Moderator: Jen Taylor, Head of Product, Cloudflare
Photo by Cloudflare Staff
JT: Tell us about what BlockChain is
JC: Going back to 2008, advent of blockchain came with bitcoin white paper.
The word Blockchain wasn’t mentioned at that point, but that was the advent of this tech.
What it solved was niche problem called double spend problem. Creation of digital cash.
What you see in a bank account isn’t digital cash. The problem in cryptography was how to create digital cash that doesn't rely on 3rd party intermediary. This is what Bitcoin created.
JB: Blockchain packs in lots of stuff: useful as brand. Like internet/web in early 90s, the meaning is fuzzy.
Properties that all of these apps have in common:
Academic definition: A blockchain is an indelible chain of blocks; once you insert information into one of them it remains.
Marketing definition: many applications have been developed over last few years, all have to do with public verifiability. Reliance on cryptographic methods to achieve goals on clearing payments and the ability to check and verify.
Across the board, removing 3rd parties from equation. Establishing publicly verifiable state of structures. Trust protocol removes trust needed from individual parties.
It points to a return to what people called for in the early 2000s. Decentralization of the power structures that control the internet.
Removing power from entrenched places.
JT: you’re both doing great work with organizations looking at moving blockchains forward. What is currently happening with this tech?
JC: I work with Tezos; a blockchain protocol and platform used to build decentralized apps. Hearkens to a concept of a hard fork of a blockchain
Hard fork: 2 different assets of bitcoin, assets and cash
Comes back to idea of decentralization. Decentralization offers many things; one problem it raises is how you push upgrades to the tech. Generally there is one centralized party. With blockchain it’s different. Lots of politicized infighting among communities and users of tech; Tesos seeks to solve this infighting and enable coordination.
If everyone here owned one Tesos token, everyone would have one vote as to how the roadmap proceeds.
Also seek to innovate on formal verification of core base of protocol to make applications more easy and accessible.
This comes back to the language we’ve chosen for the protocol and implementations on top of that. The tech will underpin trillions of dollars worth of industry, and it should be built with that in mind.
JB: we works on IPFS and …
IPFS is a decentralized hypermedia protocol.
Think of the web, and if the web itself had no notion of locations or sites but was more decentralized than now; content would not be addressed by where it is and who owns it, but instead by what the information is itself. The same information would have the same address. This isn’t how the webo works now. Today that’s not the case. We want to rethink the stack for how the web works: content addressing rather than location addressing. Peer-to-peers structure.
Think about how easy is it for content to become hypercentrialized and censored;
Also efficiency: channels of low bandwidth and so on.
If we can move entire sections of the web to a remote location and serve them at protocol level, take what we’ve learned from CDNs and build into the protocols themselves.
Finally, it’s a way of thinking if you have decentralized way of creating protocols that organize work in a public network, can you organize a system to store data for all of it. A utopian decentralized market where storage is proper commodity? Allowing ISPs to participate in cloud storage.
Today we have a hypercentralized storage system as well.
JT: The power of decentralization could really change the world. What are some of the other benefits or uses that we could apply this tech to?
You get to work with the community in such a rich way; what other use cases?
JC: Inspiration from investment bank. Started off as a bond trader.
The real innovation is just not about decentralization like bitcoin but also reshaping entire market structures.
Reshaping entire market structures that today depend on rent-seeking middlemen. Logical conclusion of this is: redefining what it means to own something in digital form.
Today we don’t really own anything that is in digital form. BoA database represents my ownership. So i get excited thinking about how completely different market structures will look in a couple of years.
JB: At its core, this has to do with establishing decentralized computing platform where you can run programs and encode business logic and where participants can’t overturn results, and there is no litigation over the events that take place. What happens to law when you can express legal agreements in a digital context.
Transactions are easier if you don’t have to draft agreements and think about them in depth every time.
The major innovation with blockchain is that law and finance were right away ripe for changes, in terms of investments and ownership.
You have the first real wave of smart contracts, finance and law are immediately being changed.
Potential is massive: you can change pretty much everything, how we reason about markets and providing services and utilities. This is the first public utility that is completely international, governing themselves.
You can run all kinds of services: cloud storage, cloud computing changed in fundamental ways.
That said, it will take a while. UX is still atrocious. Quality of platform is bad relative to modern standards. Considering migrating an application into a different context is almost a non-starter. The tech has to catch up with banking to enable developers to change how they maintain applications.
You’ll have developers able to create something like Twitter, put into into the network, and never having to worry about maintaining it anymore, because participants will. Completely different way of approaching development.
Now is the right time to get involved to help develop.
When you have a 100-line code of contract, you want to leverage all you can to make sure you get the right answer.
JC: Precisely because there are no 3rd party intermediaries to call about reversing an action.
Q: I am an investment banker, but i don’t understand what mining is.
JB: When you think about decentralized consensus protocol, where a bunch of parties are proposing values for the head of the chain, and they have to agree upon what that value is.
Mining is a way that lots of work/resource expenditure have to exert in order to propose a vote on value.
You have a whole bunch of people with computers hooked up trying to give one value weight and declare a winner.
It’s like a voting system where you use resource expenditure...
JC: proof of stake algorithm vs. proof of work system
On any blockchain network you need a validator who verifies certain things about transactions and then is broadcasting that badge or block to the network. The validator gets elected based on how much computational power they are putting into the system.
Next generation of systems will use proof of stake, where election process relies on how many tokens you have: creates new incentive structure
JB: we found a way to resource expenditure with a valuable side effect: mining is useless otherwise. It’s useful insofar as it lends weight to your proposed value, but no value outside of your company.
We found a way to use valuable storage of files and computational work which shows that resource expenditure is actually proving network that you have stored files. We think proof of stake is valuable area of research in the future.
Governance of these systems will evolve dramatically over the next years.
Q: I’m curious about how people are thinking about preventing recentralizing things, e.g. smart contracts. Like in agreeing on price of wheat over a given day, Everyone has to agree on what the price of wheat is on a given day, and certain nodes have more power. Secondly, how are you thinking about preventing recentralization as you are going through these processes of decentralization.
JB: There are many things happening that might cause recentralization; Oracle solution is solid if you have verifiability and if you know they can’t charge exorbitant fees.
Approach is to decentralize in pieces. A good enough solution for now and then go back and decentralize more along the way.
We think about it in terms of storage providers or distribution providers: how to carefully structure things to get as much value as possible.
JC: running joke in crypto space is that crypto-currency has created far more 3rd parties than it has destroyed.
A new protocol has to be very specific about the problem it is trying to solve. If i am in Venezuela using crypto-currency, the trust problem I’m solving there is different from the trust problem in file storage. One protocol won't solve every trust problem.
There are different trust problems and there is no “one protocol to rule them all”
All our sessions will be streamed live! If you can't make it to Summit, here's the link: cloudflare.com/summit17