The onchain attention economy is growing rapidly. We have a situation with Clanker in which any idea can be instantly tokenized onchain and, if it’s a hit, the creator earns. This is new, it’s exciting and it’s a way of offering ICOs at scale. The concept can be as simple as a meme or as complex as a fully-featured blockchain. In the end, the power of the audited smart contracts is growing toward the utility that was promised at the beginning of Web3.

Way back in 2020, when I first got pulled into this burgeoning hub for creative minds, things were a lot less open. Working with PageDAO, from 2021 to present, has been a remarkable experience because of the decentralization focus we insisted upon to the exclusion of everything else, even basic financial benefit for our work.

In the end, I’m not sure we were fighting our own dragons. If we’d been better-known, or if we’d had more experience, or if we’d been willing to compromise the values that motivated us, we might have ended up with better results earlier in the process. Now, however, we’re in a new situation. We’re no longer really needed to build the infrastructure itself, but rather the call has shifted to providing applications that leverage it. The infrastructure exists, and Clanker and Base prove it. Solana proves it. Ethereum proves it. ArDrive proves it.

Hence, the endgame for us is probably not a deep infra build. It just doesn’t need to be. Rather, Web3 is entering its endgame. Whether I think of myself as part of PageDAO or as a solo dev wrangling lulz and basic applications, I have what I need now.

Web3 technologies have matured remarkably. The recent Balancer hack is perhaps a moderately unnerving counterpoint to this recent movement, but across the board Web3 technologies are becoming more resilient and more accessible. The evidence is in throughput, adoption, and the improving user experience yielded by technologies such as the Farcaster Wallet.

Evidence for the maturation of these technologies is also seen in the capabilities they bring for their users. The Cosmos Web3 ecosystem used to sell itself on the Appchain Thesis, the idea that applications need their own chain to maintain control over things such as gas prices. Ethereum’s L2 network provided this control and reduced gas costs by orders of magnitude, and the appchain thesis has largely now been co-opted to scale Ethereum, which is why I’ve pivoted to the Ethereum ecosystem.

An additional component of Web3’s success in recent years has to do with AI. Web3 traditionally suffered from a relative lack of highly skilled software developers in addition to a less-trodden path, relative to Web2, where pay could be better due to better scale. AI, including Claude Code and ChatGPT, has provided a rapid way for new minds to enter the space, get up to speed, and contribute real code to the build efforts.

In general, this code is not being written at the protocol level, and it probably shouldn’t be. The role of the novice dev is to find things that work on the blockchain side, contracts and other technologies; and tweak them a bit, in a safe way, if necessary. The really open part of the equation, for the green hat developers (as I consider myself to be) in the space, is the front and back end that would represent the whole application in a Web2 build. In Web3, we have an additional immutable onchain component to consider. And frankly, that changes everything.

Builders who wear the green hat choose to accept the limitations we know we have. We lack experience in smart contract development and auditing, and these critical infrastructure pieces are freely shared in Web3 as a direct result. If you’re looking to build your first smart contract, I suggest OpenZeppelin as a source of information and as a place to find your first boilerplate smart contract to deploy.

First, you’ll want to come up with an idea for an app, of course. I sometimes crowdsource these if I don’t know what to build that day. Step 1 is fully understanding what the contract does for your app and how to build on top of it. I used Railway to provide a caching layer for https://abc.epicdylan.com because I want a quick user experience in my app, but everything that matters is decentralized and onchain.

For the front and back end of the webapp, it doesn’t so much matter what sorts of mistakes I let Claude Code build into the application on its first run or two. I understand the contract, then build the hooks, then use the hooks in the components that I show to the user on the frontend. User balance information and signatures are handled via their wallet and can be done with a public RPC, but queries from my backend use an Alchemy RPC that’s more powerful and less likely to rate limit me.

In the structures I’ve been creating lately, the onchain information is stored in the database, which is then served to the user. What you cache is up to you, but for my apps I like to cache things like transaction histories, staking information, and relational user information such as FID/Wallet address/PFP.

Payments are a very important thing to get right in Web3. If you publish the private key of a wallet that you’re directing payments to, anyone who finds that key can send the money wherever they like. You probably don’t want that, so be sure to practice good key safety. Don’t start your GitHub repository off as a public repo, if you’re brand new at this. For starters, you’re going to make a bajillion mistakes along the way and people mostly just want to see what you end up getting to work and how. But also, you’re likely to commit things you don’t intend, such as your environment variables, if you’re going fast and loose with it. So private repositories aren’t anything to be ashamed of, especially when you’re just getting started.

I build in public despite the risks and yes, on occasion I do still make these basic mistakes and publish PKs. So the defense against this problem relies upon my knowledge of myself – I make a new wallet for each build and discard it if there’s anything even remotely suspicious that happens along the way. It’s especially important to practice really good safety technique with the built-in farcaster wallets that are associated with the accounts we create there, say if you’re building a bot.

There are other best practices, of course, and we’ll probably eventually write about them here.

The way is opening for builders in Web3. Technical complexity, scammers, and a general lack of sufficient developer support for onboarding newcomers have historically made it extremely difficult to succeed in building technology here. Now, however, we have technologies such as Farcaster and the Base App to help onboard people and get them using Web3. The phase of the new technology cycle is shifting, we’re moving from the era of the dominance of the master builder to something a bit less stressful. The master-built technologies have finally reached a point where they have these connection points that, with or without AI assistance, the rest of us can reach.

At this point, the opportunity has arrived. This is what we, at PageDAO, needed, in 2021, to ship the code we were working on. Indeed, we struggled for years with things like encryption technology that has since become available independently. ABC_DAO is an attempt to build an organization that recognizes the quality of the infrastructure that has been built in the space and onboards new developers directly into the heart of this understanding to help them rapidly create their first app.