## Intro
This is the transcript derived from my opening speech made at the Arweave Day In Asia event in Singapore on the 12th Sep, 2023. Most of the ideas are tailored towards Arweave centric builders, but at the same time applicable to builders outside of the arweave ecosystem. The article is close to the exact transcript so please bear with me on the grammar.

Disclaimer: SevenX Ventures is an investor in the Arweave ecosystem.

For the full video you can access it at https://youtu.be/vg_R6J9Ku4s?t=12959.

You can check out the slides at https://hackmd.io/@Hill/BuildOnArweave.


## Inspiration

So hello everyone. I'm here from SevenX Ventures. Today I want to talk about Requests for Builders, Startups I Would Build Today on Arweave. This is a piece inspired by Coinbase so they put this out I guess a few days, if not weeks ago. That's a massive favor. Basically, they said, we can build a lot of stuff by Coinbase alone, but then however, they don't want to do everything. So therefore they want to inspire builders to build in different directions, and they list out 10 ideas. If you're interested, you can always go check them out. This is more crypto blockchain general. And then obviously we all know that bear market is pretty deep and then however the best projects are built here in the beer market. So I'm here today to share seven ideas that I will build today, but more Arweave specifically or should I say seven direction of projects?

## About SevenX Ventures

Probably before I get started, I should talk about SevenX. We're one of the earliest and largest supporter and investor in the Arweave ecosystem. Actually the second day when I joined SevenX I fly out to Hangzhou to meet outprog, the founder of EverPay. And then for quite a long time we've been focusing on supporting projects in Arweave. You can see a lot of the portfolios projects here. But however, beyond Arweave we also invest into other ecosystems of crypto. And I can see that we are obviously not entirely developing in the same place. So I would like to bring some of the external insights and showcase basically what's promising in crypto, but at the same time, what promising sectors are more applicable towards Arweave.

## About Myself & Disclaimer

And then something about myself. I'm the research lead for SevenX Ventures. I also spent some time building in a fair community project called Venture Gold, which comes out of the Loot NFT. And then I also started in crypto since 2018, founding the crypto society in the University of Sydney. And then some disclaimer, obviously no financial advice, you should always do your own research before you invest or before you go fund projects. Also if you're building the sector announced below grid, but that does not mean you will make it, and that not mean a hundred percent will invest and we might just be wrong. So take this with a grain of salt.

## TLDR

And then, so the seven verticals, or should I say actually specifically six verticals are first of all, ZK Middlewares that provides better security guarantees or some kind of privacy features and Modular UX Layer. And when it comes to applications, it'll be fully onchain games, luxury media, and actually more trading platforms. So let's get started.

## ZK Middlewares - Security

So basically, if you are building infrastructure, this is something that I some kind of recommend if you are building Arweave. So we all know how powerful Arweave is and we kind of all believe that however people outside is not really sure of it. And then in order to better serve them, we kind of have to say, hey, look, we are great, but when you are trying to come to Arweave, we need more secure ways to do it. And then so ZK technology is, I guess the buzzword right now outside of crypto, but for the right reasons and people basically, I guess a general way to express this is that ZK might be a more impactful technology than blockchain alone. For the sheer reason that ZK improves the security of trust assumption of a known network where it's supposed to be decentralized, but for some reason it remains centralized. We see that happening in Oracles, we see that happening in Bridges. And then if somehow we cannot immediately decentralize everything, we can use ZK to increase or improve the security assumptions of such platforms. And then ZK essentially do that by lowering the cost for verifiers to check on the centralized entities. So if the centralized entities act maliciously, it's easier for the outsiders to notice. And then that's applicable to many other things. VMs and computation platforms, Oracles, Bridges, Indexing and Query Services. So for example, if we can let's just say persuade somebody to run their, I guess scripts or logics on Arweave based platforms, but then at the same time if we were to introduce additional security assumptions it does not necessarily will be the best option for them. Chances are they might go to a ZK rollup or something, but however, if we tell them, hey, look, your competition on Arweave, done by us is also secured by ZK, that's a very powerful value proposition. So in that sense, I believe this will be very useful in the next coming bull market, if we were to expand the Arweave outreach to different platforms.

## ZK Middlewares - Privacy

The second part is privacy. You can achieve privacy with a lot of other technologies. But then basically why I want to mention ZK is it has some really unique features. So basically privacy is a must have for mass adoptions. And then the other two are scaling and good UX. And everybody else is trying to solve scale and good UX with different L1/L2s, different consensus mechanisms, account abstractions, smart contact wallet, people are working on it, but then privacy is kind of neglected. And then, so basically how ZK achieves privacy is it allows the separation of public state and private state, but at the same time you can verify each other from the other state point of view. So therefore you don't have to encrypt everything and then decrypt everything. You just have to verify, hey look, this state is correct, I trust it. That's it. There are other technologies like TEE, NPC and flu homomorphic encryption. But however for now, Elliptic Curves and ZK is the only technology we see feasible on the client side. So that means that you are opening yourself up towards more privacy applications where you need the information to stay on the client side. For example, my social security number, my health status, I am not supposed to trust this with another set of centralized servers. Therefore, if I can achieve privacy on my own client side, that would be the best solution. So in that case, ZK is one of the great ones. However, privacy is not the easiest thing to build So I guess a few things we can start is privacy account addresses, that would be for sending and receiving transactions in anonymous ways. As well as privacy AMMs and Order Book, obviously that's for privacy trading. If you want to make money, you don't want other people to make money in a lot of sense when it comes to trading. So yeah, privacy AMMs and order book is a good thing. And then the last thing I'm pretty excited about is private games state modules. So basically if we look into the gaming applications that's buildable on the blockchain, all the information is public, therefore you cannot put a lot of the games on there. For example, if you want information asymmetry or some kind of gambling or card game that is, it's almost impossible without a privacy layer. So that's why I believe private game state modules also make sense to be built.

## Modular UX Layer

And then probably, I guess the last thing on the infrastructure side is Modular UX layer. So I guess this is kind of already achieved in Arweave, we see a few projects already enabling that you don't have to click sign transaction every time you want to execute a transaction. However, on guest fee payment is also another thing that can improve UX. So paymaster services where users does not feel like they're paying the guest fee, rather than it's being kind of compensated by the application itself. Web2 log-in is also pretty important, but I see it's also being realizing Arweave here today as well as fiat on-ramps and off-ramps. But a lot of them are realized today I don't think they're the largest problem, but it'll be great if we can have more of them.

## Mental Model for Infra

So basically the mental model for building infra and middlewares. So as we mentioned earlier for mass adoption, we really need scalability with security and decentralization. That means we don't really compromise and then great UX and then privacy. So we need to have all of those three to have mass adoption. And then for whatever infrastructure you're trying to build, either you're trying to build a generic computation platform on top of Arweave, or you're trying to build an oracle, always try to aim to achieve all of these because if you can achieve three of them, that means you have a really great chance to capture all the potential customers who are looking forward to use this kind of middleware. However, that means it's going to be tremendously difficult. So a lot of the times people are great if they can achieve one of these circles. So if you can do a great privacy computation platform that would be great, that's awesome enough. But however, in order for you to survive and out compete competitors in the long run, you want to aim for max composability that give less reason for people to compete with you rather to, or that give more reason for people to work with you rather than compete with you. So this, I guess is the general mental model if you're building an infrastructure.

## Fully On-Chain Games

And then another thing I want to cover on the application side is the only application, you can call it vertical or narrative that's going up in the trend right now that is fully on-chain games. This has not been a new concept, basically got started around 2021 with the inception of dark forest as well as loot. And then basically by being fully on chain, it means that the entire state and all the logics of the game should be on the chain. And this is pretty counter-intuitive to how people build games on the blockchain. Usually people thought games are really competition heavy, therefore you should move those heavy competition off the chains and then only put the most important or valuable parts of the game on the chain, which is like your assets. So people put NFTs, people put tokens on the chain. However, we don't think that unleash the full potential of gaming, for the reason that there are two types of game. The first type of game gets you dopamine. That means that you get really quick, easy fun. That's it. Just like when you're swiping TikTok and there's another game. There's another type of game where you get the fun out of it when you get the sense of achievement by working hard in the game. And so we believe that fully on-chain game should look more similar to the second kind, but in a more intuitive way. We prefer the definition of game in the game theory context rather than the game for fun. So in game theory, we're trying to compete, we're trying to have some kind of sporting with another player or another group of player. So we believe fully on-chain games should involve some kind of this. Why, because this justifies the usage of a blockchain by highlighting that fairness and transparency. It's important under this context. If I'm just having fun, why should I care about my money being secured on the blockchain? Why should I care that the logic of the game is correctly executed on the blockchain? Why do I even care that the game is open source? It doesn't make sense for me. However, if I am competing, if I am putting my money, my effort, my time into it, I want to make sure that this is a fair playing ground, this is verifiable, and that's why we believe this type of game will thrive on the blockchain compared to just games for fun. And however, take this with a grain of thought because this is discussing fully on chain games. If you believe your game does not have to be fully on chains, you don't have to go for the game theory place, but however, if your game is fully on chain, it makes sense to focus on that. And then another thing is a general feeling, but then it's a very intuitive feeling, it's that players of fully on-chain games should feel like they're doing a job or as if they're working. Again, this is not purely having fun because we don't think purely having fun itself embraces as much value that justifies the cost of putting it onto a blockchain. However, if you, while playing the game, create a lot of value, then it justified that. And then the activities or behaviors of you playing the game, if that is creating a lot of value, then it probably looks like doing some kind of job. So for example, one of the full on-chain games out there, I probably forgot to put in there you guys can try it out, it's called primodium. It's basically building a factory where you got different resources on the map and then you build roads, you build factories. To move those resources into a factory, you kind of manufacture them, you generate the eventually output product. It feels like you are designing and running a factory. And then people often mock that it's not a fun game, but you are literally doing a job by designing the factory. So that's a great example. And so two types of applications or games I envision to come out of the fully on-chain game direction, the first one is autonomous world. This is a complicated concept, but basically by being autonomous you are only designing the physical mechanism of the game so that users are supposed to be able to create a different version of the game. A different game such as Metagame or like a small game that plays according to the rule of the general world you're building, or they should be building a different kind of client, a different version of client, a different mod for your game. So in that sense, I kind of aggregate all of them into this concept called autonomous world. The reason why I believe they make sense to do well in the next cycle is because they allow a game to become a platform. So how do we understand this? Think of traditional gaming as the team creating content, and then they sell the content to the players. So in that case, you either sell the game copies, which is what people do on Steam, or you sell it via some kind of microtransaction or subscriptions, which a lot of other games also implement this, or you sell it to users via advertisements. So you make the user watch advertisements and then you charge the advertisers. However, eventually you are building a content that you sell to the users. Now content has a mode, however, the content mode is not very deep. That's just to say somebody tries to fork your game or your content simply does not become popular anymore, that means bad news for you, therefore you have to create new content or create another game. However, if you are a fully on-chain game or autonomous worlds which allow users to co-build with you, then your game experience will look completely different on day one compared to let's just say a year later. So on day one, no user creates any plugin, no user creates any client-side improvements. The user is just playing your game. However, as more user come in and then build different plugins, the aggregated user experience or the user experience for another user is improved thanks to other user building on your game. Now, after a year, somebody else tried to build the same game with you, although they can achieve the same functions or features or even the graphics with your game, their user experience is not on par with you. Therefore, you eventually are able to achieve platform moat. Another kind of game we envision to arise is called Arcade Games, which is a middleground between gamified ponzi of DeFi and free games. Basically arcade games is pretty simple to understand. You go to the arcade, you want to play a game, but it's not free to play or you don't have to buy a thousand dollars NFT to play the game. You're putting a few coins, you play around. If you die, you try again, put in a few coins. This is middle ground when it comes to financialization and that we believe this will fit well as a token model or business model for a lot of the simpler cash or games out there.

## Luxury Media

Another thing I want to touch before the end is luxury media. So basically we're living in Web2 where the information is abundant and almost free. And then however, for information to be on the blockchain, there also has to be of high value. So in here we see two trends coming over. The first one is luxury content, the second one is luxury social media. So basically luxury content is content, but that's premium with rights or just exclusive assets. And then social media is basically you're trying to establish new social connections outside of the existing social network, obviously as a social connection that has to represent status and identity. However, there's one thing to really pay attention to is that the content or the social connections you are establishing should make sense for people to invest in. I'll probably explain why you need to allow people to invest into this by the last few slides. But long story short, financialization is important in crypto. It is how people identify value. It is how people measure value and it is how people scale value. So in order for you to scale your content or scale your social connection, there has to be the right reason to invest into it.

## Trading Platforms

I'll be quick on this one. It's obviously super important, trading platform because it is the only everlasting use case we see in crypto and blockchain. It has been there since 2008, it has been there today. So we probably need more of these stacks, AMMs, Order Books, NFT marketplaces, derivatives, launchpads, but this is kind of a consensus right now, so I hope we can build more of it.

## Mental Model for Applications

So I'm pretty close to the end. Mental model for building applications. So this is something that's really interesting. So we compare the application verticals with the mode that you can achieve or the amount of customers you can attract if you simply build a fork, which everything looks exactly the same and then on DeFi is supposed to be pretty easy for you. If you look at how Sushi swap did this on vampire attacking Uniswap, it's like everything is very similar, but then we just issue a token. We provide higher yield for you, naturally capital will flow over. And then however it is not an easy game with gaming. So imagine you create a game that's the same as somebody else. Chances are you get, I'll say half the players, somebody wants to play the old version, somebody wants to try something new, you get some of the user naturally. However, if you're building a social network and you achieve feature parity, let's just say you want to build a decentralized Twitter or something and then you make everything similar to Twitter, you get nobody. Social connection has incredible network effect that's really hard to break. So before you start building up an application, make sure you understand existing product's moat, then decide if you want to build something entirely due and then obviously start from the premium users because they pay up. Financialization is key, which also I guess the next logical step is that figuring out a way to convert the user's consumption behavior into investment behavior will give you an easier time in Web3. And then last but not least, take advantage of existing infras, do not reinvent the wheel.

## Tips

Yeah, probably a few more tips to go through. I'll be very quick. Yeah, we'll have to be honest to ourselves. What does my application or the asset or data is of such high value that it needs decentralization or permanent storage? After you figure that out, you need to figure out a way to represent it because you can't measure it if you can't improve it. Then it's about capturing and then it's about taking as few compromises as possible. And then yeah, again financialization and then hopefully you can make the system self-sustainable. And then premium sector, again focus on the customers who pay off first so you will have an easier time with product market.

## Holy Grail & Closing

Yeah, I guess the holy grail here is we want to see timeless applications here in Arweave. So basically Patek Philippe had this really great saying that you never actually own one of your watches. You're merely keeping it for the next generation. So think about whatever you're building here today will last basically forever on Arweave. What do you want to build? You do not own it. You're merely keeping it for the next generation. Das Ende. Thank you.