« 9/11 | Main | Little Boxes Made of Ticky-Tacky »

09/12/2008

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Gareth Nelson

Regarding trolling, what comes to mind for most people is people like the GNAA (Gay Nigger Association of America - if you google them do NOT visit their site using an insecure browser) or the less "in your face" kind of trolling with people who delibrately set out to provoke drama. Trolling is a very real phenomenon indeed, and is not simply a label used to censor people (though saying "you're just a troll" can of course be used for censorship too).

An example of the kinds of posts that would be considered trolling:
desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu desu

Or things such as rickrolling......

Another kind of trolling I once witnessed consisted of a group of trolls posting insults on a satanic forum saying they were from a christian forum, they then posted on the christian forum saying they were from the satanic forum. Pure drama-generation basically, and nothing to do with free speech by any means.

Gareth Nelson

Oh, and this looks like a fairly basic description of neural networks to me:
"- learning occurs as a distributed process in a network, based on recognizing and interpreting patterns

- the learning process is influenced by the diversity of the network, strength of the ties

- memory consists of adaptive patterns of connectivity representative of current state"

If you apply this to the human brain, that's what pretty much all the evidence we have says about how that big pile of neural tissue in our skull does it's job.

Darien Caldwell

Sounds like the whole "Ebonics" thing all over again. Someone comes up with some fanciful theory, and tries to implement it as fact without any factual backing or peer review.

Getting a lot of people to go along with anything has never been a problem throughout history. I mean, just look at Scientology. :)

Horus Vale

Gareth - If you believe intelligence is an emergent property, then you might accept the view that all the information flow, connectivity, interactivity, and stored raw data and espistemology on the net will eventually, if not aready, produce a meta intellegence upon it. Gwyneth Llewelyn already posted an essay by Extropia Dasilva recently that the web does indeed possess many of the properties of a brain and if you compare many maps of the internet with magnified samples of brain tissue, both have the same node and link branching structures to them. So, who knows? Maybe the net having its own thoughts about the the net. But of course, its just theory and only one possible outcome of connecting all these dots. Still, the hive mind would be the ultimate in collectivism.

Prokofy Neva

Here's the Cliff Notes to the chapter on the ants in T.H. White's "Once and Future King":

http://education.yahoo.com/homework_help/cliffsnotes/the_once_and_future_king/19.html

If the Internet starts thinking on its own, I'll be likely to put a stick in its eye. But it won't be thinking on its own in any human way, as much as people fantasize about the Singularity.

When people like Gwyn or Extropia talk about the Net becoming conscious with all its nodes and whatnot, what they really mean to say is, "We hope that we, as a small group of tekkies who perceive ourselves as enlightened, can make the Net function with more artificial intelligence that we ourselves will code, which will be automatic processes that we can force other human beings to succubm to, and control them."

Don't be fooled by all the talk of decentralization.

Gareth Nelson

Horus - if you know anything about neurology then you'll know it's not as simple as "throw a bunch of nodes together in a network and BANG you'll have a mind". I can't see how anything as complex as intelligence can arise in a short timeframe without some kind of input.

The packets flowing around the internet or the links between webpages may look vaguely similar to a brain on the topological level to those who are uneducated on the subject, but they're fundamentally different.

Within the brain, the fundamental exchange between nodes is electrical firing in spikes and neuromessengers. There's no information as such within that localised signal, the information is distributed throughout the system in much the same way as the transistors in your computer's processor individually possess no meaningful state. In artificial neural nets we use what's called a trigger function to cause the next firing of one particular node's outputs, then to teach the network there's several possible algorithms - all of which involve tweaking the strengths of the connections. Now, one could try to make a strained analogy to the internet's various backbones and other pipes as having different "strengths", but consider that the bandwidth/capacity of a link is independent of the information in the system as a whole.

And of course the processing happens online in single nodes, often the raw data flowing is already processed in some form and the "intelligence" lies within individual nodes. We're not going to see the internet suddenly become concious or intelligent in itself unless we first turn it into a giant neural network (rather than a packet-switching network) and teach it extensively OR implement AGI in software without using the neural approach.

My money's on distributed agent approaches using old fashioned linear programming for the most part and a few dashes of neural nets for domain-specific pattern recogonition sprinkled about running on self-contained nodes. Seems like i'm not the only one who supports this view either.

Horus Vale

Gareth, a sufficiently vast and widely distributed intelligence might not be humanly recognizable as something that can really be definitively proven at a human scale as a scientific fact. That said, most of the advocates for such a theory would point out that there is one very significant element missing from your model of a classic artificial neural net that the Internet possesses - People. The net has people who can act as sensory input and triggers and already has individual intelligence as nodes. The kind of intelligence being discussed for the net is more like that of an ant colony than that of a true brain. But hey, you are right, I do not have the credentials or the data to pose a technical argument. And, even if I did, like I said the actual nature and thoughts of such a system spread widely out over time and space might not be something that was detectable, let alone provable. So, its just a theory and its likely to stay one.

Horus Vale

Interesting essay from the notes, Prok. And, no I do not want to think and behave like an ant or live in a society that thinks or behaves that way. It leads to the destruction of free will.

ichabod Antfarm

There once was a fella named Roger Penrose who said "look, I can make a one to one model of the human brain out of beer cans and string" (I paraphrase) and, he went on to say, I can set them all in motion and they will bang and clank and do all of the things that we observe when the meat version of this bangs and clanks but, he asks, is it reasonable to ask if these beer cans and string have any understanding? Are they intelligent?
Have we done anything significant?

Well, that was him. Then there was another fella, John Searle, who talked about a Chinese Room. Perhaps you are familiar with it. Perhaps this talk of machine intelligence requires more connectivism?

Gareth Nelson

The chinese room is good, since what it makes me think of is simply that the system as a whole in fact IS intelligent, even if individual components are dumb. Single neurons possess no intelligence at all, but a few million wired together in the right way are intelligent.

A neuron in your visual cortex has no clue what a shape is. All it "knows" how to do is accept inputs in electrical or chemical form and fire off outputs in electrical or chemical form. The first stage of the visual pathway in the brain has no clue what all those shapes mean either. (Interesting random off-topic fact on that by the way: the initial shape recogonition happens in the retina rather than the visual cortex).

Gareth Nelson

"is it reasonable to ask if these beer cans and string have any understanding? Are they intelligent?
Have we done anything significant?"

Can minerals calculate things?
Yes actually, if they're silicon and copper arranged into digital circuits

Are they intelligent?
If the system as a whole shows intelligence then the system itself is intelligent, but the components are pretty much dumb

Have we done anything significant?
Yes, that'd be one hell of a model. Simulating the whole brain in any form would be a huge breakthrough

Prokofy Neva

Again, the people constantly making up models for making artificial intelligence or making the Internet start to "think" are inevitably involved in aquiring power and influence for themselves. This is a social and political problem *they* don't want to address, obviously, and constantly keep hidden from sight, but it's the main topic.

A person writing particularly clever code isn't making an AI robot isn't making some separate, discrete entity that can think. He is making a tool that extends *his* will and intruding into interactions with other human beings in particularly insidious ways so as to compel them to treat the AI as "real". The coder can never be removed from the equation as a social actor, and he is operating on his fellows by a kind of subterfuge and stealth. The AI robot may get out of his control, but he is still responsible.

I didn't know about the Chinese room story, and am just learning about it know, but guess I tend to agree with Searles as I simply believe that intelligence is a mystery that is unsolved and in the realm of God and creation, and I simply insist that those who go beyond this mystery to some theory should also concede that they are engaging in belief in religious doctrine, and not science. The problem is they don't concede it, and I do.

I think key to this is examining the meta-layers, but the meta-meta-meta layer isn't just another meta, but something grander that does more than connect, it infuses with meaning. Infusion with meaning is not connection, but a mystery of why millions of connections underlying it become "something else".

You might ask whether a man manipulating Chinese symbols at the right time isn't essentially learning Chinese. Of course, he's still manipulating and not really understanding. It's like Russia, with its various bound phrases and idioms and tag-lines from culture. I can master the situations in which Russians say the phrases "Hello, I'm your aunt!" or "Hoofs and horns!" to indicate surprise or deceipt, but I won't fully understand that they come from a popular movie of the Soviet era or from a line in Ilf and Petrov's "The Twelve Stools". Even after I am taught those connections, and study them, I still won't have a good feel for when I should and should not pop in with those idioms, which is why foreigners trying to use idioms like "I'm baaaack" or "Houston, we have a problem!" can sound "off" even when "understood". It is a delicate matter.

Prokofy Neva

and the problem with the "luminosity" argument against Searles is that it cannot go the whole way to explanation, either. It says, essentially, that if I didn't master the movie idiom, it's just that I had another meta layer of more subtle cues to learn, and they are masterable, and if I didn't master the social cues, that is merely another layer, and masterable, etc.

For example, I remember very distinctly as a child of 3 or 4 writing numbers on a sidewalk to play hop scotch, and asking my father when they would be finished. "They won't be finished," he said. "That's infinity. You can always add one more."

"But wait a minute," I said. "I can keep writing all the way to the end of the block!" "But you can always add one more," my father said patiently." "But...but...what if I were able to write numbers on all the sidewalks of the whole world!" I asked, certain that would clinch it. "But you can always add one more," said my dad. "That's how infinity works."

And yet, it is not merely that you can add one more even if you run out of sidewalks; it's that you can run out of sidewalks, and still think one more, and conceive of one more, and yet have no sidewalks -- and that is something grander, the infusion of infinity.

Fred

Quote: "Neither of the meanings are legitimate in free democratic discourse and often are co-terminous with what is in fact a) deliberate incitement to imminent action, which should be actionable even under the First Amendment or b) merely persistent or idiosyncratic or annoying criticism that those in power don't like."

Having recently read the US consitution in full, would you like to do the same and read its commentaries and reconsider your statement?

What the first amendment very specifically does is prevent congress from passing legislation to inhibit, amongst other things, your freedom of speech. It absolutely does not prevent others from limiting your freedom of speech. See, for example, http://www.usconstitution.net/constnot.html#exp In addition, unless you are part of the fourth estate, you are not permitted freedom to write what you will, only freedom of speech.

Given you can't get the details of your own legislation correct, are we supposed to consider the rest of your statements as well researched and worthy of consideration?

Eyal Sivan

Once again, your arguments are circular.

On the one hand, you say "how patently fake the claim of 'applied Connectivism is'... in sayng it breaks down old power structures, or goes around old hierarchical relationships." You accuse George of being some kind of dictator, trying to dumb down his students with "endless subjectivity" so that he can spread his doctrine and tear down existing mechanisms of establishing legitimate credentials.

On the other hand, when George tries to demonstrate his credentials, he is "in credentializing mode." His field is "the last refuge of scoundrels," even though he went through the same University process that you vehemently defend.

Why are the credentials George received invalid? And if they are invalid, doesn't that contradict your assumption that Universities are good at establishing experts? Or, if his credentials are in fact valid, then doesn't that contradict your assumption that George feels they Universities are irrelevant?

You say: Expert = good, therefore George = bad for saying expert = bad, even though George = expert, which would mean expert = bad.

Or the other way: Ideology = bad, therefore George = good for saying Ideology = bad, even though George = ideology, which would mean George = bad.

This is a paradox.

Since you have studied in Russia, I'm sure you know that the history of Russian literature (according to my wife, the "expert") is a history of dialectic, with every subsequent master questioning the views and assumptions of the last. That almost makes me think you know full well that your statements are paradoxical...

P.S. In my opinion, you are not a troll. You are an intelligent and challenging critic, and there is a big difference. That didn't mean I wasn't pissed for a couple of days. :)

Prokofy Neva

No, again, not circular, you just don't like hearing the theory debunked because you like it.

I need to catch George in "credentializing mode" to illustrate the absurdity of his premises. The entire gang of 2,000 should overthrow George and have a 17-year-old kid in Omaha, Nebraska teach the course if he has the most time to be on line, the fastest Internet connection, and oh, the highest SAT scores or something else to throw in. If the pipe is more important, those at the pipe spigots should rule. George is irrelevant now under this sort of thinking.

But I'm happy to concede George *does* have credentials. He's wielding them inappropriately for somebody who says they don't matter (as he did astoundingly by insisting I read 4 subscription-only dense neuroscience magazines or I couldn't disagree with him lol). I can surely concede George has the credentials of a university position; a higher degree; publications. Those all count. I just don't think his theory is helpful, and is in fact a doctrine requiring religious belief.

I just don't think George gets to have his cake and eat it too -- tell everybody there are no more authorities and knowledge doesn't matter, but process on the way to accessing knowledge; that the pipe's the thing, not people who are experts, and then wield his own credentials to bang this theory through.

There isn't any paradox except in what George is doing; I'm calling him on it. I don't have a problem with George being the expert, and with others viewing him as the font of knowledge; HE is the one with that problem and wishes us all to be equal fonts of knowledge. We don't have publications and PhDs, however, and they matter, to some extent. Our fonts could be enhacing and complementing or challenging George's font, but in his plumbing system, we either all flow equally or we aren't good little connectivists.

I'm not concerned if you pounce on my and declare me "hypocritical" or "paradoxical" -- I simply wanted to call attention to what is obvious: George's credentialing maneuvers which are patently suspect in his system.

Both he and Stephen of course are bowing and scraping to the four compass points daily to say "I am not worthy, I am not worthy" but it's not good enough, frankly, as it's suspect -- and you catch them at it occasionally when George wields his subscription-only thick journals and Stephen suddenly tells us his politics are based on his "science."

Prokofy Neva

Fred, I'm afraid you're full of shit. You sound like a total loon, even an *educated* total loon.

Everyone knows that the law of the land in the United States is the Supreme Court decisions interpreting the Constitution. The Constitution itself is not the only law; it is not a dead book of laws such as it might be in a civil law system. It is made alive precisely by its constant interpretation and reference to precedents.

That basic nexus is something that Europeans, even British, often fail to realize. They get obsessed with the Constitution itself, or with "commentaries" or with things on the books like the Smith Act, and don't read the entier body of *Supreme Court decisions.*

There is no qualification in any Supreme Court decision that only journalists have freedom of media.

Anyone else is free to suppress my speech? Well, no, you need to look at a variety of decisions. Freedom of speech and freedom of assembly do clash, and then the rule is on behalf of the freedom of assembly (Boys Scouts of America case). In that case, gay rights activists wanted to stop the Boy Scouts from teaching anything against gays -- to control their speech, to make their speech not free -- but they failed.

You could turn it around and say, but the Boy Scouts suppressed the freedom of gays to teach their beliefs -- and they did -- but the law would say -- "make a new Boy Scouts then, and stop whining".

There's also the Mall of America case with the anti-Bush t-shirt. The idea that non-state actors are endlessly free to suppress your speech is one of those fictions of MMORGPs and fanboyz on forums.

When Congress "makes no law" such as to suppress the freedom of speech, or elevate one religion over another, it is making it so that no law can be passed that suppresses people's speech. And so it can't be. But the freedom of association means that anyone can make a club, business, blog service, and set up their own rules. And this is the greatest danger now to free speech, as there is a shrinking genuine public commons that is not corporate-owned that can be a true Hyde Park.

Fortunately, actual blog services are nowhere near as hysterical about application of their own more restrictive TOS as wacky and litigous forums-dwellers are.

Prokofy Neva

Also, "Fred," the rule for posting on MY blog, which DOES control your speech (you can go elsewhere and make your own blog) is that you must have a first and last Second Life, real-life, or bloggers' name.

BTW, it's helpful to look at the actual credentials of the loon whose wacky opinions you are quoting

"Common Name: Steve Mount

Formal Name: Stephen J J Mount

Title: Webmaster and researcher for USConstitution.net

Degree: Bachelor of Arts, Political Science, 1989, University of Vermont

Other formal learning: Constitutional Law, Community College of Vermont, 1999 (audit)

Relevant Experience:
Reporter, Vermont Cynic, 1986-1987
News Editor, Vermont Cynic, 1987-1988
Managing Editor, Vermont Cynic, 1988-1989
Webmaster/Editor-in-Chief, USConstitution.net, 1995-Present

Military Experience:
Vermont Army National Guard 1987-1992; tank loader, driver, gunner (M60A3). Stationed: Det. 1 1/172nd HHQ, Swanton, Vermont; Basic training and AIT at Fort Knox, Kentucky. Honorable Discharge 1992.

All research done for the USConstitution.net site was done by me. All opinions are mine and mine alone, with the exception of messages posted by others in the Message Board. A bibliography is available."

I'd prefer to get opinions on Constitutionality not from a Vermont crank who is a tank loader with a B.A. and a history of writing for a contrarian local publication, but the Supreme Court justices. I don't see from a quick glance that he is grasping them, either.

Eyal Sivan

Well then help them!!!

If the theory is not helpful, then help them remove the elements which are not helpful, but you can't deny that there is a legitimate problem that they are sincerely trying to solve!

I completely agree that some kid digital native kid is more qualified to frame these tools, more in his element, than George or Stephen. You clearly are as well. Hell, I am too! I started writing about this damn stuff almost ten years ago (see http://theconnective.org/what-is-the-connective/origin-of-the-connective/), and sometimes I feel I had it more right then than George does now. But that's why I'm on the course, to challenge and compare his version to mine, and to learn what I can about things I don't know (like learning theory).

George's credentials are paradoxical. All RL credentials are paradoxical in this new model, because they can be immediately and continuously double-checked. That's the problem!

Gareth Nelson

Prokofy:
"Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances."

1 - A private entity (a church for example) can respect a particular religious establishment as much as they like (though businesses and employers in particular may not discriminate on religious grounds - human rights stuff)
2 - Congress can't make a law prohibiting certain kinds of speech, but that says nothing at all about whether they can give owners of private property the right to arbitarily decide who is and isn't welcome
3 - The right of the people to assemble (right of assembly) is one covered by the UN declaration of human rights too and is nearly universally interpreted as "you also have the right to choose who to assemble with"

So, a private service's TOS could include "we don't allow speech that promotes prok-like views" and it'd be totally legal (if not rather nasty).

Why do so many americans mis-interpret their own constitution while also being so proud of it? I've found the more patriotic an american is the more likely they are to think they have a right to freedom of speech on private property.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)