Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
Spazzle
Jul 5, 2003

So putting eripsa's nightmarish schitzo dystopia to the side, what are ways that upcoming technologies actually be used to help make the world better on a broad scale?

Adbot
ADBOT LOVES YOU

RealityApologist
Mar 29, 2011

ASK me how NETWORKS algorithms NETWORKS will save humanity. WHY ARE YOU NOT THINKING MY THESIS THROUGH FOR ME HEATHENS did I mention I just unified all sciences because NETWORKS :fuckoff:

Obdicut posted:

So in other words, it's not attention that shows the importance, but the qualities of the thing itself. In order to know how important it is that someone is paying attention to something, you have to know everything about that thing.

Again, you're showing how what you're saying is wrong, and yet you don't realize it. You're making the argument against attention showing things that are significant, and instead arguing that the actual qualities of those things show that they're important.

Wait, you are complaining is that I used the only bit of semantic information you gave me, that they were nitrogen canisters? Why is that information off limits? Where did I say that attention alone and no other information will solve all these problems?

My phone knows where I am at any given time, and what routes I typically take to get from place to place. But it also knows that one of those places I've designated as "home", and so it treats that place differently than other places. I go to work and come home every day, but that doesn't mean the two locations are equally important to me, and so Google treats them differently. That's a difference in semantic content that gives the attention data interpretive significance. I'm still looking at the things you are paying attention to, and I'm using semantic data to give that activity significance. It still all depends on capturing that activity.

Or for a slightly more ambitious case, I might correspond to lots of people in my company, but the frequency of email exchanges isn't itself enough for knowing how those exchanges reflect the position in the company. I might email mostly to peers, for instance, and to bosses and underlinings infrequently. So a single letter from a person doesn't tell me enough about what that email might mean. But if the email also contains a signature line identifying itself as the CEO of the company, that semantic information helps characterize the importance of the exchange.

I never said that semantic information is irrelevant. Its incredibly useful for building predictive models, and I'm not sure why you think I'm cheating by appealing to it. It's exactly where Google's services are headed. I'm not sure why you think I ever suggested otherwise, but you've obviously misunderstood the view, and your critique appears to be aimed at a strawman. That would explain why my articulation of the view makes you think I'm arguing against it.

Obdicut
May 15, 2012

"What election?"

RealityApologist posted:

Wait, you are complaining is that I used the only bit of semantic information you gave me, that they were nitrogen canisters? Why is that information off limits? Where did I say that attention alone and no other information will solve all these problems?

That information isn't "He has nitrogen canisters" it's "He has nitrogen canisters, and that means..." That's the system, that can put together that 'semantic' information, that is the magic hand-wave you're depending on. Attention is the least important part of your system. So why are you calling it an 'attention economy'?

You could also pay work on the other major issues with the attention model itself-- that people ignore problems, that interest groups don't equal expert groups, etc. etc.

Cicero
Dec 17, 2003

Jumpjet, melta, jumpjet. Repeat for ten minutes or until victory is assured.

Spazzle posted:

So putting eripsa's nightmarish schitzo dystopia to the side, what are ways that upcoming technologies actually be used to help make the world better on a broad scale?
An autonomous cartopia would have fewer accidents and thus fewer injuries and deaths, fewer cars manufactured, more accessibility (to old/young/disabled/drunk), less space allocated for parking lots, and possibly better traffic and less pollution/better mileage per passenger. It'd also destroy a lot of jobs though.

Mc Do Well
Aug 2, 2008

by FactsAreUseless
Eripsa, I'll do a quick statistical experiment this month.

I'll look at as many of your threads in D&D as I can (can you give me a definite N?) and get some stats about repeating repliers (like myself) and how the threads turn out.

It would be interesting if they follow a similar pattern.

Cicero posted:

It'd also destroy a lot of jobs though.

I disagree. There would be regular maintenance/cleaning routines, work done by humans with tools. The system must be administrated, probably done on a county/state level in terms of purchasing/requesting custom vehicles (work done by human designers and manufacturing robots [machines that also need regular maintenance]).

The Amazon thing evokes an image of local warehouses staffed by people and forklifts with trucks coming and going. Although I read a BBC story recently about Amazon pickers carrying smartphones and running around like rats in a maze. Give them bicycles. I'd do that job.

There will still be markets for industrial vehicles, as well (tractors, UAVs, dumptrucks), and for insurance reasons there will be a long time where a human will be sitting in the driver's seat as a failsafe.

Mc Do Well fucked around with this message at 06:51 on Dec 2, 2013

RealityApologist
Mar 29, 2011

ASK me how NETWORKS algorithms NETWORKS will save humanity. WHY ARE YOU NOT THINKING MY THESIS THROUGH FOR ME HEATHENS did I mention I just unified all sciences because NETWORKS :fuckoff:

Obdicut posted:

That information isn't "He has nitrogen canisters" it's "He has nitrogen canisters, and that means..." That's the system, that can put together that 'semantic' information, that is the magic hand-wave you're depending on. Attention is the least important part of your system. So why are you calling it an 'attention economy'?

You could also pay work on the other major issues with the attention model itself-- that people ignore problems, that interest groups don't equal expert groups, etc. etc.

I don't need to know exactly what those canisters mean to you; if I have data about your behavior around them (that is, if I have your the behavioral data of what you do around these objects, your attention profile), then even without knowing anything more about you, I might compare your use to the way the general population engage similar objects, and use this information to generalize on other aspects of your behavior. That's all an attention economy really is: a way of generating predictive models on the basis of past behavior. If your attention profile matches the profile of what we know to be, say, a physics lab tech, then we might use this to infer more precise things about how important the canisters are to you because we've made a guess at the kind of user you are.

And again, this guess isn't by having some magical insight, but is simply by comparing your profile to other user profiles, clustering the results, and interpreting them with any semantic data we have on hand. If the email signature profiles of most of the people who use nitrogen canisters in the way you do contain the title "Physics lab tech", then its a reasonable guess that you are also a lab tech, and this generates a host of inferences about how you might use those canisters in the future.

Tuning these predictions in any particular case is going to take a lot of patience and training. I'm not saying it is trivial. But surely it is possible, and surely our machines will have ever increasing semantic awareness of their surrounding and their relevance to our lives. I suppose I can talk in more technical detail about the technical and computational issues involved if you want, but I'm not assuming some computational miracle to make it happen. I'm only assuming that increasingly accurate information about the actual behavior of people will help us build more and better predictive models for anticipating their future behavior. I'm not assuming that they are perfect decision makers, I'm only assuming that we can suss out what kind of decision makers they are, and use that to engineer a world that is properly responsive to the decisions they do make. That doesn't mean giving them everything they want, but it does mean working out the procedures by which they can voice what they want and take clear actions to pursue it, and that the rest of us can try to accommodate and incorporate into the larger social fabric.

I've been putting most of my writing effort into tackling this last issue, of how everything becomes incorporated into a fabric that can operate as an organized system, so most of my answers are pointed in that direction. But again, there are technical and theoretical issues at many points in the scope of the picture. I'm not claiming that digital theory is a completely worked out alternative system that already has all the answers to all the hard social problems. I am saying that digital theory is a coherent framework of values and techniques for addressing all these questions in a consistent and methodological way, and for reorienting the political discourse towards towards these issues.

RealityApologist fucked around with this message at 06:46 on Dec 2, 2013

Eej
Jun 17, 2007

HEAVYARMS
I like this attention economy idea. It means that things like KONY gets more resources allocated to it than gay kids being bullied to death.

Obdicut
May 15, 2012

"What election?"

RealityApologist posted:

I don't need to know exactly what those canisters mean to you; if I have data about your behavior around them (that is, if I have your the behavioral data of what you do around these objects, your attention profile), then even without knowing anything more about you, I might compare your use to the way the general population engage similar objects, and use this information to generalize on other aspects of your behavior.

Right, again, this is the magic handwavey system. Right there. That you don't seem to realize this is magic handwaving is troubling. Not to mention that there's a huge amount of differences in how people who use nitrogen tanks actually use them, depending a lot on the system itself. The important information is not at all the attention they pay to them, it's that they have them in the first place. I'm pointing out that your 'attention system' requires a label on everything in the goddamn world, or requires the system be able to recognize every object in the world. It furthermore has to evaluate the uses of those object and make sense of the disparate ways people interact with the same objects to do the same work with them. That is the magical handwavey system, the actual 'attention' bit is really goddamn secondary to your magic AI system you've got there that can recognize all objects in realtime.

RealityApologist
Mar 29, 2011

ASK me how NETWORKS algorithms NETWORKS will save humanity. WHY ARE YOU NOT THINKING MY THESIS THROUGH FOR ME HEATHENS did I mention I just unified all sciences because NETWORKS :fuckoff:

Eej posted:

I like this attention economy idea. It means that things like KONY gets more resources allocated to it than gay kids being bullied to death.

The allocation of resources for major social projects is something that takes months or years to foment and coalesce. The Kony2012 video's huge quantity of attention was paid briefly, but that attention was obviously not sufficient to move the kind of resources that were being called for in the video. The Kony video was like a flash of light: it caught our attention but couldn't maintain it. We decided that there was nothing interesting here, and we moved on. If anything, it's an example of how good we are already at managing our collective attention.

Contrast with the networks and poliical coalitions building in the gay communities for decades. These groups have marshaled the dedicated contributions and lives of unaccountably many volunteers and activists who have remained passionate about the cause and motivated to do something about it. If these groups fail to attract the resources they require, it is because the systems in which they are embedded are actively hostile to the rights they defend. These interest groups persist despite the hostile environment.

It speaks volumes about how important we consider the one over the other. Y'all act as if youtube view counts are the only measure of attention, but that's certainly not the case.

RealityApologist
Mar 29, 2011

ASK me how NETWORKS algorithms NETWORKS will save humanity. WHY ARE YOU NOT THINKING MY THESIS THROUGH FOR ME HEATHENS did I mention I just unified all sciences because NETWORKS :fuckoff:

Obdicut posted:

Right, again, this is the magic handwavey system. Right there. That you don't seem to realize this is magic handwaving is troubling. Not to mention that there's a huge amount of differences in how people who use nitrogen tanks actually use them, depending a lot on the system itself. The important information is not at all the attention they pay to them, it's that they have them in the first place. I'm pointing out that your 'attention system' requires a label on everything in the goddamn world, or requires the system be able to recognize every object in the world. It furthermore has to evaluate the uses of those object and make sense of the disparate ways people interact with the same objects to do the same work with them. That is the magical handwavey system, the actual 'attention' bit is really goddamn secondary to your magic AI system you've got there that can recognize all objects in realtime.

This seems like a very petty objection. You don't like that I'm using the word "attention" as the name of the system. Okay, well suck it. I like the name, and it seems perfectly appropriate to me. If it bothers you, then get over it and talk about something that actually matters.

I'm not handwaving around this system. I've been giving several attempted to describe how it works at various levels of generality. Your objections are wild and untargeted and stem from a very unsympathetic reading of my view, so if I'm struggling to target exactly your worries it is a dialectical problem as much as it is a problem with the view itself.

But you yourself just characterized exactly the sort of thing it has to do. I think you are exaggerating; it doesn't need to know every object and label, but it should know as many as it can, and it should be able to learn and generate new labels as and when required. My brain is a cognitive system and has to do basically the same thing with my world. There's no "right way" to carve up the objects that exist in the world, but my cognitive system makes the best guesses it can in real time, and has a bunch of back-ups and fail safes it is ready to deploy in case something goes wrong. That's an impressive feat, no doubt, but obviously it is possible. I'm not asserting that such things are possible today, but only that our technology is developing in that direction and we need to plan our organizational models where such techniques are common and widely available. Because they will be.

Kalman
Jan 17, 2010

So the actual attention is meaningless and instead we need to interpret that attention through human-generated filtering and data analysis to determine how important a given piece of attention actually is.

Why, no, I don't think this will in any way reflect existing economic, social, and racial biases and systematically undervalue attention to anything that isn't important to those writing the system (I.e., middle class/upper class technocratic elites).

RealityApologist
Mar 29, 2011

ASK me how NETWORKS algorithms NETWORKS will save humanity. WHY ARE YOU NOT THINKING MY THESIS THROUGH FOR ME HEATHENS did I mention I just unified all sciences because NETWORKS :fuckoff:

Kalman posted:

So the actual attention is meaningless and instead we need to interpret that attention through human-generated filtering and data analysis to determine how important a given piece of attention actually is.

Why, no, I don't think this will in any way reflect existing economic, social, and racial biases and systematically undervalue attention to anything that isn't important to those writing the system (I.e., middle class/upper class technocratic elites).

You can't just overturn the social zeitgeist in one fell swoop. The fact is that most of our biases now are reactions to the heavy propaganda and conditioning that is imposed by the socially powerful. But with participatory digital technologies, individuals have been given a voice that partly counterbalances those ideologies, and more importantly encourages others to seek out alternatives. And through this process we've been slowly chipping away at the 20th century values bit by bit. We're living in a world where most media is consumed through sources that probably didn't exist even 10 years ago, that can be actively engaged through computers we are holding in our hands, that are cheap enough to give to basically everyone in the world, and that are being used to form actively engaged communities that are organizing across all sorts of political and social boundaries.

These fundamental changes to the infrastructure of social life offer the best hope of challenging the biases and dogmas that have become so deeply entrenched in our world. Taking advantage of the circumstances is the best shot we have to effect change. I'm really not sure what you think the alternative would be. More of the same?

Kalman
Jan 17, 2010

I think the attention economy is more of the same.

N. Senada
May 17, 2011

My kidneys are busted

RealityApologist posted:

These fundamental changes to the infrastructure of social life offer the best hope of challenging the biases and dogmas that have become so deeply entrenched in our world. Taking advantage of the circumstances is the best shot we have to effect change. I'm really not sure what you think the alternative would be. More of the same?

Have you ever studied postcolonial development? Like, do you consider any part of the world that is not English-speaking and/or not connected to the internet?

Eej
Jun 17, 2007

HEAVYARMS
So what I'm getting from this is that in the brave new world of the attention economy, Phillipine disaster relief will rank lower in importance than next generation taxi app because everyone will forget about it in a couple weeks since there hasn't been years of attention focused on the Phillipines by anyone who matters.

Cicero
Dec 17, 2003

Jumpjet, melta, jumpjet. Repeat for ten minutes or until victory is assured.

McDowell posted:

I disagree. There would be regular maintenance/cleaning routines, work done by humans with tools. The system must be administrated, probably done on a county/state level in terms of purchasing/requesting custom vehicles (work done by human designers and manufacturing robots [machines that also need regular maintenance]).
Sure, but on the other hand, all taxi and truck driving jobs in the country, and a lot (possibly a majority) of car manufacturing and design jobs.

Kalman posted:

So the actual attention is meaningless and instead we need to interpret that attention through human-generated filtering and data analysis to determine how important a given piece of attention actually is.
Perhaps we could abstract and decentralize this by using some sort of token that users would explicitly invoke to indicate which parts of their attention they consider important.

Nude Bog Lurker
Jan 2, 2007
Fun Shoe

McDowell posted:

There will still be markets for industrial vehicles, as well (tractors, UAVs, dumptrucks), and for insurance reasons there will be a long time where a human will be sitting in the driver's seat as a failsafe.

Right up to the point where the actuaries decide that self-driving cars cost insurers less overall, at which point you'll start paying a colossal "human control excess" if you want to drive your own car...

America Inc.
Nov 22, 2013

I plan to live forever, of course, but barring that I'd settle for a couple thousand years. Even 500 would be pretty nice.

Cicero posted:

Perhaps we could abstract and decentralize this by using some sort of token that users would explicitly invoke to indicate which parts of their attention they consider important.
The problem is that even if a neutral program could process this information to find a solution, we have the problem where, as far as we can tell, machines have been unable to find a solution without poring through already available information. It seems to me that this implies that the machine would be unable to work out novel, out-of-the-box solutions from available data (take Watson, for example). Shouldn't we wait to see how to mimic human problem-solving in machine form before we discuss how a machine could process massive loads of data and come to its own conclusions?
It appears to me that the AI needs to be able to act and think independently like a human, devoid of the bias of its creators, but we don't completely understand yet how to make a machine like that.
As Tokamak stated about Reality's algorithm:

Tokamak posted:

The full letter you wrote that explains the algorithm in more depth has the same problem. If you are saying that you have unpublished work that solves the 'obvious' infinite looping problem in your layman's algorithm/procedure, then computer scientists and AI researches would be VERY interested in reading about them. I'm sure google would be pleased not to spend millions of dollars on a boondoggle quantum computer.

Care to elaborate on why the professionals seem to be so 'behind the times' (using coarse-grained tools and not in real time)? If you have come up with a way that can get the same results but much faster, then why haven't they thought of it themselves? They are paid to sit around and think about this all day, surely they could have thought up some sort of consensus decision making model to solve complex social problems. Why is your suggestion 'too early' for the field?
(And this post is about facilitation, not independence).
I don't think you, Reality, responded to this post by Tokamak.
VVVV Doesn't change anything though. If we don't understand how a computer can act by itself yet, it seems premature to try and model how our benevolent dictator will operate. It's like we're trying to figure out how to build a rocket without understanding physics.

America Inc. fucked around with this message at 08:41 on Dec 2, 2013

Cicero
Dec 17, 2003

Jumpjet, melta, jumpjet. Repeat for ten minutes or until victory is assured.

Negative Entropy posted:

The problem is that even if a neutral program could process this information to find a solution, we have the problem where, as far as we can tell, machines have been unable to find a solution without poring through already available information. It seems to me that this implies that the machine would be unable to work out novel, out-of-the-box solutions from available data (take Watson, for example).
The post you quoted was me making a joke. What I was describing was money.

OXBALLS DOT COM
Sep 11, 2005

by FactsAreUseless
Young Orc

RealityApologist posted:

You can't just overturn the social zeitgeist in one fell swoop. The fact is that most of our biases now are reactions to the heavy propaganda and conditioning that is imposed by the socially powerful. But with participatory digital technologies, individuals have been given a voice that partly counterbalances those ideologies, and more importantly encourages others to seek out alternatives. And through this process we've been slowly chipping away at the 20th century values bit by bit. We're living in a world where most media is consumed through sources that probably didn't exist even 10 years ago, that can be actively engaged through computers we are holding in our hands, that are cheap enough to give to basically everyone in the world, and that are being used to form actively engaged communities that are organizing across all sorts of political and social boundaries.

These fundamental changes to the infrastructure of social life offer the best hope of challenging the biases and dogmas that have become so deeply entrenched in our world. Taking advantage of the circumstances is the best shot we have to effect change. I'm really not sure what you think the alternative would be. More of the same?

Nothing you've said here gives any idea of why "giving people a voice" will somehow change their biases or counterbalance the prevailing ideology. People already have voices, as in literal voices. Turns out they're mostly used to spread and enforce orthodoxy and biases, not challenge them. How does yelling into the technological void somehow change this? Do you really think socially powerful people who already utterly own and control the means of communication as well as production will somehow allow any threats to their power to survive? There's just a total lack of understanding of how human society works being shown here.


Also, in that vein of fundamentally not understanding society or humanity, I'm quoting from a page or two earlier this other example of classic eripsa idiocy:

RealityApologist posted:

I honestly think that in a world run by software, all the NSA cameras and tracking systems are operated on a publicly accessible, open and transparent website, something like Wikipedia, except for security.

I think we'd do a much better job of handling security, and the process would be overall less invasive and arbitrary, and much harder to exploit for petty political purposes.
HAhahahah yes the NSA will crowdsource national security surveillance. This is an actual thought had by a supposedly intelligent adult human.

OXBALLS DOT COM fucked around with this message at 08:49 on Dec 2, 2013

America Inc.
Nov 22, 2013

I plan to live forever, of course, but barring that I'd settle for a couple thousand years. Even 500 would be pretty nice.

Cream_Filling posted:

Nothing you've said here gives any idea of why "giving people a voice" will somehow change their biases. People already have voices, as in literal voices. Turns out they're mostly used to spread and enforce orthodoxy and biases, not challenge them. There's just a total lack of understanding of how human society works being shown here.
And this is why the machine needs to be able to think for itself, because otherwise we have a super-Watson which would merely facilitate human biases in catastrophic fashion. Which only raises more questions. Should it have a goal independent of human behaviors? How does it find a goal?

OXBALLS DOT COM
Sep 11, 2005

by FactsAreUseless
Young Orc

Negative Entropy posted:

And this is why the machine needs to be able to think for itself, because otherwise we have a super-Watson which would merely facilitate human biases in catastrophic fashion. Which only raises more questions. Should it have a goal independent of human behaviors? How does it it find a goal?

Pretty sure this is the plot of the Terminator documentary films series.

Seriously, what are you talking about? You are talking about creating a hypothetical perfect unbiased computer supermind to rule all of humanity. We're entering Star Trek levels of handwaving here. (PS the specifics of the handwave dictate the outcomes).

America Inc.
Nov 22, 2013

I plan to live forever, of course, but barring that I'd settle for a couple thousand years. Even 500 would be pretty nice.

Cream_Filling posted:

Pretty sure this is the plot of the Terminator documentary films series.

Seriously, what are you talking about? You are talking about creating a hypothetical perfect unbiased computer supermind to rule all of humanity. We're entering Star Trek levels of handwaving here. (PS the specifics of the handwave dictate the outcomes).
I'm not trying to handwave. I don't defend Reality at all. All I'm saying is this: If the attention economy merely reflects human biases, and the AI will reflect these biases and not be an improvement over existing systems of government, then the AI needs to be able to act independently of human biases. In my mind, that would require the AI to be able to have a sense of right and wrong. Of course we don't even know how to create an AI which can act by itself yet on a human or greater level of understanding. Thus my saying that this whole thread is like trying to build a V-2 rocket without any understanding of ballistics.
Edit: I'm sorry if I have frustrated you, Cream_Filling. This thread is frustrating enough.

America Inc. fucked around with this message at 09:02 on Dec 2, 2013

OXBALLS DOT COM
Sep 11, 2005

by FactsAreUseless
Young Orc

Negative Entropy posted:

I'm not trying to handwave. I don't defend Reality at all. All I'm saying is this: If the attention economy merely reflects human biases, and the AI will reflect these biases and not be an improvement over existing systems of government, then the AI needs to be able to act independently of human biases. In my mind, that would require the AI to be able to have a sense of right and wrong. Of course we don't even know how to create an AI which can act by itself yet.'
Edit: I'm sorry if I have frustrated you, Cream_Filling. This thread is frustrating enough.

What do you mean by "biases"? This seems to imply that there's some objective truth or morality or other measure of human value that a computer can somehow magically winkle out. This is an utter impossibility.

Humans can't even agree on what's right and wrong. If ever a computer is created that can determine right from wrong, it will reflect the standards of its creators or else reflect its own structure and nature in a way that may or may not be comprehensible to people (or some mixture of those two).


edit2: nevermind

OXBALLS DOT COM fucked around with this message at 09:38 on Dec 2, 2013

America Inc.
Nov 22, 2013

I plan to live forever, of course, but barring that I'd settle for a couple thousand years. Even 500 would be pretty nice.

Cream_Filling posted:

What do you mean by "biases"? This seems to imply that there's some objective truth or morality or other measure of human value that a computer can somehow magically winkle out. This is an utter impossibility.

Humans can't even agree on what's right and wrong. If ever a computer is created that can determine right from wrong, it will reflect the standards of its creators or else reflect its own structure and nature in a way that may or may not be comprehensible to people (or some mixture of those two).
By biases, I mean from my own leftist perspective (national, ethnic, race, class, sex, gender, sexuality), which of course someone else could disagree with. I agree with your post.
I'm not an eripsa alt. How about you look at my other posts instead of creating a straw man.
Edit: Cream_Filling, you're calling me an eripsa alt despite the fact that I'm criticizing his distorted logic, or at the very least trying to mangle through it.

America Inc. fucked around with this message at 09:31 on Dec 2, 2013

Adar
Jul 27, 2001
Re: heroin addiction, does the attention drug market run on what the TCC consensus decides is the flavor of the month or does it take overdoses into account? Drugs falling out of favor just because their most frequent users die off is wildly unfair to the rest of us who just want to steal money attention to get our fix :(

OXBALLS DOT COM
Sep 11, 2005

by FactsAreUseless
Young Orc

Negative Entropy posted:

By biases, I mean from my own leftist perspective (national, ethnic, race, class, sex, gender, sexuality), which of course someone else could disagree with. I agree with your post.
I'm not an eripsa alt. How about you look at my other posts instead of creating a straw man.

That's not what straw man means. The term you're looking for is "witch hunt."

And I'm not a fan of the term "biases" here. Sounds more like you're describing ideology. How would you separate value judgments from ideology? It's just not possible.

OXBALLS DOT COM fucked around with this message at 09:34 on Dec 2, 2013

Adar
Jul 27, 2001
The forum rules force anyone trying to reply to Eripsa to pay attention to his posts. Is all of this simply practice for when the attention economy makes Eripsa into the wealthiest troll of all? Discuss.

OXBALLS DOT COM
Sep 11, 2005

by FactsAreUseless
Young Orc

Adar posted:

Re: heroin addiction, does the attention drug market run on what the TCC consensus decides is the flavor of the month or does it take overdoses into account? Drugs falling out of favor just because their most frequent users die off is wildly unfair to the rest of us who just want to steal money attention to get our fix :(

The attention economy name is the cherry of stupid on top of this poo poo sundae because attention really has very little to do with it. Basically the whole scheme sounds like a magic black box that is supposed to derive an objective measure of value for all things everywhere based on big data and google glass and, uh, the cloud and stuff.

It's utterly worthless.

America Inc.
Nov 22, 2013

I plan to live forever, of course, but barring that I'd settle for a couple thousand years. Even 500 would be pretty nice.

Cream_Filling posted:

And I'm not a fan of the term "biases" here. Sounds more like you're describing ideology.
Which is why I said that "someone would disagree with me", i.e. it is ideological.

Cream Filling posted:

How would you separate value judgments from ideology?
I don't know. I don't really want to philosophize on the subject because I don't see it going anywhere. Coming to some sort of conclusion through the scientific method isn't possible.

Adar posted:

Eripsa, just out of curiosity, did you pay any attention to the last thread where evilweasel pointed out that applying your crowdsourcing traffic plans to the US would result in doubling the death rate in the country as a whole? As you've immediately brought up another car analogy my guess is you did not. Is this perhaps a crack in your theory that the wisdom of crowds rewards good ideas?
And thus Eripsa's internal logic falls apart. He ponders the futility of it all before his head explodes.

America Inc. fucked around with this message at 09:57 on Dec 2, 2013

Adar
Jul 27, 2001
After three months of careful global deliberation, the smoke rising from the chimney of the Neo Papal Conclave has finally turned white. The Catholic world has focused on this for most of the past three months; candidate after candidate has been hotly debated, scrutinized from all angles, and deemed unworthy. But consensus has unified around the only candidate able to sustain global attention for the entire time.

All hail His Holiness, Ceiling Cat The Third.

Adar
Jul 27, 2001
Eripsa, just out of curiosity, did you pay any attention to the last thread where evilweasel pointed out that applying your crowdsourcing traffic plans to the US would result in doubling the death rate in the country as a whole? As you've immediately brought up another car analogy my guess is you did not. Is this perhaps a crack in your theory that the wisdom of crowds rewards good ideas?

OXBALLS DOT COM
Sep 11, 2005

by FactsAreUseless
Young Orc
Hey guys, your friendly NSA community liason here! I'm uploading the latest elint captures and also last week's satellite and drone footage to the NSAcloudwikibase as we speak (or rather as I type XD ). As always, a reward of 50 attention bux to anyone who spots a terrorist or seditionist. Have fun out there! And as always a big welcome to our friends from China and around the globe!

Adar
Jul 27, 2001
ITT, the problem with Occupy Wall Street is that the smelly annoying guy who talked the loudest and blocked consensus on every issue for three months was not talking loudly enough

Adar
Jul 27, 2001

Cream_Filling posted:

Hey guys, your friendly NSA community liason here! I'm uploading the latest elint captures and also last week's satellite and drone footage to the NSAcloudwikibase as we speak (or rather as I type XD ). As always, a reward of 50 attention bux to anyone who spots a terrorist or seditionist. Have fun out there! And as always a big welcome to our friends from China and around the globe!

Posting to claim my attention bux for quoting this post

Is there still an attention bux reward for everyone who bumps an NSA thread over 500 replies or was that last week?

Adar
Jul 27, 2001
Don't mind me, just testing the theory that posting a bunch of times in a row is worth more attention bux than only posting once

here's a mildly NSFW picture of a porn star to make sure you're all looking forward to the next one

America Inc.
Nov 22, 2013

I plan to live forever, of course, but barring that I'd settle for a couple thousand years. Even 500 would be pretty nice.

Adar posted:

Posting to claim my attention bux for quoting this post

Is there still an attention bux reward for everyone who bumps an NSA thread over 500 replies or was that last week?
I'm sorry, that reward is only available to NSAcwb premium members. To become a premium member, you must spot 20 terrorists.
Get going kiddo!

(Advertisement)
Is the size of your tool not getting enough attention among your peers? Don't worry, because now It'll Go Far! automatically assigns a set amount of attention to your tool size! Never feel inadequate again!

America Inc. fucked around with this message at 10:11 on Dec 2, 2013

Adar
Jul 27, 2001
Remember, to register your opinion about this thread, it is very important that you not rate it at all

Tokamak
Dec 22, 2004

Negative Entropy posted:

The problem is that even if a neutral program could process this information to find a solution, we have the problem where, as far as we can tell, machines have been unable to find a solution without poring through already available information. It seems to me that this implies that the machine would be unable to work out novel, out-of-the-box solutions from available data (take Watson, for example). Shouldn't we wait to see how to mimic human problem-solving in machine form before we discuss how a machine could process massive loads of data and come to its own conclusions?
It appears to me that the AI needs to be able to act and think independently like a human, devoid of the bias of its creators, but we don't completely understand yet how to make a machine like that.

I don't think you, Reality, responded to this post by Tokamak.

They won't respond because they're lacking the ability to articulate their own thoughts in a technical way. Its easier to chat about attention economy, because it's 'just an idea' and it can't really be critiqued (unlike, say a decision making algorithm).

Besides that was not the point of my post, if you read the link then you would know that individual people will be arguing on the proposed solutions. It is similar to consensus decision-making and the 'brains' behind it are people splitting into groups/camps and coming to agreement in the smaller group. I guess the idea is that the 'best' solution will bubble up from a small group and bring about a consensus. If the group can't come to a consensus, then the suggestion is to try it again; this obviously does not work for divisive issues (so almost all of them).

It's really just a variation of utility function iteration with a bit of simulated annealing thrown in to try and get the utility function iterating towards its goal. It's not really novel, and the pitfalls of the algorithms the poster is cribbing from are well understood. It will only work if the bar for consensus is relatively low (~50%-66%, similar to how governments operate), or if one group suppresses enough dissenting views to form an artificial 'consensus'.

"Boston Massacre" style debate is actually a descriptive name for it and illustrates why the poster is a crank and shouldn't be taken seriously.

Adbot
ADBOT LOVES YOU

Tokamak
Dec 22, 2004

Adar posted:

ITT, the problem with Occupy Wall Street is that the smelly annoying guy who talked the loudest and blocked consensus on every issue for three months was not talking loudly enough

According to Eripsa, it is because he did not massacre his opposition and become the defacto consensus.

  • Locked thread