John leads NEO Global Development in Seattle, WA, with a focus on the developer experience and tools. Previously he built and successfully exited two Machine Learning start-ups. Earlier in his career at Microsoft, John incubated and built Microsoft Digital from zero to $0.5B in revenue, and he led Architecture, Product, and Developer Engagement for the .NET platform v1. He has published a number of books on Software Architecture, and is currently working on a book on Decentralized Systems.View the profile
About the talk
John deVadoss, a blockchain expert, spoke about the benefits of decentralized systems and the emerging patterns and practices for decentralized platforms and systems. He also suggested some ways to use these new capabilities in real business scenarios. Formerly with Microsoft Digital, John was able to cover both the technical aspects and business case. This session offered an interesting perspective on blockchain.
John David. Typically, when I give a presentation on a discussion and blockchain that are two very regarding questions. The first question is a good night make money off Bitcoin. No, I did not because I'm a techie from not a trailer or Speculator. And the second question is why do I care? I mean to use blockchain systems is the second most often question. I get asked all the time so quickly about myself that you're wondering. Why am I why am I here? What am I listening to this guy? I spent many years in the IT industry to work in machine learning back in the 90s long before it was cool
back in the 90s. Nobody care about machine learning. I spent many is Microsoft Microsoft digital. Which is the part of the company that's working with Fortune 500 customers and helping them with their DT efforts much like a previous session. Build a couple of startups in a few exits and currently I lead the new blockchain in North America. I'm here and maybe they'll be some use for you to session the first the second question, you know, if anybody doing anything with blockchain and so I figured just over the last ten days, I would take
some snapshots of news articles and Venice to spend a couple of minutes giving you a flavor but what's going on across the industry. It can be somewhat easy to step back and say really this is the crypto guys doing something silly over that why do I care if you look at this this light here. Top left? 4 days ago JP Morgan Announce, they're launching a coin to enable currency transfer on a global level. It'll be easy for you to manage your your current across the globe.
That's a non-trivial obviously step forward examples of blockchain usage looking Facebook weeks ago. They acquired a blockchain startup in London in the UK. You look at the bottom left the HSBC. So apparently he just busy that's about 350 billion dollars of transaction volume every day on their private blockchain infrastructure again of a non-trivial number if you will at the top right you have The Usual Suspects Microsoft is working on blockchain Amazon is working on blockchain systems and databases on the right across Google as well on the Lower Lights sing about a
week and a half ago. So probably didn't send Joseph health is a very large Healthcare entity anybody in the room from Providence Saint Joseph. I don't want to offend anybody here. All right. So you ask yourself it on this is a this is a healthcare provider and and they bought a blockchain startup working in the healthcare space with you for for what might be going on there. So I can do the question of why do I care? Is that somebody out there doing something relevant or useful? There are a lot of people out there right now this to me feels like the
cloud back in 2006 or Circa 2007 Talk about public Cloud private Cloud security. Can I only put my stuff on the cloud what happens if there is a compromise about hybrid clouds this feels very much like that stage in terms of blockchain platforms give you a flavor. So alright, so After the clock what is next? What what why do I have this light if you let me step back and look at the pendulum of computing and I began my career in the early nineties. If you look back at the pendulum of computation of computing. You know, I'm through the seventies I would argue the
focus and the attention was squarely on the Mainframe on centralized Computing and computational models. And then in the eighties some of the 80s Microsoft in the PC and IBM came into the picture and the Pendulum swung slowly away from centralized to put up the same or on the edge more on the client more towards the individual. Do the early 90s? And once the game is swung back in the late 2000s back into the cloud computing models. So once again the shift in terms of energy and
focus and and money and attention back to more of a centralized model. So I cannot predict the future. However, if you look at history. History tells us that these ships have been periodically and since you asked yourself, so what is going to happen next? Will the cloud be the end game will it will cloud computing be the final stage? Probably not most likely they will be a shift once again from the centralized cloud computing models do something a little more on the Edge Wraps decentralized. And I believe what's happening right now
with the blockchain infrastructure and the blockchain platforms and DS is the canary in the coal mine for attention focus and energy shifting once again away from the centralized model tomorrow for decentralized model. So what does it mean to be decentralized probably two perspectives broadly. What is Waterford technical perspective, which is in a decentralized system going to truly decentralized system. You would expect there to be no single choke point in the app or the network. That could be
attacked which would lead to the system of Italy been compromised. That is also an organizational or you could say more if I may be a social Dimension as well in that there is no single person or entity that could be compelled in to compromise the system. So let me talk about decentralization that is very clearly like you all know I technology perspective but that is very much an organization perspective as well. I want to step back until I give you my own journey in summer self how it for me. This this blockchain space when he came into the picture.
Like I said, I'm not going to speculate that. I did not buy Bitcoin back in the early days the first time around I saw blockchain was probably about 7 8 years ago and my foot reaction was look this is about cryptocurrencies and the dark web and such and imma take you so I stepped back. This is this is not relevant to be of no interest. the second time when I went back in I said, okay, you know what? This looks like a database. And I spend a lot of time with this with SQL databases. This looks like a
database. It's a ledger. Robert Lee, it's not a centralized Ledger. It's more of a decent the last Ledger but still is database. And and why would I want to go look at another database SQL? I know how to build these apps to design your diagrams and so on I grew up in the 90s, right so so it wasn't interesting. What was the big big idea at the bigger? How about a database or a ledger the second time around us about two years ago. I was working on my second startup and I happened to delve deep into the blockchain world.
And what I discovered was the big for me was that in the world of blockchains blockchain platforms? Yes, that is a crypto basis. The underlying basis is based on cryptography and secure protocols. Yes, that is a ledger a database. It is decentralized. But what is Maurice every node in a blockchain platform actually has a virtual machine. And so probably thinking what is a VM? It's a platform. It's an app platform and that to me was personally my own Journey the big guy and it said hang
on. This is not a crypto perspective. This is no more a database perspective a platform perspective. And for me as an app developer as an IT person as a platform guy, that's when they was the very big of how does it look I've been stupid about this. I thought it was very smart in terms of being able to comprehend and so the put the pieces together, but clearly I was just I was just stuck in my own little box. I did not know what I did not know. And those are the big on having to look that is a new platform pattern here. I need a platform
you have apps you have tools if management security and so on and so forth. And that is what I would ask you to think about it when you look at the blockchain world C does an application platform. So it's a quick walk-through inspection certificate and I want to just give you a quick summary typically digitized assets. Are these lights readable the color soft if I if if so, I apologize. Thank you working as a medium of exchange that's based on very strong cryptography to
secure transactions and the transfer of assets. This is the cryptographic base. Call blockchain platforms. Do you typically have a ledger that is distributed which acts as the underlying infrastructure of the plumbing? So the next day just okay. So we agree. The second level is the DLC the so-called DLT the decentralized distributed Ledger technology. Michelle Ledger is a consensus of replicated shared and synchronize data across multiple sites multiple geographies of countries, and so on and so forth and there is no centralized administrator. There is no centralized data
storage. And I was in this face probably bored. What do two and a half years ago. It was nothing. There's no big deal. No big bass. Then you come onto okay. There's a guy called vitalik buterin vitalik probably heard the name metallic was the founder of aetherium. Is helium South Indian was different because metallic said look if you look at Bitcoin. The think of Bitcoin as a single scenario platform on a single-use Case platform. What do you transfer? And you know, you could
store Bitcoin and you could obviously have transaction was a single-use case stack what metallic did with its helium was to add this virtual machine to every note. And then yeah, what's going to be at platform? Orbi satellites application platform an apple run by many users obviously on a on a d network with a dress less protocol. And obviously the goal is to try to avoid or minimize any single point of failure. Need the way I wouldn't ask you to think about the the architecture conceptually is much like, you know, we've all bills clients over apps. We've
built SOA systems in architecture. We've built obviously, you know, we're back by Babson and in global global global scale web sites and services you obviously have storage. It is decentralized. You always lie have communications. It's decentralized. You obviously have identity also decentralized. Then it was called contracts smart contracts, which is a fancy name is a marketing name for the code the drums on this VMS in the notes. I'm not certain who actually chose to
use the term smart contracts. I believe it goes back to the early 90s and then selecting cryptography. But the phrase smart contract really means is called application Level code running on every node because there is a VM in the in the infrastructure. So then and you can run them on on a custom Client app that are also browser add-on. For example, if you use Brave or fire frogs that I don't like for example metamask as opposed to having a custom client,
you can actually engage with the D a platform like it's EDM or Neo and so on using a browser I don't like for example of matter. That's really good synopsis, if you will there is a new platform at play. Did I knew I was sleep when there is a platform that are new apps. The new apps meaning that are clearly new app patterns. And what they seemed as far as a few of those patterns, we've seen the single use case File a Bitcoin single-use case stack and application. We've seen folks in terms of JPMorgan
what's called a stablecoin? We've seen all this leading health-care. We've seen the end of the big block from players Amazon obviously clapping to Google and so on investing building these Stacks Lofts Thompson tools Turn the exposure for a few minutes and I'll be talking about perhaps maybe 10-12 minutes only supposed to stop here. Maybe questions comments. Maybe you agree disagree. Yes, sir. What is it? What do you catch? So what do I do with it? Okay, but a good just hold on for a second. I have a few more slice Donovan question.
Obviously. Yes. Yes. Absolutely. Great question. So yes think of blockchain platforms AS application platforms. Think about Cloud platforms clutch bags as Container Store apps going back up this late to websites and web services all the way through now to serverless Lambda and Azure functions. Please do think of blockchain platforms AS application platforms. We see at me a sing emerging app bathrooms, but I don't live too early in this game in the face to truly understand all of the different kind of apartments.
Public blockchain transaction rate is extremely low because of private blockchain the advantages that you have with the cryptographically secure so I can kind of go away is like one of the music database the volume and the latency of these transactions and the company was a private blockchain it what's the point? Anyway, I could just use Oracle and be done with it. Right? So I questioned hold on. I have a slight so in terms of why the apps and obviously, why do I care about public vs private as well?
Any of the comments about the way I just talked about the journey in time. So the shift in the platforms. Otherwise, I'll keep moving on. Are you going to talk about the consensus is absolutely I will talk about this as well. And what is the advantage does the consensus question in public versus private? So so let's keep moving up in the Enterprise. So why is there this early interest why are people looking at this? And by the way in the Enterprise space from what I've seen thus far it's mostly pilots and proof of Concepts. The big players of eclipse in
Morgan and Providence and so on are building our building and deploying their systems, but at the mainstream enterprise-level, I'm seeing a lot of pilot project plc's Executives of the business level. I think a struggling to comprehend the business value the other why I live at the same time. There is also a concern about being left behind and so they compromised his let's go do a pilot. Let's go do a POC. Let's see if we can understand what's going on what example of JPMorgan I believe mr. Was a demon was their
CEO of JPMorgan two years ago was publicly scathingly critical of Bitcoin blockchain two years ago last year. Is this View? And then a month ago he announced Morgan claim. And probably that is some positioning into market in terms of not giving out obviously your strategy possibly possibly going to Simone some element of learning and journey around obviously for him to understand what's going on. But that's what I'm seeing. The Enterprise mainstream is not quite that which is a good what is a good point in time because clearly there is
opportunity and the potential for us to be able to influence to create value. So a lot of Pilots almost any Enterprise customer I'm talking to right now. It was at least one or two pilot projects inside the organization. So why what is it about the apps that actually maybe help me in terms of the benefits of technology and Technical benefits, but I would not discount the organizational benefits as well as a way to look back at SOA and a couple of books. My passion for Service Plus
architectures. What's the hope that Not only was it about technology advancements but also economic change what I learned to my own detriment this that SOA and even microservices have still remained as that technology or a technical Advance things like causes law and such a dumbass of transaction Cox the way has not done much in that context but this particular platform shift promises that that's far and we shall see that's far that not only is it a technology ship that is also an economic model ship
transparency lettuce cryptographically secure. That is very clear transparency of death of the log. So for example Amazon SQL DB which is their own in-house Ledger now, it's in some sense. It's almost pointless to use something like qldb with Amazon because as the provider that's the controlling Authority if they choose to they could do all kinds of things would be with the log even cryptographically because the keys are there, but if you truly have a decentralized blockchain
infrastructure the fact that you have this business on cryptography gives you a transparency. Which is based on map meaning there is no way around it. I think it when it comes of automation. Then it's a lot of focus on contracts... What does Smart contracts with contracts in a sense of you have multiple endpoints and being able to use this diab? So smart contracts to be able to automate. And very precisely based on mathematical cryptographic protocols to be able to validate. To enter guarantee execution autonomy, and this is where the
line between technology and be a question of service technical advantages and asked me so I'm a solution architect which I am as well. So, why would I want to go away from a centralized model which gives me plenty of benefits to a decentralized model M answer is at this point in time purely from a technical perspective. They might not be for your scenario. A centralized pattern may be the most optimal technical pattern quite possibly. However, if you step back and look at the the the overall be the culture of the organization the system for example
as a Consortium, if you're working as part of the Consortium and you have multiple entities multiple organizations, perhaps there is value in this autonomy in not having a centralized decision-making body taking my private supply chain. Exchange And being able to have this level of transparency. Automation and autonomy without being at the behest of a centralized entity or an organization that as well. Security obviously need the fundamentalist the crypto fundamentals. Give us a very high-level people say in math
We Trust if you trust in that then clearly the blockchain infrastructure is something that you didn't ever go you would thrust the underlying infrastructure accountability. Once again, having contracts across the endpoints across the that the entities the participants and being able to validate verify execute guaranteed execution the some perspective in terms of thinking about the apps. So I would say from a purely as a Solutions architect depending on the domain in a scenario you may or may not because sometimes life is simpler in the in the central world is much
simpler moving parts, and we all know in the Enterprise space the happy Pappy's always happy. It's only going to be gone off to happy pad. The lights gets interesting. And so we want a Minnie Mouse Moving Parts minimize, obviously, you know that the need to stay off the happy path and so centralized models. I'm not going to tell you that that you shouldn't have to think about shifting no much like the Mainframe the PC the cloud is about coexistence. But the shift the focus will shift and the energy will ship. So some possible use cases so I can
I will not drain the whole slide we could talk for a long time. But I want to just give you a Flavour of what what I'm seeing what's going on, but it's really banking and finserv is very much a hotspot. If you are in the banking space or conserve our insurance would strongly recommend you take a look perhaps embark on a proof of autopilot. This this this is extremely extremely prolific in times of the opportunity and the word going on things like stablecoins. We all know yesterday. Gobierno Swift
and ripple is competing with that sort of this .62 getting worried obviously for good reasons, you know, if anything it's the good thing is lippolis pushing Swift to have to innovate rapidly. Because they've been quite comfortable. So triple is something to wash blockchains like stellar. Intime self at money transfer and Times of exchange example of the Facebook very very popular in Asia South Asia the Middle East and such rice almost pervasive. You have a huge number of individual accounts Antonio
have immigrants you have for people who work in foreign countries there limits back. So I believe that they're working on that essentially a money. Let me send some I can listen to WhatsApp. Possibly possibly like in Speculator yet or someone using a Facebook stablecoin. You can think about the possibilities the middleman today or the middle. I should be careful the middle of people who are in the middle of money transfer send remittances in such obviously then have to ask themselves. What is the value I provide if it's truly something like WhatsApp at the push of a
button and able to transfer across countries across continents the latest issue and that we won name. Do you think the company the organization for intensive tracking of food items you can think of Pharma and dumps of drugs and and the components of drugs being able to have the audit Trail be able to validate be able to regulate but also the flip side is in the mood for food items be able to guarantee and say look a certain way. A certain part of the world and
so was he has a higher cash value to be able to guarantee to the buyer that the does that the process of getting the tea from maybe you know somebody in Northern to get my last back to Seattle is actually legally verifiable today. It's almost impossible to to validate to guarantee across the supply chain public sector, so things like a identity. So if you look at that the UN and nations are the European Union in terms of being able to provide identities being able to then provide services seeing a lot of interest in these public sector organizations
in particular in Europe right now in particular across Europe. So check out what the EU has been doing in insulin the infrastructure energy being able to fractional ice assets like real estate and then have the money CFB democratized so they could be for example of I think an education just picking example, there's somebody a condo summer in Johannesburg South Africa You can basically have the owners have fractional ownership of the condominium and be living in Russia and China and maybe but you
can use healing and because we have the underlined infrastructure what's called off of a security token sto you able to bench provider level of acid digitization. And enabling new economics in areas that today a very difficult if not impossible and dumb stuff of enabling let alone enforcing economics is a quarter like a lot to comprehend the school of Economics. You have to know what happened before it. Do understand canes. You have to know about the depression the Great Depression. Do comprehend marks you have to understand what was happening with respect to labor before his time.
Do you understand what's happening with QE like Melvin and all this obviously we look to the Great Recession. The same way as we look at economic models that is talked about things like UTI. So yeah, basically Universal, you know, so we are sitting across the number for the government space and suddenly, you know, it's a cross to list a lot of interest on a potential. I can't see the clock. I have so few minutes. Alright, thank you Public public blockchain private
be able to host the blockchain. So for example, given supplies in scenario, you can host as helium or Leo my flight from inside your organization across a variety of notes and then Consortium, this is a very interesting scenario where you have multiple identities legally. Working together and that they want to have a stable basis infrastructure to enable us Corporation public option. North is totally private Seller said three broad recurring patterns in terms of the public. Obviously, there are challenges because you obviously have everything out there in
the public and that you might not want to share a nice, you know, it was a blockchain platform all the transactions or at least the log is publicly viewable not the contents were certainly the log and very clearly for Benny Enterprise cenarios. You will not want to have it publicly be a little private is completely closed you could say what do I can afford private wealth because you have a certain level of of verifiable auditing and the transaction log, which is not at the base of a single player currently for the most part. He look at consorcia. There's probably one player in
the middle who has the control and is able to ensure that the discorporation and obviously maybe there's a big buyer in the supply chain. But this gives every participant liability in terms of working together in times of the visibility dump set the vanity and lots of the construction as well as most folks do not know. Why do we care? Because when you have multiple entities multiple nodes multiple systems, how do they come to some agreement? I eat consensus on a transaction being committed or not in the back in the early days to
pay stomach right to Face camera. Is an algorithm for consensus? A simplified Algona for consensus and I have 16 entities. How do they come to some agreement and consensus on this is this about a transaction or do we turn it back? And that's what these things coming? So then I probably about three categories of this one is spelled p o w proof of work like Bitcoin. And the way this work is basically depending on how much computational power you have. You
can solve this cryptographic puzzles and that gives you that they the power of the ability to have a role to say in committing a running back a transaction is the very big problem here to big problems because of the amount of energy Electricity and power being used is exorbitantly expensive is a bigger problem. Proof of work has what is called probabilistic finality? Meaning that it is always the probability that the chain might be formed. Is a main chain of a fork maybe another Fork Bitcoin Bitcoin
classic Bitcoin cash. Those are the forks. So proof of work might work for possibly Bitcoin or some dark of snow together. Yes, sir. A fork is basically we're yes. You too have the ability to control that's a 51% off the competition capacity. You could actually get some under the road map for the chain. If you could ask yeah, yeah, you can actually make it go down a different path. Exactly. Exactly. So previously committed blocks. Obviously, you don't have to
appoint right? Because what do you say a fork? It doesn't mean there is a reset there is no reset up the chain what it means is beyond a certain point South Sudan blocks at eight 4 coffee beans that you know, so that's how I feel about pick a number 12 14 hours every two days you have enough commits in terms of the chain last subsequently 4th would happen only for the recent window of time. If you don't go back and you completely have the whole thing. Aquatic, that's good. But the currency is that something that is
probably Stickley final does not quite work for Enterprise in oreos. Cuz you can have a supply chain business blockchain, which says well perhaps this might actually not being the final chain. That's a big issue with proof of work. Like which often do I speak to about 10 people seven or eight? I'm not familiar with this notion that there are different algorithms another person. Can I I would say my opinion using something like proof of work for Enterprise scenario would be absolutely just be a total oxymoron. I just stopped and white that I put a
substance about, you know, the usage and the scenarios and the patterns of proof of stake which is basically not about complication. What about taking your assets if you have enough idiom or are you not near enough a Bitcoin? You can stick this asset and say look I'm going to back that. This is the chain the final chain what's called a conomic finality. Once again, not quite satisfying for an Enterprise scenario are acceptable perspective. You're probably mostly all at some point taking the algorithms class
you probably remember the Byzantine generals. I'll give them or the Byzantine generals problem of a city and you know, six seven or eight generals problem. You got to say 6 generals. They have a plan to attack a city. We have to exchange messages. They could lose their messages. The messenger could be shot could be compromised could be a traitor some generals to be traitors. And so what do you think is an emergence of a new class of algorithms for consensus-based on B ft
ft. And so for most of you always says you look at the Block infrastructure looking BST base algorithm for consensus b o w n p o s probably most likely will not be will be a waste of time for you in terms of if you don't Focus ninjas. Misting. So what is not so apparent is depending on the stack or the platform you use like for example, its idiom their own custom languages and and Frameworks and so on which makes it quite expensive in terms of Pi LaSalle proof of Concepts or building is apps
Islip decentralized. There's no Dogma that cannot be Dogma. There's a place for both but there is an emerging class of this application platforms with a decentralized. Let's use them they will not displace existing platforms Instax, but there's a place and a time for them second the block it infrastructure. If you look at the album of them is actually computer science. It's just this is just going to CS 201 and 301. And if you had a long weekend free, I would say please go take a look into it. Look at some of the code bases. Its cs201
301. It is the algo smell like the designs with them. You know, we've all come earlier Note 10, 15 20 years ago. We we we we did these things we solve these problems. It's nothing that's nothing new Under the Sun here. The last thing is there personally what I see as the potential is that you look at the you look at us as a nation or even at the world, you know in terms of inequality. significant levels of inequality which I think is not going to be sustainable. We have to have new wave in terms of how we work new ways of enabling economic incentives or
Sibley things like basic income and so on. Really good death the economic fundamentals of asleep sitting at this point with arcade games in or bb bb extreme cases to go back and think about how we ensured some level of equity and and and some level of you know, the basis for us as human beings as Citizens and lastly technology. So I would say seeing blockchain platforms as a purely technical stack. I'll give you one perspective. Bucyrus right now we're seeing have impact in terms of economic basis in terms of
organization cultural systems. Maybe we have some time for a couple of questions for Epson. Yeah. I know it's late in the evening. I know you for a long day, you're probably tired and bored and so on but that at my email by the way, is that John? You can always send me I have a question. Just want to add to one of these cases gaming. Yes. I'm familiar with me. Oh, that's right example, where is going on? Like Barcelona has got a program the Connery make the city and the services it offers to the citizens of the
City Brew Pub public permission was watching The work going on Barcelona right now around the city concept inventing. How how are we living in talking to me to get to play in the gaming world? Any surprise with any questions and it got me so let me quickly. So the question of what why would I care as that as a technologist? Maybe you don't as a purely technology play possibly maybe maybe the world is simpler in a centralized pattern public versus public or private something to Solutions basically as a hybrid, so it's not just an absolute public or private
in a public chain. You can let one go to later to suction. And so the later to is private to you. It is visible only to you. And so you can get the best of it was like it's fun. It's like a hybrid cloud like back in the days and it was very big like you could actually take it public saying and have a later to solution which is the best of both worlds. But consensus, please take a look. These are fundamental CS algorithms but be very cautious about things like Theo w m p o s t ghost they might work for Bitcoin, but certainly for the Enterprise. I'm going to
be probably the most relevant the most applicable. I want to thank you again for being nicer. Stellar is a bft base the blockchain platform of BFG. Not certain about that one. All right. So, thank you been long day. I hope there was some years maybe some value in this very some takeaways. My email is here, please you can always reach out to me. I want to thank you again and have a good evening.
Buy this talk
Access to all the recordings of the event
Buy this video
With ConferenceCast.tv, you get access to our library of the world's best conference talks.