zlemflolia

First off I want to reiterate that this thread is not about "making money via cryptocurrencies" but instead the social and economic implications of them as technologies. Now that that is out of the way... All altcoin prices are highly tied to Bitcoin's price. When BItcoin crashes everything crashes basically, and there is generally a correlation between Bitcoin pumps and altcoin pumps, though random altcoins routinely make much higher percent gains than Bitcoin. It is speculative, but it's a sound speculation -it's not just some shot in the dark or a dart throw. First off, there is a limited supply of Bitcoin. There are only 21m of them. And the population of the world is growing along with the user base of cryptocurrencies. Therefore the amount of Bitcoins one person can reasonably be certain of acquiring lowers every day. People have been calling it a bubble since before it was $1, and they were always just as confident as they are now in its uselessness, it not being a "real" currency, it being a failure, etc. https://99bitcoins.com/bitcoinobituaries/ Don't invest in anything based on some random autist's advice on the internet - do your research. But if your investment portfolio does not involve some cryptocurrency, you are missing out on the most important financial invention since written language, this is no exaggeration. If you're poor and can't afford to lose money then don't do it or only invest a little. But if you have some money in the stock market already, I suggest you say "fuck it" and put a few thousand into cryptocurrencies, mainly Bitcoin (BTC), Ethereum (ETH), and Monero (XMR). You may think such grandiose language is too good to be true, but it's not - we're merely experiencing what insider investors have been experiencing for decades - being able to invest in a sound asset before the general public and financial institutions become heavily entrenched in it, because they are involved in the centralized institutions associated with making such investments. The internet removes that reliance upon centralized authorities. Make no mistake Wall Street is highly involved, that's for sure. But there are no cryptocurrency commodity ETFs yet. That is when the boomer 401k money comes in and that is when we see 100x gains overnight

It's kind of like the ultimate philosopher's stone time paradox of game theory and economics in that once it's invented it will always dominate because decentralization is such a superior system because of the massive profit-taking inefficiencies in society it's capable of exploiting and therefore eliminating, making us more efficient and free from centralized control. The only risk of these not transforming society is some sort of mass government oppression taking place like the dark ages, but in the end it will prevail because all that is needed is one copy of the whitepaper or one person who understands how to reimplement it, and even if they are all destroyed it WILL be invented again. It exists and is real, it will exist forever, this is a way out of the matrix of greedy despots and rulers

Also it's the first 100% literally completely uncounterfeitable object that humanity has created, in this case as money. Sounds farfetched but it's true. It just ironically isn't even an object in the traditional sense, and exists everywhere and nowhere at the same time, and if 99.999% of blockchain copies are destroyed, all you need is one of them to restore the entire network because you can verify that the copy is real independently of any authoritative source through mathematics (backwards rehashing the chain and checking that the hashes match up)

It looks like that initially but you're ignoring how much electricity the current financial services and banking industry uses to build its buildings, keep its lights on, pay its employees, and things like that. Cryptocurrencies are aiming to eliminate all of it, yes, all of it completely. When you take that into account the energy expenditure is worth it, and on top of that this also removes the inefficiencies inherent to the way those existing institutions work, and removes the ability for centralized corruption, and the gains society makes on that can easily be invested back into the environment Ignoring all of that completely though, there are still other cryptocurrency implementation methods that are far more environmentally friendly and we are luckily moving closer towards them as I said in the OP

You need an understanding of intro computer science, distributed networks, cryptography, economics, game theory, and monetary policy to understand cryptocurrencies fully right now. A basic explanation is this We're all trapped on an island We have a piece of paper where we write down how much money everyone has You can only append new changes to the end i.e. "Bob gave Joyrex $500" How do you make it so nobody can write fake entries for themselves or other people, and anyone can write entries whenever they want? Satoshi solved that through distributed proof of work consensus

State issued currency - central control, they issue as much as they want. In places like the US they have a decent currency, the USD which only loses 3% in value yearly. In other places like Venezuela they have 100000% yearly inflation meaning their currency is absolute garbage compared to even the biggest shittiest cryptocurrency, quite literally Fiat = printed by government Banks = fractional reserve lending = multiplication of money supply which forms a feedback loop since some lent funds result in deposits into the bank AGAIN which they can then lend out fractionally. Fractional reserve banking is a massive bubble - deflationary cryptocurrencies eliminate the possibility of this happening since you can't create currency out of thin air to lend to people

It's based on your ability to take part in the global, uncensorable, pseudonymous, indestructible money transfer protocol that Bitcoin is. Its monetary value is based on demand to enter such a highly useful network. It is a deflationary currency where only 21m bitcoins will ever be created no matter what, as opposed to fiat currencies which are continuously produced and degraded in value every year. Nicehash was likely an inside job. I personally was monetarily affected by a large amount but it's not a serious issue - it's still the wild west right now, once things get more stable and we have user friendly security methods this won't be as common

XRP is a garbage centralized shitcoin, he should not be holding it.

Replies like this are expected, this is what most people think at first. But literally everything I said is true.

In the same way that the internet brought free, uncensorable word-speech to the world, cryptocurrencies will soon bring free, uncensorable financial-speech to the world. This is the most important point. The internet let anyone say anything they want to anyone securely. It lets anyone receive any information they want and transmit any information they want. But before it was just text and raw data. There was no linkage to the real world i.e. physical objects Now that there is an unsensorable means of financial information communication, that linkage exists - a political prisoner can merely publish a cryptocurrency address and receive as much financial help as donators want to give, without anyone ever being capable of stopping that from happening. The same goes to any organization. This of course allows illegal things, but it's impossible to stop that by definition All new freedoms allow bad people to do new even worse things. That's the nature of freedom. But the benefits always outweigh the cost. Free speech allowed criminals to communicate secretly. Is that worth censoring it? No. Similarly free financial speech will allow the same, but the benefits for humanity vastly outweigh the drawbacks, especially since much of the "drawbacks" are merely defined to be drawbacks by those in power instead of being actual legitimate drawbacks (for instance drug illegality which makes no real sense and clearly isn't based on an interest in public health)

There are serious environmental concerns to cryptocurrencies but they only apply to Proof of Work (PoW) blockchains which require heavy computation to create. As described above, DAG based cryptocurrencies, and even in the future Proof of Stake (PoS) blockchains, will cut down on environmental costs significantly Furthermore and most important, the goal is to eliminate all centralized monetary corruption inherent to fiat currencies. If we have cryptocurrencies as widespread financial assets, anyone in any country can take part in a global economy. We don't need permission from any government, we don't need support, we don't need shit. The cat is out of the bag, it's too late. Never again will people be dependant upon predatory banking systems and the corrupt centralized fiat producers of the country they live in. This will eliminate so much market efficiency that we can eventually Eliminate all predatory money transfer fees Eliminate all predatory banking fees Eliminate all predatory purposeful-inflation economic value extraction perpetrated by centralized authorities Eliminate many financial service industries such as brokerage firms and investment firms. Your stocks can be cryptocurrency tokens whose ownership you can prove with your private keys instead of through an agreement that a centralized authority will just say you own the stocks. Dividends can then be distributed in the form of some other dividend distribution cryptocurrency automatically, programmatically, without any human intervention or service fees

Cryptocurrency discussion thread. Anyone heavily invested? Invested a little bit? Obsessed with it? Think it's a scam? It's most definitely not and once you understand the value it provides to society you will never go back Unseizable and invisible (brainwallet) Indestructible (all we need is ONE copy of the blockchain to survive and we can restart from there and know for a fact it's a real copy by seeing if the chain hashes backwards correctly, it's like a horcrux) Uncensorable (use Tor and Monero) Sound money, can implement any monetary policy trivially, such as deflationary i.e. holding gives you gains by definition Decentralized, can't be controlled and manipulated by individuals at the expense of others (except 51% attack which is unfeasible for heavily decentralized coins) Cryptographically secure I can go on and on. If you know you know, if you don't I suggest you research. Don't miss out on the next stage of human social organization. Decentralized autonomous organizations (DAOs). Trustless smart contracts. Identityless store of value. This isn't even just about finance or currency. It's also not just about Bitcoin, that was just the first step. Bitcoin = first cryptocurrency, most battle tested, decentralized, and integrated with society Ethereum = first robust smart contract platform Monero = first robust anonymous coin It will continue to get even weirder. There are now coins that don't use blockchains in the traditional sense, but instead Directed Acyclid Graphs (DAGs) such as Byteball and Raiblocks, which can be visualized as many blockchains split off at arbitrary points and forks, but always moving forward and never looping backward. They aren't quite as robust yet but they will be soon. Who knows what else will come. Regardless of the technology, the most important impact will be in third world countries without sound monetary policies, and instead hyperinflated currencies ruled by dictators. Most of us don't see the need for cryptocurrencies because we live in first world utopias in comparison, but these people can be holding money one day and have it be worth 10x less the next day. They need access to the global economy free from their country's exploitative shit currencies, and cryptocurrencies give them that Ironically, if centralized fiat currencies were cryptocurrencies, they would be the type called "shitcoins" i.e. garbage coins pre-mined and controlled by a few people.

anyone who doesn't like pineapple pizza is basically just a loser

so a genetic algorithm would be stateful? http://boxcar2d.com/ different results guaranteed. :) It would still be stateless by my reckoning since it keeps no memory of past generations, only the current one. But that's state I guess, but not very useful state. I think my distinction doesn't work actually and needs more refining. There are many types of state, I would say the current generation of a genetic programming algorithm is analogous to the current state of the board, not the current state of the AI. There is nothing traditionally associated with AI in genetic programming apart from being a tool to search parameter spaces for useful outputs. Genetic programming is not in itself useful for decisions, it's useful for creating tools that can be used in the future, essentially a training method. The act of training the AI would be done offline ahead of actual usage and this is what genetic programming in the midst of occurring is, like this simulation

Fair enough, I'm by no means an expert here. My reading was that actual game data and therefore player data is not used by and therefore during actual gameplay no modifications to state occur either. And it's a matter of update refresh time really, I highly doubt Google translate is literally trained in realtime where if I make a query, training differences are propagated immediately to the point where the servers have to keep track of a realtime state of incoming queries. It would be batched and retrained, then new models would be propagated outwards through the network at varying time intervals. So it would be stateless still in this case. So this ML model refresh cycle could be analogous to the organic sentient being's flicker flash rate potentially

EDIT: Basically I'm using "state" here as the "personal memories" of the AI which influence future behavior. Take two identical AIs and put them in different situations and leave them for a few hours. Then bring them back together again. Are they different from each other? Then they are stateful. But most "AI" systems we have now would not be different, because they are simple stateless models being executed on a discrete time quanta worth of input data, maybe a single voice command from a user (siri) or a single web request (Google search) which is then later processed into new state and updated after a larger discrete time interval A self driving car keeps track of lots of external state and maps that external state to transitions in internal state, but it would depend on the implementation whether the cars continue to learn forever on their own, or whether the cars send their observational data back to the car manufacturer, which is then processed into new ML models which are then sent back to the car and executed statelessly Basically I'm using "state" here as the "personal memories" of the AI which influence future behavior. Take two identical AIs and put them in different situations and leave them for a few hours. Then bring them back together again. Are they different from each other? Then they are stateful. But most "AI" systems we have now would not be different, because they are simple stateless models being executed on a discrete time quanta worth of input data, maybe a single voice command from a user (siri) or a single web request (Google search) which is then later processed into new state and updated after a larger discrete time interval

I'm probably using "stateful" in a non-rigorous way. But for instance instance simple neural nets are stateless because there is no internal state to keep track of while executing them i.e. they can be executed while stored in const memory. There is no state to keep track of and each scoring is deterministic. If it's trained ahead of time once and then an immutable instance is executed from then on, it's stateless If it's trained ahead of time and also continues training itself during use, it's stateful AlphaGo Zero in this case seems stateless by these definitions because it was trained ahead of time using first principles, rather than using actual game data (and presumably therefore also not using realtime in-game data apart). Note the "state" of the board is not related to the state of the AI, it's simply an input and in this case it would be a stateless function of BoardA -> BoardB A self driving car keeps track of lots of external state and maps that external state to transitions in internal state, but it would depend on the implementation whether the cars continue to learn forever on their own, or whether the cars send their observational data back to the car manufacturer, which is then processed into new ML models which are then sent back to the car and executed statelessly

I don't think the Chinese room experiment encompasses this topic, but it is clearly related. I think the room is analogous to specialized, domain-specific AIs as opposed to general self aware AIs which would be a different class entirely, the type I think would "magically" obtain emergent sentience by virtue of their computational structure, if it was made correctly. The Chinese room is just a stateless function much like Chess AIs and I do not think those would qualify for sentience since sentience is inherently stateful in the moment and through memories, both short-term and long-term. Maybe I'm wrong though, maybe it isn't and there are many forms of sentience, and I'm sure you can make a stateful Chess AI (the latest ones may even be stateful) I guess it depends on how you define stateful. Does learning while playing count as newly added state? And if so does the function itself have a continuously self modifying state? It's quantized each turn but does that take away from the fact that it's dynamic? Waterbears can be frozen and have their speed of execution, so to speak, slowed down to a crawl, yet they still function afterwords. Are they sentient? I think it's a spectrum most likely not a binary condition, and the domain and range of this spectrum are emergent qualities of the universe itself as a result of certain configurations of, stuff. I feel like these questions should have answers that can be discovered through reasoning but it always ends up coming down to a matter of definitions themselves obscuring things. Or maybe I am just a brainlet. I think it will end up being that anything which appears sentient is, because there's no reasonable way for one sentient organism to be sentient, and another identical copy not be sentient, so consciousness would be then sort of a continuous field emerging as a result of certain topological computational structures. Okay I will stop.

Your link is unrelated to philosophical determinism and is really a misuse of the term and an abstract theoretical conception of computation that is dissimilar to practical applications. In reality these algorithms have to be implemented making use of either -pseudo-RNGs -heavily obfuscated and chaotic state such as that of disk activity and OS scheduling like the article references (and in general a lack of synchronization across individual parts, whether they be on one system or on a distributed system) with regards to the apparent non-determinism of concurrent computational systems. None of these are true non-determinism unless you want to talk about "approximate and apparent" non-determinism instead of hardline non-determinism. You may have a point though about introducing outside sources of randomness but even in that case I'm extremely skeptical about whether apparent external randomness inherently kills determinism. These random parameters are still sculpted and interpreted by the software itself and do not "break out" of the software in any way I think probably determinism vs. non-determinism is just a confusion of language a la Wittgenstein, causing people to mistake the map for the territory. As for the rest of your post, agreed that intelligence is ill defined

Be aware that there is much drummed-up fodder about the AI, with a goal to create steady and positive acceptance of the AI research. There is this unilateral force of opinion that the AI is a logical continuation of our development as a species, with all sorts of utopian idealism about how the AI would make our lives easier and streamlined. Everything can be replaced with an algorithm. Nothing is mystical, wondered at, and sacred anymore; everything can be calculated and replicated, over and over again; improved, personalized, accessible anytime you desire. No more longing, working to achieve, failing to achieve. I say force of opinion, as the recent scientific and rational radicalism is presenting itself as the only solution - a perfection, without the burdens of spiritual ways. A ghost in the machine as the allegory... the idea that sufficiently powerful computational machine is spontaneously capable of emerging as a sentient being. However, there is really no proof. It is an idea perpetuated to the level of a dogma. No one really knows. The singularity is a theoretical construct. In the most positive light I can see this as not understanding what a spirit is, what is it that makes us human. It has achieved a level of political propaganda. It needs to in order not to delay and revise the aspirations that lead to the want of creating the AI in the first place. The true "why" lies somewhere else. If it's going to make our lives easier and better and whatnot will be a mere byproduct. The primary "why" is to have power and supremacy. I say propaganda, because it's really just a play of words. Your feelings are "mere" electrical signals interpreted by your brains? Well of course, there must be some medium, but that doesn't mean that we are just a bunch of biological wires. Love is "just" high brain activity? Of course it is, if you look at it from only a point of being able to measure things with lab oscillators. But measuring is not understanding. Defining reality with such an approach is seeing with tinted glasses. What else is there? Well, because I don't understand, I can only say what I know. Brains are computers, so strong that they are now sentient. Just like that. Well, that's quite an unilateral idea, devoid of any other possibilities. Let me take one strong example from 2049 to prove one of my points: A scene where K meets the huge holographic ad for Joi. It is one of the definite examples of the practical use of the AI (among strong, subdued and efficient killing machines). It's basically a billboard, sensing your presence, evaluating your emotions and aspirations in order to access your very personal space. "She" is nude, beautifully rendered, speaks to you in a soothing voice, saying things in a manner your closest friend or lover might, with a tagline "I can be whoever you want". It's an advertiser's wet dream. This is the reality. 2049 was a horror film for me, as I said before. It depicts that world where everything is replaced with algorithms and computers. You can fall in love with a hologram and don't need genuine biological partner, because you are nothing more than a set of wires. To fall in love you only need a beautiful face, puppy eyes and attention (audition) of your particular way of life. Horrible. My read of your post though is "if you're right that would be sad and take away our value and our special place in the universe where our mutual biological emotions are valuable and spiritual" which is just the old "but if evolution is true then we aren't special anymore" argument for another topic. Take a look at dysfunctional relationships and mental illness and you can see that often even human relationships can be like two mutually "attracted" AIs each trying to maximize their value function in a way that doesn't manifest itself very wholesomely. I agree that this is a horror story, but horrible in a different way. I'd say the AIs created by Wallace corp are legitimately sentient but the ultimate slaves and proof of determinism. It's more obvious because their programming guides them towards certain actions, whereas our programming is more abstracted away from ourselves so we can't observe it easily, but it's there just as much.

Damnit guys... Wallace got her eye color right, but this was not some narrative mistake. Deckard lied and said they were wrong as a way to underhandedly attack Wallace and reject his implicit claims to superiority. Also interestingly, notice how Wallace looks actually surprised when Deckard says this and almost has a face showing that he's thinking "oh, damn" then he nods to Luv to kill her and immediately moves on to telling Deckard about how he can torture him to get what he wants, when his plan to flatter him fails Anyway, autism time: -This movie is great, it combines so many important philosophical themes that are, to me, the most important ones we have to answer. What does it mean to be a human? -Does it even mean anything special, or are self awareness, intelligence, and the other "higher concepts" like honor and sacrifice the only qualities which determine the value of an object? -Are artificially intelligent objects sentient? -If you feel and experience something, but it ends up being "fake" according to sterile human categorizations because the fundamentals upon which the abstractions are based can be explained away in some way (is human and replicant love just neurochemical reactions? Is AI love just NAND gates switching on and off?), is it still a real experience that the individual should value and cherish? -Regarding the last point, I haven't read this book yet but I bought it and will once my reading backlog is cleared but Nabokov's "Pale Fire" which is where the baseline phrases come from and which K keeps on his desk that Joi picked up and asked him if he wants to read, is supposedly about a guy who reads some story written by a neighbor which resonates with him on a spiritual level and he thinks they have some sort of connection, but it ends up being that the neighbor did not intend that all and it's just a big coincidence. Was his experience "real" and what does real even mean? -K is a supposedly soulless replicant who in the end lives an authentic life in the real world, acting just like an ideally heroic human would -Stelline (memory girl) was a supposedly soul-endowed born-replicant who lived a completely synthetic life, isolated from the outside world and interacting only with holograms except when visitors occasionally come to visit, who from what we see only use her to ask questions instead of actually be her friend -Deckard sees a "synthetic" (replicant) copy of Rachel and rejects her as "fake" -In the very next scene, K sees a "synthetic" copy of Joi and we don't get to see inside his mind. This symmetry between "copies" of organic beings and AI beings being shown to try and attack the value of the originals with which personal experiences were formed, and the applicability of the metaphor in both cases, I think destroys any lines which can be drawn between organic and AI. It's the same exact situation just on different levels of familiarity to us. My personal interpretations and conclusions on these topics themselves are as follows: I think sufficiently advanced AI is sentient, but we will have a difficult time as a society coming to grips with this or even accepting it as the reality. This entire area of epistemology is a really tough one since technically none of us, including myself, can actually even prove that other humans are sentient, or that anyone else exists, or that even if other people do exist, they aren't just p-zombies i.e. blobs of matter walking around with nobody inside observing the signals coming into their eyeballs. Consciousness is romanticized and "spiritualized" by people as a sort of new-age buzzword, but it's very clearly an inherent aspect of the universe itself - consciousness is real. I'm sentient, you're most likely probably sentient unless the aforementioned philosophical problems have very counterintuitive answers - and we're just created through sexual reproduction so there's nothing super weird happening here, it's clearly an emergent property of certain arrangements of matter. I would go further and say that consciousness is an emergent physical property of certain types of self-referential computational systems such as ourselves and animals, and of course, if this is the case, sufficiently advanced AI as well. We're learning more and more as a species that our language based categorizations are not binding upon nature itself but instead tracings we draw on top of it, trying to make sense of it. There is no inherent reason why AI cannot be sentient. There is no inherent reason why "synthetic" experiences of said AI are not just as significant as any other experience, unless we want to denegrate experience as a whole Anyway I will stop here

What part of the movie is this about? Maybe I missed some seconds, because I was explaining many things to my younger sister who didn't watch the first Blade Runner, so I might have been too quick to write that... but at some points it wasn't clear to me how K got his clues about where to go next (like going to the LA landfill, finding Deckard)... and planting the tracking device thing was a bit lazy (in terms of writing) in my opinion, but ok. The movie and art in it is fucking great. I think I'm going to go watch it again... alone this time. Empty theater, and smoking a marryjane while watching it would be a blast (in a perfect world) :) edit: yeah, I definitely need to watch it again... scenes were absolutely beautiful

SPOILERS AHEAD I've now seen this movie 4 times in theatres and I think it's absolutely excellent. I have complaints most definitely, but this movie is just great. The threesome scene, the advertisement scene, even the Elvis scene was really good I think and added to the suspense of attacks being mixed in with bursts of sound, very cool. The ending fight, the ending snow scene, the prostitutes scene with the creepy music, just damn great. I don't like K's boss at all she can basically be removed 100%, only good scene involving her was when Luv smashed her hand into the glass. I think Wallace was okay, I didn't even know it was Jared Leto until after I saw it and read reviews (not very familiar with him but lots of people seem to hate him) I think he did the whole delusional quasi-religious evil villain thing pretty decently. I don't like the flashbacks that hold your hand and I don't like the "explain the plot indirectly through K's boss repeating things a lot" thing during the autopsy. Also didn't like Joi saying "1s and 0s" she should have left it as "I have just two". Not nearly as many beautifully "dirty" and atmospheric street scenes as original, and in general the visuals are not nearly as appealing as the original. But overlooking this and looking at this as an independent movie instead of a sequel this can be ignored Advertisement scene is the best scene in the movie by far, and this is why Joi is essential. Mandatory, the main theme of the entire movie 9/10

"holo-girl subplot" she's a main character and one of the most important scenes in the entire movie is K's realization with the advertisement coming right after Deckard's "realization" (seeing a fake Rachel) with Wallace. The movie would be way worse without Joifu It's bringing in another dimension of artificiality of consciousness Blade Runner 1982 = Humans vs. Replicants Blade Runner 2049 = Humans vs. Replicants vs. Holographs (computer simulations) It was great and if they didn't have the hologram aspect this movie would have suffered a lot and brought nothing actually new to the table except the twists.