part must at least be eliminated. "Synthetic life" is not so annoying. Create an intermediate layer, called "Artificial Life Examples" with links to the various examples and their categories: digital organism, alternative chemical analog, etc.

Again, this relates to my ignorance in wet alife. Perhaps you could add to the wet alife page some examples of accomplishments made in wet alife. --Numsgil 11:41, 28 December 2006 (UTC)

Look into the work of Norman Packard. He is a good starting point. Then search the web for similar topics. It is sure to yield a dozen projects, perhaps more. William R. Buckley 20:47, 28 December 2006 (UTC)

Bottom line - alife studies life as it might be (the various forms of alife) while biology studies life as we know it (and in particular, the specific case of life here on Earth). William R. Buckley 02:41, 28 December 2006 (UTC)

There are issues to be discussed on the topic of synthetically produced life that do not neatly overlap the topics to be discussed on the objects of study in alife. Issues of ethics, obtainability, political rights, and treatment in fiction. AI, genetic engineering, robotics, and probably a host of other fields seek to create something that would probably be called a synthetic life form. These would be poorly discussed on a page for the objects of study for the alife field. Lastly, supposing a page for "artificial life examples" were created, would it be sufficiently unique from what the page at digital organism simulators discusses? Perhaps digital organism simulators needs to be expanded from a list? Actually, the majority of the current alife article could probably be labelled as "artificial life examples". --Numsgil 09:32, 27 December 2006 (UTC)

Agreed, and more strongly, the majority of the current article should be moved to one labelled as "Artificial life examples." Only the most general information should appear in the root article, "Artificial life."

The issues of ethics, etc., do need to be discussed, and the root article is the best place. There tends to be a bit of personality-cult activity on wikipages, such as that demonstrated within the article "Cellular automata" by the praise given to Wolfram. All that kind of stuff needs to be buried within an article sub-tree, and not be placed in the root node of an article. William R. Buckley 02:41, 28 December 2006 (UTC)

I reverted most of your edits, but, on my computer at least, it's not showing in the history page. I don't know if this is a random issue on my computer or if wikipedia is having an aneurism. It still shows the right text on the main article, though. --Numsgil 01:55, 18 December 2006 (UTC)

I understand where you are coming from. But when artificial life actually exists (and you will acknowledge that a few already think it does and that many think that one day it will) what will you call it? "Artificial life" is a type of life, however far-fetched that concept may be. The argument I make above - life vs biology - is unanswered by you. Paul Beardsell 08:59, 18 December 2006 (UTC)

We will call it *life* and will recognise it as having an artificial, as opposed to natural, origin. Biological life, even when derived by artificial (read, human directed) means, will still be biological. Living systems borne of non-natural (artificial) mechanisms will still be alive. We will not change the usage of a field name, nor rename the field, so as to give a name to our product/progeny. Instead, we will recognise that the field has resulted in the production of that which the field studies, living systems, whether natural or artificial, and call it life. My colleagues in this field will likely support my position regarding nomenclature. William R. Buckley 19:31, 18 December 2006 (UTC)

The journal "Science" is not science - it is a journal about science. The journal "Artificial Life" is not artificial life - it is a journal about artificial life. The Earth is not a field of study - it is what geologists study. The body of knowledge about something is not the something. The somethingology is not the something. Artificial life is not a body of knowledge - it is the subject of a body of knowledge. Paul Beardsell 09:03, 18 December 2006 (UTC)

Feel free to create an article about artificial life forms. We can link to it. But that's not what this article is about. Do a google search for "artificial life". What do you come up with? How many of the top 10 links deal with anything but artificial life the field? 0, that's how many. I don't care if you like it or not; the term "artifical life" applies to a very specific scientific discipline (and artistic discipline). It's a compound word that does not refer to life that is artificial, but to a field of study that examines life and evolution in the abstract.

I didn't invent the term, I offer no apology for it. I don't know or care if it's a good label for the field, but that's what the field's label is. Neither you nor I are in any position to make wide sweeping changes to the nomenclature of existing scientific fields.

Again, if you want to write an article that discusses artificial life forms, that's fine. We can add a disambiguation link at the top. But I vehemetely refuse discussing both topics in the same article, because they are so very different and only loosely related to each other. --Numsgil 13:25, 18 December 2006 (UTC)

Declaring that artificial life and the study of artificial life are "only loosely related to one another", is profoundly wrong. "Vehemently" so. Paul Beardsell 23:01, 18 December 2006 (UTC)

Usually in ALife literature, the subject of study is in general called either a digital organism or, in the case of sociological studies within Alife, an agent (though the term agents applies equally well to digital organisms). Most simulators define this definition further. Darwinbots, for instance, calls its agents "bots". And wether these bots attain "alive" status, become sentient, or solve cancer, they'll still be called "bots" (or perhaps biobots, or sapienbots, since Darwinbots likes to add short adjective prefixes, but that's beside the point). This is done to neatly avoid the connotations of calling something "alive" within ALife. The weak vs. strong ALife position is still an open question, and the term "life" is loaded with all sorts of conceptual baggage. --Numsgil 04:32, 19 December 2006 (UTC)

I too will call artificial life, when it exists, "life". But how will we distinguish, on the occasions we need to (e.g. in legislation or in plain old casual conversation, perhaps with our artificial buddies), between naturally evolved life and synthesized life? You will call the latter "artificial life". Despite what you say above, all the journals / publications cited already do so. At best there is an inconsistency. The article starts out by saying "artificial life" is a field of study, and then it (or the references it cites) goes on to refer to the (potential) creations of this field of study as "artificial life". You cannot have it both ways. Or, if you must, you must reflect both uses in the topic's definition, not just the one. Paul Beardsell 21:19, 18 December 2006 (UTC)

Can you point to a specific reference this article links that does this? Does it do this in a way that can be construed as definitive for the entire field? --Numsgil 04:32, 19 December 2006 (UTC)

Paul, you begin by agreeing with me, then do an about-face on the issue. On first expression, we will call it life, period. If there becomes the need to distinguish between examples, then we will add qualifiers to our speech, but only then will we so do. Further, any such distinction drawn will reveal an underlying prejudice, a bigotry between life derived of non-directed means versus that derived of directed means: "I'm biological, and you're not!" William R. Buckley 22:25, 19 December 2006 (UTC)

William, I am not being inconsistent (as if that matters - I will not stick to my position in face of good contrary argument). We are addressing two different questions: I accept that artificial life will be regarded as truly being life by you and many others, and me, too! But not all people will agree. In law, for example, a distinction is likely to be made, at least initially. And within the field there will be occasions when it will be useful to distinguish between the e-coli that landed on the petri dish next to the manufactured wet alife organism. One will be the foreign natural bacteria and the other will be the locally manufactured example of artificial life. The question is not the philosophical one you insist on answering i.e. "Will artificial life truly be life?" but rather "What do we call artificial life to distinguish it from naturally occurring life when we need to make the distinction?" Paul Beardsell 22:47, 19 December 2006 (UTC)

Paul, you use the word *occasion* which certainly implies other than always, and generally implies less than usual. And, consistency does matter, as any physicist will tell you. It shall be only a few who will not follow the norm then applicable, to call all seemingly alive things as being members of the category life. I urge that you are assuming an a priori requirement that distinction be made. I see no such causal implication for human social action. However, it is clear that for the special cases which concern you, where it is absolutely necessary to draw distinction on the basis of primary origin, then we shall call the two life and alife, and these are short for biological life and artificial life. Finally, it may even come to pass that more subclassification be imposed, so that alife exists in several occassionally specially identified forms.

Your question is thus answered - when not important, all is life, and when otherwise important, there is life and there is alife. William R. Buckley 22:59, 20 December 2006 (UTC)

Yes! Thank you. "Artificial life" or "alife" is, according to the article, "the study of artificial life [forms]"...Paul Beardsell 03:09, 21 December 2006 (UTC)

I would disagree. I don't think the article ever claims that the field of artificial life" studies artificial life forms. --Numsgil 07:36, 21 December 2006 (UTC)

...But you have used the terms "alife" and "artificial life" above to refer to the instances of life themselves, not to "the study of" or "the modelling of" artificial life forms. This, I imagine, to Numsgil's considerable dismay! Now, this is where I started this discussion and why I started it: The term "artificial life" does not mean the academic or practical discipline of theorising about or manufacturing devices which are alive. Or, if it does mean that, it does not mean that exclusively, as you have ably demonstrated. So, can I change the definition at the start of the article back now to encompass both meanings? Paul Beardsell 03:09, 21 December 2006 (UTC)

I have changed the opening paragraph, Numsgil and I having thereafter worked to improve accessibility. See what changes you find desirable.

Coming next is a need to revise the second section, as it makes a great many unfounded claims. Also, we need to archive this topic. I think it is now resolved. William R. Buckley 07:57, 25 December 2006 (UTC)

No, not exclusively. Life is a class, made up of living systems. Alife is a class, made up of living systems having non-natural origin. Blife is a class, made up of living systems of natural origin, which we also call *life* in honor of blife having been the first kind of life that we came to know. Artificial life is a synonym for alife. Artificial life is also the name of the field that is dedicated to the understanding of living systems whatever their form. There are a great many words and names that serve multiple purposes.

The primary point which follows is not likely to reflect future term usage. Artificial life is equally well applied to the field, and the subject therein studied. This is as with the word life, which is used to refer to the sum total of one's experience (one's life), to refer to the category of living things (life), a term of temporal change (life time), and other variations of meaning. I do not see a need for different terms within Wikipedia just to satisfy some wild and idiosynchratic semantic hair. It is of no value to Wikipedia. William R. Buckley 07:28, 22 December 2006 (UTC)

My primary point has been and will continue to be: A) The term "artificial life" is best applied towards the field, not the organism. B) The two are sufficiently distinct to warrant seperate articles. There are usages of the word to apply towards synthetically created life, but they are not in any way cannon or common compared with its use for the field. We could rename synthetic life to artificial life (class of life) or something equally wordy, but that just seems to be overly verbose. Synthetic life seems the simplest, distinctess way of distinguishing the two ideas from each other. Synthetic and Artificial don't even have any differences in connotation, so it's not like by calling it synthetic life instead of artificial life I'm degrading it. With a disambiguation link at the top of the alife article, this should really be all you need. --Numsgil 05:52, 21 December 2006 (UTC)

From the journal Artificial Life: "By extending the horizons of empirical research in biology beyond the territory currently circumscribed by life-as-we-know-it, the study of artificial life gives us access to the domain of life-as-it-could-be." Let's parse that. "the study of artificial life" cannot mean "the study of the discipline of manufacturing life". That would be a meta-study, a sub-section of the philosophy of science. Paul Beardsell 21:41, 18 December 2006 (UTC)

The "study of" prefix is redundant, as in "the study of biology" or "the field of physics". If anything, the quote proves my point since it calls the object of study "life-as-it-could-be" instead of "artificial life". --Numsgil 04:32, 19 December 2006 (UTC)

Let's examine some dictionaries:

dictionary.com: artificial life –noun: the simulation of any aspect of life, as through computers, robotics, or biochemistry.
American Heritage Dictionary - artificial life n. The simulation of biological phenomena through the use of computer models, robotics, or biochemistry. Also called Alife.

Note that artificial life is defined as being the simulations themselves, not the study of the simulations, not the creation of them. Paul Beardsell 21:41, 18 December 2006 (UTC)

Simulating somethings is the same as studying something. The simulation is the final product of a theory going from conception to proof. Note that these articles do not say "artificial life n. Life created through artificial means". Again, these would seem to prove my point instead of yours. --Numsgil 04:32, 19 December 2006 (UTC)

"Simulating something is the same as studying something." Not it is not! Citation required. Paul Beardsell 06:10, 19 December 2006 (UTC)

"The simulation is the final product of a theory going from conception to proof." No it is not! Citation required. Paul Beardsell 06:10, 19 December 2006 (UTC)

"The simulation of" is not the "act of creating a simulation" - it is the simulation itself. Paul Beardsell 06:10, 19 December 2006 (UTC)

This is getting tiring. I won't argue semantics like this. Scientific modelling is the cornerstone of alife research, and I won't spend the time to argue wether or not modelling something is the same as studying it. It's a word game, and ultimately it doesn't matter. I'm sorry you disagree with the current label alife has. You don't get to change it. It doesn't matter if you're right or not, you don't get to change it. It's not for you to change. It's not for me to change. ALife existed (though perhaps not named as such) before I did, and probably before you did. Feel free to write another article, and we can add a disambiguation link. Perhaps Ersatz life or Artificially created life or Synthetic life (my preferance). The compound word "artificial life" is taken, and applies to a field of study that studies evolution and life through the use of human artifacts. This is true wether you want it to be or not. --Numsgil 08:35, 19 December 2006 (UTC)

Semantics is important in an encyclopedia. We cannot afford to be sloppy like this. The issue is this: What will readers of WP expect when they see an article entitled "Artificial Life". An article that doesn't discuss anything which is or could be artificially alive? No. Paul Beardsell 09:29, 19 December 2006 (UTC)

That depends on why you're looking at the article on Artificial life. If you're coming here from Darwinbots or Avida, a discussion of Frankenstein isn't going to be appreciated. Again, my creation and disambiguation link to synthetic life should meet all the issues you're after. --Numsgil 09:37, 19 December 2006 (UTC)

My remaining issue is that the definitions seem to be yours alone. We aren't supposed to make it up as we go along here at WP. Paul Beardsell 21:14, 19 December 2006 (UTC)

I have had exactly this same issue at artificial intelligence. There, too, the topic was defined as being "the study of ...". An example I used there was, what is meant when one says, as is often said, that "artificial intelligence is impossible"? What is held to be impossible is not "the study" but "artificial / synthesized / manufactured intelligence" itself. Similarly, here, what is meant when one says, as is often said, that "artificial life is impossible"? (Incidentally, my argument was accepted at artificial intelligence.) Paul Beardsell 21:14, 19 December 2006 (UTC)

What "artificial life" is

Let's have another go and a fresh start. I quote the first sentence of Mathematics. "Mathematics ... is the body of knowledge centered on concepts such as quantity, structure, space, and change, and also the academic discipline which studies them." I suggest, in similar vein to the Mathematics article, and weakening my position expressed above in an effort to find a compromise, that this article should be both about artifical life and the study of artificial life. Paul Beardsell 06:20, 19 December 2006 (UTC)

While I appreciate efforts to compromise, I will not compromise on this, and will fight any efforts to encorporate the two in the same article to my dying breath. They are not related except by the most cursory overlap in the labels we might apply to them. What you call "artificial life" would include things like androids, cyborgs, genetically modified bacteria, and concious AI. It would involve discussing their use in fiction, possible political issues, etc. This is not what this article is about. I would not try to discuss Adolf Lu Hitler Marak in the article Hitler (though the reverse might not be true), even though they have the same name. Again, feel free to right another article, we'll add a disambiguation link. But this isn't the article to discuss these issues. --Numsgil 08:35, 19 December 2006 (UTC)

I think any hope you have of persuading others of the correctness of your view is ruined by overstatement. By Godwin's law you lose the argument but that law need not be invoked: Your two Hitlers really are unrelated but to claim that X and The Study Of X are unrelated is gobsmackingly outrageous! Paul Beardsell 09:04, 19 December 2006 (UTC)

Hitler was probably a poor example, it was just the first article that sprang into my mind that would probably have a disambiguation page. You can exchange links to Hitler to links to anything with a disambiguation page if you like.

At present, artificial life is not the study of synthetic life. Artificial life is the study of life in the abstract, and its evolution through natural selection or other means. Supposing for a moment that artificial life did study synthetic life, it would be concerned with synthetic life's evolution through time. This is why the two don't belong on the same page. Artificial life concerns itself with agents and digital organisms or even synthetic life only in so far as they are necessary to study evolution. --Numsgil 09:13, 19 December 2006 (UTC)

I suppose then that one can DO 'artificial life' in the same sence as one can do mathematics, but one cannot create 'artificial life' just as one cannot create mathematics? It might be so, but it just seems a bit counter intuitive. --moxon 10:45, 3 January 2007 (UTC)

Whether mathematics is created or discovered has not yet been resolved to anything approaching a consensus. But we all know what is meant by "mathematics" in the preceding sentence. Similarly we know what is meant by the phrase "creating artificial life" - we are not referring to the creation of "the study of artificial life", but of the little beasties themselves. Paul Beardsell 06:08, 6 January 2007 (UTC)

Link to artificial consciousness

The article now contains, in a comment, an argument as to why there should be no link to artificial consciousness. The opinion that artificial consciousness is not readily attainable or achievable and therefore is not worthwhile linking to (I paraphrase the argument) is an opinion held by some but by no means all of those prominent in the field. Artificial consciousness is considered practically inevitable by those who hold the strong AI hypothesis. They hold the view that artificial consciousness will be real consciousness just as artificial intelligence will be real intelligence and, as expressed by others above (including he who does not want a link to AC), just as artificial life will be real life. Paul Beardsell 06:48, 19 December 2006 (UTC)

The comment is mine, artificial conciousness is in the realm of AI, not ALife. The old article linked to it and its brethren, and that was the reason it was so terrible. AC comes from the top down, while ALife comes from the bottom up. AC has absolutely nothing to do with modern ALife research. Things such as the Cambrian explosion, the advent of sex, altruism, and the baldwin effect are the current focus of Alife research. AC has nothing to do with ALife, except as perhaps an extremely long distant goal for some participants in ALife. I feel that the current link to AI fulfills any needs for links to AC. --Numsgil 08:35, 19 December 2006 (UTC)

The WP NPOV policy demands that all mainstream opinions are reflected. Numsgil's insistence on one POV to the exclusion of others obviously denies other mainstream POVs exposure on WP. Paul Beardsell 09:53, 19 December 2006 (UTC)

What specific POV within the ALife community am I excluding? AC isn't an POV within the ALife community (though some overlap of participants might exist). --Numsgil 09:58, 19 December 2006 (UTC)

I removed Numsgil's comment. I'm reposting it here: User:150.176.202.5

No link to Artificial Consciousness on purpose. At the present, Artificial Consciousness is a pipe dream, a metaphysical field, and is only extremely loosely related to the current methodolgies and realities of ALife. That is, Alife first needs to achieve and is working towards proper "life" of comparable complexity to a bacteria. It is not yet ready to or working towards something that is sapient. That is the realm of AI.

Any particular reason why the removal of my comment requires reinserting a link to artificial consciousness? Since you're the original contributor of the link, what made you feel it (and the comment you placed with it) should belong on the page? --Numsgil 11:46, 19 December 2006 (UTC)

Again, the only way the two are related is through their relation with Artificial Intelligence. I just don't feel that artificial consciousness has anything to do with artificial life (though it has alot to do with synthetic life). --Numsgil 11:46, 19 December 2006 (UTC)

This seems profoundly wrong to me. Artificial consciousness is surely a subset of artificial life? Paul Beardsell 21:02, 19 December 2006 (UTC)

My argument is that artificial conciousness is not artificial life. The assertion that being alive does not connote conciousness is my axiom. William R. Buckley 18:08, 19 January 2007 (UTC)

I am without doubt wrong here. I agree: Alive things are not necessarily conscious and v.v. Paul Beardsell 20:51, 19 January 2007 (UTC)

Artificial life meaning synthetic life, yes. Artificial life meaning a field of study, no. There have been no published studies that I'm aware in any alife journals that even broach the subject. --Numsgil 05:20, 20 December 2006 (UTC)

Synthetic life

--Numsgil 22:54, 11 January 2007 (UTC) I've set up an article synthetic life which I'm currently in the process of adding a disambiguation link to. Issues relating to artificially created life forms should go in that article. That should prevent the mixing of purpose that discussing both issues in the same article causes. --Numsgil 08:59, 19 December 2006 (UTC)

A reference in the literature showing that in the field itself the term artificial life refers only to the study of artificial life [forms] and that the term used for artificial life [forms] is invariably synthetic life would usefully support your view and also shut me up. Otherwise we are inventing terminology and usage and this would constitute a [mild form of] original research which, as I am sure you know, is forbidden at WP. Paul Beardsell 21:19, 19 December 2006 (UTC)

Come on Paul, inventing a name for an article is not original research. Save the hyperbole. Any faux pas in using a name contrary to that in research is easily adjusted in later editions of Wikipedia. Until such research usage is established, there is little harm from Wikipedia using a term internally. Coining of article names is hardly a profound issue.

Come on William, I wrote "a [mild form of] original research. But the coining of article names DOES MATTER. A lot. Giving something a new name, here at WP, is not allowed. For good reason. Paul Beardsell 23:13, 11 January 2007 (UTC)

As I pointed out in an edit a few minutes ago, I do not agree with the context switch inherent in the name change, from artificial to synthetic. So, Numsgil should consider carefully a renaming of his article, to say, 'Artificial life examples.' Use of the term 'digital organism' is discouraged, as then the article would seem an awkward place to explore examples of wet alife. William R. Buckley 07:34, 25 December 2006 (UTC)

Before I begin, I admit only to inventing the term "synthetic life" (although I doubt I'm the first to use it). I think it's the best choice, and is an equally valid way to describe artificially created life as artificial life is. If there's a better location to write that article, then we can move to it, but it seems the most appropriate to me.

With that out of the way, here is a (hopefully) comprehensive list (not in any real order) of all reasons why synthetic life and artificial life should be discussed in seperate articles, with the present alife article keeping the namesake "artificial life":

1. Artificial life (the field) existed in the present article's location, so it has seniority.

2. Artificial life has a category named after it. The contents of this category reflect primarily alife the field.

3. A google search for "artificial life" returns links to artificial life the field exclusively for the first several pages at least (I never managed to find a page that wasn't about the field). Thus the term when applied to a field instead of synthetic life is far, far, more common.

4. Assuming for a moment that we could say that artificial life the field studies artificial life (synthetic life), we would also have to assume then a strong ALife position (that is, that the object of study is alive). Using this definition makes a weak alife position make little sense, and would seem to add bias to the field on this core philosophical issue. Hence why the usual term is digital organism.

5. Given 4 above as a partial support, I would say (and other alife participants should chime in) that artificial life is the study of life and evolution in the abstract. To back up this claim, I cite [2] and [3], the latter is written by Christopher Langston who coined the term to describe the field in the first place, so it's as definitive a source as any I can think of or find. Specifically notice that this definition neatly side steps the issue of wether the simulated agents are "alive" or not. In reality, artificial life is not the study of synthetic life, but the study of life processes and properties abstracted away from any single medium. Thus trying to say that artificial life studies synthetic life misses the point entirely.

6. The field is called artificial life by all those organizations and participants that form(ed) the core of the field. It should be fairly easy to agree that the field's proper name is artificial life.

7. Assuming for a moment that alife the field does study artificial (synthetic) life, we would still be well behooved to discuss the two in seperate articles anyway, so we can seperate data about the history of the field, its current direction, etc. in one article and the sorts of "artificial life" forms that exist, their potential political rights, etc. in another article.

--Numsgil 05:12, 20 December 2006 (UTC)