DFINITY organized a meetup in Technopark, Zürich on June 22. I’ve used the occasion to capture Dom’s talk. We had a great turn up – given that there was a Swiss world cup game after the meetup. Dive in for the talk and some additional thoughts on DFINITY!
Video Transcription
I am in Switzerland and today I have a doctor’s appointment. And then the rest of the day is going to be filled with DFINITY calls and tonight we have a meetup in Zurich which I’m super excited about.
We are running a DFINITY meetup in Zurich today. Dom is also here. And I’m just on my way there now.
The Internet Computer – a talk by Dominic Williams
What’s the purpose of all this, why would you want an internet supercomputer? We’ve already got Amazon Web Services and Google Cloud and things like that so why do we need an internet computer?
The basic purpose is to upload to cyberspace software, data, computation, and money. So when I say “upload to cyberspace” what I mean is that the internet computer doesn’t have a geography. It doesn’t reside in some country. With all these different computers interacting using a protocol, they generate this virtual Internet computer. And it looks like a computer, it smells like a computer, you can interact with it over the Internet, you can install software on it, the software will run. To all intents and purposes, the internet computers is a computer.
Also, it means, you know, you can create things free of the direct control of a government, right? If the internet computer is in cyberspace, what regulatory purview and jurisdiction apply to it? It’s just a computer in cyberspace that you can interact with. It doesn’t reside in a country so the laws can’t apply to it.
Of course, we want people to upload you know software, data, computation, and money to the internet computer partly because we want to see an open decentralized internet. So, in the same way, the Internet’s decentralized, we want computation of data and so on to be decentralized too.
Applications of the internet computer
Now it’s part of the original vision of the internet you know. But the internet computer is also a new kind of technology stack. So it brings features and properties that don’t exist currently. An example would be internal execution. The internet computer’s unstoppable. You can write software that runs forever. There won’t be any storage APIs or files and configuration files in the internet computer. The idea is that you just write software and that software runs forever.
1. Private Software Systems
You know, this is the kind of non-technical general introduction to the internet computer. You
know there are four main applications of the internet computer. The first one is private software systems. So that’s kind of what we’ve been talking about already. You know E-commerce, HR systems, automating inventory, order management, accounting, marketing… Business software operated by the owner. It could also be government software, right. Yeah, it could be just some software like a website to pay your parking ticket or something.
2. Decentralized Business Infrastructure
The second quadrant is decentralized business infrastructure or DBI. So this is something that a lot of people aren’t really well familiar with, this sort of corporate application of this kind of platform. The idea here is that an industry will create an open shared platform. An open platform that’s shared by all the participants in an industry in order to create some synergies or derive some efficiencies or releasing liquidity. Serving the supply chain is something that’s talked about a lot about. Foxconn has a big ambition to create a supply chain for its electronics industry because it has like 20,000 suppliers. You know a lot of these guys want more liquidity. So if you can put everything into a supply chain that’s trustless, you can invite people in to finance invoices.
There are lots of other applications like this. There’s a tracer chain involvement which is a big Tobias Group of companies and ROX’s joining so that looks like about 85% of the world’s diamond suppliers. The purpose here is to track the movement of diamonds from you know a mine in Botswana through to a wholesaler through to a cutter in India through to a jewelry factory in Paris through to Tiffany’s in London.
3. Decentralized Internet Services
The third quadrant is decentralized internet services. This is really about creating open sourcing businesses. So the idea is you know you take a service, you tokenize it, you create autonomous software that exists independently of any organization. You know the tokens can be used in the governance system to update the software. So it’s like as an open-source version, a business version of an open-source software project, right. So the actual service itself is open sourcing. And you know the governance system can you know put out bug bounties or you know badges for the addition of new features. There are a bunch of reasons in the first instance why that will be attractive. Now, this will take a bit longer to happen but it’s gonna happen.
In Silicon Valley, there’s a thing called platform risk. Platform… you know VCs are certainly lookout out for platform risk. What it means is that you know you build some startup, it’s going gangbusters. But you have a dependency on some other organization. And that other organization is a cause of concern to you. And this happens all the time. So a well-known example is a company called Zynga. Who’s heard of Zynga? Like a social games company that created Farmville. Know Farmville? It plays a huge part in like social gaming on Facebook – they have 100 million users.
Zynga IPO’d and it was going great. Unfortunately then, for them, Facebook decided they wanted to change the mechanisms, the rules that governed how social games – companies like Zynga – could promote their games. And the moment they did that, of course, Zynga fought much harder to promote its games. Now it’s at like a 1 billion valuation, something like that. Zynga was going gangbusters, it had a platform risk. In this case, it was building on top of Facebook and Facebook pulled the rug from under them. And there are so many stories like that in Silicon Valley. Yeah, I could just go on listing them hour after hour. It’s a huge, huge problem.
The reason that open source, tolerant software services you… so open source businesses that are running on top of software will be interesting is that those services could provide much better guarantees, for example, about their APIs.
So imagine you have an open source version of LinkedIn. I can tell you the story about a company I knew from Palo Alto – it’s called RelateIQ. It was a unicorn, there was this really cool service that would look into your company’s software, email and messaging systems and create like a graph of all the communications inside your company. You could like hover over the vertices in the graph and a little profile pop up. So you could see who was talking to who, who was waiting to get an email from somebody and this worked outside the organization too. So it was a really useful tool. But of course, they depended upon the LinkedIn API to get information about the person. So when you hovered your mouse over there vertex in the graph, a profile popped up. Billion dollar company everyone is super excited about, LinkedIn decided that it was going to withdraw the API and the only people that can query the LinkedIn API were gonna be Microsoft and Salesforce. So you know RelateIQ did the only thing that it could and it sold itself to Salesforce for 240 million.
So if you imagine a world where all of these internet services, or open-source businesses, right, create using autonomous software… Well, if you’re an entrepreneur or an innovator where would you want to build your service? Would you want to build your service on the traditional internet, on Amazon Web Services and integrate with the APIs of Linkedin? Or would you rather use and build on the internet computer and integrate with the APIs of some open source business that gave you much, much better guarantees about the availability of that API.
So what you get is a kind of mutualized network effect that will become very, very powerful over time. Once the internet computer, or if we don’t succeed, some other, you know, equivalent really takes off, the network effects will become overwhelming. All of the innovators, all of the entrepreneurs are going to want to build on the internet computer. They get guarantees about the rug not being pulled from underneath their feet. A much better environment for a startup or someone who’s entrepreneurially minded, right.
The other reason that the internet computer will be attractive for building actual Internet services is that we have technology that will create crypto lock boxes. What that means is you can design software that keeps its data or parts of its data within a crypto lock box. Now there are different ways of designing this. If you want to, you can design it in such a way that no human being can ever get the data. Inside the crypto lock boxes is plain text but you can’t get in.
A simple application – imagine an internet dating application. Right, people forget that something like Tinder very likely will possibly or lots of them, right, the data could be actually and hugely embarrassing. This kind of technology can put that data in a crypto lock box. So they get… the user gets guarantees. That’s another reason why people will prefer to use internet services that aren’t hosted by a company, right. Where your private data is residing on some server that that private company runs.
All of these internet dating sites – if their team makes one misconfiguration error in the network, the hackers will be in, the database will be out. If there’s one disgruntled employee, hackers will be in, the database will be out. And it’s all just waiting to happen. People could be on it in their twenties and then in their forties and then you know all of a sudden the thing’s out.
And at least in my view, you know, people should have a right to a private life. So I think you know the next thing we’re gonna see is that you know users begin to demand, as there are more of these hacks taking place and there’s more misuse of user data, people are gonna demand guarantees, stronger guarantees about how their personal data is used. And also about privacy, right. They’re going to want to know why their private data isn’t going to become public entertainment photo or worse in the future.
4. Decentralized Finance
So finally we have like decentralized finance. And you know obviously DFINITY’s not built for ICOs but no doubt people want to run ICOs on DFINITY. It’s a really popular application of this kind of technology. But you know we’re also interested in much more serious applications like borrowing in London, insurance…
I think the… an awful lot of this will come after some of the other stuff, right. Because to do decentralized finance you really need what we call crypto fear. Which are… which is a currently you know a token, are a token that mirrors traditional fear. There’s a project called Fi and it’s designed to create through borrowing and lending – works the same way as commercial banks – creates what we call crypto fear to you know FiUSD, FiCHF…
Governance
Although you know there’s a case for not having a governance – like Ethereum, you know, the code is law – DFINITY does have a governance system. But you know DFINITY is, remember, a decentralized network. It’s meant to be the internet computer, it’s not meant to be some company’s computer. We need to have a system that doesn’t run counter to that aim. You know we can’t have governance depending on the DFINITY foundation. The DFINITY foundation is just a not-for-profit foundation, it was funded to develop a technology. It’s to help develop the network and so on but it’s not our job to make decisions about who’s running on the internet computer. The whole point is that it’s an open… it’s an open public resource, right.
Our solution is to use algorithmic governance. So if you ever think of the Blockchain Nervous System, it’s not nearly as complex as it sounds actually. Ironically, the Blockchain Nervous System is probably by far the simplest bit of DFINITY.
The Blockchain Nervous System
You can use it as a liquid democracy. You guys can imagine starting off with something like you know where you’re delegating your vote to somebody else. It kind of will work something like this: you can create a neuron which is like a voting entity. Create a neuron by depositing some DFINITYs. The voting power of the neuron is proportional to the number of DFINITYs that you deposited in it. We create an incentive to make sure the DFINITY network’s well run by requiring that you can only get your DFINITYs back, your stakes inside the neuron, by dissolving a neuron which takes a year. It’s not going to be three months but it’s gonna be a year.
So, you know once you’ve created a neuron, so you can participate in voting on proposals, you can’t get that… you know voting power is proportional to the number of DFINITYs you’ve staked. You can’t get them back without dissolving a neuron which takes a year.
So you sure as hell want to make sure that every proposal that goes through increases the value of DFINITYs, right. Which is a kind of proxy for adoption. But that’s the objective of the Blockchain Nervous System.
Why would you do this? Well, you actually get paid for voting. Nonetheless, many of the things you’ll be voting on – you don’t or will not have the expertise to vote on or have no interest in voting. So you can configure your neuron to vote automatically. Process all kinds of different topics. You configure the neuron differently for different topics.
So let’s say there’s a cryptography topic. Not many people have the expertise, inclination or interest to try and vote on cryptography. So what yo do is configure your neuron to follow some other neurons. The idea is that your own neuron follows a public neuron or a private neuron. A public neuron has just one DFINITY in it. You would advertise the address of your public neuron. It’s a bit like a public key company, right.
You can advertise on your Twitter, on your reddit… You can say “right, here are some people, here’s a world of cryptography, you know here’s some academic, here’s some works at DFINITY, here’s an investor in this kind of thing…” And I’ll say “right, I’m gonna configure my neuron to watch these five other neurons”, right. On the cryptography topic. And if three of those neurons vote “Adopt” on a proposal, my neuron will follow an “Adopt”. If three of them vote “Reject”, my neuron will following a “Reject”. If neither of those things happens after time out, I’m gonna vote “Reject“. Simple enough. Of course, the effect is that you know generally speaking proposals will come in and then the Blockchain Nervous System will kind of cascade to a decision.
The role of the DFINITY foundation
Anyway – who is pursuing this? It’s the DFINITY foundation, a not-for-profit foundation. It’s all scientific research, system implementation and of course, you know when the network’s up and running, we’re gonna stand behind the network and we’re gonna be trying to mitigate threats, provide network support… We’ll also have an Ops Center that works with people that need to submit Blockchain Nervous System proposals. So if you’re a corporation and you wrote all this software and installed it on the internet computer and it went wrong… Because you know you might have done it in such a way that it’s very locked down for security reasons and you need help from the Blockchain Nervous System to fix it, right. The DFINITY foundation will help out, help people create proposals and help you get them through.
Currently, now we’ve got the biggest research center in Palo Alto. We do some work with some guys in Lausanne. We’ve got admin… admin central in Neuchatel. Most people are in Zurich. We’ve got a new lab going up with Jan Camenisch – he was a principal research scientist at IBM – just joined, he will be running that. We’ve got a guy in Cambridge, again we’re gonna try and build a lab there. In Germany – this guy Andreas Rossberg designed the WebAssembly. We’ve got three guys in Tokyo of all places.
So we’ve also got our own venture capital fund. So we invested in people… you know things that have been built on top and sort of generally you know we’re trying to build up what we call… there’s a bit of a silly expression but it’s the NASA for decentralization. Because we believe that if this internet computer thing’s really gonna succeed, in order for anybody to trust it, there needs to be a very substantial organization that stands behind it. So people can be sure that you know that in every time zone in the world there’s a big team of people monitoring the network ready to jump into action the moment that any kind of issue arises. And there needs to be a team of people helping people submit proposals to the Blockchain Nervous System, it’s a completely new thing, right. It will take some years before people become familiar with the process. There needs to be an organization that you know organizes developer conferences and things like that.
So I actually think that you know when the internet computer is running at scale, you know the foundation itself if this project works, could have thousands, could employ thousands of people. So you know we’re scaling up as quick as we can, we’re growing all the time and there’s a jobs page on the website.
You can listen to the audio version here:
