 Okay, welcome everyone to CUNY Economics University in CUNY. I'm John O'Brien and I'm the Associate Dean at CMUQ and it's my pleasure to welcome you to today's lecture. The discussion is part of the Gloriana Sinclair Distinguished Lectures in 21st Century Librarianship here at CMUQ. The series is named in honour of the Dean of Library, Emerata, who served as Dean of Carnegie Mellon University Libraries from 1998 to 2013. Gloriana Sinclair dedicated her career to building the digital library of the future. Today's guest continues that work of using technology to reinvent a 21st century library. Jason Griffith is an affiliate fellow at the Berkman Klein Center for Internet and Society at Harvard University and the creator of the library box project, an open source portable digital file distribution system. He is also the founder and principal at evenly distributed a technology consulting and creation firm for libraries, museums, education and other non-profits. Jason works on technologies that provide open and robust access to information for the future, such as blockchain and other decentralized technologies. For many of us, blockchain is a fascinating topic. In fact, it's hard to go by a date without sort of blockchain grabbing some sort of press coverage there. And you find that there are two extremes. One count would say blockchain is the greatest disruption that's ever happened to date. It's the biggest thing that's happened at the moment. And the other one would say, the other count would say it's over hype. So it's interesting, it's fortunate that we have Jason Griffith here today to share his perspective on the blockchain and how it can influence the modern library. So please join me in welcoming Jason Griffith. Hello everyone. Thank you so much for agreeing to spend your afternoon with me. Thank you to everyone who was involved in bringing here to Kearney Mellon-Kitar. I'm incredibly pleased and excited to be here to talk with you today. I'm going to do a little bit of both hype and reality, I hope. My goal is to get you a little excited about the blockchain and about the potential for some of the new decentralized technologies that are coming along, as well as try to illustrate some of the opportunities for libraries in this space, information systems in general. The thing that most people think of, I guess, when they think of blockchain is Bitcoin, and that's because it was the original instantiation of this technology. And Bitcoin came along almost a decade ago at this point. It was revolutionary in that it was the first time that we had an opportunity for digital scarcity. It was the first time really where we had an opportunity to have a digital good that was provably scarce. That is, I have a Bitcoin and I can give you a Bitcoin and I can prove that I had it and I can prove that I gave it to you and you can prove that you have it and no one else can own it. But that's not the way digital stuff works. The initial excitement about digital media was that it can be shared with everyone at all times and in all ways forever. You can have perfect copies that everyone can participate in. And Bitcoin was the first example of a system that we had where we could kind of avoid that digital multiplicity. I'm going to avoid that's the one and only time I think I'm going to use the word Bitcoin in this presentation. At least that's going to be my goal. So if you have questions about Bitcoin, no, I'm just kidding. No, that's fine. You can ask me questions. But I am going to concentrate much more on the underlying technology and on an exploration of the services that can be built on top of blockchain rather than the instantiation of blockchain technology that is Bitcoin. This is the hype cycle. We talked about it being overhyped. This is the cycle of public perception that accompanies most new technologies. So at this point in blockchain, we are well and off the top of peak of inflated expectations. Everyone at this point assumes it is a panacea for nearly everything in the world that it will fix all of our digital problems. It will be ubiquitous and it will be everywhere and it will do all things. We are, I think, just about to fall into the trough of disillusionment where there is a huge blowback against technology. Everyone will decry it as false and fake and there will be a huge falling off of all of the money that is currently going into blockchain companies after which we'll kind of crawl back up and move on into what I hope will be a very long and fruitful plateau of productivity. This is the peak, though. I mean, the end of 2017 into 2018 blockchain technology in the U.S. is probably the most funded area of technology at this point. Adventure capitalists are pouring money into blockchain companies. In my opinion, for very little good reason, it is showing up in nearly every aspect of technology as well as non-technology. We'll look at some of the more humorous examples of blockchain hype a little later in the presentation. The first thing I want to do is actually do a little bit of definitional work because one of the issues with talking about blockchain is that it is a fairly technical thing. It is a very kind of basic technology that relies on cryptography and some very fancy math that I will fully admit I do not understand nearly as well as I should. I am not a mathematician, I am a technologist, and so my interest in this is usually at the service level, but in order to understand what sorts of services are good for the blockchain, you do need to understand a tiny bit about what it is, how it works, what it can do at the technology level so that we can then start thinking about the services and the things that might emerge from it. If you will indulge me, I am going to try to do a very small amount of explaining of the blockchain so that we can then have a common understanding when we start talking about services. How many here have absolutely no idea how blockchain works? You just know that it is a thing. That is awesome. I love all of you. That is a fantastic thing. Okay, so I am going to give you, for those of you that did not raise your hands, feel free to browse Twitter for a few minutes. You will come back during the services part, it will be fine. The thing that interests me mostly about blockchain, there are a bunch of little pieces, but the thing that I am most interested in is that it enables fundamentally new types of networking, of communication networks to occur. For those of you who are not computer science or technologists, this is the classic network diagrams. We have three sorts of communication networks that can be built in different combinations, but this is the general networking 101 diagrams that you would see if you were in an intrader networking class. You have centralized networks where you have a central server and everything, all of your devices call that central server for information. This is much like Facebook, Twitter, Snapchat. Any of the titans of communication at this point in history are more or less a form of centralized communication. The truth of the communication lies in the center at the company. You have decentralized communication where you have multiple nodes where the communication happens from your device into a central node. The locus of truth can be distributed, but there are still individual points of failure. If one of those nodes fails, the clients then don't have any recourse. They just simply don't work anymore. Then you have distributed, which is a web of connections wherein any single client, any single user of a network is both a consumer and a producer for that network. If I were in this final diagram to smash one of those nodes, the network really doesn't care. Everything else keeps working, regardless of if I just smashed one of them. Now, for those of you that know a little bit of Internet history, the Internet itself was originally designed to be a distributed system. The US government originally designed the ARPANET, which moved on to be the backbone of the Internet in order to survive catastrophe. It was designed in a distributed way. Forces of both economy and some forces of technology have pushed that model. Now, at this point, we are much more in a decentralized or even, in some cases, a centralized system. Here in Qatar, you have one node that takes you out to the rest of the Internet. You have one line. If that gets cut, what happens? Nothing works. That is an example. It shouldn't work like that. We should have a more distributed, decentralized communication system. Unfortunately, economics and technology have not provided that for us. Blockchain can. By its nature, technologies built on top of a system like Blockchain, and I'll actually mention a handful of others. There are other technologies in this space, are fully distributed systems. If I am participating in a system that is built on top of Blockchain, typically, not always, but typically, I am both a producer and a consumer. If I stop participating in the system, no one else cares. It continues to work for everyone else. Half of the system could go down. It wouldn't matter. It still keeps chugging along. This robustness, this ability of Blockchain systems to resist control, to resist damage, is, I think, one of the things I am most interested in. A lot of the examples I'm going to give in a little bit have other advantages over centralized systems. This one, I think, is key to understanding why Blockchain may be something that we should all think of and care about for the future. Now we get into the what is a Blockchain? What is a Blockchain? Blockchain technology is effectively a very fancy ledger. It's just a very fancy set of transactions. Those transactions have, behind them, again, some fairly advanced cryptography, some fairly advanced math, some hashing systems, other things that verify who put the transaction there, that verify when the transaction happened, that verify if the transaction is a part of another transaction. There's some math hiding in my explanation. It's okay. You can trust me. It's there, but it's just a ledger, really. It's what it is. It's just sets of information that are gathered together. We have different transactions. Those are gathered together, wrapped in a little wrapper called a block. They're put together in a block. Blocks can be different sizes. Famously, this has been an argument among all of the existing Blockchain implementations, how big should that block be, how much information should be packed into it. The key is that it gets packaged in a specific way. It gets attached to other transactional units, other blocks. One block, the block 44, has in it the mathematical history of block 43. And block 43 has in it the transactional history of block 42, et cetera, et cetera. This is why it's called a chain, right? A chain of blocks is literally what's going on here. Each of these has the history of the entirety of the chain attached to it, which means that for any given transaction, we can verify all of the things I mentioned earlier, we can verify who, where, when, how it's attached to another transaction. This turns out to be, and this happens transparently just as part of the system, mathematically, it is very difficult then for anything in the past of the system to be changed. So if I wanted, for instance, to go back, let's say there's an entry on block 34, and I wanted to, I'm a bad actor, I'm an enemy in some way, right? And what I want to do is change the transaction. I would really like to spend my Bitcoin twice, right? Like I gave it to somebody that I'd really love to give it to somebody else. I can't do that, because in order for me to change block 34, I have to change every block after that as well. And that becomes mathematically impossible for reasons we'll get into in a minute. So this means that as the chain grows, as these transactions continue to be added, we have a system that can be verified independently by anyone that wants to look at it. This can be publicly verified in one form of blockchain at least. It's public. These live in public. Anyone can look at them and verify transactional histories. Anyone can look at it and tell who did something when. So this means that we can have transactional actions without a central source of truth. I know that's a little weird, but that means that any situation where in the current model of business there is a central broker that verifies the truth of a transaction, think a bank, an insurance company, a legal unit, a court, a lawyer, anytime there is a center that is the broker between two individuals, suddenly we have a system that can do much of that verification work without having that central broker be involved. And that's the power of these sorts of systems, because they disintermediate all of the world we have built of middlemen. This system allows anyone, as many authors as we want, we can have anyone write to this. All of that writing is verifiable by anyone in the world that wants to look at it. Because of this, because of this public nature, all of the transactions are fully transparent. There's no hidden transactions. All of the data that's being written and read can be verified for integrity because there is a history of the entirety of that piece of data. There is no central server, so it's decentralized, which means that if I stop participating or you stop participating, nothing stops working. And because of the math involved, it's very, very difficult in most cases mathematically impossible for anyone to manipulate the data that is blockchain. So how does all of that kind of get managed? So how do, just as an example, how do we decide who gets to build the chain? So I told you that pieces of information get gathered, they get put into a central kind of little unit called the block and get wrapped into some fancy math, and then it gets added to the chain. Well, how does that happen? How do people then add pieces to the chain? Well, there are some very, again, I'm using the word fancy too much, but it is fairly fancy. I'll hand wave the math again. But there are some things that happen called consensus algorithms that incentivize people to participate in this building of a blockchain. And historically, that incentive has come in monetary form and currency form of some sort, right? This is why you have probably heard the term ICO, initial coin offering. This is a lot of companies are participating, they expect participation, they reward participation in the building of their chain by rewarding people with tokens that ostensibly are worth things, whether or not they are really worth things we can argue about. But there are a number of these consensus algorithms that incentivize people to be a part of the system. So the one that most people are aware of is, and I guess I have to use the word at least once more, bitcoin mining. This is an example of bitcoin mining. And bitcoin mining is the process by which new blocks get added to the blockchain. This is the reason that, again, if you've read anything about blockchain and bitcoin, you've heard that it uses more electricity than the entire planet and soon will consume us all in a fire of fossil fuels. The way that the bitcoin blockchain adds new information to it is by generating specific hashes that take lots of computing power, lots and lots and lots and lots and lots of computing power. And that work increases in difficulty every several hours, like 10 hours, something like that. So every 10 hours it gets harder and harder and harder to add things to the blockchain. But it turns out that computers get faster and faster and it all works out in the end. This is only one. So when I start telling you how libraries should be using blockchain, I want you to immediately forget proof of work, because it's a terrible system for libraries to use. If we do instantiate anything in a blockchain, this is absolutely not the right way to do it. And we should do other things. And it turns out we're lucky there are other things. There are other proofs that you can use in order to both incentivize and generate the mathematical rigor necessary to build the chain such that it is robust, robustly resistant to manipulation. One of these is proof of stake, wherein participants of the participants in the building of the of the chain put up some of their coins or tokens or something in a vote effectively to be able to be the next person to add a block. This is used in some existing situations. If you've heard of Ethereum, which we'll talk about in a few moments, they are switching to proof of stake as a proof mechanism for their blockchain. Proof of authority is a more centralized version of a blockchain. This would be like certain members of the community that are participating are given the authority to mint new blocks. They are the ones that are blessed to be able to add new blocks to the blockchain. This doesn't take computing power really at all. There's a little bit, but it does not result in the heat death of the universe, which is a plus. Proof of weight is a certain type of proof of stake, where instead of giving up tokens, you give up something else. One of the systems that I'm going to talk about in a minute uses proof of storage, where you basically promise that you will donate X amount of disk space to the system. You're going to say, you know, I have a terabyte drive and I'm going to plug my terabyte drive in and you can have that old blockchain system. You can use that. And then that storage space is used as a good in the economy of the blockchain. And then two, that I am literally not going to explain because I do not understand them, but I wanted to put them in front of you because they are important, Byzantine fault tolerance and directed acyclic graphs. Both of these, if you are interested in the math behind how proofs can happen in ways that are not directly related to blockchain, Byzantine fault tolerance and directed acyclic graphs are two of the ways that happens. So there's some Google homework for you. So we come back to the ledger. We come back to, now that we've kind of gone through what a blockchain is, how it gets built, let's come back and look at the ledger. I described it as a ledger because that's really what the original instantiation blockchain was designed to be. But as it turns out, a ledger is really just another fancy way of saying database because a ledger has described entries, a database has structured data, anything you can do in a ledger system, you can also do in a database system. And so one of the major turns in the history of blockchain technology is people realized, oh, we can actually use blockchain, not just to store transactions of coins, we can actually put anything we want in it. And so one of the major turns in the history of blockchain is actions of coins. We can actually put anything we want to in there, including full programming languages. Because if you can write data to something and read data from something arbitrarily, you can make that a programming platform. And so the second largest of the current blockchain systems is a system called Ethereum. Ethereum is also a coin based system. It does use tokens or coins as its backbone, as its way to generate both money and to the economy of the blockchain that Ethereum is attached to. Ethereum, however, doesn't want to be a currency. Ethereum wants to be a general programming platform for everything in the world. And it is an open programming platform. You can write anything that you can write a program for. You can instantiate that on the Ethereum blockchain. And it would have all of the advantages of the blockchain system that I described to you earlier, that is robustly resistant to manipulation, distributed across all nodes, impossible to censor or shut down. All of the advantages that come with blockchain, on top of that, you get universal computability. The first use of the Ethereum system was for something called smart contracts. They were really interested. The people that designed Ethereum were very interested in disintermediating the legal profession. They really wanted to tell the lawyers to go jump. So the first thing they did was try to figure out how you could write contracts into this system, wherein I would agree to something, you would agree to something, and the system would enforce it. Not a lawyer, not a judge, not a legal system, but the system itself. This is a fairly simple sort of thing to understand if you reduce it to like a common sort of monetary transaction that would have some legal overtones. If I wanted to, for instance, sell you a car or sell you a house. We could write a contract up, we could instantiate it on the Ethereum blockchain, it could have requirements for both sides, only when both sides were satisfied with all of the money change hands, and all of that could happen without any sort of judge, really. The first example of this, the biggest by far, was a smart contract that was created very early in Ethereum's history called the DAO, and the DAO stands for Distributed Autonomous Organization. This was an idea wherein some fairly misguided people thought they could create a sort of mini society writ large on through calculation and blockchain. DAO was open source, it was a platform by which people would invest and then decide where money went. Unfortunately, because this was a smart contract, it didn't care what people wanted, it only cared what the programming said, and someone that was smarter than the people that wrote it realized that there was a programming error and they could steal everyone's money, so they did. I was looking to see if I had the number, it was $17 million, something like that, that they ended up stealing. There was really nothing anyone could do because they didn't actually break the rules of the system, they just manipulated the system that was poorly designed. This was a warning for a lot of people that maybe this was not such a great idea. Another system that I am particularly interested in that is being instantiated on the Ethereum blockchain is called Gollum. Gollum is a system that is designed to be what it says, a worldwide supercomputer. It is a programming system on top of Ethereum that attempts to use everyone's computer to do simultaneous computing tasks distributed across the entire world, across everyone's computers. It handles all of the data sharding and everything that comes with supercomputing across a blockchain-driven system, kind of fun. My absolute favorite, only because a friend of mine did it, is an art project that is based on Ethereum and smart contracts called Plantoid. Plantoid is an art project that was created by Primavera Deathly Bee, who is also a fellow at the Bertman Klein Center and a researcher for a number of other very well-known research centers across the world. Plantoid is a mechanical plant, for lack of a better term. This picture you see is the literal Plantoid. It's a mechanical art object that, at its base, has a computer that contains a smart contract that's based on the Ethereum blockchain so that you can, if you think the plant is beautiful, give it money. You can send it Bitcoin. Once it hits a threshold of Bitcoin, once it gets enough money, it will, as a part of its smart contract, contact artists to see if they are interested in building another copy of itself and it will pay them. The smart contract will negotiate with the artist as to how much that would cost and whether they would like to build a copy of this Plantoid and then promise them payment for it. Once a copy is built, the Plantoid will pay for itself to be reproduced. The idea is that you have these autonomous self-reproducing artificial life forms that are all run on a smart contract on the blockchain. Let's talk about a few more platforms that are rising up as a part of the explosion of new blockchain systems. You have systems that are trying to build identity systems on the blockchain, so this is an idea where instead of having a Facebook login or a Google login for multiple systems across the net, you would have a login that you control that lives on the blockchain and then other systems can come to it and request authentication for you. Again, because blockchain systems are in public and verifiable and resistant to manipulation, it turns out they're actually fairly good arbiters of this sort of identity information, authentication information, and so having a secure identity management system that lived on something like a blockchain system could disintermediate Facebook, Google, and all of the other arbitrary login systems. You have systems that are trying to break the ad-based economic model of online digital media, so rather than paying for eyeballs, rather than having a system in digital media where when I look at something the the advertiser pays the platform, you would have a basic attention token where through this micro-currency blockchain I could pay for media that I like directly through microtransactions. You have data networks. This is safe. It's being built in order to do distributed data across the world via blockchain. Cloud storage. This is a system called STORGE, which is attempting to beat Dropbox on the blockchain effectively. It's disintermediating the central server by distributing the data across systems around the world. Sony. It's not just these little startups, really big players are also getting into this. Sony is developing an education and testing platform powered by blockchain. We'll talk a little more about educational systems on the blockchain in a bit, but just to say this before you're like, what in the world does that mean? It turns out that mostly what people value educational institutions for, aside from the fantastic education, is the credentials. And if you can disintermediate the credentialing from the place, you can have some interesting disruption of education. And we talked a moment ago about identity systems. Microsoft thinks that this is so important that actually this was what, 10 days ago, they announced that they are actually taking their existing Microsoft authentication systems and instantiating them now on a public blockchain. They're actually going to move their existing infrastructure over to a public blockchain system. This is, as far as I'm aware, the largest company actively promising to go to a public blockchain system for something that is intrinsic to their business. Authentication is an intrinsic part of what Microsoft does. And this was a very big deal. This was just announced February 12th. So two decentralized systems that I am particularly interested in, and that I think we should be watching. The first of these is a system called Blockstack. Blockstack is a system that uses, again, decentralized tech. It uses blockchain on top of some other things to have an entire application ecosystem. So for those of you that may be familiar with Chromebooks or Chrome browser where you have apps that run inside a browser, this is that except with no Google, no central server. Completely decentralized app store for the web. For those of you who may be scientists or working in a field of science, there is a system called DatDAT that promises to be a decentralized shareable data repository for very large-scale scientific data. And this is attempting to take large-scale science data out of the hands of commercial publishers, take it away from some of the large international publishing units, and put it back into the hands of the scientists themselves to be able to share. And my personal favorite, this is the, if I had to like, if you asked me like to put a bet on who I think was going to be one of the bigger movers in this space over the next 20 years, this is the company I would bet on. This is IPFS. It stands for Interplanetary File System. They have grand dreams, I guess. But this is a system that uses blockchain-like systems, merkle trees, and some other fancy math to do the web, but in a fully decentralized, trackable, hashed media way. Peer-to-peer hypermedia is probably as close as I can come to explaining it. But for a librarian, this type of information system is fascinating, because every single file in the system has a unique identifier, unique metadata, it can be tracked as it moves, it's incredibly interesting what they're doing here with IPFS. Some companies are actually now offering blockchain as a service, as of course they are wanting to do, they are interested in making money in this new technology, so you have all of the big players, IBM is offering blockchain as a service, you can sign into their service and create your own little blockchain company using their computing power, Microsoft is doing the same with Azure, you can, if you have an idea for building something on the blockchain, you don't have to do it yourself, you can leverage their tech, and of course Amazon as well, because they will offer you everything as a service, because that's their job, so they will do Amazon Web Services, does blockchain as a service as well. So you now have not only the ability to build a blockchain system through open source technology, you can actually go to one of the major providers of web services and build things on the blockchain. So what does all of this mean for information technology? What does all this mean for libraries specifically? Because that's the area that I'm the most interested in, as I'm a librarian. So I have some ideas, I haven't built any of these yet that I would really love to, I haven't found anybody willing to pay me to do that yet, but I would really love to. So here are a handful of possibilities that I think would be interesting, disruptive potentially, for the world of libraries and library work. So the first is a very simple one, and that is, in archives and museums, we worry a great deal about provenance, about the history of a physical object, who had it, when did they have it, where did it come from, what is it doing, etc. We have this idea that we need to track that, doing that in a blockchain system, super simple, really easy. The thing I'm interested in is, now that we have systems that can do this sort of thing automatically, that can do automated tracking of movement of digital goods, including the limitation of them in ownership, we can actually think about digital files having provenance in a way that we've only previously thought about physical goods having provenance. I'm not aware of anyone doing this, I don't know of anyone that's doing serious work on trying to track the movement of digital files across owners. It's probably not that interesting in most cases, I can think of a few where it might be, like government, a few where it would be super interesting to know who had that file before I had it. So the ability of blockchain systems to automate this sort of digital provenance is a big deal. The one that I am most interested in probably is the idea of taking our existing models for bibliographic metadata and expanding those out into a distributed decentralized system via something like blockchain. In the US, we mostly rely on OCLC as a arbiter of truth for our bibliographic metadata. The Library of Congress is another of those sorts of central arbiters of truth. We don't need to do that if we have a public system like a blockchain. We could have libraries arbitrarily writing their all libraries everywhere, all of them in the world for free, all writing their bibliographic metadata to a central system, all of which was trackable, all of which was verifiable, all of which could be then built upon over time. You could have metadata updates that you could track across a single work. You could, when you are reading data out, choose the library that you think does a particularly good job and only read their metadata and ignore all the rest. A lot of interesting things could come from libraries banding together to build a blockchain system to handle their bibliographic metadata, one of which would be we wouldn't need to pay people arbitrarily large sums of money to be the central source of truth for our metadata. Super interesting. I really would love to find a few libraries willing to experiment with this. In the US, we have a legal concept called first sale. First sale rights are a thing that are part of our copyright system effectively wherein if I buy a physical good, if I buy a book, a physical book, I am allowed to then sell that to someone else. I'm allowed to give that to someone else. I'm allowed to loan that as a library because it is a physical object and there's only one of them. I can't take the book and magically make two and give them to two people. I only have one. That has a long legal history, a lot of case law in the US that supports that with blockchain because suddenly we can have digital scarcity. It is theoretically possible that we also have a case for digital first sale. This would mean that if Amazon sold me a book and I could prove that I owned it on a public blockchain, I could then sell it to someone else, which we can't do now. This would need to be arbitrated in the courts. There's a lot of work that would need to be done there, but blockchain systems could give us a legal background to actually defend this position, which would be super exciting for libraries. We talked a moment ago about credentialing through education, distributed verifiable credentialing if there was a universal system by which all kind of recognized experts in education credentialed people onto a public blockchain. You could actually have a university that didn't live in a spot or a place, but instead you could be credentialed individually by professors gather the credits necessary across time and space instead of across place because blockchain has this again, verifiability and robust resistance to manipulation. It's fairly interesting and aggressive, but it's possible. This is where I live, by the way. I put that in there specifically so I can show you a picture of where I live. Distributed verifiable software and identity. There's a lot of buzzwords in that for those four. Those are all buzzwords. All I mean is there, if we had if libraries again banded together and built a universal system of authentication, we could provide people with a library card that literally worked in any library that you happen to walk into without having to verify who you were and bring an ID and everything because that would all be stored in an authoritative authenticated place that's distributed and not doesn't have the ability to be controlled or centralized. We could have a universal library card where you could go anywhere in the world maybe and just walk into a library and borrow a book and of course it's not all roses. There are problems. There is a risk with these systems even though they are designed to be centralized, to be decentralized, they're designed to be distributed. The way they are governed has a lot to do with spreading power among people. Even then there is a great risk for centralization of power. For bitcoin, this can be found by miners who control a certain amount of mining in the system. For instance, China has a huge amount of the bitcoin mining potential in the world. They are sort of overly influential from a democratic point of view in what happens with bitcoin. Any system that has some governance model has a threat of being centralized through politics or economy. So you have to manipulate the system in order to try to be resistant to that. Not all blockchains are. So that's something we have to worry about and think about. There is a huge gold rush right now going on. I used this term earlier, ICO initial coin offering, but cost coins and tokens and currency is a lot of the way people think about blockchain systems. Even though it's not necessary, even though you can build blockchain systems that have nothing to do with them, it is the way that most of the news is being spread. So there is a gold rush mentality. If I can build the next big thing and get people to give me lots of money for it through an ICO, then I don't have to worry about it and I can just run away with the money. There is a lot of hype and threat in that setup. One good example of the hype curve that we are on is this happened in the US. There was a company that made iced tea, a drink company, literally bottled drinks, who, this was their stock price. Then right here, they changed their name to the Long Blockchain Corporation. They didn't change their business. They didn't change what they were doing. They still make tea. They just changed their name. That's what happened to their stock price. That's ridiculous. It is a good example of the sort of thing that we are in the risk of if we are not careful when we talk about this technology. You have blockchains again being instantiated for the, come on, banana coin, come on. That's all I've got. It's just the world's first blockchain option for investing in the production of organic bananas. We do run the risk of systems being co-opted by large existing companies. For anyone who reads the blog site Salon, so this is a pop-up from Salon, they have decided that if you block their ads, they're going to use your browser to mine for cryptocurrencies using blockchain. We run the risk of just people being turned off because companies are doing creepy things with other systems. This is really creepy. No library presentation would be complete without talking a little bit about cats. This is a game that was instantiated on the Ethereum blockchain called CryptoKitties where people were paying real money for digital collectible cats. They're cute, but this does not strike me as a robust way of selling the idea of blockchain to the world. Maybe it is. Maybe cats are the absolute best way to sell blockchain to the world. That may actually be the case, but it strikes me as a somewhat silly use of the tech. In conclusion, we've seen how it works. We've seen some potential for the world and some potential for libraries specifically. There are places that are working on better understanding this. If you are interested in this, specifically how it applies to libraries, San Jose State University Library School is doing two different gatherings, one virtual, which obviously would be easy for you to hopefully attend, and one in person that I'll be attending later this year, where they are studying how libraries can use blockchains. This is funded by the Institute for Museum and Library Science in the U.S., and I think we'll hopefully get the right people in the U.S. together to try and talk about how we can move forward with this tech in responsible ways. If you want to follow along, obviously there is the URL. There's a blog where we regularly update information about what we're doing. Blockchain is not a panacea. It doesn't solve everything. There is a wonderful website, douneedayblockchain.com. That allows you to answer some questions, and it will tell you, do you need a blockchain? The answer is usually no. No, you do not. You actually only need it if you meet a certain series of needs. But I do think that those come up in library science. We deal with information in very interesting ways, and I think blockchain is an answer for some of the things that would make us more future-proof. Anytime I talk about technology, I try to remind myself of this quote that, in the short term, we tend to underestimate the effects and underestimate in the long run. So overestimate in the short and underestimate in the long. I think there's lots of hype, but I actually do think blockchain could change things. It can disintermediate huge players in our world. Kevin Kelly, one of the founders of Wired Magazine and a technologist of Renown, said we've just started making a technological society. The technological changes in the next 20 years would dwarf those of the last 20. I think this is right. We are changing exponentially at this point, right? We're on the hockey stick part of the curve. We're on the uptick of the curve of technological change. If you think about the things that you've seen happen in the last 20 years, the next 20 are going to be remarkably faster, remarkably different, with many more things changing. And then finally, my last employment to you as a quote from one of my favorite authors, Douglas Adams said, anything that's in the world when you're born is normal and ordinary, and just the way the world works. Anything that is invented between 15 and 35 is new and exciting and revolutionary. You can probably get a career in it. And anything invented after you're 35 is against the natural order of things. I will admit that I am over 35, and I rage against this every day where I don't want to fall prey to the belief that new technologies are not going to change the world, right? So those of you that are between 15 and 35 have good careers. Those of us that are over 35, let's try not to stomp on their dreams too hard. So I think that blockchain has the potential to change things. It can remove the centers that consolidate power and consolidate control of our communications. It can democratize the way we want our systems to work, and it can reduce the friction to enable us to build new and interesting things. I think all of those are powerful and can change the world. So this is me. All of my contact information will be there in a moment. Thank you very much for your time, and I will be happy to take some questions. Yes, sir. If I may, I'd like to break your attention to a couple of potential threats. Sure. One, a big, huge one, and the element of threat of the state of death of a thousand cats. The big, huge one, quantum computing, which is a very powerful new type of computing that would make all of our computing capacity at child's play now, could easily break and break apart any kind of system created on the blockchain. How would you address that? That the 1000 cut is okay. I'm a records manager. This is like evidence of transactions, and when you destroy the theory of how you destroy the theory of how you get this position, how does that apply to a blockchain environment? Both of those are fantastic questions. You are absolutely correct about quantum computing. Quantum computing has theoretically the ability to render effectively all of our encryption useless. I assume that we would use quantum cryptography then to move forward, right? Keeping secrets in a world of quantum computing is a thing I am interested in. I don't know much about it. Be honest, I don't. But you're right. I mean, that is a huge threat to our existing sorts of hashing cryptography functions. It just kind of wads it up and throws it in the trash. Blockchains, in general, there are a few that have been claimed to be anonymous in hard sorts of ways of the word, but they're not typically used for privacy, right? If what we're worried about is the privacy of communication, which I also worry about, I think quantum computing is far more of a threat because it does break kind of all of our existing models of privacy for encryption. For the purposes of authentication and the other kind of bits that public key cryptography does in the blockchain, I think quantum is a threat. How quickly we get to the point where it breaks our entire world? I don't know. The second piece of that, would you remind me? Yes, that's right. The records management. There are ways to render old information useless. If you destroy the private key that is encrypting a piece of information, it's gone. We would still, you're right. We would still risk a future quantum unlocking of the data. In the current world, our current computing paradigm, if you destroy the private key, as far as we are aware, unless the NSA or someone has a computer that's capable of it, then it's unknowable. Quantum changes that. It is potentially possible that, again, even the destruction of a private key is not enough. I do think that you keep, so blockchain systems can be engineered in different ways. I can imagine a situation where you would have a blockchain system where the blocks at a certain length would fall off, but that the length would be designed such that the new transactions were still protected, that there was still a history enough so that they would be protected while maintaining this falling off of the end, deletion at the end. I've not seen one of those. I can imagine it, but I don't think there's anything necessarily that would prevent such a system from being robust in a blockchain. Yeah. It's a trigger of an event of a business transaction. Yes, but the blockchain systems are synchronous. They generally synchronize themselves. You could have rogue nodes that shows not to synchronize and thus didn't participate democratically in the system. There's problems there, but both of those are great. Anyone else? Yes, in the back. Yes, sir. I was wondering if we as the library community want to build a blockchain system, for example, for the metadata. We would need to think about how to incentivize miners or how to keep our own server parts and so on. All this probably also costs a lot of money. Servers are pretty cheap these days. Servers are pretty cheap. I run a number of different nodes of a variety of blockchain instantiations on Raspberry Pis. You wouldn't do that for a production system, but it illustrates that they can be run on incredibly inexpensive hardware. They do not require the mining part, right, the actual hashing function for proof of work is computationally expensive and thus expensive in a power sense. It requires a lot of actual electricity, but as I said earlier, I would not use proof of work if I was designing a system for libraries. I would use some other proof mechanism, so proof of storage or something that was less harmful to the universe as well as less difficult to instantiate. So I would architect it not using proof of work for one. Second, it does, so that is another potential problem is rather than paying a central service we are now simply absorbing that cost in a different way. It may not ultimately be cheaper to host your own blockchain than it is, for instance, for a library to pay OCLC an arbitrary amount of money to participate in their collective. I still think it would be worth doing even if there was no net gain because of the decentralization and the democratic nature of the system. So I think we would gain in other ways even if not monetarily. Yeah. I would also want to point out a bit of skepticism for that vision that it will help democratize and so on. Because I think the key point we started with was that idea of scarcity and how it would supersede that idea that everything digital can be dispersed and so on. So once maybe the publishing world gets into that, then we are back to books, that digital books now, that can be tracked and monitored and so on. So we will probably end up with again that lending system where you hand out one copy and track it and so on, and then we are back in a different world. I think publishers, so I do worry about publishers jumping into this with a new DRM system that could be more limited than the existing DRM systems that we have. On the other hand, that is one of the reasons that I am researching and interested in the idea of digital first sale. Because if publishers create a system that does that, we may then have legal recourse to make a claim about that good that we did not have previously. And so we may actually again, maybe, no promises, but we may end up ahead because we now have a right to deal with this digital good in a way that we did not previously. We would not be bound, for instance, necessarily by licensing agreements in the same way that we are now if we had a first sale sort of right. Now that is a very U.S. perspective. I do not know enough about international copyright, specifically Qatari copyright law, no idea. So I don't know what the exact kind of equivalent would be here. But in the U.S., we may gain more than we lose even in a draconian DRM system, which I hate and would not like to see. But if we see it, I think there is a potential we may still come out a little ahead. I will be around, though, at the launch after. So again, thank you so much. Thank you, everyone.