Haseeb’s Thoughts On Crypto Universe

On adversarial thinking, blockchain consensus, and the merits and flaws (yes, flaws) of decentralization Thank you Haseeb, for your time today. I’d like to ask you a few questions about your past and then dig into a few deeper blockchain topics to pick your brain. Would you mind telling us, first, about your background and what eventually brought you into blockchain? Yeah, absolutely. So my background is a little bit helter-skelter. In a past life, I was a professional poker player. I did that for about five years, pretty successfully. Haseeb: _In a Nutshell: I first stumbled into poker when I was 16. I started with $50 and within a year, turned it into $100,000…_haseebq.com About I quit poker when I was 21 and went back to school. From there, I meandered my way into the tech world where I took a coding bootcamp. I then became Director of Product at that school, App Academy. After that, I got a job at Airbnb as a software engineer, working on payments fraud. That’s when I actually started looking seriously at Bitcoin. I first heard of Bitcoin in 2014, I think — so relatively late, compared to some people who have been in the space for a while. I sort of shrugged it off as a curiosity, learned about Merkle trees and blockchains and how it all works. But I didn’t really get the “big idea”. I “got it” after working on payments fraud, thinking a lot about adversarial environments and networks — in part from working at Airbnb, but also from my past as a poker player. A lot of the ideas behind crypto really started resonating and making a lot of sense to me. So, when I left Airbnb a little while back, I realized that was probably the most interesting thing going on right now in and that as a technologist, I want to jump into the fray and get into it. blockchain technology So, for about the last nine months I've been going super deep into blockchain, reading and getting my hands on anything that I can, building, and getting into the ecosystem. Since August, I've been at , a blockchain company run by , who you know. That's what I'm doing in a nutshell! Earn.com Balaji S. Srinivasan Fascinating. Thank you for that background. I'm curious to dig into a couple of things. First, it would be interesting to hear about your poker background and the things that you bring with you from poker that remain useful today. Yeah, it's funny. Poker has a lot in common with blockchain. They’re both very multi-disciplinary. In order to be a good poker player you have to be thinking about strategy and mechanics, which is the way the poker “protocol” works, so to speak. That's the base layer of any interaction. Then, you have to think about game theory. If I do set of moves, how is someone else going to respond? And, x how is this game going to play out as we iterate it over an entire match? Also, just in terms of economics — there's an enormous amount of “table selection” in poker. Poker players are constantly thinking: at this stage, on this site, at this time of day — what are the conditions like? How does poker change in terms of the kinds of players and the different dominant strategies? All these things that play into the ecosystem of poker are similar to the way blockchains are analyzed. Blockchains are also these iterative games with incentives that a lot of people are playing with each other. There's some perfect information, some imperfect information. Imperfect information meaning information that you don't directly have, but may be able to indirectly access. Imperfect information refers to an economic concept where one party has more information than another in a transaction (for example, a patient knows their body better than a health care provider). When designing systems, it is necessary to build in strategies which reduce risks resulting from imperfect information. [More ] Side note from V: here Most importantly, poker teaches you adversarial thinking. This helped me a lot when I was doing work on anti-fraud stuff, and is especially useful in blockchain. Thinking from the perspective of an adversary is exactly what you do as a poker player. You have to be thinking: What is this person trying to trick me into doing that I don't want to do? That is part and parcel of how the psychology of poker plays out. Yeah. That's fascinating and makes a lot of sense. Now that you've explained how poker is similar to what you’ve experienced in blockchain, I’ll take a step back and ask: what is a blockchain to you? And what are situations in which a blockchain solution is merited? Sure. That's an question, so let's start with the first one. What is a blockchain? A blockchain, to speak most literally, is simply a linked list where each block in this list points to the hash of the previous block. enormous From Mastering Bitcoin, Chapter 7 That's the blockchain. It's not some groundbreaking revolutionary idea. Blockchains have been around for a while. It was the 1980s that someone first invented the idea of a blockchain. What people often do is use the word ‘blockchain’ to be an emblem of the innovations behind Bitcoin. The real big idea behind Bitcoin is not the blockchain. The big idea behind Bitcoin is Proof-of-Work and Nakamoto consensus. Proof-of-Work Proof-of-Work (“PoW”) is basically the idea that you build a blockchain where, instead of just having a cryptographic hash of the previous block, you require that the cryptographic hash of the previous block is sufficiently hard to compute. The way you make it hard to compute is by generating a random number (a nonce) and trying different numbers until you can get a small enough hash for the block (with enough leading zero’s). That shows proof that you've expended computational work to mine this block, or solve this puzzle. On the face, it's not obvious why this matters. What Proof-of-Work gets you is . some pretty interesting security properties in a peer-to-peer system A peer-to-peer system is not your normal ‘client-server’ computer model, where one computer serves as a client (your PC / laptop) and leans on a server (or server farm, like Google Drive or Amazon Web Services) for computational power and memory. In a peer-to-peer system, each computer (or ) plays both client and server and contributes both ways to a ‘leaderless’ network. If you’ve ever downloaded music via Napster / BitTorrent, you’ve seen how this works. Peer-to-peer systems have opened up possibilities in computing, but have not traditionally been . With PoW, that began to change. Side note from V: node the most secure So what are the security properties you get? To answer, here’s a quick thought experiment: What does a blockchain look like without Proof-of-Work? Let’s say anyone could add a block to the blockchain at anytime. What would happen? The answer is the network would constantly be diverging. Nobody would be able to agree on anything. It would be very easy, 300 blocks in, for someone to broadcast 500 blocks from the beginning of the blockchain and completely wipe out the original 300 block history. That's undesirable if we want to have certainty that this block structure is going to be stable and we’re going to agree on history. really Proof of work 1) so that we can steadily agree on the state of the network in a way that's unlikely to reverse and 2) to secure the network. slows down the system uses computational power If somebody does not have sufficient computational power, they're not able to re-write history, and they're not able to produce long-ranging forks in the network (re-write a branch of history). So PoW creates desirable properties for any system which is both peer-to-peer and decentralized. It turns out a great candidate for this is money. Abstract Bitcoin White Paper That’s what Bitcoin ended up doing: it serves as a blockchain for money. Since then, many other ideas have been come up with for how you can use blockchains. Some of the same ideas of Nakamoto consensus / Proof-of-Work consensus to build up cryptographically secure systems in a decentralized way, which can still — fairly quickly and reliably — coalesce to agreement about the state of the network. Ethereum is a good example. Instead of building a currency, Ethereum is building a computer. Ethereum is a gigantic computer that everybody in the world gets to run and they are able to pay in the native currency of that computer, which is Ether (ETH), in order to get the computer to run the program. _Ethereum is a decentralized platform that runs smart contracts: applications that run exactly as programmed without any…_www.ethereum.org Ethereum Project That's a big idea. And there's a lot of creative stuff being built on that idea and a lot of other applications people are coming up with to use the blockchain. So again, it's not really the blockchain. The blockchain isn't a big idea. The big idea is all the consensus that goes around the block. The combination of , the blockchain, , and Proof-of-Work / Nakamoto consensus. People are building interesting things on top, but that's what a blockchain is. public key cryptography peer-to-peer gossip protocols Awesome. I'm excited to link to a lot of those primitives and allow people to dig into your definition. I hadn’t heard the thought experiment of a blockchain without proof-of-work. What would happen where you just put a blockchain out there, you didn't make it hard computationally to validate the blockchain - what would happen? That was helpful. I think this segues well into another conversation. We talked a lot here about proof-of-work and Nakamoto consensus. Before this conversation, we've talked about other consensus mechanisms — some traditional consensus mechanisms, pre-Bitcoin era, and some newer consensus mechanisms that people are thinking about and trying to understand. Would you mind digging in a bit further into Nakamoto consensus, then some prior consensus mechanisms, and where they all sit today? Yeah, sure. So the feature of proof-of-work which is so desirable is Byzantine Fault Tolerance. First, we should define what fault tolerance is, and then disambiguate between fault tolerance and Byzantine Fault Tolerance. Fault tolerance is a term in distributed systems, which basically means your system can tolerate some number of faults. Well, ? what is a fault Fault tolerance is a quality of a computer system where a system can continue to function without a hitch to the end user, despite errors in hardware, software, or other inputs. The fault likely requires the system to re-direct resources elegantly in order to accomplish it’s task. Side note from V: To use an example, Google runs a few data centers. They host all your data, the search indexes, all the other stuff. They're running a lot of hardware, and sometimes that hardware is going to fail. It's going to go down and die. Google knows this happens. It happens every day at the data center. Statistically, several computers are going to fail every single day. So, what Google wants is for any single node (or computer) in their distributed system to go down — have — without a hitch to you. So the process dies, but the system stays up and keeps working and you don't notice it as a user. a fault So that's fault tolerance. The algorithms have been studied for very long time. They're now fairly well understood. There are a lot of things you can do to achieve fault tolerance and it is a desirable property. Fault tolerance is relatively cheap to get at this point. There's another form of fault tolerance which is stronger, which is known as . Byzantine Fault Tolerance The Byzantine General’s Problem (1992) Fault tolerance is your server going offline and the system not dying. Byzantine tolerance is where — instead of a server dying, it does something totally random. Totally arbitrary (or even malicious). So normally, you might about a hard drive dying... If it's going to die, it's just going to do the normal hard-drive thing and die. That's a normal fault. But what if the hard drive like was acting normally and then started lying about it’s contents before dying? It says I have no more files on me, my hard disk space is full, I’ve replicated my information — when it actually hasn’t. It’s simply lying, and then goes offline. Especially if the system started maliciously lying, which is most important to think about in a peer-to-peer system, when you have a lot of individual peers who have incentive to lie. How do you tolerate bad actors? And what threshold of bad actors can be tolerated in the system? This is the measure of how Byzantine fault tolerant you are. How can your system tolerate liars? So there's a lot of literature of what are called Byzantine Fault Tolerance (“BFT”) algorithms that can achieve consensus despite some number of Byzantine faults. These classical BFT algorithms came around in the 80's. Traditionally, BFT algorithms can generally tolerate up 33% of the network being Byzantine. That’s pretty good** ** right? A third of all your nodes are just outright lying maliciously to you and you can tolerate it. That’s pretty good! However, one downfall of a lot of these algorithms is they are very slow. So that would obviously not be doable in a serious peer-to-peer network. , There are some algorithms that are faster for Byzantine Fault Tolerance. The problem, though, is that you must know in advance all of the nodes in the network. So you need to say beforehand: these 1,000 people are in our peer-to-peer system. If a third of them are offline at any given time…Well, that's already your 33% potential Byzantine faults. “Offline” and Byzantine are basically treated the same. Also, in a peer-to-peer protocol, things generally change fast. Nodes are offline all the time, it's hard to know who's a member. It’s all over the place. These PBFT algorithms don't really work that well in a peer-to-peer environment. One of the chief innovations of Nakomoto consensus/proof of work is that it . doesn't require you to know who's a part of the network Bitcoin allowed us to say: just show this Proof-of-Work. Doesn't really matter who made it and who’s online. The proof of work algorithm is self-adjusting, so it gets easier / harder depending on how many nodes are online. This auto adjusting difficulty level + PoW in genereal made it nice for a peer-to-peer system, that anybody can join, at anytime. You don't need to register everyone beforehand. That being said, there are other approaches. One of the downsides of PoW is the speed. We discussed why this slowness is desirable, as it allows us to coordinate better — but, it also limits our throughput. So blockchains, even the fastest, highest level blockchains are still quite slow. Ethereum can only process about 7-10 transactions a second, which is a pretty trivial amount of throughput relative to VISA processing thousands up to 10,000’s of transactions a second. So blockchains can’t right now can't really support meaningful throughput using PoW. So, there are some blockchains that use . For example, if you look at , , EOS, Steemit or Bitshares… they all use some kind of leader election, such as “delegated” proof of stake consensus. other forms of consensus Ripple Stellar Delegated Proof-of-Stake The idea behind delegated proof-of-stake is having a smaller pool of people — elected officials, essentially — who are democratically selected. If those elected officials achieve consensus, a block can be mined. Now, you literally have only a hundred ‘elected’ nodes, and you can use some of those old school BFT algorithms. There's no slowing of messages, because you don’t have to wait for PoW to go through. For a relatively small network, those networks are much, much faster and you achieve much higher throughput and scale and than PoW can. So if you , you can get a lot more throughput. But then, there's this risk for delegated proof of stake algorithms. One of the criticisms is that the early people who join the network very often have the most stake, and they also end up having the most power — they end up becoming the validators (i.e. ‘elected officials’) and the miners. This means founders and close friends are a big part of the consensus protocol, which is much less decentralized than a full proof-of-work. trade off some centralization Proof-Of-Stake That's what brings us to Proof-of-Stake (“PoS”). PoS is one of the of the newer consensus mechanisms being explored today. First, let’s talk about the total hash power that underlies the Bitcoin network. Bitcoin’s nodes collectively produces the same amount of energy as the entire country of Morocco. _As Bitcoin continues its stride towards mainstream adoption, it turns out that its surging price rates are not the only…_thenextweb.com Bitcoin mining consumes more electricity than 20+ European countries That’s quite a lot of electricity. Its extremely expensive to run Bitcoin. One question that people have been asking for a while is: One of the ideas was Proof-of-Stake. how can we reduce the waste / environmental footprint of running a peer-to-peer network? So as we discussed, Bitcoin is essentially secured with computational power (which is burning all that electricity). Without as much computational power as the rest of the Bitcoin network, you’re not able to launch an attack against Bitcoin. What does computational power cost? Money. Well, what if instead of having a proxy for money — because buying computational power is ultimately a proxy — what if you just vote with that money directly? Right now, in order to develop the power to attack Bitcoin, you need money out of band. You’d have to go buy some computers and electricity / computing power in the real world. What if instead, you put money up for stake Within the blockchain protocol. After all, we have money in the protocol. in band? We can use that money to come up with a voting system on what block should be mined next. This allows us to adopt the best ideas from some of the traditional BFT algorithms, while also moving everything into the protocol instead of depending on extra, computational work. I’m glad you clarified the differences between the mechanisms and why it’s environmentally important to consider alternatives to PoW. It’s also interested that the move to Proof-of-Stake actually gets us closer to traditional BFT mechanisms. Outside of proof-of-stake being more energy efficient, could it be considered even more powerful than Nakamoto consensus? It’s contentious. Advantages of PoW is that by using some of these more older school BFT algorithms you do get higher throughput. We talked about before: the limited number of nodes validating transactions happens quite a bit faster than PoW can. That's one of the factors at stake here. That said, one of the downsides is that you get some more centralization. The other nice thing about PoW — which is somewhat underrated — it's simple to understand. There's something about it which appeals to a lot of people and make them trust in Bitcoin. You can explain how Bitcoin works to a teenager and they would understand, so long as you explain the constituent parts. pure A currency secured by Proof-of-Stake is much harder to understand. For some reason, something about burning energy and converting that into money... People are like okay, got it. With PoS, it feels like black box magic. Why is this money? What makes this valuable? There are other problems with PoS — which maximalists or people who are skeptical of alternate forms of consensus don't really like. One issue with proof-of-stake is that the This is certainly true, because the more stake you have, the more you can ‘stake’ in a PoS system, and the more money you can make from validating. This freaks people out — the money is going toward people who already have a lot of money. rich get richer. In reality, this also happens in Bitcoin. The miners who have million dollar rigs get all the newly minted Bitcoin. But, the narratives persists. Bitcoin mining is decentralized and anybody can do it, just open your computer and you can do it. Which is just completely false at this point. Because ASIC’s are so much more efficient than CPU’s / GPU’s. However, the narrative does a lot of work. Money is not just about the technical property of the system. It's also about this game that we're all playing with each other and these points of agreement. All these beliefs about each other is a large part of what makes money, money. PoS is tricky for a lot of people to wrap their heads around. And that's a real weakness. But it does have a lot of nice properties such as being much more energy efficient having high throughput. The Ethereum Foundation has recently stepped up efforts to explain Proof-of-Stake, perhaps to help alleviate some of the concerns Haseeb aptly mentions about the importance of it being easy to understand in order for people to accept it as a consensus mechanism securing their currency. Side note from V: _tl;dr Casper will implement proof of stake in Ethereum. We begin with a review on why proof of stake matters and…_medium.com Ethereum Casper 101 Great. Thanks for talking us through consensus. That was important historical context on all approaches available. We’ll switch gears a bit to some more philosophical questions from here. There’s a narrative in blockchain summed up well as ‘decentralize everything’. We’ve talked before, and you’ve shared an interesting perspective on the merits of decentralization but also the issues with decentralized systems. Is this wave of decentralization the future of everything? In what cases is centralization better? Yeah, that’s a good question. From the issues we’ve discussed, you should have some idea of the challenges of using the blockchain. To sum up, decentralized systems are a) slower, b) considerably more complex and thus, more broadly susceptible to attack. Side note from V: If you have a system that needs a lot of throughput, tough luck. Blockchain is not going to work out well for you, today. Another disadvantage is that blockchains are complicated, and very hard to build. If you want to build most applications, it is much easier to do if you control the infrastructure than trying to make the application work on a blockchain. Many ICO’s say “we’re going to build A, B, C, and D in the next few years”. Most of those businesses would be fairly trivial to build with traditional technology (Server, MySQL database, a nice front-end). Easy to build most of these things centralized. Decentralization introduces an enormous amount of technical overhead and complexity to make these applications work. Those are some of the most straightforward disadvantages. It is useful when you have a system with a lot of participants where those participants don’t trust a centralized party. Now when is blockchain useful? There are not many situations where there’s no trusted central party. A lot of people point to a market place, like Uber, and claim there is going to be a decentralized Uber. Uber on the blockchain is going to be much better because nobody has to trust Uber. Well, a problem with that line of thinking is that people actually don’t have a problem with Uber. People may not like Uber’s corporate practices, but then they’ll just take Lyft instead. The problem, however, is not that I don’t trust Lyft / Uber to steal my money or pair me with bad drivers. People are largely concerned with: Nobody is not using Uber because they are concerned about the fact that Uber is centralized. is it good and fast? Does it get me to where I want to go? People often assume that ‘decentralizing’ reduces rent seeking ( ). There’s no way to get rent, great! In reality, there are a lot of things that centralization makes more effective. Dispute resolution, customer support, economies of scale — all of these things are easier with a centralization authority. Good business strategy — quickly making decisions without coordinating with thousands of people — is much easier when centralized. There are a lot of benefits to centralization. Note from V: like Uber’s 20% fee on each ride Where Decentralization Is Better There are very few things that people say, “I’d really like to use this, but it’s centralized.” Because it’s centralized, it’s a no go. To me, it’s obvious that money is actually one of those things. A lot of people, especially people like the Cypherpunks, are exactly the people who don’t like centralized money, controlled by governments. _From bitcoin to blockchain to distributed ledgers, the cryptocurrency space is fast evolving, to the point where it can…_www.coindesk.com Bitcoin and the Rise of the Cypherpunks — CoinDesk In their eyes, we need to find some way to get this out of the hands of corrupt governments, politicians, and states that all have their interests at play. I can see why money is exactly that kind of thing which needs decentralized. Another good example is Mastadon, sort-of a decentralized Twitter. Mastodon doesn’t use blockchain, but Mastodon is a place where people who are censored on Twitter, or have interests/communities against Twitter’s policies, they can use Mastodon. This is a pretty large use case. There are a lot of communities that can’t really conduct themselves over Twitter. In evolutionary terms, it would be surprising if decentralization had so many benefits, but nobody has done it yet. The limits of decentralization is not that we didn’t have blockchains. There are many ways to decentralize projects without blockchains. But people don’t do it. Why not? Either there’s a real stigma against decentralization and people are irrationally deciding to decentralize or there are a lot of advantages to centralization. There might be better decision making, management, and execution via centralization. a) not b) I suspect that when a lot of these decentralized versions of centralized applications start trying to get users, and actually try to get market share (not just selling tokens), they will find out decentralization is hard. They will realize a hand and a leg are tied behind their back. It will be a rude awakening to these those ICO’s who will realize that doing this on a blockchain is a value subtraction, not a value add. That’s roughly my argument when it comes to decentralization. There are a lot of different issues that you can disagree with, but by and large, there are relatively few applications suffering from the problem: we want to completely eliminate rent-seeking, we need a totally cryptographically secure, peer-to-peer system, with no leader. We don’t trust anyone, Yes, fascinating. I think this is an important topic. It’s a reality check. It’s good to have a perspective takes into account both positives and negatives. The question this begs is: you still have a lot of excitement about what’s happening. What gets you excited about blockchain — even though there are many challenges — that continues to push you to build in the space? Yeah. I think if you’re not willing to look bad ideas in the eye and call them bad ideas, you’re not thinking deeply enough about what’s going on in space. Blockchain is one of the most cutting edge and uncertain spaces in technology. For machine learning, or AI, or VR there’s a fairly well understood path. For blockchain, no one has any idea. There are a lot of different guesses that you can make. I’m fairly confident that decentralized Airbnb, Uber, whatever is probably not going to be a thing. But there is going to be a lot of stuff massively disrupted by blockchains. It’s easy to say okay, well money… now what? But money is huge. Money is first. It is the biggest thing in terms of market size and value that you could possibly imagine. It’s naive to assume that the way we’ve done money for the last 100,000 years is the way we’re just going to keep doing it. Money is paper, which we hand to each other and then eventually we go take the paper to the bank. The guy who made all the piece of paper decides when to make more paper. The idea we will do this forever is absurd. It’s naive because everything else about the way that we have lived has changed so dramatically with technology. The idea that money would be impervious to this — and that the first system we established is the best one — is silly. So I don’t know if blockchains are going to win — I certainly don’t think Bitcoin is going to replace money — but it’s the first shakeup of a system that’s been in place for a long time. Almost all the fin-tech built on top of money (like Paypal and Stripe) seem cool and futuristic, but are still abstractions over the piece of paper. It’s still using a piece of paper at the bottom layer. The whole idea about Bitcoin and cryptocurrencies is: If you just have technology, how do you do it? what if we start over? Bitcoin is the first solution we’ve had to this problem. The first solution for: how can you replace money, entirely digitally? Bitcoin solved this problem. I believe technologists give too little credit to what an enormous innovation just that is, even if nothing else happens. This is a watershed moment in human history, if the way we think about money and currency changes after the advent of Bitcoin. That’s fantastic. What’s tempting to ask is, “What are the twenty five different things that blockchain can disrupt?” When an answer is simple as money comes up, it’s hard to remember that just saying money disrupts thousands of things and you don’t really need to go too much further. You can dive deep there. I’m glad you remind us of that perspective. Definitely. That said, the exciting things enabled by digital money is that we’ve built all these abstractions on top of money which can be re-imagined. Contracts, handshake agreements, banks — all these expensive things that we come up with to adjudicate problems from paper money. There are a lot of things that you can do if the contract natively interacts with your money. Right now, it doesn’t. You don’t staple dollars to your contract, right? If you could do that life would be a lot simpler. Turns out with smart contracts, this is possible. There’s a lot of amazing things you can do when your money interacts with your software / business processes at the protocol level. Who knows what it’s going to look like… I think it’s way too early. Ethereum was just invented two years ago. We have no idea. We are taking baby steps right now. We’re still drooling on ourselves when it comes to our sophistication when it comes to blockchains. We’re right now in the very early innovation stage. We’re trying different things, seeing what works. That makes me excited. I think you have to call a spade a spade and say, there are a ton of bad ideas right now, but the potential for innovation is enormous. There’s going to be a lot more stupidity before we eventually figure out: here are the real killer apps for smart contracts. That’s part of the process. I’m excited, too. In a year or two years, we can have another conversation about smart contracts and the innovation to see progress. Last question: You’ve been in the space for nine months. You had strong software development experience and prior experience at Airbnb, a great company. How did that transition go? What would you recommend to two different parties: a) somebody who has no software engineering experience looking to get into blockchain and b) somebody who is already a software engineer by trade who wants to get rolling? Yeah, good question. So let’s say you want to get into the crypto space. First things first, you learn what Bitcoin is. Bitcoin is the granddaddy of cryptocurrencies. In order to really understand what’s going on in space, you need to understand Bitcoin. Luckily, it’s the easiest to understand, so it’s a good way to dip your toes in the water. medium.com A Rabbit Hole Into Blockchain Universe Note: While listed a 3 minute read, this article is simply taking you down the rabbit hole. From there, there’s a really great reading list called by Scanate. There’s great reading on Bitcoin, blockchains generally, and then transitions into a reading list about Ethereum. It’s developer focused, but also has non-technical stuff for non-developers. EthList There’s also a talk I gave a couple of weeks ago on YouTube which basically talks through how to build a blockchain from scratch, how to build a cryptocurrency from scratch. And it’s fairly simple. Even if you’re not developer, there’s a lot of value to get from this talk. ( This talk is fantastic and extremely underviewed. It’s worth chunking and watching over a few weeks, if you don’t have time in one sitting). Note from Vivek: Generally speaking, if you want to really get into cryptocurrencies, there are two ways. You could approach from the mindset of a) a trader or b) a builder. There’s really no space in the middle. If you think about crypto as markets, and you want to make money off the markets — the information you’re looking for is quite different than if you are a technologist. The meetups in the space, the blogs you want to be reading are very different and separate in those two cases. So I think for any person who’s looking to get into cryptocurrencies — decide which of those two sides of the aisle you want to be. If you want to make money, probably trading. If you want if you to plug yourself in for the long term, or make your career in blockchains, you probably want to be on the other side. That’s generally the way that I’d draw out the system. For me, I took about 2–3 months just learning stuff and studying. There’s a great Princeton course on . It’s a Coursera course and all the videos are on YouTube. cryptocurrencies by Dr. Arvind Narayanan There are a couple of good books that have been written: one also by Dr. Narayanan ( ), and the other by Andreas Antonopoulos ( ). He is also coming out with Mastering Ethereum, which should be quite good as well (right now, still pre-publication). Bitcoin & Cryptocurrency Technologies Mastering Bitcoin There are a lot of good lists of influencers. Most of the blockchain people live on Twitter. Given this is a ‘decentralized’ thing, there’s no city that has a monopoly on it. Unlike a lot of other technologies, that most think Silicon Valley is a place where almost all the action is happening, but for blockchain not really true. The SF Bay Area community is big, but certainly far from a majority portion of what is going on with blockchains. So given it’s very international, people all around the world are talking about blockchain. So if you want to get involved in the conversation, find influencers on Twitter, and follow everybody that you see and pretty soon you’ll have a Twitter feed ready great to go. That all is great stuff. Any last thoughts to share? I’ve written a couple of blog posts about blockchain. The next post is about what I think the killer apps are for blockchain. Hopefully that will be interesting. I’ll continue to write, so follow me if you’d like to see more on , , or . Twitter Medium my website _It wasn’t too long ago that Silicon Valley scoffed at cryptocurrencies. All over coffee shops in Mountain View and…_hackernoon.com We already know blockchain’s killer apps Awesome, this has been fantastic. Thanks again for your time! Of course man. My pleasure. Thanks again to Haseeb for taking the time to review blockchain with us at Kryptos. If you have any questions or suggestions on future topics, leave them in the comments below or find me on Twitter !