Jump to content

Talk:Technological singularity/Archive 1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1Archive 2Archive 3Archive 5

Old talk

Arvindn, the difference in speed is more significant than "hertz" precisely because it affects any form of computation. Nonclocked computers already exist, and in these, Hertz is meaningless. Even so, human being clock at about 10 Hertz (that's why movies run at 24 frames per second). The 200Hz cycles are because neurons use frequency modulation to encode analog values.

So, I'm going to restore the "speed" text. I appreciate the grammar corrections, by the way. User:Ray Van De Walker

Fine. BTW, you should use [[User:Foobar|Foobar]] rather than [[Foobar]] for user pages. -- Arvindn

I wonder what that meant: "technological progress rising to "infinity."" I have never seen this particular misinterpretation. Since progress is the process, it can't rise to infinity - only its speed can, or the level of technological development. Could the person who wrote it rewrite this part into what he/she wanted to say. Paranoid 21:01, 9 Oct 2003 (UTC)


Loose note: Sid Meier's "Alpha Centauri" game featured, as one of its three endings, "Ascent to Transcendence", involving the humans becoming one with the native intelligence of the colonized planet. The game also relied on beyond-state-of-the-art tech trees with such things as "Self-Aware Machines" and "Secrets of the Brain" -- essentially, one of the ways of winning was establishing a technological singularity.

MAZKORE(RATCAN303@YAHOO) I feel that further investigation should be warranted on the emerging existence of neo nanotechnology and neo micro-cellular philosophy. I KNOW that most people feel that this paranoid talk has n relevance to our world today, but I feel that the world we have been shown in film and reel, the world of the future, is already in existence or as a whole we are on the verge of this new existence. The truth of the matter is hidden in the infinity of fractals and the madelbrot set. I feel that we are going towards a world where war is no longer fought on the oceans and deserts of what we call the ‘real’ but on the planes we forgot existed. The microcellular doctors and investigators of nano technology fight the true war of this era in a world all too forgotten by the masses of this era. If you want to discuss this topic further, contact me because I am scratching the skin of something I do not fully understand. Thank you MAZKORE



Hope im not sounding pedantic here but Moore's Law actually discusses the number of transistors doubling every 18 months, not the speed of the processor. Thus the speed of the processor will not neccessarily double. There soon reaches a physical limit to the size of thetransistor (the minimum amount of molecules it can be made from) -> thus eventually to double the number of transistors one would have to increase the size of the processor which helps to slow it down (as electricity has further to travel).



OK, I thought I'd check up about this Murray guy and google does indeed throw up evidence of spamming. Even he's a spammer and a "prominent theorist", I'd still say keep him out. -- Arvindn 17:51, 27 Jan 2004 (UTC)


Hello, I added this paragraph but I think it is far too negative. Also I find the article a little scattered, I am thinking of the best idea to divide it into sections (i think a singularity in culture (including novels, fiction and games) section should be included, as well as talking about AI and superhuman intelligence section (one possible form the singularity will likely take).

My paragraph:

The singularity is often seen as the end of human civilization and the birth of a new one. As Vinor Vidge, a futurist who has written on the Singularity, predicted: "Within thirty years, we will have the technological means to create superhuman intelligence. Shortly after, the human era will be ended.". Why should the human era end? Humans will likely be transformed in the process of the singularity to a higher form of intelligent existance. In any event, after the creation of a superhuman intelligence, people will necessarily be a lower lifeform in comparison to it.

ShaunMacPherson

von Foerster addition

There is much confusion over the concept of "singularity" vs "exponential" in the history of "the technological singularity". I've attempted to clear this up with appropriate cites in adding the real origin of "the singularity" as it pertains to models of humanity's future. Heinz von Foerster et al were the originators of this "absurd" mathematical model -- which was nevertheless published in the prestigious scientific journal "Science" in 1960.

People, please sign.
Agreeded, it is an interesting addition. But you state that: After 1973, however, the model ceased being checked against hard statistics and instead entered the realm of popular culture, via such books as Alvin Toffler's Future Shock (1970). What is the source of that? I Googled, couldn't find any but found several scientific articles still assuming this is a valid equasion, for example:

--Piotr Konieczny aka Prokonsul Piotrus 19:36, 27 Aug 2004 (UTC)

The work of von Foerster certainly is interesting, but I don't think it's relevant to the Singularity as the term is used today. A technological singularity is just that: technological. von Foerster's work seems to concern only human population growth, not technological progress. von Foerster gets credit for being the first person to use the word "singularity" in a sociological context, but his definition is, as far as I can tell, otherwise unrelated to the topic of the article. The initial conception of the Singularity should be credited to Vernor Vinge. -- EHS

Bill Joy and other criticisms

Bill Joy should be mentioned somewhere im not sure wether as a critic or supporter, also Bruce Sterling did a very good criticism of the subject in a talk for the Long Now Foundation which is avalible here:

http://seminars.moose.cc/

and there is some coverage of the topic by Stewart Brand:

http://www.longnow.org/about/articles/ArtSingularity.html Htaccess 08:52, 26 Sep 2004 (UTC)

I inserted mention of Bill Joy's article in the criticisms section, but I haven't yet listened to Bruce Sterling's comments. As for Stewart Brand's coverage, I don't see anything novel or interesting here. It seems to be a pretty basic overview with lots of wishy-washy metaphors about progress and society, and I don't think it warrants specific mention. -- EHS

Concepts and terms

I'm removing the "Concepts and terms" section from the article, as the ideas it contains are rather unordered, and in some cases slightly POV due to wild unattributed speculation. Having a section for word definitions is not a bad idea, but as a Singularitarian I have to admit that, while I recognize most of these words, I do not commonly see them used in Singularity discussions. Most of the words listed are not even used anywhere else in the article. Defining singularity neologisms as they're used should be sufficient. -- EHS

Original research

I'm about to suggest that this article, interesting though it may be, really doesn't belong in its present form in Wikipedia. I think it should be moved to Wikisource or somewhere else more appropriate for original research, opinions and the like. If it were simply an encyclopedic account of a set of speculations, that would be different, but actually it seems to be nothing but a series of speculations presented in an uncritical manner. The references to "Moore's Law" are particularly irritating to anyone who understands the origins and application of the phrase. --Minority Report 00:50, 23 Nov 2004 (UTC)

I agree. Some of the speculations need to be trimmed down and attributed. I'm removing the second section on Moore's law, as it seems especially speculative (I've no idea where those numbers are from). Moore's law itself is too significant to cut out entirely, but I'll try to ground it more in reality. I'll look for some sources for the bits in the "Singularity technologies" heading. As for the article being "uncritical", I'm just not seeing it. There's decent-sized section of attributed criticisms. — Schaefer 03:16, 23 Nov 2004 (UTC)
Your main qualm about this article seems to be the misrepresentation of Moore's law. Yes, I realize that Moore's law really only concerns transistor densities, but it's very often generalized outward beyond that. I believe Ray Kurzweil actually gave some fancy name to the more general form to distinguish it from the old Moore's law, but most transhumanist keep using the latter because it's more well-known and thus more convincing. I'll try to rewrite the infringing sections to reflect Kurzweil's law instead of Moore's. — Schaefer 03:26, 23 Nov 2004 (UTC)

Well the language of the piece isn't really encyclopedic. The Moore's Law section was a particularly egregious example of a very contentious assumption--that growth in technology is reliably predictable and headed towards some predicted state. This is really just a reiteration of a claim first seen in Von Neumann over fifty years ago, repeated more recently by Vinge. An encyclopedia entry should certainly cover these ideas if only because Von Neumann and Vinge aren't just amateurs, even their speculation often merits serious consideration. However the assumptions must be handled with care, and clearly identified. --Minority Report 10:43, 23 Nov 2004 (UTC)

I definetly think this is an important subject that deserves much expantion. Note: since I will be preparing a presentation for the end of April on this subject, I will likely work on the article over the coming days. --Piotr Konieczny aka Prokonsul Piotrus Talk 13:40, 18 Apr 2005 (UTC)

New Scientist article

Just so you know, the lead-in quote to this week's cover article in New Scientist was Wikiipedia's definition of the Singularity. [1]. — Asbestos | Talk 16:40, 24 Apr 2005 (UTC)

The definition they use (without specifying article version date) is: 'In futurology, a technological singularity is a predicted point in the development of a civilisation at which technological progress accelerates beyond the ability of present-day humans to fully comprehend or predict. The singularity can more specifically refer to the advent of smarter-than-human intelligence, and the cascading technological progress assumed to follow.'. --Piotr Konieczny aka Prokonsul Piotrus Talk 13:03, 26 Apr 2005 (UTC)

Definition

I have rewritten the lead, so it summarises the current article. I am however worried that the old definition we use as our first sentence (see above) is not correct. What especially worries me is that our old definition defines the post-Sigularity times as those when the technological progress is beyond understanding of present-day (not a very specific term, that) humans, but it sais nothing about the rather obvious fact that the very world and civilisation will also not be understandable. I.e. it gives a rather false and limited impression that singularity will only affect the technological progress - but won't change anything else. As Wiki is no place for orginal research, I would like your help here: let's collect existing definitions of singularity and see if we can improve our Wiki definition based on that. --Piotr Konieczny aka Prokonsul Piotrus Talk 13:03, 26 Apr 2005 (UTC)

Existing definitions

I can't find a clear defintion in the classical Vinge article. Patched together pieces of the article gives this partial definition: 'change comparable to the rise of human life on Earth. The precise cause of this change is the imminent creation by technology of entities with greater than human intelligence. (...) It is a point where our old models must be discarded and a new reality rules. As we move closer to this point, it will loom vaster and vaster over human affairs till the notion becomes a commonplace. Yet when it finally happens it may still be a great surprise and a greater unknown. (...)
Ulam and von Neumann quoted by Vinge above and by our Wiki article: ' the ever accelerating progress of technology and changes in the mode of human life, which gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue'
From Kurzweil 2001 article: Singularity is 'technological change so rapid and profound it represents a rupture in the fabric of human history'
From Kurzeweil AI net definition: 'Defined by Vernor Vinge as the postulated point or short period in our future when our self-guided evolutionary development accelerates enormously (powered by nanotechnology, neuroscience, AI, and perhaps uploading) so that nothing beyond that time can reliably be conceived. The Singularity is a common matter of discussion in transhumanist circles. There is no concise definition, but usually the Singularity is meant as a future time when societal, scientific and economic change is so fast we cannot even imagine what will happen from our present perspective, and when humanity will become posthumanity. Another definition is the singular time when technological development will be at its fastest.'
From Singluarity Institute definition: 'The Singularity is the technological creation of smarter-than-human intelligence. ' or from their homepage 'What is the Singularity? Sometime in the next few years or decades, humanity will become capable of surpassing the upper limit on intelligence that has held since the rise of the human species. We will become capable of technologically creating smarter-than-human intelligence, perhaps through enhancement of the human brain, direct links between computers and the brain, or Artificial Intelligence. This event is called the "Singularity" by analogy with the singularity at the center of a black hole - just as our current model of physics breaks down when it attempts to describe the center of a black hole, our model of the future breaks down once the future contains smarter-than-human minds. Since technology is the product of cognition, the Singularity is an effect that snowballs once it occurs - the first smart minds can create smarter minds, and smarter minds can produce still smarter minds.'

Proposed new wiki definition

  • Old Wiki defintion: 'In futurology, a technological singularity is a predicted point in the development of a civilisation at which technological progress accelerates beyond the ability of present-day humans to fully comprehend or predict. The singularity can more specifically refer to the advent of smarter-than-human intelligence, and the cascading technological progress assumed to follow.'
    • Votes to keep:
  • Piotr Konieczny aka Prokonsul Piotrus Talk proposal: 'A technological singularity refers to a predicted point or period in the the development of a civilisation at which due to the acceleration of technological progress, the societal, scientific and economic change is so rapid that nothing beyond that time can be reliably comprehended, understood or predicted by the pre-Singularity humans.

Was Skynet a Seed AI?

Perhaps I am being obtuse, but I don't see the connection between Terminator's one self-aware AI and the technological singularity as cited in the Singularity and Culture section. Isn't is more all of the AI's becoming self-aware, and human computational ability so vastly enhanced that it is impossible to tell who is organic and who is AI? I feel Skynet would fail a Turing Test, and so not qualify for Seed AI. I suggest the reference be removed to avoid confusion (like what I just experienced). - Thanks, Rlacy853 6 July 2005 23:48 (UTC)

Rose: Be bold. I agree it is not perfect - the Terminator's world does indeed fail the 'rapid changes' definition - so try to fix it. On the other hand, we know little about Skynet to be sure it would fail the TT, but as it can design apparently sentient cyborgs I would err on the side of caution. Same argument could be used for example on 'Sentient's from Odyssey 5. Now that I came to think of it, the world's destruction (episode 1 so no spoiler :D) could be viewed as a singularity, don't you think so? --Piotr Konieczny aka Prokonsul Piotrus Talk 7 July 2005 08:07 (UTC)
Piotr: Thank you tremendously for the reference to Be Bold. The rest of the Culture section is so strong I feel it doesn't need any reference to Terminator. Also, good point that Skynet designed sentient cyborgs, but I have to disagree about whether they are sentient. It voluntarily self-destructed based on a chain of logic that served its master. I wish I had seen the Odyssey 5 series. I don't have cable, but I will order the series from Netflix.
Cut section Although their treatment of the subject is superficial at best, Hollywood's "Terminator" films all have a technological singularity as a major element to their overall plot in that a defense supercomputer dubbed "Skynet" attains sentience and, as its first "conscious" decision, begins the global extermination of humanity.
Rose Rlacy853 8 July 2005 00:48 (UTC)
I'm putting reference to Skynet under "Sentient AI and baseline humans" Rose Lacy 00:00, 2 August 2005 (UTC)
Well, last time I checked, the Turing Test was not necessary for a seed AI :-) but rather the fact that seed AI produces more seed AI beyond its own class. --Ihope127 18:09, 17 August 2005 (UTC)

Well I'm not going to cut Terminator again. The new reference is phrased to relate to Colussus so I'm not sure I could put Piotr's section back. This has gotten me wondering though.

I was introduced to Singularity by reading stories by Marc Stiegler and Charles Stross in Analog Magazine. If someone comes across this page and connects Singularity to doomsday visions of Sentient AI would they even read the rest of the article?

Part of me feels the culture section should be limited to Seed AI and transcendent cultures. Which pretty much eliminates anything out of Hollywood, but would probably give a better idea of Singularity concepts. Rose Lacy 17:04, 11 August 2005 (UTC)

Well, we are supposed to be neutral and comprehensive. The doomsday singularity should be mentioned, but of course only as one of the possibilities. As for Skynet, and other post-apocalyptic fictional settings, I wonder - does simply adding AI to the plot makes a book a singularity one? I think not. --Piotr Konieczny aka Prokonsul Piotrus Talk 18:20, 14 August 2005 (UTC)

I added the term local Singularity to the Sentient AI/baseline human listing after reading the discussion on the chessbots. Rose Lacy 18:32, 21 August 2005 (UTC)

Singularity Reached in Chess?

I'm wondering if it could be correctly held that a 'singularity' was reached in chess, when the machine Deep Blue defeated the then World Champion Gary Kasparov in a match.

Dear anon, I very much doubt it. That a computer has bested a human in one area is not a singularity. For many years, computers have done things better then humans - fast math calculations, for example. This is just another such area. See above definitions - singularity will be reached when progress will accelarate to such speeds it will change the world beyond our recongnition. One cannot apply singularity to parts of reality, such as chess. Also, even if we made such a metaphore to chess only, I don't think that chess have been changed beyond recognition. --Piotr Konieczny aka Prokonsul Piotrus Talk 21:14, 5 Jun 2005 (UTC)
If Deep Blue could create a new, superior Chess-playing AI, and thereby improve upon itself without human interaction, then this would create a (very local) singularity. Of course, if a human could create a Chess AI that could do that, they could probably also create a general AI with similar capabilities, and that would cause a real singularity. Until a Chess AI can do that, like Piotrus said, this is just another case of a computer simply being better at one task. --CoderGnome 15:28, 8 Jun 2005 (UTC)
Heh, aye: a chessbot that created chessbots would be a local singularity. However a chessbot actually cannot create chessbots; the only thing it can do is play chess: it's a chessbot after all. The exception would be if it created chessbots by playing chess. (Now there's a concept.) A chessbotbot:chessbot could create chessbots, however, so chessbotbot:chessbots would be a nice thing to go for. (Chessbot; not cheesebot, heh.) User:Ihope127
It should also be pointed out that Deep Blue's programmers were, um, not entirely ethical, imho. They designed Deep Blue to defeat Gary Kasparov, and not just any chess champion. In fact, this has been true of high-profile computer chess matches in general. The chess singularity isn't going to come until a chess-playing black box computer can beat any and all humans, and that has not happened yet. Functce,  ) 00:16, 18 August 2005 (UTC)

A *MAJOR* Omission

This article needs someone (who knows more about the subject than I) to include some discussion of Alvin Toffler's famous book Future Shock. →Raul654 08:21, Jun 13, 2005 (UTC)

Good point. I will do so soon. --Piotr Konieczny aka Prokonsul Piotrus Talk 09:34, 13 Jun 2005 (UTC)

Author/Title of short story with nano and Singularity?

About 20 years ago I read the greatest short story about humanity acheiving Singularity through nanotechnology. The touching part was humanity's need for an emotional anchor as they explored the universe with their minds. The heroine of the story was a simple woman who had an elemental connection with her own humanity, and through her mind being a part of the collective, provided the emotional safe harbor and anchor needed for the whole race to achieve Singularity.
The story was in Analog. Naturally I went to their site to search their archive. No luck! I couldn't find a searchable archive. Any ideas? I'd love to add this story to the Singularity and Culture section.
Thanks, Rose_Lacy 21:01, 11 July 2005 (UTC)

This is a type of question that you may want to ask at Talk:Science fiction or perhaps the relevant sf usenent group. As a side note, one of the best stories about singularity I read have been written by Jacek Dukaj - unfortunately majority of his works is untranslated into English :( The entire fiction section could use much expantion. We should mention the works of Iain M. Banks, Greg Egan, Greg Bear - just out of the top of my head. --Piotr Konieczny aka Prokonsul Piotrus Talk 23:09, 11 July 2005 (UTC)
Thank you again, my Wiki mentor! Rose_Lacy 22:03, 12 July 2005 (UTC)
A reader on the Analog discussion forum thinks the story is "The Gentle Seduction" by Marc Stiegler. I've ordered a copy of his anthology containing the story to confirm. Rose Lacy 18:16, 15 July 2005 (UTC)
Marc Stiegler? Never heard of him. Do let me know if he is good. --Piotr Konieczny aka Prokonsul Piotrus Talk 18:28, 15 July 2005 (UTC)
I finally got a copy of the collection The Gentle Seduction. It's just as touching as I remember. The description on Marc Stiegler's page has been expanded. Rose Lacy 04:00, 2 August 2005 (UTC)

Going overboard with the images

I think there's more graphs on this page than are needed to visually convey the point that technology progresses exponentially, and would like to remove some of them in the name of aesthetics. Anyone else have strong opinions on this? -- Schaefer 15:26, 14 July 2005 (UTC)

Well, as it was me who uploaded and added them here, I will elaborate on this. I certainly think most of them should stay in the article - visual images do contribute greatly to the text. Perhaps some may be *moved* elsewhere, but I would like to make sure that they are linked from some article - they are pretty enough for it to be a shame if they were wiki-orphaned. I have already linked some of them to other articles. If you have any thoughs which specific graphs need to go, do tell. Constructive critisms is always welcome. --Piotr Konieczny aka Prokonsul Piotrus Talk 16:36, 14 July 2005 (UTC)
Instead of removing them, I'm going to shrink some of them down to be true thumbnails. As it stands, they're difficult to read without zooming in anyway, so we might as well shrink them to the point that it's clear to the user that they weren't meant to be read without being zoomed in on. Also, I'm going to move them around to areas where each graph is more relevant. -- Schaefer 16:58, 14 July 2005 (UTC)
Thumbnails look good. Rose Lacy 22:34, 14 July 2005 (UTC)

The smaller images make it look better, but there still is some redundancy. For example, the expontential growth in computing image and the moore's law image are basically telling the same data. They could either be merged or one of them deleted. (I would rather the dark grey borders not be there either: they take up space and the boxes around images in the HTML already give us the borders an image would need.)

If one had to be cut I'd suggest PPTSuperComputersPRINT.jpg, which is speculative at best and has very low information density.

Another redundant pair is PPTCountdowntoSingularityLog.jpg and PPTCanonicalMilestones.jpg: these can be merged or one of them deleted. MShonle 01:44, 15 July 2005 (UTC)

  • Agreed, I would vote for eliminating PPTCanonicalMilestones.jpg, the events in the other one seem much more relevant to me (in general). Rbarreira 13:56, 16 July 2005 (UTC)
    • The article is not image heavy. I see no point in deleting those graphs - they are similar, but not duplicate. One shows the trend expanding into astronomical scale, other is for living matter - both represent somewhat different philosophies. I am happy to see some activity here after months of none, but try to expand this article, bring it up to FAC status, not reduce it. I worked three months to get those graphs to Wiki... --Piotr Konieczny aka Prokonsul Piotrus Talk 23:50, 16 July 2005 (UTC)
      • I don't see that distinction you're making - both include technological and living matter evolutions. But fine by me, the images don't bother me at all :) Rbarreira 23:55, 16 July 2005 (UTC)

Well, if we're talking about eliminating images, I think the cartoon one is pretty irrelevant here... Rbarreira 13:52, 16 July 2005 (UTC)

A screenshot for the culture section, which currently has no images. What else would you suggest? --Piotr Konieczny aka Prokonsul Piotrus Talk 23:50, 16 July 2005 (UTC)
Does every section need an image? Rbarreira 13:48, 17 July 2005 (UTC)

Schaefer, why did you remove Odyssey 5, Matrix, Collossus, Akira and NGE from the culture sections? Having seen all but Collossus, I'd say they do represent Singularity (or near/post) scenarios. --Piotr Konieczny aka Prokonsul Piotrus Talk 17:54, 16 July 2005 (UTC)

The end of NGE is a Judeo-Christian religious apocalypse, not a technological singularity. The cause of all the big events in Akira are humans with psychic powers, not runaway technological progress or AI. Colossus took over the world by threatening it with nuclear weapons, and there was no runaway technological progress or even mention of Colossus finding ways to improve its own intelligence. The Matrix comes closest, being weak cyberpunk at least, but the AI society is stagnant and incapable of bringing about a singularity. They can't even kill off a bunch of rebel humans or clean up their own atmoshphere. Any respectable general AI with the intelligence of an Agent, access to its own source code, and a healthy desire for galactic domination could rip apart the existing AI government, clean up the skies, and get on with important things like converting all the matter in the Solar system into quantum supercomputing substrate.
But, as fun as it is to discuss, I don't feel very strongly about any of this. If you want them in, feel free. It's not terribly productive for me to be worrying about fiction stuff anyway. -- Schaefer 21:38, 16 July 2005 (UTC)
I'd say that the end of NGE is an apocalypse reached through advanced, incomprehensible technology - so its the 'pessimistic singularity'. In Akira, those psycho-powers are again reached through scientific experiments. Haven't seen Colossus, but as it was AI taking control, it has *some* similiarity to AI, although I see your point that there was not much tech progress beyong AI. As for Matrix, I wonder if the time when AI took control of the world wasn't *the Singularity* already? Or at least something close to it - perhaps some kind of failed Singularity... although yes, I agree that those AI seem fairly stupid. And it's best to forget about the 3rd movie in a series...how the mighty have fallen, right? :> As you say, culture section is not that important, but on the other hand - it is how many people meet the singularity idea first - by reading sf. --Piotr Konieczny aka Prokonsul Piotrus Talk 22:00, 16 July 2005 (UTC)
The Animatrix talks about the rise of the machines in the Matrix, and indeed they export new technologies and more efficient ways of making old technologies that they collapse the rest of the world economy (I think that's how it went; in real life, though, the humans would be no worse off than if they didn't trade at all, but it wasn't exactly written by economists). So, I think there's a strong case for a singularity in the Matrix world. That the aliens didn't have any space exploration program just shows you how weak the Wachowski brothers were at writing, which is already evident in the poor quality of the sequels. MShonle 01:55, 17 July 2005 (UTC)

Biased Images

I believe this article is rather biased for the theory. Not so much because of the text, but because of the massive presence of professional impressive graphics showing events in a unscientific way. That´s the false authority argument: you show the public a lot of things that he cannot have the time to fully understand (logarithmic0what?) but feel compelled to believe, as you clearly know more in the matter than him. That´s why I made this graphic, with the same hype language, to show prove that indeed the end of times are around 1500, when the lord will save us from all those sins we committed since we came out of middle age. Oh do ye dare to enter that new discovered land...--Alexandre Van de Sande 03:42, 18 July 2005 (UTC)

I think the graphs go well with the text (debate on if there are too many of them or if four of them can be merged into two notwithstanding). Simply showing a graph of progress is completely appropriate for the article. I've included a link to Logarithmic scale to explain further. MShonle 05:14, 18 July 2005 (UTC)
The graphs, FYI, were made by Ray Kurzweil, a reknown scientist. If somebody doesn't understand what a logarithm is, he may read up on this on wiki, or go to Wikipedia:Simple English Wikipedia or repeat the secondary school. As far as your graph goes, striking any unfounded religious implications, it is correct (assuming you did use historical dates for it?): as the underlying trend is expotential, if you cut it in 1500, 1000, -5000 B.C or whenever you chose, it will still show the singularity a'comin' - i.e. the incresing speed of changes. However, as the singularity means changes more rapid then we can understand it, it has not happened - yet. --Piotr Konieczny aka Prokonsul Piotrus Talk 10:23, 18 July 2005 (UTC)
Piotrus, the point isn't that you can cut the graph anywhere you want, it's that you can hand-pick the historical events so that they are more and more frequent as you approach a pre-specified point of singularity, and there's no objective way of measuring the importance of these events. The first graph on the page doesn't suffer from this problem, though, because Kurzweil compiles many lists of key events in human history written by other people not worrying about the Singularity. Also in interest of NPOV, I think the title of PPTCountdowntoSingularityLog.jpg should be changed. -- Schaefer 13:08, 18 July 2005 (UTC)
Yes, this is why I chose this pic for the lead. I would like to see all those events from this graph listed, though - compiling sources seems fairly NPOV, but it would be nice to know what those souces are saying, exactly (i.e. which events to they consider important). Feel free to rename the files, I uploaded them under the same name I got them. --Piotr Konieczny aka Prokonsul Piotrus Talk 21:32, 18 July 2005 (UTC)
I don't think the graphs are biased at all, for the reasons other people have pointed out. And of course, the criticism you had put in the page should have never been in the article itself. Marking the edit as minor could also be regarded as not-nice, but I'm not saying you did it with intention :) Rbarreira 23:20, 18 July 2005 (UTC)
And this discussion is also taking place Talk:History_of_the_World#Graph_of_Singularity - should we use one of the pics there or not? --Piotr Konieczny aka Prokonsul Piotrus Talk 10:02, 19 July 2005 (UTC)
On the graphs: They vary in quality. The one you parody above is questionable, certainly and I'd vote to remove it on the basis that it's a POV construction of history. However, the one with the "operations per second" of a human brain was completely unfounded, unsourced and just plain non-encyclopedic. As such I was bold and removed it. If someone specifically disagrees that that image is unsourced and uses facts which have not been demonstrated, please reply here, and give me some references to the research which has quantified human thought in those terms. Thanks. -Harmil 12:17, 19 July 2005 (UTC)
It comes from Ray's article [2]. I think we can add it back with a note it's based on his calculations/article (as are all of the graphs, in fact). --Piotr Konieczny aka Prokonsul Piotrus Talk 17:12, 19 July 2005 (UTC)
(i am sorry i have all edits minor by default- i just let his one go)
Of course to understand logarithm is a pre-requisite to understand tech. sing. but my false authority statement goes that when you call "ray kurzweill", Logarithm, and a nice looking graph (nice looking in a way edward tufte would call chart junk), you get a graph that informs little but imposes a lot. The graphs is biased because the Y axis is non-existent: measuring the next step is not a precise date, because you need to hand pick your next step. And Unless someone cite Kurzweill´s source, I think the other graph (the first on the page) is also biased, because we cannot check it´s accuracy. But now there are a lot less graphs on the article (and it does not looks like rays personal page), so this discussion was productive.--Alexandre Van de Sande 16:38, 19 July 2005 (UTC)
Nonetheless Kurzweil is one of the top voices in this field. Regardless what we think of his ideas and graphs, they are notable. I restored this with expanded caption about problems with this image and his choices, feel free to improve this so it is more NPOV - but don't delete useful info.--Piotr Konieczny aka Prokonsul Piotrus Talk 17:12, 19 July 2005 (UTC)
How are we measuring "this field" exactly? This is an encyclopedic entry that could at most be narrowed down to social science as a field, and I'm not sure that Kurzweil ranks among the top voices in the field, no. What's more, this is simply bad science and POV to boot.
Seriously, is it that hard to tell that these graphs are simple measures of the quality of recorded history? We record one data-point for "techniques for starting fire" and separate data-points for "modern physics" and "nuclear energy". Why don't we list indoor stoves? Coal-fired smelters? Gas as a utility? These are major, world-changing events, but they seem less important because they are removed from us.
I guess we just have to shoot them down one at a time to make the point:
  • Milky Way - Clearly there is no logic in associating this with the technological singularity, since this has no bearing on the fate of human technology, nor of its singularity, but let's go with it.
I doubt that our singularity would exist without our galaxy, but I think I see what you mean. Rbarreira 02:04, 20 July 2005 (UTC)
  • Life on Earth - Ok, this is a reasonable item to include, since it sets a starting point for the activity that will lead to the signularity. This makes some sense.
  • Various types of cellular and multi-cellular life emerge - Should be one data-point.
  • Asteroid collision - Has nothing to do with anything.
I think it does, it affected life on Earth to a great extent. You may see it as a new "beginning" for Earth, since it (theoretically) ended the dinosaurs domination and allowed mammals to take over. Rbarreira 02:04, 20 July 2005 (UTC)
  • First stone tools - Ok, here we go. This is useful, but ignores the explosion of tool types that happened after the emergance of Homo sapiens. Various types of spear-head, arrow-head and kife were critical to the history of man. These events happened in a rush as man exploded onto the scene. Of course, there were probably thousands of important discoveries between the two. They may have included the ability to preserve hides, the construction of simple fortifications against animals, hunting techniques, rudimentary weather prediction, food preservation through salting (salted fish) and/or fermentation (fermented milk and yogurt) or both (brine pickling). These are all key discoveries, but the initial date for each one is lost. This explains why they don't show up on lists such as this.
So you think it would affect the graph in a negative way? Maybe I should have understood what you meant, but please explain. Rbarreira 02:04, 20 July 2005 (UTC)
First of all, there is only that much detail one can put on the graph and still have it readable. I suggest thinking along the lines: 10-15 key events/milestones, then making a rank. This is why they are summarised like this. Second: no matter how many of those early tool types or abilities can you list, I am sure that a list of components of lets say, a modern computer, would be at least as long. In quality, singularity is very visible. Thord. Whether invention of first tools or homo-sapiens tool explosion is more important, or should both of them be listed, I don't know. Stone tool is a poor article, but note that History of technology has an interesting list. --Piotr Konieczny aka Prokonsul Piotrus Talk 09:04, 20 July 2005 (UTC)
  • Invention of agriculture - Here's a tough one. You're going to have to allow for the fact that each stage of this process was a major step forward on-par with the step between modern physics and nuclear energy. Crop management involves pest-control, tool invention and refinement, domestication of animals, harvesting technique, crop storage, etc.
You did lose quite a few milestones from the Image:PPTCanonicalMilestones.jpg image. Why - or are you basing this discussion on some other list? And I won't agree with you. The agricultural revolution was composed of those smaller inventions, just as modern physics is composed of various milestones (newtonian, einstein, quantum...), same with nuclear energy (theory, pile, controled reaction, reactor, etc.). Again, let me remind you you have to lump the details together. Of course, in the end this is a kind of POV: you may think that technique of crop storages is as important as modern phyciscs and that the entire agricultural revolution was much more important then modern physics - but I will disagree. Btw, domestication of animals is often seen as the milestone separate and as important as the agricultural revolution. --Piotr Konieczny aka Prokonsul Piotrus Talk 09:04, 20 July 2005 (UTC)
  • Development of the wheel, writing - Here is where the nature of the graph becomes truly evident. Certainly these two things did not happen at the same time, but because their exact origins are lost in time, we put them on the same data-point. Of course, if we were to assume that they *might* have occured 50 or 200 years apart, that would certainly look wrong on our graph...
200 years in 10^4 years is only a 2% difference. Rbarreira 02:04, 20 July 2005 (UTC)
Agreed. I don't see much difference here. Wheter they are just 'next to each other', lumped together or even spaced significantly won't change the graph much. Say, I suggest we play a game: let each of one list 10 key events in the development of humanity, then see how they fit on the graph. It would be interesting if there was any kind of a pool on the net somewhere, to get a statistically significant sample. And as I wrote earlier, it would be great if we could check sources and collect specific data on Image:PPTParadigmShiftsFrr15Events.jpg. --Piotr Konieczny aka Prokonsul Piotrus Talk 09:04, 20 July 2005 (UTC)
  • Democracy - Why are other forms of government not shown? Shortly after democracy would come the republic, but that would throw off the graph.
In what way? Again I may be really stupid for asking. Rbarreira 02:04, 20 July 2005 (UTC)
Actually, I have written my MA on the 'history of democracy', and I was quite suprised to see the singularity in the evolution of the political systems. Consider, for start, Max Weber's Tripartite classification of authority: charismatic authority ('I got the biggest stick') determined the ruler of the primitive, even pre-human societies for more then hundreds thousands of years. Traditional authority is as best a little over 20 thousands. And rational authority that's the last few hundred. If you want to put patriarchy/autocracy, monarchy and modern systems like democracy, single-party state, military junta, communist state (yes I know it's problematic, but see [3] for my favourite list of gov's), again you would see that those types have vastly multiplied in the last hundred years or so. Btw, republic is no replacement for democracy. --Piotr Konieczny aka Prokonsul Piotrus Talk 09:04, 20 July 2005 (UTC)
  • Renaissance (printing press) Sorry, but there are MANY highly critical steps between that and the invention of writing. These include the invention of plant-based writing materials, the development of the scroll tube for preservation of knowledge and the bound book.
Important? Yes. Critical? Perhaps, but unless you have some publications proving this...which is why I suggested gathering some data above. You could add ink, and both the graph and you forget about the 'other side' of history of communication - for example, I would say that development of alphabet was definetly a milestone as important as the printing press. One could argue about the development of symbols, pictographs, ideographs, but... suprise...when plotted on the graph, they would again form, suprise, and expotential line. Check the dates in my article (unfortunately, it is in Polish, but dates don't change): [4]. PS. I didn't write this article thinking about the singularity, but after I checked all the dates, the conclusion was, again as with my MA, suprising. --Piotr Konieczny aka Prokonsul Piotrus Talk 09:47, 20 July 2005 (UTC)
  • Industrial revolution (steam engine) Nice to be able to put this in on its own, but it's composed of dozens of very rapid developments which include the invention of interchangable parts in manufacture, the steam engine, mechanized harvesting, textile techniques, etc. I would argue that the rate of development during the height of the industrial revolution rivaled our current rate of development per-capita.
I wrote about the 'details' above. From what I read, the 'rate of development' was slower then. Not the best measure, but check the graph plotting number of patents. Anyway, until we get better data, it is POV ('I would argue otherwise....'). --Piotr Konieczny aka Prokonsul Piotrus Talk 09:47, 20 July 2005 (UTC)
This is pure hogwash, and demonstrates only that humans keep history in a logarithmic fashion (note that I'm a proponent of the singularity theory... I just think this graph is POV nonsense) -Harmil 18:49, 19 July 2005 (UTC)
Well, I think the graph makes sence. After all, isn't the acceleration of progress the backbone of singularity theory, and isn't a logarithmic graph the best way to visualise this? --Piotr Konieczny aka Prokonsul Piotrus Talk 09:47, 20 July 2005 (UTC)
After clearing my mind for a few days, looking back I can see what´s iching me about all the logarithmic countdown graphs (including the first one). Both of the descending graphs are dangerously misleading, and should be used wuth caution. First let´s separate the theory from an statement:
1-The amount of information available, discoveries and technological advances have acelerated extraordinarely fast in the last century, specially the last ten years. That is a fact and we do not need a complex theory to prove it: just open the newspaper or connect to the internet.
2-Because of this, in a specific point of time in a near future (the singularity event) something will happen that will change everything (and what is that something can be interpreted as just information overflow or total doomsday by AI overlords). That´s what this article is about, to discuss wether this event will or will not happen is not into our scope, our goal is to gather all information objectivelly. And that´s precisely where the graphs fail.
By accepting statement #1 we can fairly imagine that if we put 100 random wikipedia articles about events in a log timeline, they would be fairly homogeneous, the quantity of articles concerning the last century would be about the same of those concerning the 1st millenia DC. That´s perfectly fair.
Now the given image starts to be bias, the moment it picks this timeLINE and add another dimension, making it a graph. Note that by putting on the Y axis the interval between each event, Ray does not adds any new data, just duplicates the X axis: plotting any graph this way would result in a descending line. That is, a sequence of Logarithmic intervaled events, in a logarithmic 2D graph will, of course result in a straight line.But until now, he just did something awkward, but not wrong.
Actually, what I think the graph is showing is that DESPITE using a decreasing log scale, the visual density of the events is more or less the same now as it was many years ago. That shows progress is accelerating. OK, this graph's points were chosen by Kurzweil, but the ones in the first graph that this article shows weren't... Rbarreira 03:12, 20 July 2005 (UTC)
The catch goes, that gestalt makes our brain see things in the most simple form, and when you see a line going down to the x line, our natural reaction is to guess that it´s going to hit it down. We are not used to think logarithmically so it´s hard to understand what happens when the line reaches the bottom - so the most natural conclusion is that something that we do not understand is going to happen, and since Ray is a renowed scientist ("Kurzweil is one of the top voices in this field" remember Carl sagan, in science there is no such a thing as Authority) we hear what he wants to say.
The graph does not says that the line will touch the bottom, but it´s finely drawn to visually give us that clue, and since the left represents the past we can guess that the future lies to the right. Of course, it does not: the more we walk to the right on the graph the more it approaches the moment of now: one tenth of a year, one hundreth of a year and so goes on. If the graph was to be extended to both sides, we would see a graph similar to a upside version of

where it would never touch neither axis, declining fast as if coming from the infinite up, and then going almost parallel in the infinte left direction
So, the graphs are extremelly misleading: they are drawn in such a way not to help you navigate in a complicate information sea, but the opposite, to cover up with unnatural language what is a rather simple information. And given that this was meant to be a quick slide among many, then our friend Ray Kurzweill does not deserves the respect I used to have for him. I still admire his [www.kurzweilai.net/brain/ frame.html?startThought=Artificial%20Intelligence%20(AI) The Brain interface] thought.
Given all that, I do not think the graph should be deleted, but in the same way a political propaganda poster fits a certain article, not to ilustrate the article but to show a new reader what kind of false arguments advocates of singularity use. If anyone manages to nPov my comments, feel free to copy them anywhere... --Alexandre Van de Sande 02:22, 20 July 2005 (UTC)
A interesting way to do it, for example would be to plot a real log timeline, with the following data: infinite years before now (x=0), beggining of universe, formation of solar system (...) the internet invented, one year before now, 36,5 days before now, 3,6 days before now, 8 hours before now, 52 minutes before now, minus five minutes, minus 31 seconds, -3.1 seconds, -0.3 seconds, -0.031536 seconds, and so on, never reaching the Now moment (to the infinte right) --Alexandre Van de Sande 02:22, 20 July 2005 (UTC)
What about Image:PPTCountdowntoSingularityLinear.jpg? --Piotr Konieczny aka Prokonsul Piotrus Talk 09:50, 20 July 2005 (UTC)
Just proves that the other axis of the graph is there just to confuse, not to plot any new data. The only thing that matters is the base ten line, that approachs menaciously to 10 to the power of 1, or even less. What does it shows? what conclusion can you take out of it? None!
If you don't understand logs, or don['t care to take time to think about it, you conclude "wow it's going to hit! it's approaching! Probably around 2012" If you do any thinking, if you extend the graph, to 10 to power of zero, 10 to the power of minus one ten to the power of minus 42 you only get a fraction of seconds before right now. 2012 is nowhere in that graph, nor any event in any place in the future! That's using mathemathical ignorance to awe the masses! Piotrus, you seem obsessed by this: I am not saying that singularity theory is false or that it's not going to happen, but you have to see that those graphs show nothing about it!--Alexandre Van de Sande 00:13, 21 July 2005 (UTC)
(message reposted from above, where perhaps no-one didn't even see it) Actually, what I think the graph is showing is that DESPITE using a decreasing log scale, the visual density of the events is more or less the same now as it was many years ago. That shows progress is accelerating. OK, this graph's points were chosen by Kurzweil, but the ones in the first graph that this article shows weren't... Rbarreira 00:22, 21 July 2005 (UTC)


Kurzweil's notability and representativeness

Piotrus wrote: "Nonetheless Kurzweil is one of the top voices in this field. Regardless what we think of his ideas and graphs, they are notable." As of 20 July 2005, Googling for "singularity" shows that Kurzweil's Law of Accelerating Returns doesn't pop up until result number ten. This doesn't say much by itself until you consider that of the ten results, only two appear to be unrelated to the topic of this article (it's used twice as a brand name), and the first three results are Vinge's original essay on the Singularity, a collection of Singularity links that makes no mention of Kurzweil, and the homepage of the Singularity Institute. Granted, Google searches aren't the be-all and end-all of ways to determine notability, but I think this should at least suggest that the article should focus more on Vinge than on Kurzweil, and I think as it stands the article gives the false impression that Kurzweil's opinions represent all Singularity promoters. Yes, Kurzweil's graphs are notable enought to warrant inclusion, but I think the context of "this is what Kurzweil thinks" should be made more clear, instead of presenting them as fact, which appears non-NPOV and in favor of the Singularity in general and Kurzweil's Law of Accelerating Returns in particular. -- Schaefer 15:32, 20 July 2005 (UTC)

Agreed. I wrote that Kurzweil is notable, not that he is the most notable. Vinge, as the 'father of singularity' is definetly more notable, this is why he is mentioned in lead. Btw, this brings me to this quote: "Predictions range from 2007 (Dan Clemmensen 1996) and 2012 (Terence McKenna 1996) " can we have a source for their publication with those predictions? Otherwise, as they are also not that notable, I'd like to move this sentence out of the lead somewhere into the main body. --Piotr Konieczny aka Prokonsul Piotrus Talk 15:42, 20 July 2005 (UTC)
I rewrote the intro (before seeing your comment here) and removed McKenna in the process for the same reason, but left in the 2007 figure as a lower bound. But now that I Google for Dan Clemmenson I see he's not really notable at all. (He is, however, humorously optimistic -- in one 1996 message I found he schedules a reunion party at a specific crater on the Moon on May 1st 2006, fully expecting the Singularity to have happened by then.) -- Schaefer 16:35, 20 July 2005 (UTC)
As for the lead, I started a discussion about it some time ago (see talk above) but nobody replied then. While I don't mind you shortening the references to other singularities, I would like to stress that I spend a lot of time working on the every word of the first para of the definition. Let me go over your major changes: 1) you changed ' point or period in the development of a civilisation' to 'future event' and 'societal, scientific and economic change' to just 'societal change' - both of make the definition shorter, and I could live with them, as it wise to make the definition as KISS as possible but limiting it just to societal change is I think a bit to specific, I would suggest droping the adjective and just go with 'change' or 'change in our enviroment' 3) most importantly you changed the logic by replacing 'due to the acceleration of technological progress, the societal, scientific and economic change is so rapid that nothing beyond that time can be reliably comprehended, understood or predicted by the pre-Singularity humans. ' to 'where technological progress and societal change accelerates due to the advent of superhuman intelligence beyond the ability of present-day humans to comprehend or reliably predict.' Where the older definition had the 'acceleration of the technological progress' as the cause of the singularity, you limit to 'advent of AIs', and from what I read, not everybody agrees AIs will be invented first - some speculate that nanotechnology will be required to reach that level. In the definitions I cited above, not that von Neumann, Ulam and Kurzweil do not mention inteligence, just the acceleration of changes. 4) Finally, you replaced the 'pre-Singularity humans' with 'present day humans' - this is an error, as many theories argue that some humans will be able to 'accompany the singularity, transcendt, etc. - thus some of those present day humans may in fact be able to comprehend or understand the Singularity world, if they undergo some kind of transition. I would recommend replacing your version with this: 'due to the acceleration of technological progress and likely advent of superhuman intelligence the change in our enviroment is so rapid that nothing beyond that time can be reliably comprehended or predicted by the pre-Singularity humans.' How does that sound? One last thing: why do you think that 'evolution of society' should be a subsection of the Kurzweils' law section? I see little connection, at best I'd do it the other way around. --Piotr Konieczny aka Prokonsul Piotrus Talk 19:10, 20 July 2005 (UTC)
I'm continuing this in a new heading below, as it no longer concerns the Kurzweil log charts. -- Schaefer 21:13, 20 July 2005 (UTC)

Graphs used elsewhere

There are three other places that the graphs have been used:

  • Game complexity (removed for being off-topic)
  • History of the World (removed for being "one of the worst examples of pseudoscience ... speculation and bizarre graphs created with a specific end in mind")
  • Logarithmic timeline (where the logarithmic history plot makes a useful example of a particular way of compressing history from the modern point of view)

it's already happened

I think the singularity has already occured. See Duffy's law. -- 216.234.56.130 21:29, 18 July 2005 (UTC)

Asimov's Three Laws

Under The desirability and safety of the Singularity, the article states:



"However, in one of Asimov's novels, despite these laws, robots end up causing harm to individual human beings which brings about the formulation of the Zeroth Law."



I think the causation has been confused, surely the Zeroth Law led to individuals being harmed (presumably to humanity's benefit)?

Although not explicitly specified as the Zeroth Law, it is implied that the reinterpreted First Law of the 'Machines' (which is effectively equivalent to the Zeroth, if a replacement) was their rationalisation of The Laws, not because of people being harmed (which it was said to allow, not prevent).


Also, perhaps it should be emphasised in the article that the novels generally demonstrated the fallibility of The Three Laws; given the subject at hand, it would seem important.

By all means, if you feel confident in your knowledge about Asimov's laws, Be bold. --Piotr Konieczny aka Prokonsul Piotrus Talk 15:45, 20 July 2005 (UTC)

Evolution of society

Comments appreciated at Peer Review: Wikipedia:Peer review/Social evolutionism/archive1. --Piotr Konieczny aka Prokonsul Piotrus Talk 15:45, 20 July 2005 (UTC)