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.

 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 The procedure above should all count as one timestep in the network; its all the computations necessary to calculate out the payment transaction, which is a single step in time. That said, the "coupling" relationship (at steps 9 and 12) is obviously very simplified in this procedure. If X and Y are coupled at c=cxy=cyx, with income δx and δy repsetively, then coupling results in ΔBx += cδy, and vice versa, so that each is changed by a factor of the other's global income. Notice that X and Y only "couple" once in the transaction; it is just one computation for each balance. That's fine, but the spirit of the proposal is a deeper tie between the two accounts, and given that this is all taking place within a discrete time step, we can be more sophisticated than this. In particular, we should be able to convert steps 9 and 12 into computing not just one coupling relation, but an infinite series of them, with the result being the convergence of the series. So ΔBx += cδy, but that changes Bx in a way that changes By, which in turn changes Bx again, until the each value converges. I'm fairly sure this can be done without too much more complexity, but after about an hour messing with it I couldn't figure out how to express it correctly. I left the simplified version in the procedure, but if anyone can help with that, it's an obvious place to extend the procedure to be more in line with the spec. # ? Apr 6, 2014 19:50
 Adbot ADBOT LOVES YOU # ? Aug 3, 2024 00:53
 JawnV6 Jul 4, 2004 So hot ... This is the kind of detail that should have been in the original. The fact that you even began the conversation of the higher level effects without figuring out these details shows a complete disrespect for the audience. The fights you had with me and others who bothered to spend the kind of effort you're just now scratching on without doing the legwork yourself is yet another slap on the face. I only began because you indicated these equations and effort had already been done. You lied about the equations. You excoriated people who'd obviously put more thought into it than you. These are not the hallmarks of good faith debate. # ? Apr 6, 2014 19:58

#### RealityApologist posted:

Step 4. Determine ΔB for all users implicated in endorsements of X. // ΔBy=eyx*p, for all users Z endorsing X, ΔBy=ezx*p
Step 5. Given ΔB from Step 4, calculate new B for all users implicated in endorsement. // By+=ΔBy; Bz+=ΔBz

Should that be ΔBz=ezx*p at the end of step 4? Otherwise you're constantly overwriting ΔBy. Also, if I'm reading this correctly, endorsement can create and destroy money - is this the intended behavior?

Your description of δBx in the preamble (global change on Bx over the lifetime of the network) and in the description of how to process a transaction (change in Bx over the transaction/payment being processed) aren't consistent. I'm assuming the second description is the correct one, since you never use δBx in the global change sense.

(Also, try to keep your subscripts consistent with your variables - you lowercase user variables when they're used as subscripts. At the moment, your equations are still understandable, but that's not guaranteed.)

#### RealityApologist posted:

I dropped distinct income/expenses as distinct abs functions; now everything is put in terms of Balance and the its changes (B and ΔB), with an additional value δB representing the average change in balance for users since the beginning of the network.
So you ripped out the tangible approachable Income/Expense that were easy to represent in the code. And replaced it with a value I can't even read in the preview window that you don't specify how to calculate. Yes, yes, I know you think you specified it, but you just give a few sentences, slapped it in an equation, and never once touched how the value actually changes in response to stimulus. The equation looks like a second derivative, the sentences make it sound like a rolling average, and in the actual worked example it's nowhere to be seen. I'm talking about 948B, fwiw. I'm imagining it's proving difficult to get a value for it at each time step if you're actually working the example as you said. This equation:

#### quote:

Let δBx represent the (global) average change x's balance over the lifetime of the network, or δBx = ΔBx/Δt.
It's what's going to screw you up. I gave you three possible definitions to pick from up there, let me know which one you meantdecide to go with.

There's also the words 'modified' and 'unmodified' that get tossed around quite liberally, especially when the TUA is kicking around. It's kinda used as "payment value before any of the recursive transactions kick in" and TUA is never modified, but no definition for these terms has ever been given. Wanna go ahead and ground that one for us or just keep it floating out in space?

I felt bad about tossing barbs while pissed off on a tarmac waiting for bad weather in a city I hate to let me get the hell out of OK/TX, but coming back for yet another 'honest' shot with someone who isn't burns me again.

#### SedanChair posted:

Yeah, interesting that he understood after you clarified by asking "what's SC mean?"

Do all of Eripsa's fans have some kind of neurological damage? Have you spent the last ten years huffing gas?

I seem to recall Bernie being a literal schizophrenic, so he seems like a likely candidate to understand this.

 Who What Now Sep 10, 2006 by Azathoth Jawn I just want to say that you go above and beyond what Eprisa deserves or even acknowledges. You're a good guy. # ? Apr 7, 2014 00:18

#### Who What Now posted:

Jawn I just want to say that you go above and beyond what Eprisa deserves or even acknowledges. You're a good guy.

I understand the motivation. Learning how to work with clients who overestimate their own abilities is a good skill. So is being able to pick apart a bad and incomplete argument. If you look at the essays they give you to analyze for the MCAT, they're either bad arguments, or good arguments that have had big sections removed.

 Vincent Van Goatse Nov 8, 2006 Enjoy every sandwich. Smellrose Jawn, even if you can manage to get a workable spec from Eripsa, do you think Strangecoin will ever be a useful model of anything at all besides Cargo Cultism? # ? Apr 7, 2014 04:21
 Lehrer Mar 6, 2004 I know that there are people who do not love their fellow man, and I hate people like that! I didn't really read this, but are you really equating google glass with black people? Really? Regardless of anything else, that is absurd. # ? Apr 7, 2014 04:22

#### ALL-PRO SEXMAN posted:

Jawn, even if you can manage to get a workable spec from Eripsa, do you think Strangecoin will ever be a useful model of anything at all besides Cargo Cultism?

A working strangecoin implementation in code would make a pretty fun addition to an indie game, I think.

#### Lehrer posted:

I didn't really read this, but are you really equating google glass with black people? Really? Regardless of anything else, that is absurd.

You just don't have enough knowledge about the philosophy of science

#### ALL-PRO SEXMAN posted:

Jawn, even if you can manage to get a workable spec from Eripsa, do you think Strangecoin will ever be a useful model of anything at all besides Cargo Cultism?

I feel like you could run a real monetary simulation and see what using the new Strangecoin monitary axioms do to a simulated economy vs regular money. All edivence points to Strangecoin loving literally every part of an economy up, but it could be fun.

#### Dusseldorf posted:

I feel like you could run a real monetary simulation and see what using the new Strangecoin monitary axioms do to a simulated economy vs regular money.
How, though? You could define some arbitrary goals, such as:
• "an individual shall maximize his throughput" [fitness coefficient 0.6]
• "an individual shall strive to minimze his Δbalance and keep it under 5% of his total throughput" [fitness coefficient 0.3]
• "an individual shall grow his economic network and link to as many users as possible" [fitness coefficient 0.1]
and then create some rules for inter-agent discovery and transaction negotiation. You'd then run a sandbox economy which periodically resets itself, does some genetic recombination on the most successful agents, and proceeds through another round of simulation. However, the results would almost certainly confirm the criticisms already raised: "money can be wished into existence", "it is infeasible to distinguish a sufficiently-complex money-laundering conspiracy from a legitimate trading network", "everybody presses the 'Defect' button as hard and often as possible", "High-Frequency Trading algorithms (and/or Twitter bots) will outcompete any human-like economic agent", "TUA stability is a hopeless dream." The one interesting outcome I can foresee is that the most successful agents might rely on relationship patterns which are unfamiliar to humans (e.g. because they show no clear evidence of either dominance or trust). But it would be very difficult to draw out any useful conclusions for real-world phenomena - you might as well speculate about the psychology of a glider.

Strangecoin is not, in itself, a complete model of human society. It's supposed to exist alongside actual human relationships (e.g. kinship, locality, affinity, community), which would define the shape of the initial economic networks. The customs and mores of those groups would help to identify and restrain any obviously-abusive behaviour, while gradually (or spontaneously?) evolving to accomodate the emergent economic behaviour of Strangecoin itself. Strangecoin does not satisfy all of a person's needs, and there's no guarantee that a Strangecoin-wealthy person will actually be happy (in fact, they might be consumed by the stress of managing their hyperactive economic web and maintaing their exalted position). The spec clearly indicates that each account must be tied to a unique human identity, notwithstanding the challenge of actually enforcing that rule.

If you try to strip away the actual social framework then you're left with a bunch of HFT bots competing against each other under a bizarre and (thus far) poorly-defined set of rules, until the system crashes because of an overflow or infinite loop (or one agent manages to gather all of the simulated money as "king of poo poo hill" and the simulation is unable to proceed). The simulation isn't going to reveal any great insights about economics (as it relates to human concerns and the distribution of scarce goods), and - in my opinion - wouldn't be worth the effort of implementation. Especially because it would take weeks to program the thing and Eripsa would rewrite the spec dozens of times during that interval

#### GulMadred posted:

If you try to strip away the actual social framework then you're left with a bunch of HFT bots competing against each other under a bizarre and (thus far) poorly-defined set of rules, until the system crashes because of an overflow or infinite loop (or one agent manages to gather all of the simulated money as "king of poo poo hill" and the simulation is unable to proceed). The simulation isn't going to reveal any great insights about economics (as it relates to human concerns and the distribution of scarce goods), and - in my opinion - wouldn't be worth the effort of implementation. Especially because it would take weeks to program the thing and Eripsa would rewrite the spec dozens of times during that interval

Well that would be a more realistic model of Strangecoin in the wild than some hypothetical world where everyone checks their Facebook to see how many likes they have before they can know how expensive coffee will be at Starbucks.

This was written:

#### RealityApologist posted:

The marble economy was meant to illustrate an attention economy, where units of attention were treated as discrete particles. Its more natural (but less memorable) to treat the phenomenon as waves, or more precisely to model behavior in field-theoretic terms. We'd ultimately like a master equation to describe the collective activity of agents. But we're not there yet; the ideas I was proposing were probably a decade ahead of the science. The cutting edge attempts at field-theoretic models of goal-directed behaviors are still dramatically underexplored.

A central planned economy is centrally planned. I'm explicitly arguing for distributed planning and self-organized coordination, which would be like using standardized algorithms as the "planning committee" and sampling directly from reddit, twitter, 4chan, et al. for feedback in training the system.

@eripsa I have a unique proof of the Master Equation which this tweet is too small to contain #fieldtheoretic

#### moebius2778 posted:

Should that be ΔBz=ezx*p at the end of step 4? Otherwise you're constantly overwriting ΔBy.

That's right, my mistake. I'm working on rewriting the original proposal so this is all clear upfront and I'll be more careful at keeping the notation consistent. I also left out the inhibition transaction.

I understand completely why Jawn and others are frustrated with me, and I can only say that I didn't know how to do this, and this is teaching me how. I did write up equations and think through them, but the notation wasn't standardized and a lot of the details (like OoO) were left implicit. It's my sincere intention to put in the work to improve it and fix the problems while minimizing overall frustration for everyone.

#### quote:

Also, if I'm reading this correctly, endorsement can create and destroy money - is this the intended behavior?

In the normal case, Endorsement and support shouldn't be creating money in the network overall; every coin drawn in the transaction is taken from other accounts (user Z) who is offering endorsement/support to X/Y in their exchange. If the user accounts are empty or full then they draw from or deposit to TUA. So for the overall network, these transactions are zero sum.

The coupling relation does create money. The way I did the calculation above, the trade between X and Y made the whole network .007 coin richer. This doesn't work properly yet; I'm pretty sure the way it is written now, the thing just blows up too fast and everyone hits their account caps. Again, I think the solution is to also give TUA an account cap much lower than the overall number of Scoin, and then have a distinct account that only TUA can trade with which has much more coin than TUA's cap. But I need to work this out. The goal is for the overall value of the network to stay roughly stable over many transactions.

The key to this is figuring out the coupling relations. It has something to do with Taylor series, but I need to learn more before I figure that out. I think it would be more productive for me to update the rest of the spec first, so if anyone can help with coupling I'd appreciate it.

#### quote:

Your description of δBx in the preamble (global change on Bx over the lifetime of the network) and in the description of how to process a transaction (change in Bx over the transaction/payment being processed) aren't consistent. I'm assuming the second description is the correct one, since you never use δBx in the global change sense.

Yes, this needs explanation. I did not introduce this to abandon the spec, I did it to make meeting the spec easier and more straightforward. But I need to make this explicit.

So yes, when X pays Y p coin, p is the unmodified transaction, and all the additions and subtractions laid out in the procedure are modifications to the actual exchanged coin in the transaction. The procedure lays out the steps for making these modifications.

δB is the global change as a rolling average (there's your answer Jawn) over the growth of the thread. If δBx > 0, the users has more income than expenses on average, and vice versa for expenses. This is to keep track of both overall income and expenses as stated in the spec. δBx is the value that the network considers to be that user's income/expenses at this timestep. δB is updated a few times in the procedure, because coupling modifies the results of support/endorsement, and coupling with TUA modifies the result of coupling, and so on, so steps 8, 11, and 14 I update δB to reflect these changes. Step 14 is the final update of the value for this procedure (at least, this version of it, I'm updating soon), and that value is carried on into the next timestep to calculate the next transaction.

ΔB is the purely local change, used to keep track of changes being made within the various steps of the transaction. This is just to keep the order of operation straight. ΔB is set to the value set to change by that step in the transaction, and then that balance is applied to the account by the code B += ΔB. This is to sort out any confusion about which values are being modified when. After the step where ΔB is applied to the balance, the value is reset and used again, and then discarded at the end of the procedure.

So to calculate the modification, some value of ΔB is calculated for any user expecting a balance change for that modification, and then all the user balances are updated to reflect that change.

The order of operations for this procedure is:

support
endorsement
inhibition
update δB
coupling
update δB
coupling with TUA
update δB

One more note: in the upcoming revision, I want to deal with the issue of generations of coupling. If X and Y are coupled and either X or Y's value changes as a result of the transaction, then both X and Y will modify each others balance by their coupling relation. If Y is also coupled with Z, these coupling relations must be calculated in some order. I'm thinking that this is resolved as follows, although suggestions are welcome.

First, assume they are ordered X, Y, and Z.

In timestep n at the coupling step, first calculate ΔB for X coupled with Y. Then calculate (the new value of) Y coupled with Z. (then the rest of the coupling in the network). Remember the difference caused by the coupling of Y and Z not accounted for by X.

In timestep n+1, perhaps before step 1 but maybe just before coupling, adjust all balances in the network to reflect the changes not accounted for at n, as an upkeep step. Then proceed as normal.

But this might still not make sense. I think coupling is still broken. I could use help.

RealityApologist fucked around with this message at 08:49 on Apr 7, 2014

#### GulMadred posted:

Strangecoin is not, in itself, a complete model of human society. It's supposed to exist alongside actual human relationships (e.g. kinship, locality, affinity, community), which would define the shape of the initial economic networks. The customs and mores of those groups would help to identify and restrain any obviously-abusive behaviour, while gradually (or spontaneously?) evolving to accomodate the emergent economic behaviour of Strangecoin itself. Strangecoin does not satisfy all of a person's needs, and there's no guarantee that a Strangecoin-wealthy person will actually be happy (in fact, they might be consumed by the stress of managing their hyperactive economic web and maintaing their exalted position). The spec clearly indicates that each account must be tied to a unique human identity, notwithstanding the challenge of actually enforcing that rule.

If you try to strip away the actual social framework then you're left with a bunch of HFT bots competing against each other under a bizarre and (thus far) poorly-defined set of rules, until the system crashes because of an overflow or infinite loop (or one agent manages to gather all of the simulated money as "king of poo poo hill" and the simulation is unable to proceed). The simulation isn't going to reveal any great insights about economics (as it relates to human concerns and the distribution of scarce goods), and - in my opinion - wouldn't be worth the effort of implementation. Especially because it would take weeks to program the thing and Eripsa would rewrite the spec dozens of times during that interval

The first paragraph is the first time in probably three threads that I've read a genuinely constructive and insightful elaboration on the nature of the network I'm trying to describe, whether or not I've been successful. I appreciate that.

I think the account needs to be attached to a consistent identity, but I don't think there needs to be any serious worry about it being strictly one person:one account, especially if the implementation allows the script kiddies to set up botnets and play among themselves without doing any real damage to the network. Some people might find it useful to operate disjoint nodes in the network for any number of reasons, and can handle the extra cognitive load it requires without difficulty. I think the limit is cognitive load: maintaining multiple high-throughput nodes isn't something a person could do on their own.

But I'm going to shut up about speculating on what the thing will do now that I'm actually doing the work to build it so we don't have to speculate.

 JawnV6 Jul 4, 2004 So hot ... Okie dokie. The simulator's getting beefed up. On my last flight I forgot to grab the coupling description, but I wanted to start to build the abstraction layer above the simulator itself. I'm trying to comment the functions as I go now, but the gist is that a list of users is randomly populated with balances. Each step, a new transaction is randomly generated. Two users are pulled from the list. For payments, the amount is selected from (1,balance-1), for Support and Endorsement the proportion is between (.05,.15). Fixed duration of 5 steps for those. I cleaned up the output so it generates a .csv. I didn't have the matplotlib docs handy, so I had to kick things over to Excel to actually chart, but .csv should be easily consumed on most people's systems. Automagically plotting the results would be really neat, but I doubt I'll get around to it. Here's a couple runs: It's a lot of data. Excel kinda chokes up on it, autochart has made me quite lazy (or good at generating chartable data when I need to?). There's an interesting side effect of how payment balances are selected. It's still implementing a lazy check_user function that only looks for negative balances. So, what's happening to kill the simulation (nearly always under 500 steps with 50 users and 10x inequality) is a support or endorsement transaction making a user go under. I'll admit I hadn't really thought of a user endorsing someone else and going broke from it until it happened on every run. This whole idea was sparked when Slanderer was starting to describe potential collusion scenarios. There was a bit of back and forth if it was possible, and I had a sneaking suspicion that the gap might not be in behavior, but in possible implementations. So if a sim was built, someone could prove their point by running it and posting the parameters. I also like the idea of using it to understand potential proportions. I kinda guessed Support should be in the 5~10% range, and that's what I coded. But you could potentially use it to figure out what a 'good' Support ratio is. Run simulations with 5%, 10%, 25%, 50%, 80%, etc. and watch which ones go off the rails quicker and which are more stable. "User balances" is a pretty poor charting solution. The only narrative I can really build on them is "1~5 people get a lot of money, most are in the middle, a bunch are really poor". Gosh, if you think handwaving is bad when it's just words, imagine the kind of plausible bullshit you can sell with pretty pictures to distract. I really love charts. So that's another question. What's a good chart to condense strangecoin activity? So far I've implemented pure balances as well as 'income/expenses', which is very sparse given the generation patterns. # ? Apr 7, 2014 08:43
 JawnV6 Jul 4, 2004 So hot ... Yeah, y'all got ahead of me and are already imagining a simulator with agents and real decisions. I can assure you, it goes off the rails quite easily on it's own without anyone trying to game it. I really don't see the reasoning behind collapsing income and expenses. They're quite different quantities, and collapsing it at each step is just hiding information from the future steps. If I make sure my payments in and payments out match, I get a big fat zero for the others. I'm not sure that's what I want, but I was told I want to be moving a lot of coins around. So I'm confused again. # ? Apr 7, 2014 08:51

#### RealityApologist posted:

That's right, my mistake.

Holy poo poo folks, it's a red letter day.

 CheesyDog Jul 4, 2007 by FactsAreUseless As soon as someone's Strangecoin rep (which will have to act as a metacurrency because the actual currency is unlimited and valueless) falls enough that people start refusing to trade with them, they'll either have to remain in-system and be exploited, circumvent Strangecoing through bartering or other exchanges, or both. You can give an impoverished person foodstamps now so that they can feed themselves. What happens in the Strangecoin economy, where TUA provides "universal welfare" but no one will trade with them? # ? Apr 7, 2014 13:31
This may be of interest to people in this thread:

Do Bitcoins make the world go round? On the dynamics of competing crypto-currencies
http://arxiv.org/pdf/1403.6378v1.pdf

#### quote:

Bitcoins have emerged as a possible competitor to usual currencies, but other crypto-currencies
have likewise appeared as competitors to the Bitcoin currency. The expanding market of crypto-
currencies now involves capital equivalent to 10^10 US Dollars, providing academia with an unusual
opportunity to study the emergence of value. Here we show that the Bitcoin currency in itself is
not special, but may rather be understood as the contemporary dominating crypto-currency that
may well be replaced by other currencies. We suggest that perception of value in a social system
is generated by a voter-like dynamics, where fashions form and disperse even in the case where
information is only exchanged on a pairwise basis between agents.

#### CheesyDog posted:

As soon as someone's Strangecoin rep (which will have to act as a metacurrency because the actual currency is unlimited and valueless) falls enough that people start refusing to trade with them, they'll either have to remain in-system and be exploited, circumvent Strangecoing through bartering or other exchanges, or both.

You can give an impoverished person foodstamps now so that they can feed themselves. What happens in the Strangecoin economy, where TUA provides "universal welfare" but no one will trade with them?

They trade with them anyway because why the gently caress wouldn't you, it's just that it will be on the barter system, in the black market or at a 20% discount or all of the above. Also the black market would quickly overwhelm the regular "economy", using that term loosely, as people quickly realize that they don't really *want* to calculate the exact benefit of coupling with the guy who brings them their coffee or the guy who inspected their sewer pipe prior to every single time they make a trade for something and other people realize they can launder every transaction made by the -rep crew and become billionaires until they themselves fall into -rep 200 billion richer and recruit the next guy to take their place ad infinitum. At some point before -rep becomes standard and total societal collapse because, again, why the gently caress wouldn't it, you can rest assured I will run bartertown on my throne of starving orphan skulls.

Adar fucked around with this message at 18:04 on Apr 7, 2014

#### RealityApologist posted:

This may be of interest to people in this thread:

Do Bitcoins make the world go round? On the dynamics of competing crypto-currencies
http://arxiv.org/pdf/1403.6378v1.pdf

#### quote:

Bitcoins have emerged as a possible competitor to usual currencies,

No, Bitcoin is not, has ever been, nor will ever be, a possible competitor of usual currencies. I highly doubt the article says anything credible if it starts with that sentence.

#### Who What Now posted:

No, Bitcoin is not, has ever been, not will ever be, a possible competitor of usual currencies. I highly doubt the article says anything credible if it starts with that sentence.

Eripsa didn't read the article. It ends with:

#### quote:

Overall, our consideration serves to emphasize the crypto-currency as a good model-system for the study of human folly...

#### RealityApologist posted:

This may be of interest to people in this thread:

Do Bitcoins make the world go round? On the dynamics of competing crypto-currencies
http://arxiv.org/pdf/1403.6378v1.pdf

I mean,yeah, it's interesting. But only in the way that the freak show at a circus is interesting.

#### quote:

Overall, our consideration serves to emphasize the crypto-currency as a good model-system for the study of human folly...

So what you're saying is, it fits right in this thread?

#### Adar posted:

They trade with them anyway because why the gently caress wouldn't you, it's just that it will be on the barter system, in the black market or at a 20% discount or all of the above. Also the black market would quickly overwhelm the regular "economy", using that term loosely, as people quickly realize that they don't really *want* to calculate the exact benefit of coupling with the guy who brings them their coffee or the guy who inspected their sewer pipe prior to every single time they make a trade for something and other people realize they can launder every transaction made by the -rep crew and become billionaires until they themselves fall into -rep 200 billion richer and recruit the next guy to take their place ad infinitum. At some point before -rep becomes standard and total societal collapse because, again, why the gently caress wouldn't it, you can rest assured I will run bartertown on my throne of starving orphan skulls.

Well good to know I didn't take one sleeve of my leather jacket for nothing! Now to find a bitchin' muscle car and a sawed-off.

#### Obdicut posted:

Eripsa didn't read the article. It ends with:

"Listen, if we ignore the last line then it makes some really great points that I just don't think you guys are mature enough to consider. "

#### Who What Now posted:

"Listen, if we ignore the last line then it makes some really great points that I just don't think you guys are mature enough to consider. "

There's more beyond it, that's why I put the ellipses in, but the ending is basically "cryptocurrencies are good at modeling the way that people are loving insane about the way they assign value to things, even though assigning value is a really important part of social order. Cryptocurriences model how much people suck at this."

So basically the exact opposite of Erpisa saying "If people value furry porn, then resources should be diverted to furry porn".

Looking up S. Bornholdt's publications, it kind of seems like he's the dude that Eripsa sort of wants to be. He's a theoretical physicist applying complexity theory to various domains. The problem is, from his publication record, his main theme is that emergent markets suck and are awful.

http://www.researchgate.net/publication/259268464_Spontaneous_centralization_of_control_in_a_network_of_company_ownerships

http://www.researchgate.net/publication/51990232_No_need_for_conspiracy_Self-organized_cartel_formation_in_a_modifiedtrust_game

#### Obdicut posted:

Eripsa didn't read the article. It ends with:

I read the article. I never suggested altcurrencies are plausible candidates for economic revolution, and said plenty of times that I have no interest in using them as such.

But I'm terrifically interested in using the cryptocurrencies to study human organization (folly and all), which is precisely what this article was about.

By the way, I'll be rewriting the proposal here if you want to help or harass me while I do:

http://meetingwords.com/DXFWsQK3ZH

#### Obdicut posted:

Looking up S. Bornholdt's publications, it kind of seems like he's the dude that Eripsa sort of wants to be. He's a theoretical physicist applying complexity theory to various domains. The problem is, from his publication record, his main theme is that emergent markets suck and are awful.

I "don't want to be this guy", although I'd sure like to learn enough math so I understand this stuff better. But I entirely agree with his position. I've pointed to other articles and books in this thread coming at economics from basically the same perspective which I also find compelling in order to give the thread a sense of my perspective. For some reason you find this sufficient reason to attack me again and I'm not sure why. I'm not a libertarian and have never defended the "free market" or any of that Hayek bullshit. I'm not saying an emergent market will bring utopia. I'm trying to study the dynamics by which they emerge in the first place.

edit: GulMadred made clear that the Strangecoin network depends on antecedent social ties that motivate developing the network in certain ways (parents supporting children or whatever). I'm making that structure explicit so we can talk about ways of managing that network, which is entirely the opposite of freemarketism.

RealityApologist fucked around with this message at 18:46 on Apr 7, 2014

#### RealityApologist posted:

I read the article. I never suggested altcurrencies are plausible candidates for economic revolution, and said plenty of times that I have no interest in using them as such.

But I'm terrifically interested in using the cryptocurrencies to study human organization (folly and all), which is precisely what this article was about.

By the way, I'll be rewriting the proposal here if you want to help or harass me while I do:

http://meetingwords.com/DXFWsQK3ZH

This guy, in his papers, is finding that there are relatively simple explanations for why emergent markets suck. It isn't a highly complex thing, it's explainable by aspects of the markets themselves.

In addition, in your OP, you say:

#### quote:

The result is a model of the complex interdependencies within a community of economic agents, and the dynamics by which those networks develop and decay. For this reason, Strangecoin might have implications for quantifying the role of individual choices and responsibility in the context of corporate action, and for resolving other difficult issues in the management and ethics of collective economic action.

...

#### quote:

Users with large throughput also occupy central hubs in the network, and their role and influence in the economic activity of the network is given quantifiable significance with Strangecoin, which can be used for making high-level policy decisions concerning the management and ethics of collective economic action.

With these two, the paper you cited is opposed to you because it's saying cryptocoins are useless for making policy about the management and ethics of collective economic action, except to the point that it says "Never use a cryptocurrency".

In your OP, you also say:

#### quote:

In general, Strangecoin users seek out semi-persistent networks of support and coupling-- similar in some ways to investment, but investment in people-- and the overall stability of these networks account for the economic well-being of both users and the network as a whole. The incentive towards cooperative, stable, interdependent economic relationships takes place through an interface that bears an intuitive relation to traditional currencies, especially as it pertains to marketplaces of competition.

I wanted to highlight that sentence because it's an affirmative prediction. You are saying something about the way that people in your model operate before the model even really exists. Again, this is cargo-cult behavior on your part. You could say you hypothesize that users would seek out semi-persistent networks, and that there would be an incentive towards cooperative, stable, interdependent economic relationships, but instead you state it as a fact. You are saying that your model does something before it is tested, and, really, before it even exists. That you don't even appreciate why this is terrible is the sad part. But, in so far as your system is a cryptocurrency (which it really isn't, another aspect of your cargo cultishness is just ramming the *coin appellation in there) the paper you just cited is arguing against crytpocurrencies being stable or cooperative.

Finally, in your OP, you say:

#### quote:

There's more work required to put flesh on the bones of this idea. There are many conversations to have about the dynamics of the network and the particular constraints it should operate within, both as a matter of practical operation and as a matter of sensible public policy.

You are clearly, from the above paragraphs and this one, proposing strangecoin as public policy. Pretending it is just a model for folly is silly. In so far as this policy is a 'development' of your attention economy weirdness, you fit what this guy is criticizing: the assessment of value not through any rational or sensible means.

I am not going to 'harass' you there because there is no need. You are self-defeating, as your flailing around in talking to Jawn has shown.

Please consider the following course of action: Throw away literally everything you have done so far. Literally everything. Consider it all wrong, all to be discarded. Start humbly, by reading some basic texts on economics and human behavior. Then, after you actually get those basic concepts, reread that paper you just cited, read that other guy's papers. He actually does nifty stuff, in that he shows repeatedly that complex (destructive) behavior in a system can generally be traced back to a simple interaction of the rules of the system itself. He uses tools responsibly to look at small examples, in order to make larger points. It is the opposite of what you are currently doing, and it is why what he is doing is useful.

#### RealityApologist posted:

I "don't want to be this guy", although I'd sure like to learn enough math so I understand this stuff better. But I entirely agree with his position.

Saying that you don't understand it but that you agree with him is weird.

#### Obdicut posted:

With these two, the paper you cited is opposed to you because it's saying cryptocoins are useless for making policy about the management and ethics of collective economic action, except to the point that it says "Never use a cryptocurrency".

The article I linked says:

#### quote:

A consideration that should be at the heart of multiple aspects of social activity, social hierarchies, and thereby also be part of maintaining overall social order.

The author's position is that considerations about the "the history-dependent randomness in assigning what is valuable and what has no value" should be a fundamental consideration in the maintaining of social order and is necessary for considering aspects of social activity.

He's not saying this is useless at all, he's saying it's a fundamentally important consideration for social organization. I've not said anything that contradicts this, and everything I've said has been in virtually the same spirit as this paper, albeit without the technical competence.

I think you're making a mistake. That's okay, buddy. I won't hold it against you.

#### quote:

I wanted to highlight that sentence because it's an affirmative prediction. You are saying something about the way that people in your model operate before the model even really exists. Again, this is cargo-cult behavior on your part. You could say you hypothesize that users would seek out semi-persistent networks, and that there would be an incentive towards cooperative, stable, interdependent economic relationships, but instead you state it as a fact. You are saying that your model does something before it is tested, and, really, before it even exists. That you don't even appreciate why this is terrible is the sad part. But, in so far as your system is a cryptocurrency (which it really isn't, another aspect of your cargo cultishness is just ramming the *coin appellation in there) the paper you just cited is arguing against crytpocurrencies being stable or cooperative.

No, it's a design proposal. I'm not saying "I've built this thing that already works this way", I'm saying "I'd like to build this thing and its supposed to work this way." So, you know, others who might be interested in helping know what I'm trying to do and can correct the technical errors.

That's not cargo cult mentality, that's seeking help as an amateur enthusiast. Seriously, Obdicut, I know you're having fun on your high horse and all, but if everyone had your attitude then no one would take the risk of learning anything. You're just a bully. If you seriously want to square off in a hangout about academic integrity and intellectual ethics, bring it on. I have no doubt I can take you.

#### quote:

Saying that you don't understand it but that you agree with him is weird.

I understand some things, which I agree with. I don't understand the full thing in its technical detail, because I don't have the requisite knowledge. Neither do you, but you seem to feel perfectly comfortable interpreting it for this thread, which is all I'm doing.

It's not weird for lay enthusiasts to provide informed interpretation of academic research. There's not a loving goddamn thing weird about it except that you want to pick on me and you're looking for reasons to do it. I'm not complaining about it, but I'm not going to indulge it either. I'm calling it out so that you are aware of it so you can stop doing it. Please.

#### Obdicut posted:

Please consider the following course of action: Throw away literally everything you have done so far. Literally everything.

If you look at the proposal I'm working on now, that's literally what I've done.

I'm writing a new proposal, from scratch, with my notes on the procedure I wrote up earlier, and Jawn's code in another tab. I don't know python and I haven't coded anything for a decade, but I know enough about coding that I can read what's going on and what commitments he's made, and I'm writing the new proposal in such a way that he'll have to make little (and hopefully no) basic changes to his framework.

I'd like to be done by the end of the day because I have to catch a plane early tomorrow.

#### RealityApologist posted:

The article I linked says:

The author's position is that considerations about the "the history-dependent randomness in assigning what is valuable and what has no value" should be a fundamental consideration in the maintaining of social order and is necessary for considering aspects of social activity.

Considerations should be a consideration. Yeah. But what that means is: "People assign value randomly and horribly. Therefore, in any system where you let value emerge, it gets assigned horribly because of social stuff". Strangecoin, and above that the 'attention economy', is a system where you let value emerge. So what tha paper predicts about Strangecoin is that it will be hosed up and assign hosed up values to poo poo.

#### quote:

I think you're making a mistake. That's okay, buddy. I won't hold it against you.

Trying to do this is sad.

#### quote:

No, it's a design proposal. I'm not saying "I've built this thing that already works this way", I'm saying "I'd like to build this thing and its supposed to work this way." So, you know, others who might be interested in helping know what I'm trying to do and can correct the technical errors.

You still aren't getting it. You are claiming strangecoin is a model, and you are making predictions--not hypotheses, but predictions--about how that model will act. The bit about 'coupling', for example, is a prediction about a function that exists in your model. You don't recognize that doing this is putting the cart before the horse.

You are also admitting that strangecoin is not a model, because it doesn't even exist. What you mean is that you want a model that will have all the attributes you listed, but you have no clue how to design it and you want other people to do it for you, even though (you now claim) this model would not actually be a better way of doing things than our current system. Even then, for some bizarre reason, you've decided that you do know how to start the model off and flail around with random equations. You think that this stuff with 'coupling' and 'endorsement' and the rest of it is the way to go, that that will produce your desired results, and you don't accept everyone saying "Yeah, those things won't do that."

#### quote:

Seriously, Obdicut, I know you're having fun on your high horse and all, but if everyone had your attitude then no one would take the risk of learning anything. You're just a bully. If you seriously want to square off in a hangout about academic integrity and intellectual ethics, bring it on. I have no doubt I can take you.

For christ's sake. My main point to you is that you should bother to learn something before you just charge ahead babbling about it. Go learn something: that's my main thesis to you. Go learn. Stop trying to do poo poo you have no clue how to do and go learn. Go read stuff and learn from it.

Also, this may come as a surprise to you, but issues of academic integrity and intellectual ethics aren't generally resolved by 'squaring off' in a hangout.

#### quote:

I understand some things, which I agree with. I don't understand the full thing in its technical detail, because I don't have the requisite knowledge. Neither do you, but you seem to feel perfectly comfortable interpreting it for this thread, which is all I'm doing.

Authors write these handy 'conclusion' sections which allow people with non-technical backgrounds to interpret what they say. It's nifty! I'm pointing out that you didn't interpret that conclusion section correctly.

#### quote:

It's not weird for lay enthusiasts to provide informed interpretation of academic research. There's not a loving goddamn thing weird about it except that you want to pick on me and you're looking for reasons to do it. I'm not complaining about it, but I'm not going to indulge it either. I'm calling it out so that you are aware of it so you can stop doing it. Please.

I'm not going to stop calling you out on your bullshit, Eripsa, or on your weird habit of 'supporting' your argument with irrelevancies and appeals to authority. And again, if you want to actually engage seriously in the academic world, you should really, really be open to your stuff being criticized, you should embrace and welcome it, rather than ranting about bullies and how we're all morons and assholes.

#### RealityApologist posted:

If you look at the proposal I'm working on now, that's literally what I've done.

I'm writing a new proposal, from scratch, with my notes on the procedure I wrote up earlier, and Jawn's code in another tab. I don't know python and I haven't coded anything for a decade, but I know enough about coding that I can read what's going on and what commitments he's made, and I'm writing the new proposal in such a way that he'll have to make little (and hopefully no) basic changes to his framework.

I'd like to be done by the end of the day because I have to catch a plane early tomorrow.

I meant throw away the idea of doing this poo poo in the first place, dude. You need to learn the basics of economics first. At a minimum.

And a quick look at your proposal shows the same 'support', 'endorsement', 'TUA', and the rest of that, so yeah, you haven't 'literally' thrown away everything, either. I don't know why you think you have.

Obdicut fucked around with this message at 19:26 on Apr 7, 2014

#### RealityApologist posted:

The author's position is that considerations about the "the history-dependent randomness in assigning what is valuable and what has no value" should be a fundamental consideration in the maintaining of social order and is necessary for considering aspects of social activity.

He's not saying this is useless at all, he's saying it's a fundamentally important consideration for social organization. I've not said anything that contradicts this, and everything I've said has been in virtually the same spirit as this paper, albeit without the technical competence.

Eripsa, that's not exactly a groundbreaking thing to say. They mention it in the paper exactly because it is obvious. It's obvious to people with the technical background in economics, hell in anthropology, in a million other disciplines. In living life it's obvious, that's the spirit in which they state it. And it's apparently new to you, and for either that reason or because it uses terms substantively that you use as buzzwords, you have picked it out because it's colorful and woven it into your nest.

e:

#### quote:

No, it's a design proposal. I'm not saying "I've built this thing that already works this way", I'm saying "I'd like to build this thing and its supposed to work this way." So, you know, others who might be interested in helping know what I'm trying to do and can correct the technical errors.

In the OP you heavily implied that you had run your own equations to the point that they behaved in the way your narrative stated. Jawn plugged in your numbers and even made adjustments when you admitted certain things hadn't been accounted for; until you finally admitted that nothing had been accounted for, you ran nothing. If you were holding yourself to academic standards, that'd be fraud of course. But you're not holding yourself accountable to those standards, you've said. You only hold yourself to the standards of the "mean streets of the internet," the mean streets, incidentally, that according to your vision will dynamically and adaptively self-organize to provide a better future. But I guess your contribution to it is to pretend and lie and act like you've done math when you haven't.

woke wedding drone fucked around with this message at 19:32 on Apr 7, 2014

#### RealityApologist posted:

Seriously, Obdicut, I know you're having fun on your high horse and all, but if everyone had your attitude then no one would take the risk of learning anything.

I'm just addressing this one sentence because both I feel it's a good indication of your self-harmful beliefs and it's total bullshit. Obdicut can handle your post better than I, but this one thing jumped out at me as especially egregious.

If everyone had Obdicut's attitude then people would want to learn as much as they could so as to minimize the risk of looking like a complete moron. Or, to put it another way, the majority of people in higher education already think like Obdicut does, otherwise they wouldn't even be there at all!

See, you want to be lauded and patted on the back for being a forward thinking man without actually doing any of the work yourself to see your ideas become a reality. You just want to make sweeping proposals that will "change the world" and then poof! have it appear before you on a silver platter. And when people tell you how you could very easily go out and do the things your asking for yourself you lash out at them, because you think why should you? And when people tell you how to at least approach others' help in a manner that would be more conducive towards actually getting the help and attention you so desperately want, you lash out again because you don't understand why someone wouldn't immediately see your genius and want to help. And so we become bullies and bigots, children, too-haughty intellectuals, and more in an effort to write our criticisms away because you find them too difficult to contemplate, much less execute.

Obdicut has given you the best advice of anyone in this thread: scrap this whole thing. You're too emotionally invested in it for it to go anywhere. You refuse to change any part of it in a meaningful way. You're fighting tooth and nail because you so desperately want to believe that if we just looked at it from this angle instead of that one we'll see your true genius shine through. Your ideas are stagnant and you refuse to refresh them, and it's no one's fault but your own.

If you'd have listened to people from the start of this thread, StrangeCoin would be almost unrecognizable in a good way. Instead it hasn't changed in the slightest, and it suffers for it. All you have to show for any of this is a monument to your own hubris and idiocy, and you can't even see it.

#### Obdicut posted:

Considerations should be a consideration. Yeah. But what that means is: "People assign value randomly and horribly. Therefore, in any system where you let value emerge, it gets assigned horribly because of social stuff". Strangecoin, and above that the 'attention economy', is a system where you let value emerge. So what tha paper predicts about Strangecoin is that it will be hosed up and assign hosed up values to poo poo.

You are completely misunderstanding both the paper and my proposal.

The author is saying "people assign value in random, history-dependent ways, and this is a fundamental consideration in understanding the development of social value". Their article proposes a model by which people trade between various alt coins as a predictive model of the way their assignment of value develops.

Obdicut, you're interpreting the author as claiming that agents assign value horribly, and you are concluding from this that the author is arguing that it's a waste of time to study different networks of transactions by which they assign value.

But the authors are producing a model of transactions in a network very much like the strangecoin network, and using it to model the way agents assign value in the network. Far from thinking it is a waste of time or a fool's project, they think studying these kinds of (random, history-dependent) transactions is a fundamental part of studying social activity and human socioeconomic organization.

I'm proposing another network of transactions very much like the network proposed in the paper. I'm studying different transactions, modelled on various kinds of social support and dependence that might occur in a community. It's a different network, but its designed in the same spirit as the network in the paper. I'll just point out that the arxiv paper isn't a paper in economics, it's a paper studying multiagent systems. It's basically a voting process. It's a very simple model, much simpler than Strangecoin. But it's really not doing anything different than what I'm doing. The author's conclusion is that it's important to understand these kind of dynamics and the way they influence the assignment of value. Everything about strangecoin is in alignment with this conclusion.

I'm pointing this out in detail because you're arguing explicitly that the author's conclusion is precisely contrary to the model I've presented here, and that's simply a bad interpretation of the paper, and of the work I'm doing. It's simply false. You are wrong. I am actually very familiar with this literature, which is not economics literature, and again while I'm an amateur I'm quite sure I'm interpreting the conclusion correctly. It's not a trivial conclusion; again it's exactly in line with the claims I've been making in this thread.

 Adbot ADBOT LOVES YOU # ? Aug 3, 2024 00:53

#### Who What Now posted:

Obdicut has given you the best advice of anyone in this thread: scrap this whole thing.

Your opinion has been noted.