(1) Composability Series: Web3 Products, Protocols, and People
The future of work is in crypto/web3 due to the composable nature of people and protocols on Ethereum.
I know the article title is a very strong statement to lead with, but over the last few years I have witnessed and personally experienced how building products and businesses in crypto (namely on Ethereum) is unlike anything else I’ve ever seen. Technically, decentralized applications (dApps) or decentralized finance (DeFi) are the best examples of everything I’m going to discuss — otherwise called Web 3.0. Don’t worry if you have no idea what that is, I’ll build up to it in a non-technical fashion.
Also, my point here is not to say crypto is perfect (it’s still far from it), but what crypto is starting to enable is absolutely game-changing. While this is a non-technical article, I’ve linked many technical explainers throughout I highly recommend you read.
What do community and product typically look like?
I’ve worked in a variety of different industries over the last six years. This includes business roles at a Big 4 consulting firm and bulge bracket investment bank, and also developing data science models/products and AR/VR products. I’ve even spent time freelancing on Upwork, delivering on Postmates, and teaching/mentoring college students.
Getting into the community and then finding others to work together with to build up something new are two constant experiences no matter what industry I’ve jumped into.
In consulting and finance, it goes without saying getting into that community is incredibly difficult — I made my way in mostly out of privilege. Even once you’re in, creating and getting a “product” or service to market takes a ton of buy-in from all sorts of parties (both internally and with clients or public entities). Most of your time will be spent speaking with or reading reports written by “thought leaders” to get to a new insight or framework.
Data Science is a large improvement, the community is open to anyone who has access to and internet. And the nature of open source and platforms like Medium and YouTube means you can learn pretty much anything you want and built up to the basics very quickly. But moving up from there is insanely difficult — for Data Science you need access to quality, labeled datasets, sizable budgets for training and monitoring your models, and deep expertise in mathematics and statistics that takes years. The path can honestly be lonesome at times.
For AR/VR it’s a little hard to say since the space is still pretty young and communities are small and scattered, but for a great product you need to know (or know people who know) Unity/Unreal Engine, spatial design, have someVR headset, and be familiar with blender 3D modeling and rigging — which for the last one you do have people likeIan Hubertto help (thankfully). For AR you have some community aroundAR Spark, but it isn’t at a largely collaborative or open-source point yet either.
For most industries, it is hard for a newcomer to go farther than surface-level/education in terms of community, and building or contributing to a product requires a lot. Lastly, most industry communities are not all working towards one goal — that is to say, they are not aligned and don’t really have a need or force to do so. That isn’t to say they aren’t making progress or the community isn’t strong, but you’ll see what an aligned community provides as you read onwards.
Crypto products are just built different (literally)
It all starts with the architecture of the product. If I wanted to build a new Reddit, I would have to start from scratch with my own backend, frontend, and hosting. The good news is along the path to Web 2.0 we’ve heavily built out open-source software and IaaS/SaaS/PaaS, meaning building Reddit now would be a lot easier than building it ten or more years ago. This leads to a high level of flexibility in decisions and speed of development, such as choosing a ready-to-go tech stack like this:
Ethereum has the same flexibility since we leverage a lot of the tools from above, but we’re all building applications on the same base layer — you can think of it as a shared backend. For reference, the application stack for decentralized exchanges (think currency/FX exchanges) looks like this on Ethereum:
Here the “Base Layer” is the AWS hosting and data storage equivalent, the “Smart Contracts” is the backend, and the “User Facing” portion is still the frontend. Those are not one-to-one comparisons but helps create a mental bridge. However,noneof these components are siloed or owned by one person or entity — it belongs to anyone who owns a drop ofEther (ETH).The code for Ethereum base layer and any smart contract can be seen and verified by everyone —so this isopen source on steroids.Anyone can build a new frontend by connecting to thesmart contract APIs, anyone can re-deploy an existing smart contractwith tweaks and call it their own, and anyone can suggest and develop an improvement to Ethereum throughEthereum Improvement Proposals(EIP). Ethereum is “hosted” through a network of nodes, so there’s no messing with AWS or Heroku for you.
In terms of mutability, it is easier to redevelop the user-facing interface compared to the smart contracts layer, and easier to upgrade the smart contracts layer than the base layer of Ethereum.
All together, we like to call this product composability (or legos).
You could start from scratch on any layer, but also directly plug into whatever others have already built. I do want to note parts of Web 2.0 are composable, such as map apps built upon OpenStreetMap. However, a large majority of them are not fully open and require using custom-built APIs where the provider can change the rules of access at any time. So while Web 2.0 may look like legos, it’s like building with legos that are sometimes flat on top or have a constantly shifting available surface.
Smart contracts are where most of the app functionality is built, and every public function on it automatically becomes an API endpoint anyone with an Ethereum address can call — even by other smart contracts. These can’t be changed or controlled by even the deployer, what you see is what you get.
This composable architecture changes three key things:
No one owns the data.
This is huge, and one of the ideas that originally brought me to the space. Say, for example, I was a walker for Wag. If I wanted to start walking on Rover instead, I’d essentially have to start all over in building up a walking history and client reviews — none of the data would carry over. If these apps were built on Ethereum instead, then I as the walker can carry my services to whichever platform suits me best since the data exists on the blockchain itself. If you’re curious about how this would be done, I actually built this out with a few friends already over a few weekends.
Data interoperability for social networks has been a big point written and discussed for years now, with Facebook, Google, Twitter, and others owning the network data (on top of a lot of other data points). It’s not to say we wouldn’t end up with winner-take-all monopolies, but starting up a competitor would not be nearly as difficult. No one owns this “backend”, we all have full access to the data and use of it on and off-chain (you can explore data on Ethereum through blocks and transactions or even run SQL queries or GraphQL queries).
Yes, some things like personally identifiable information (PII) should remain private — and the space is evolving to allow for that while building solutions for decentralized identities (DID). Injecting/calling data not stored on-chain comes with nuances, and this is a problem of centralization the community is still focused on solving. For example, you could bring credit scores on-chain by externally pushing the data to a smart contract — but then the data provider still has control over that flow (though not on any data which has already been pushed). Some solutions for sharing private datasets include Ocean Protocol.
2. User and developer experience is markedly improved
Plenty of products on the internet that don’t suit all of their users’ needs simply because there are too many users to satisfy by one company and product. On Ethereum, frontends/platforms now compete to provide the best user experience for specific customer segments. If you ask a hundred people what apps they use to manage their crypto, you’ll get dozens of different answers based on their needs and preferences. The three user-facing exchange examples from above is just one example of that (Metamask, Zapper, and Zerion all have different approaches and benefits for different types of users)
On the developer side, you’re now all building together no matter where your product sits. That means there is a hefty focus on good documentation, constant engagement and help (on Twitter, Telegram, Discord), and education incentives. But it isn’t just developer experience, but developer opportunity that is enhanced. In addition, you don’t need to be full-stack to build something on your own. Whichever component you think you can make better, from designing UI components to smart contract gas efficiency, you build it and thus immediately improve the product ecosystem.
3. Everyone actively wants you to build and experiment with them
On top of high availability of help, everyone in the ecosystem will literally pay you for building with their product (across all above layers). For now, this comes in three main forms based on the scale of contribution: bounties, hackathons, and grants.
Gitcoin is a great place to find bounties, where product teams will pay around $4000 in ETH for help with issues/features. Hackathons hosted by the likes of ETHGlobal run anywhere between 3 to 30 days at a time, with sponsors offering upwards of $100,000 across dozens of winners. And most products that make it big tend to have ecosystem funds, where you can apply with an idea and paid a grant or investment to build it out (See Uniswap Grants and Gnosis GECO Grants as just two great examples).
Essentially if you build something faster than me or better than my product, I can directly leverage it at no cost to build an even better product in the future — not just zucking an idea and rebuilding it from scratch like everyone is doing to clubhouse right now. This incentive structure is likely what has led the most this grant/sponsor-based build and experimentation trend.
Most of the ideas I’ve worked on started just as fun thought experiments, so you don’t have to come with a full business case ready or anything of the sort. And if you want more money, there’s always VC.
Overall, the boundaries between products are less clear-cut than on Web 2.0. People have made the argument Ethereum is just one giant, decentralized autonomous organization — otherwise the same as a conglomerate public company on a stock exchange. I personally think that is overgeneralizing, but understand the sentiment. It’s unusual to see a community building so many different products in such a synergistic manner.
When a community is aligned to the core, growth is both well thought out and explosive.
Product composability is a big part of the value in developing on Ethereum, but it is not the only source of value. As a derivation of how these products are built, the community around them has become purposefully aligned. At the core, everyone is focused on making Ethereum better in every single way.And I mean everything: fromprotocol infrastructuretodeveloper toolingtoend-to-endeducation.
The key is to remember we’re all on the same base layer, so we’re all uniquely tied by the value and security of ETH.
Because of the Lego-like product architecture, one product’s failure can affect the whole ecosystem either directly or second-hand by reputation. This has led to a culture where everyone is highly focused on quality user experience and setting development standards. Unit testing, security, monitoring, and transparency (DevSecOps) are heavily focused on in a way that isn’t seen in the development of other products until late-stage/cycle (or once the CEO is in a Senate Hearing). Everything that is discussed, developed, and implemented is carefully documented and standards are widely agreed upon and shared. Over time, we’ve arrived at a high degree of trust in one another and the system, and call-outs through both tweet threads and long explainer articles happen quickly if someone or some team rushes development or seems to manipulate users.
All that discussion happens openly with ease of access and high availability in mind as well, not just through emails or obfuscated forums. Most of this occurs on Twitter, Slack, and Discord. If you want to discuss problems with the protocol layer itself, feel free to drop into Eth R&D to talk to Vitalik himself. If you have questions about literally any product on Ethereum, just google the product’s name followed by “ discord” (See links for Uniswap, Zerion, Chainlink). These servers end up having thousands of people in them discussing a large variety of topics across users, developers, and product owners. Can you imagine being able to talk to the team and listen in on conversations about products at Facebook or Google? I sure can’t.
This open discussion format also means less wasted time figuring out who to network with (no more waiting on cold connections on LinkedIn) and also a larger variety of voices instead of a political or professional echo chamber. I’ve talked to everyone from artists (ArtBlocks) teaching me the history of generative art, to lawyers (OpenLaw/LexDao) talking about smart contract infused property agreements, to business/finance analysts (Bankless) discussing the DCF of DeFi products like Yearn and other trading opportunities. These aren’t just one-off conversations — they are continuous discussions that happen every single day leading to newly built products in the ecosystem.
If you love to learn, share ideas, and then actually build them out, this space enables that tenfold. Work is no longer tied to a single corporate entity, but shared across the entire space. It is still fiercely competitive, but in an ultimately collaborative way.
Wow, that was a lot.
If you’re skeptical or just overwhelmed by this point, that’s totally fine. It honestly takes being part of the ecosystem to internalize this. As you go about your day, think about crypto like this:
Crypto provides strong and tight feedback loops and development cycles because products are closely tied together and funnel directly to a community of users, creators, builders — and can be rebuilt or built upon by anyone on any layer.
If you’re curious about how to find a job in crypto then stay tuned — I’ll publish something soon on that 😊. In the meantime, go join some Discords and follow leaders in the space on Twitter.
Or if you just want to discuss or politely argue with me, feel free to reach out to me on LinkedIn or Twitter.