Here's the content from our Tampa Bay Blockchain Devs meetup last weekend. You can find the group at https://www.meetup.com/Tampa-Bay-Blockchain-Developers-Meetup/. Github at https://github.com/TBBD/crowdsale.
In the process of thinking and building Pocket Network I have been obsessed with learning about proper cryptoeconomic design.Reading through the recently announced Plasma white paper inspired this post. It has nuggets of cryptoeconomic definitions and analysis like this:
This assert/challenge construction allows one to assert a particular state is correct, and if the value is incorrect, then a dispute period exists where another observer can provide a proof challenging that assertion before a certain agreed time.
About a week ago I came across this talk by Vlad Zamfir -
I watched the video, and he mentioned literature in cryptoeconomic design. I was curious, and asked him about the kind of literature he was talking about:
Vitalik never answered back, but I started to compile a list of academic papers, books and blog posts that involve blockchain, cryptography, game theory and distributed systems research. This is really just scratching the surface - every academic paper has a wealth of references that lead to more academic papers.
I'm finding that reading the papers closely (It usually takes me a couple read-throughs to internalize) has been the most comprehensive way to learn this stuff. Here is the list of work that I believe can give you a decent base in at least thinking about cryptoeconomic design for your project (in no particular order):
Blogs (These are prolific writers - reading through everything will take many hours)
- Richard Gendal Brown
- Ethereum blog (Mostly Vitalik, start from the beginning)
- Nick Szabo
- Vitalik's blog
- Werner Vogels
Distributed Systems Papers
- The Seven Deadly Sins of Distributed Systems
- Theory and Practice in Large Distributed Systems
- Dynamo: Amazon’s Highly Available Key-value Store
- Paxos Made Live - An Engineering Perspective
- Spanner: Google’s Globally-Distributed Database
- Bigtable: A Distributed Storage System for Structured Data
- The Google File System
It's a lot. It would take many years of study and practical application to reach any kind of real proficiency in this field. Since we're standing on the shoulders of giants we can design novel cryptoeconomic experiments on Ethereum at a small scale, iterate and learn from our own and others mistakes.
I wrote this as a resource for everyone. If you think I should add anything you can find me on twitter or email@example.com.
I host the TBBD meetup for people interested in learning blockchain development. Right now it's focused on Ethereum, but excited to explore other blockchains as well. This is a video of a demo I did last Sunday deploying a hello world contract to the Rinkeby testnet. Credit goes to @abuiles and his series of blog posts starting with https://blog.abuiles.com/blog/2017/06/13/smart-contracts-for-the-impatient/.
Link for the meetup is here - https://www.meetup.com/Tampa-Bay-Blockchain-Developers-Meetup/
Did my first public talk on blockchains. Crowd was about 400 people. Ignite is an awesome event in Tampa with all sorts of inspirational talks, not just tech stuff. It's a great organization that aims to support entrepreneurs in the area.
Below is a quick demo I made the other day, explaining how a Relay will interact with Oracles in the network.
In my previous post I explained what kind of on-ramps there are into crypto, and how one of the major bottlenecks right now is lack of mobile development in the crypto world.
There are several legitimate reasons for this:
For security reasons iOS and Android developers are primarily limited to HTTPS requests
Running and syncing light nodes on a mobile client is slow and a terrible user experience
Centralized solutions for companies managing keys and wallets have high security and KYC/AML costs for indie developers or small teams
As a result the promise of decentralized peer to peer applications have not been fully realized on our mobile devices yet. If one were to imagine a network enabling a world where we can easily interact with cryptocurrency on our mobile devices today, a mobile developer would need all of the issues above abstracted away.
If a mobile developer downloads an SDK to leverage any blockchain, they need -
Strong client side security
Proper wallet functionality
Easily construct, sign and send a transaction
Reliable endpoints to send transactions to
The interesting problem in the above list is having reliable endpoints to send transactions to. What incentive do people or organizations have to host those endpoints? Right now, there are two major centralized services that offer endpoints to propagate Ethereum transactions - Infura.io and the SOFA protocol.
Although free for now, these services likely will act as a tax on developers wishing to use them. Not only that, these services are not flexible to what a developer may want. Since they only support Ethereum, what if another blockchain supports some sort of functionality that Ethereum doesn’t?
What if you could pay a node to relay any cryptocurrency for you? They just need to be running a node for your chosen cryptocurrency. This begs the question, what would incentivize a node to do so?
The incentive alignments with a tokenized model is perfect for this. A dynamic market where fees for a given cryptocurrency are based on how popular it is (Ethereum and Bitcoin would be the most popular). Nodes receive the token as payment for running a node in your chosen cryptocurrency.
This is powerful because it is decentralized and flexible for any cryptocurrency. A mobile developer can just pay a fee, and send any kind of transaction in simple JSON format.
By abstracting the need to hold any cryptocurrency in a centralized database these properties alone could allow a flowering of thousands of different applications leveraging the blockchain to be created. With no need for a traditional backend to write to, the developer can write to the blockchain and authenticate via the client side private key. Not only that, but such a system is blockchain agnostic - as long as there’s an endpoint running a node in your selected cryptocurrency, any can be sent.
Imagine logging into Clash of Clans and earning real tokens while playing. The developer announces a huge new release that will make the game 10X better. The next time you log on the amount you've earned has appreciated 100% or more as a result.
Making it as simple as downloading an SDK for mobile developers would increase the surface area of entrance points into crypto, without users of their apps necessarily realizing it.
Onboarding the average person into crypto is hard. There are a limited number of ways to purchase your first cryptocurrency or interact with a decentralized application. Right now, anyone who owns any kind of cryptocurrency is part of a self-selecting group. To have the intent to download a wallet program and have the technical awareness to not be overwhelmed when buying crypto is a fairly large leap for most people.
Current ways to onboard people into cryptocurrency -
Exchanges (Coinbase, Kraken, etc)
All in one messaging style apps (Token, Status.im)
Ethereum DApp browsers (MetaMask, Parity, etc)
Sending crypto directly to someone (word of mouth)
Coinbase is the gold standard as the onramp for non-technical users and is absolutely needed in the ecosystem. Token and Status are ambitious projects but are single entry points into the Ethereum DApp ecosystem with an uphill battle to gain traction competing against WeChat, WhatsApp, Messenger, etc.
MetaMask is one of the projects that I’m most excited about - they are widening the surface area of potential entry points into the ecosystem. The more browsable DApps that get made, the more useful MetaMask becomes; a compounding utility that just needs time to come into it’s own. Regular people are accustomed to downloading a chrome extension, creating a password and you can instantly interact with Ethereum-based DApps. Relatability is important when trying to bring crypto to the masses.
Outside of word of mouth, most people will not be onboarded into the crypto economy unless they are an early adopter and explicitly seek it out themselves. Much hay has been made about the lack of a killer app for crypto. This is untrue. We have two (primitive) killer apps:
Holding bitcoin enables self-sovereign control of your money
Ethereum has enabled a thriving funding environment for hundreds and soon to be thousands of radical experiments
With all that said, there is one important key missing to unlocking the mainstream crypto puzzle - mobile. Inability to create truly peer to peer applications on our mobile devices using crypto is one of the biggest bottlenecks holding up mainstream adoption. When everyday people can start using cryptocurrency on their mobile devices without necessarily even realizing it, we’ll have broken a major barrier in adoption.
Lack of mobile access is the biggest hinderance of adoption in the blockchain space right now. Status is a group that’s got the right idea - so many incredible ICO’s and protocols being built, but at their core they cannot serve a world that is mobile first.
There’s a couple reasons for this -
- Frankly, we cannot reasonably scale to a mobile, global world (a good thing right now). Lightning network, sidechains, sharding, proof of stake.. We’ll get there.
- P2P transactions on these new protocols are impossible in the iOS and Android stores. The apps that offer services (Coinbase, Blockchain.info) have centralized servers that serve as a central point of failure.
As an iOS developer, I want to be able to download a library that gives me the ability to create a decentralized app using any blockchain I deem useful. I don’t want to depend on a centralized service to be the payment processor. I don’t want the responsibility of managing wallets, security, auditing, KYC etc.
The question is, how can we enable Decentralized AirBnB, Decentralized Uber, Decentralized Tinder or Decentralized Soundcloud; By decentralizing would bring costs down for the customer or money into the pockets of the creators?
This is a problem that my friends and I have been deeply thinking about solving. We’re starting with one basic fact - Any mobile device can sign and send a transaction over HTTPS. Client security is incredibly important, but should any security leaks happen, it will be contained to the device itself.
I’ll continue writing about this as we flesh out the model a bit more deeply, but we’re excited about the possibility of enabling the ability for any mobile developer to create a decentralized application in today’s existing infrastructure.
I've been interested in the cryptocurrency space since I read Satoshi's white paper about 4 years ago. I was mainly an observer and hodler while being generally excited about all the amazing things the core developers were producing. I remember being at the Bitcoin Miami Conference in 2014 and watching the announcement for the pre-sale and thinking to myself, "This must be some vaporware scam". That was not uncommon at the time (Side note - incredible to see how much the industry has matured since then!).
When I learned to program a little over two years ago I gained a new appreciation for Apple. The ease in which someone with zero coding experience or technical knowledge can get a basic app running on their phone is pretty incredible. I believe this went a long way in the incredible growth for the language for new and experienced developers alike. It's a big reason why I believe Chris Lattner when he talks about Swift's world domination plan.
To be fair, Ethereum and cryptocurrencies in general are far from Swift/Xcode levels of drag and drop simplicity. That being said, when it comes to getting your hands dirty, creating meaningful code and seeing the results quickly the Ethereum community is leading the way. In the spirit of this, I'd like to share the resources that I've come across that have been most useful for me in learning over the last month.
I'll be under the assumption that you know how to work your way around the Terminal and are familiar with how cryptocurrencies work at a high level.
- You need to explore the Ethereum.org and Solidity websites first. They have great documentation with code to copy pasta and use yourself. I want to highlight the create your own cryptocurrency page in particular. Reading through that really helped me understand how Ether and gas are used to create and send custom Tokens.
- So you've read through all the docs and have an understanding on with how Solidity works. What next? I learn quickest by watching videos - I've watched almost all of the create your own token Youtube videos. By far the most useful one I watched was Jordan Leigh's two part video called Introduction to Ethereum Smart Contract Development With Solidity. You get a great primer on having a good workflow for developing DApps along with great random pieces of information.
- On his Youtube page Jordan has a few videos that end up linking to a Decypher.tv website where he created a web series that you need to pay 3 Ether to have access to the entire series. They are fairly short - between 7-15 minutes each. He flies through the content so between pausing sometimes it would take me up to 45 minutes to finish a video. It's useful content but 3 Ether seems a bit pricey for me. He also says he'll do 20 videos (10 are up) but I'm not sure what his timeline for following through with that is. Using Metamask for the first time is a cool experience as well.
- You've started coding and compiling your Solidity contracts and having issues debugging? Etherem Stackexchange has already been invaluable for me. You should read through the most popular questions and answers too. I've been lurking there hoping there are questions that are simple enough for me to answer, but have not completed that quest yet.
- The advanced examples on the monax.io website are incredible and mind blowing. The series from Jordan pointed me there and it's one of the best resources for advanced contract building you'll find.
- It's important to keep up with the community and for me, Twitter is the best way. I am extremely excited for Casper to finally be released. I have curated a broad cryptocurrency list with almost 300 people primarily from bitcoin and Ethereum. If you don't want to deal with the noise (there is a lot) I recommend subscribing to my Nuzzel Newsletter or curating your own. Nuzzel is an amazing product and the newsletter shows the most shared links on a given Twitter list so you won't miss much.
These are all the tabs I have open when I'm in Ethereum mode. Feel free to reach out on Twitter.
Figuring out on-boarding for your product is hard. It’s the first time someone experiences it and first impressions matter. It’s one of the things we keep coming back to — I believe that it’s one of the most important parts of an app. It’s in those first 30 seconds or so that you need to convince someone that there’s a reason to come back.
Building a social network, some of the factors that my team and I have talked about are —
- How long should the on-boarding process be?
- Where are they at that moment they download?
- What kind of investment do we ask of our users to give them a reason to come back?
- How are they feeling? Are they sober, drunk, sad, excited?
- How much do we teach them about what we think makes TimeSet special?
- Should we encourage them to create content for us right at the start?
- Do we emphasize our network, and the people using it?
- Should we encourage them to subscribe to BucketLists and/or follow users?
- How much does the user know about our product when they download it?
So far we’ve tried several things with our on-boarding. At first it was a simple sign-up. No real on-boarding process, but when you reached the first screen we had some quick tutorials pointing to various parts of the app that you could read or tap through.
We realized we weren’t asking anything of our users when they first logged in — so we decided to go to the opposite end of the spectrum.
Our core functionality is taking a photo to complete a goal in a BucketList. We created an elaborate explanation of what a BucketList is, and heavily encouraged our new users to post to our “On-boarding BucketList”. It had 3 goals:
- Take a photo of what’s in front of you
- Take a screenshot of your home screen
- Take a screenshot of what you’re listening to
We assumed that because these were such simple, easy goals that it wouldn’t be too much of a hurdle to complete when someone is using the app for the first time.
We were wrong.
We noticed immediately that encouraging someone to post to a social network that they know nothing about, with no understanding of the context or consequences of that post was a bad idea. It was too much of an ask. It made people uncomfortable, and many times said, “Wtf why am I being asked to post something?”. So we axed it and went back to the drawing board.
Our next iteration will be much more simple and much less of an investment for new users. We don’t have many users, and our strength is the small network that we’ve created. Most of us post every day, and check the feed constantly.
We asked ourselves,
We ARE a social network, so why don’t we bring users into ours? They can see the awesome, funny stuff we post and feel like they are a part of a community right when they log in.
It made sense after 2 other tries of messing it up. In our next update, when a new user signs up we will show them two screens:
- One of our most popular, active and interesting BucketLists where they tap subscribe to the ones they are most interested in
- Our most active users, where they can tap follow so their posts populate the feed as well
On top of this, when a user visits or creates a BucketList for the first time, we will play a quick tutorial that is easily digestible and highlights the core functionality of a BucketList. We believe this approach provides several benefits to a new user:
- Most importantly, it highlights our community of people posting who have created interesting BucketLists.
- It allows their feed to be populated with goal completions and posts right from the start.
- Subscribing and following are familiar, comfortable actions.
- It gives a new user a reason to come back a few days later when someone has completed a BucketList (We send a notification when someone has completed one for the first time).
We hope that this new on-boarding will increase the odds that a new user “gets” TimeSet, regardless of where they are experiencing it.