Hyperledger Foundation Events, 2 Jan 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Keynote: Security, Trust, and Blockchain

Description

Bruce Schneier, Fellow and Lecturer, Harvard Kennedy School

Blockchain is fundamentally a technology that enables trust. Trust is much more than verification. It is a complex social system that underpins all of society. Understanding how blockchain distributes trust, how it shifts trust from people to technologies, and where it doesn't, is critical to understanding its security and usefulness.

----
Hyperledger – Open Source Blockchain Technologies

Hyperledger is an open source community focused on developing a suite of stable frameworks, tools and libraries for enterprise-grade blockchain deployments.
It serves as a neutral home for various distributed ledger frameworks including Hyperledger Besu, Fabric, Sawtooth, Indy, as well as tools like Hyperledger Avalon, Cactus and libraries like Hyperledger Aries, Ursa. Learn more about Hyperledger projects: https://www.hyperledger.org/use

Case Studies: https://www.hyperledger.org/learn/case-studies
Training & Certification: https://www.hyperledger.org/learn/training
Tutorials: https://www.hyperledger.org/use/tutorials
Webinars: https://www.hyperledger.org/learn/webinars
Events: https://www.hyperledger.org/events
Vendor Directory: https://www.hyperledger.org/use/vendor-directory

Subscribe to the Hyperledger Newsletter: https://www.hyperledger.org/newsletter

Follow-us on Twitter @Hyperledger

Learn about Hyperledger Membership: https://www.hyperledger.org/about/join

#Hyperledger #HyperledgerGlobalForum2018 #Blockchain

A

Hey. Thank you thanks for waking up early to to see me and everybody else who is this guy.

A

I guess I'm supposed to ignore him all right. So all right, all right, you were hard to ignore all right, so I'll get started, I'll be okay, all right. So if you read the 2008 white paper, Satoshi Nakamoto concludes with the sentence. Like the first sentence of his conclusion. We have proposed a system for electronic transaction without relying on trust all right. So that's just not true.

A

My Bitcoin is not a system that doesn't rely on trust. It eliminates certain trust intermediaries, but you have to trust Bitcoin whatever that means, and in general, what block chains do is they can change the nature of trust they can displace trust. They can reshape trust and really to the extent that block chains are a interesting or be valuable. It's this and to the extent that they are uninteresting and useless.

A

It is also this so in order to understand any of this, you need to think in terms of trust, which really means thinking in terms of systems and I'm, a big fan of systems. Thinking I think that's really in too short supply in the tech world right now, and then we need a lot more of it all right. So, let's talk about blockchain I want to get away from the hype. I want to get away from the current price of Bitcoin and I want to talk to something very specific here.

A

The data structures and protocols that make up a public, blockchain, private blockchains are I, think 100% uninteresting right there right there Merkel chains they're easy to create they're easy to secure they.

A

They don't they don't need any special properties. You can achieve the same effect in many other ways, and we've been doing consensus protocols and distributed systems for like 60 years. Github is a fine consensus protocol without blockchain right only Ledger's, no problem, nothing interesting there, so public block chains, our what's new and that's three elements that make them work right. The ledger, the record of what happened and in what order the consensus algorithm and ensures that all copies of ledger are the same and the token the currency right.

A

The digital thing of value that aligns incentives to make the whole system work and all these pieces fit together as a single system, and it actually is kind of cool question. Of course, is it good for anything and that gets back to trust so I think when I, when I read blockchain enthusiasts writing I think they have a very unnaturally narrow definition of trust. I'll read things like encode. We trust in math. We trust in crypto.

A

We trust, so that's trust as verification, I trust as a verification mechanism and verification, aids, and trust that that is true, but you cannot replace trust with verification all right. So, let's talk about trust for a bit. This morning, I woke up in hotel room I trusted the locks on the door. Everybody in the hotel had a key I went down to the first floor, guess the ground floor. I'm in Europe had breakfast trusted an enormous number of people involved in food distribution, food preparation.

A

There were other diners, maybe some of you I trusted and none that none of them would leap up and attack. Me left my hotel trusted all sorts of people on the streets got onto a tram a whole lot of trust. Their cars on the streets got here. All of you I'm an enormous amount of trust went into my morning and it's like he's. Barely past nine o'clock and Trust is essential to society.

A

Humans as a species are very trusting. We are all sitting here basically strangers and nobody has lept up to attack the person sitting next to them right. Some of you laugh, but we were the only species on the planet that can get away with that kind of behavior. If we were chimpanzees, this would be impossible and that that trust right. We trust many thousands of times a day and society can't function without it and the fact that we don't think about it most of the time is a measure of how well trust works.

A

So, okay, so Trust is a real, complicated concept. The word is overloaded with a lot of meanings, so this there's a personal, intimate form of trust. When we say we trust a friend, it's less about their specific actions and more about them. As a person right, some general reliance they'll behave in a trustworthy manner. I trust their intentions. I know their intentions well form their actions right.

A

You know we can talk more about that kind of trust, there's also a less intimate, less personal trust right, you mean I, might not know someone personally or know their motivations, but I trust their actions. This is more like the trust going on in this room. I mean I, don't know if somebody wants to steal, but maybe I trust that they won't and, and that trust maybe is more confidence, and that trustworthiness is more compliance in another sense, I'm, reducing trust to predictability or consistency and there's a real leap of faith.

A

Here we have Beyond confidence because it really is no confidence, it's hard to quantify and in today's society we trust people, institutions and systems. All right, I flew here yesterday. I trusted the pilot except kind of not really I I trusted that air france has a system to put well trained and well rested pilots into cockpits on schedule, when I got to the one from the airport and into Basel I trusted a taxi driver, but again not really I trusted the system that produced licensed taxi drivers.

A

It's. Similarly, when I put my ATM card into a machine yesterday and got out Swiss francs I trusted some very complex international banking system. That would draw a similar amount of money from my account the United States and then give me the cash here and I, don't even know how that system works, and so this trust is relative. It's fluid it's multi-dimensional and I can trust Alice to return and ten dollar loan without a ten thousand dollar loan.

A

I might trust Bob to return a ten thousand dollar loan, but not to babysit an infant I, can't trust Carol to babysit an infant, but not with my house keys. My trust Dave with my house keys, but not with a ten dollar loan right I can trust. No one with my computer passwords I.

A

Can trust my brakes to stop my car ATM machines to dispense money from my account and in the United States is something called Angie's List that I use that I trust to find me a plumber, even though I actually have no idea who Angie is so in 2012, I wrote a book about trust insecurity and in it, I listed four very general security systems that we use to illicit trust. First, one is morals very general term. A lot of our trust. Behavior comes from inside our own heads.

A

Most of us don't steal not because of laws and door locks, but because we believe it's wrong, and then these are feelings of guilt and shame, and we do wrong senses of fairness and altruism desires. To be honest, feelings of group belonging differences to Authority, there's biological, those cultural factors to go into this. It's complex, but actually works really. Well, it's most of us working in this room right now.

A

What's also working in this room is another system we use, which is reputation wholly different and stronger.

A

So a lot of trust behavior stems from how others respond to our actions. We get praised for good behavior, we get snubbed for bad behavior and we are very sensitive to reputation information, but how we are perceived, how we perceive others by gossip information about each other. There's a strong theory that language developed as a way for us to transmit reputation, information and humans are unique here. There are other species that that deal in reputation. There are other mammals.

A

Actually some birds too, then recognize individuals and use reputation, but we are the only species that can transfer it right. I can tell him about him and that allows us to scale reputation. These two sis is very old. There is old as humanity you can find them in some. You know non-human primates, but of course they don't scale very well and we have more trust systems that allow siteís scale to larger communities.

A

The first one I'll call institutions, but basically its rules and laws, laws, basically formalized reputation, their norms that have been codified, whose enforcement has been delegated and, of course, they induce people to behave in a certain manner and a lot of our trust comes from these these laws.

A

If it's a government but rules of it's you know a non-government organisation and, lastly, security systems, right mechanisms that do things like induce cooperation or prevent, prevent bad behavior induce trust, compel compliance, and we could list lots of these door locks, tall fences, alarm systems, guards, friends, again, auditing systems, mitigation, recovery systems, these technologies, change the human history, basically doing the same sorts of things and they all work together.

A

Most of us don't steal because we know it's wrong. A lot of us don't steal because we're what our friends think very few of us don't steal, because what, if we get caught and way at the top, a tiny percentage of us don't steal, because we can't get through the door lock, but that door lock only works because of everything else. In that stack.

A

Right, you know: modern banking is the same way: morals prevent most from theft and fraud, financial institutions, merchants, individuals, Orchestra reputation, lots of laws and regulations and lots of security systems again all working in concert.

A

So in his really good book blockchain in the new architecture of trust, Kevin Werbach, who I don't think is in the room. He outlines four different trust, architectures and, let's think about them. Relation to what I just said it appeared appear, trust right. The most intimate I trust you, you trust in me- works great, doesn't scale what he calls Leviathan trust, which is I, think more institutional, I think about contracts media by the government.

A

We don't trust each other, but there is a dispute resolution system that we both trust, so we can enter into a contract. Third is intermediary, trust. Think of you, know PayPal or credit card I, don't trust the merchant. The verse doesn't trust me, but we both trust visa to make the transaction work, and the fourth is is what blockchain enables which is a distributed, trust which is an emergent trust in the system without any event. Individuals in the system, trusting each other and in this way, blockchain reduces trusted.

A

Intermediaries doesn't make them go away, doesn't make it easier institutions go away, but it helps coordinator crossed trust boundaries, I blocked change, shifts trust in people and institutions to stress on technology. So you have to trust trust, the cryptography. You have to trust the software. You have to trust the computers. You have trust the network.

A

You have to trust the people who are making all this work and a lot of these are single points of failure. But if a Bitcoin exchange gets hacked, you lose your money.

A

If you forget your login credentials, you lose your money, it's even worse for smart contracts and it's interesting question to ask: would you rather trust a human legal system or the details of some computer code that you actually probably not have the expertise to audit in the first place and in a lot of ways, trusting technology's harder than trusting people blockchain doesn't necessarily reduce the cost of trust it shifts it around. I mean the way to think of block chains are three sided markets, the buyer and the seller.

A

Those are the obvious ones, but the miners and the miners are paying the cost of trust in those proof-of-work systems, and that can be a very expensive cost.

A

The cost of the Bitcoin mine, the cost of the transaction fees, I'm thinking of a Bitcoin here tracks and fees, the environmental waste. That is an expensive amount. We are paying for the trust in that system and it doesn't remove the need for institutional trust. You saw you still need it, because the technology naturally centralizes, my Bitcoin, might theoretically be based on distributed trust. Yet practically that's just not true. You have to trust the wallets and there aren't many of them. You have to trust the exchanges and there aren't many of them.

A

You have to trust the software, the operating systems and the computers that everything is running on and, if you think about the attacks on Bitcoin, this is where they are. They don't go against the math. They go against the computer. Science need institutional trust also, because people still need to be in charge of these systems. There always is a need for governance outside outside the system, and you see the debates about changing the Bitcoin block size or fixing the etherium Dow attack.

A

Those are examples where, where governance has to come back into the system to deal with problems, there's always a need to be able to override the rules.

A

There's always a need to be able to make permanent changes in the rules when things change and I think about the IETF as a processor, making permanent rule changes as a consensus based system. There's an example of that and blockchain systems always have to exist with other more conventional systems.

A

My Bitcoin will always need to interoperate with the rest of the financial world and that interface with its laws and norms, often who require breaking the trust architecture of the blockchain system, and you can't have a Bitcoin system where transactions clear, immediately work with a credit card system which one's actions clear in three days.

A

This is how steve wozniak got scammed out of $70,000 it's across that trust, boundary and also in any distributed trust systems. There are back to our methods for centralization to creep to creep in in Bitcoin. There really are only a few miners. Only one company makes mining hardware. There are only a few dominant exchanges and to the extent that normal people interact with Bitcoin it's through those centralized systems and Trust is why regular people like normal ways to exchange value like credit cards.

A

It is a key feature of trust that if the transaction goes bad, you get your money back that if your credentials are stolen, you get your money back right. That is a feature of the system.

A

Regulations that protect users from fraud are a feature, so are regulations that disrupt terrorist financing, organized crime, money laundering, traffic and illegal goods I mean yes, these systems can be costly and there's a lot of talk about the cost of intermediaries and that summer, that's the inherent middleman problem and because trust can be expensive, I trust isn't free.

A

So, in order to succeed, blockchains have to be trusted to the extent that people don't use Bitcoin it's because they don't trust Bitcoin, and that has nothing to do with the cryptography or the protocols right. In fact, a system where you lose your life savings.

A

If you forget your key is not particularly trustworthy, I mean a currency that is volatile, is not particularly trustworthy and that's the human way of looking at a trust, I mean look at aetherium and you know: try moving, try and move to proof of stake, and that is an interesting example of the way Trust is working.

A

We need the fact that we have hard Forks means we still need trusted people, and this Trust is a lot more complicated than transaction verification, block chains of trusted based and reputation right. We will people, not us regular people will choose Bitcoin or not based on reputation, they'll choose an exchange or a wallet based on reputation. They'll read an article they'll get a recommendation or a friend there'll, be a reputation based trust decision.

A

Even in crypto we evaluate crypto algorithms based on reputation. It's mathematically informed reputation, but it is still reputation. We use this algorithm because it's been around for twenty years. No one's broken it yet, and it's got a good reputation.

A

I'll, look at the various blockchain supply chain, security systems, I think they're, interesting here. Okay, so of course, blockchain is not necessary for any of these systems and the real reason they work is. Everyone has a single software platform enter their data and even though the systems are built on blockchain distributed trusts, there are companies, just don't accept that there are companies that don't trust the IBM Musk supply-chain blockchain system, because it's I be immersed and not them, and that's how Trust works.

A

You can't replace with algorithms and protocols, because it's much more social than that right. It's a human thing, so truly understanding this requires systems. Thinking and I really want everybody who designs the implements block changed to understand the systems they're working in I mean these are ridiculously large distributed systems again, not just the tech but also the social parts, and we really need to know how they work.

A

Ok, so do we need a public, blockchain I?

A

Think the answer is almost certainly no and by this I'm really answering the security question, not the marketing question block chains likely don't solve the security problems. You think they solve the security problems they solve are probably not the ones people have manipulating audit data is rarely in the top list of threats. Systems face the security problems they cause are probably more than people want to deal with.

A

The inefficiencies, especially scaling, you know, are another problem and there actually are almost always simpler and better ways of a cheat achieving the same security properties.

A

So that's the bad news too, but to really answer the question you have to look at the trust architecture. It does the blockchain change it in any meaningful way, or does it just shift it around to no real effect? Does it just try to replace trust with verification? What aspects of trust does it try to fix and fail?

A

Does it strengthen existing trust relationships or does it go against them? Are the trust intermediaries of the new architecture that are worse than the old architecture?

A

How can trust be abused in the new system and is it better or worse than in the old system? And lastly, what would the same system look like if it didn't use blockchain at all?

A

And my guess is that if you ask those questions seriously in most cases, people will choose solutions that don't use a public blockchain, because all the problems they bring I'm not saying that they're useless. They do do things, but I've really. Yet to find an example where the things they do are worth the problems they bring.

A

A