Sept. 30, 2021

Gubsheep: Dark Forest, and building crypto-native games.

This is a conversation with Gubsheep about Dark Forest and learnings from building a uniquely crypto-native game.

This is a conversation with Gubsheep about Dark Forest and building crypto-native games.

Dark Forest is a uniquely crypto-native game, and embodies patterns that we can all learn from. It's a game that takes place in a procedurally-generated universe that's shared by all players, where the smart contracts enforce the rules of the game, but where they don't leak information about what each player is up to by using ZK SNARKs.

In this conversation, we talk about how Dark Forest uses ZK SNARKs to enable this core game mechanic, how it has organically built a community of players and developers, how early DAOs are participating in the games, how the game might interoperate with other games in the near future, and we also have a good discussion on Layer 2s and the trade-offs involved in choosing a scaling solution.

Check out our newsletter for other episodes and writing:


Sina [00:00:20]: Hey everyone. Welcome to Into the Bytecode. My guest today is Gubsheep the builder behind Dark Forest. Gubsheep and I met at the research workshop hosted by the Imperium Foundation early on in his journey of working on Dark Forest. And then I had the good fortune of being a counterpart for him and the team as they've built Dark Forest into the game that it is today. Dark Forest is a uniquely crypto native game. And I think embodies patterns that we can all learn a lot from. It's a game that takes place in a procedurally generated universe. That's shared by all players where the smart contracts enforce the rules of the game, but where they don't leak information about what each player is up to in this conversation. We talk about how they use ZK Snarks to enable this core game mechanic. And then we talk about how Dark Forest has organically built a community of players and developers, how early DAOs are participating in the game, and how Dark Forest might interoperate with other games in the near future. And we also have a good discussion on layer twos and the trade-offs involved in choosing a scaling solution. I really enjoyed this conversation and hope you do too.

The first thing that I was curious to talk about was interoperability and I know you wrote this piece about it. And to me, it seems like this applies to gaming, but also to crypto products and protocols. Otherwise, there's a weak version of interoperability, which is saying that, okay, I have my product, I have a web UI and it's not using the blockchain for all that it can do because you're still requiring the person to use the same front end of the same client. On the other hand, like what you've been doing with Dark Forest is the only thing that's anchoring this game together is these core smart contracts. And you're almost purposefully encouraging people to like to build different clients and whatnot. So, I'm curious, was this clear to you from the beginning that there's a spectrum, and like how has thinking about this evolved?

Gubsheep [00:02:41]: Yeah, for sure. I really like the word that you use here, spectrum to describe the different ways that we can think about interoperability. I think interoperability, or like the full promise of interoperability that a platform like Ethereum offers is really going to be much more than just I write my smart contract, and like technically anyone can connect to its technically permissionless. It really involves all these other components, like building and designing for third-party development from day one and treating third-party developers as first-class users as well. And things like adhering to not just like the strict standards that have been set up for things like NFTs or whatever else. But making it easy for players to integrate that, or like other third-party services or applications to integrate what you're doing with what they're doing.

So I think for us with Dark Forest, one of the things that were exciting to us from day one was this idea that you could have a game that is client agnostic, a game where things like bots and scripts and automation and other utilities that players might write on their own custom front ends are first-class features of the game. You know, the game can be played in multiple ways. It can be played from ZKGA.ME. It can be played by locally running source code that we provide. It can be played by like rolling your own client or modifying any of the above. So yeah, for us, we've always felt that we would achieve some of the most interesting outcomes if we could make that a priority from as early as possible, we didn't quite hit this with the first few releases of Dark Forest but over time I think we've been getting better. And there are so many more dimensions to this as well, like smart contract interoperability that we're wanting to explore too. And it's an open question what are the best patterns to expose those things?

Sina [00:04:37]: Yeah. So this the idea of people writing their own clients or developing plug-ins around it, like that sounds easy when you're saying it verbally, but just to like actually tease apart what's happening here, people are literally running local XDAI nodes, which is the side chain you're running on and then they're running a local version of the web client hosted on localhost and then they're writing scripts right there to communicate with them.

Gubsheep [00:05:13]: Yes. So, there's again a spectrum of things that players do depend on the level of complexity and customization that they're looking for. So, in the first versions of the game, back in 2020 one of the first patterns that we saw was that people were opening the Chrome dev console and you know, we expose on the global winDAO object, an interface into you know you can programmatically, for example, trigger Dark Forest moves and upgrades and other transactions. So, we saw people starting to write things like automation or scripts that would do things like oh you know send a hundred energy from this planet to that planet on an interval of like 3,600 seconds or something like that.

Sina [00:06:01]: Like they would just have a timeout loop with like copy-paste it into the Chrome console.

Gubsheep 00:06:07]: Yeah, exactly. And so, you know, when we saw that happening, one of the first things that we did was we wanted to, I think a lot of players just like, don't even realize that this kind of thing is possible. Like, you know, sort of like the conceptual language isn't even there yet to realize that like, oh yeah, like this is a game that's encouraging me to script and automate. So, we also just included this cute little terminal at the side of the game UI as well, that was basically just like a worst Chrome dev console. But just like, you know, plugged in some instructions there to be like, yeah, you can type in like DF. Move into this terminal. And then it'll spit things out. It'll like trigger and move. And whenever you make a game move in the UI it'll also print that to the little game terminal to encourage players to do that.

Sina [00:06:51]: Interesting. So, people were doing this on the Chrome dev tools, but then you saw that happen it's like this pattern of seeing your users misuse the product, or like basically leaving enough degrees of freedom so people can do weird things that you didn't expect and then like taking them and codifying them into like actual patterns.

Gubsheep [00:07:31]: Yeah. I mean, I would say that with this particular pattern actually, it was more like we would prefer actually that people use Chrome dev tools rather than the Dark Forest built-in terminal UI but what we saw was that players were using the Chrome dev tools and we like the bleeding edge of players were using Chrome dev tools. And we were like, oh, how do we make more people realize that this is the thing that you can do? And the answer to that is like, we'll build a really shitty you know like a [7:40 inaudible] like in UI terminal that you can use and, you know, you start writing some automation and then you realize like, oh, wait a second. I can just open the inspector and I'll get like all these other nice things like auto-complete. And then like some people start moving into like actual like writing JavaScript code and actual IDEs and like pasting back and forth. And then we got a whole other host of ideas from that as well. So actually, I can speak a little bit about like, kind of the plugin system and how that's evolved too because that was sort of the next iteration that we saw.

Sina [00:08:10]: Totally.

Gubsheep [00:08:11]: Yeah. So, to move forward chronologically a bit we saw that people wanted the Chrome inspector workflow to be really, really clunky. And for example, one thing that it doesn't lend itself very nicely to is writing shareable code. You know, and one of the goals here has always been to build a developer community to get people excited, not just about playing the game, but also about writing code, sharing that with others, getting feedback on their own code and scripts, and whatever else. So, in December we introduced a very simple plugin system where essentially, we have an environment where you can write code into and load custom code into the game. And it has access to programmatically interact with all of the objects in the Dark Forest clients. So, for example, it can interact with the core Dark Forest game manager object which allows plugins to do things like schedule moves in the future.

Sina [00:09:16]: So, there's if I remember there are two objects like one DF object and one UI object.

Gubsheep [00:09:21]: Right, right. And those are the ones that we build out with the intention of them being accessed by developers. But a lot of developers go even deeper like kind of pull apart or reconfigure the internals of the DF or UI objects if they want more advanced functionality which is really interesting to see as well. So, you know, now you've got this plugin system where you can basically copy-paste a snippet of code into one client and then pull that out and share that with someone else. And they can have like similar functionality. And now I think when a lot of developers are writing plugins, they're writing plugins with the intent of those plugins, being able to be consumed by other players, which is in my head kind of like the next step here from just hacking around in your own Chrome dev console.

So you have these nice repeatably usable utilities, such as the remote minor, which allows you to connect to a web server that hashes the fog of war faster than your own CPU. You've got like the remote Snarker, which allows you to offload the burden of snark proof computation onto another web server and things like, you know, repeat attack all the way up to like very complex plugins that are essentially using like heuristics to conduct a war against another player in an entirely automated fashion.

Sina [00:10:44]: Like it's like war strategies baked into plugins.

Gubsheep [00:10:47]: Right. Right.

Sina [00:10:48]: I feel just speaking about the snark prover, like maybe it's worth just briefly talking about how Snarks are used in the game it's basically that the universe is being mined or I mean you explain it, you'll probably do a better job.

Gubsheep [00:11:03]: Yeah, for sure. So, the core mechanic behind Dark Forest is that when you initialize into the Dark Forest universe, which is like a really big infinite procedurally generated universe of planets and other celestial objects you are initializing into a massive cryptographic fog of war. So initially when you load into the game, you don't know where other players are. You don't know where other planets are. You only know the location of your own home planet. And what happens is over time, you're essentially going to be doing proof of work, something analogous to the proof of work mining to compute what are the locations and points of interest in your neighborhood? So, you know, essentially what's going on is that you're, you're like running computation in order to lift the cryptographic fog of war. Now the smart contract is also, and anybody just like looking at the blockchain is also not going to be aware of the exact locations of planets or players or anything like that.

All the smart contract has, is it has commitments to locations uploaded by players. So, when you initialize into the game, you submit a commitment to your home planet. Now, the game still has to verify though, that the actions that you're taking are valid. So, for example, planets have a range stat, and you can't move energy from one planet to another, beyond the range of the origin planet. So how it checks that is that even though locations are hidden, the only thing that the chain sees is commitments to locations. Players can publish zero-knowledge proofs that move between the locations underlying two commitments are indeed valid. And that allows us to preserve the cryptographic fog of war mechanics while still ensuring that all moves actually are happening accord in accordance with the rules of the game.

Sina [00:13:06]: Yeah. Yeah. It's easy to not realize how crazy this idea is, right. That there's a shared universe that there is no central party keeping this it's literally stored on an Ethereum smart contract and anyone looking at it can't see anywhere else in the universe other than where they are themselves. And then the moves like the patterns of expanding from where you are, are governed by these like snark proofs.

Gubsheep 00:13:38]: Right. Right. It's really interesting. I mean like another framing of this sort of thing is that it's an uncheatable game where even the core Devs can't know where you are. It's a protocol for location and movement that everybody gets to buy into.

Sina [00:13:56]: Yeah. So then back to the remote snarker is it called? Plugin so basically the way the game worked out of the box was you would open your web browser and as your just Chrome instance would be like hashing would be, I don't know exactly how the mining like works but it…

Gubsheep [00:14:18]: It hashes in a spiral around

Sina [00:14:19]: Right. It hashes in a spiral. And that takes a lot of time. And basically, people have been spinning up AWS servers and like running this process remotely there. And so maybe this is an interesting foray into obviously this game is on Ethereum and items that are inside this game will be valuable right. For one...

Gubsheep [00:14:44]: I mean, I guess well we have seen things like artifact NFTs start to transact for small to medium-ish amounts of value but that's not something which we try to deliberately have a hand in. It's something that I think is organically happening.

Sina [00:15:02]: Given that like a remote snarker plugins exist already will the same kind of like GPU mining, like ASIC mining wars that happen on blockchains happen here in the game basically like reaching to the Asim tote of what the things you're mining are worth. Like people will spend that much money mining the universe. Right.

Gubsheep [00:15:25]: I think it's really interesting we are seeing people spend like hundreds of dollars on like AWS or digital ocean droplets to get more compute for being able to mine the universe faster. I think that like, well, one difference here is that once one person has minded an area of the universe, if you trust them, they can just like sell you that data and you can resell it. And so I think it's like, it's somewhat different from just like, you know, hash power is worth a certain amount of dollars. Like now there are all these interesting questions about like the topology of the map and the location and like what pieces of information do you want more than others or like more what sellers are more trusted than others. But yeah, I think that there's this natural feedback loop that we're starting to see happen because this is a crypto game built on an EVM-compatible system. Players are empowered to start assigning economic value to assets in the game, whether that's information or literal NFTs. And because of that, players are in turn also more willing to spend money on things like mining power or even buying patches and plugins from other players too.

Sina [00:16:42]: Yeah, totally. There is a large space of interesting interactions to explore and like almost any new plugin can change the dynamics. And the one that I'm particularly thinking about now is this one where you reveal a portion of the map where you sell a portion of the map. What is the idea there?

Gubsheep [00:17:07]: Yeah, so there's a couple of plugins actually in this under this sort of thematic umbrella. So, one is like very early on. I know that some folks built a map import-export tool that allows you to snip just a search and rectangle out of your map and send it to someone else.

Sina [00:17:25]: Oh, just like a screenshot.

Gubsheep [00:17:26]: Yeah, basically. And another really exciting plugin that we saw from a recent round was the broadcast marketplace, which basically allows players to put bounties on certain planets coordinates to be revealed. So, you might say like, oh, I know that I'm in a war with Cena and I see that Cena has some you know, Cena's got like some big level nine planet, and I don't know where it is, but I know that it's sort of funneling energy into the front of our, like battlefield here. So, I'll put a bounty on that planet to be revealed. And yeah, so we're seeing information basically as like a first-class resource in this game, that's also getting transacted and that's something that's made possible by the ZK stuff. But you know, and this is why I think the ZK stuff is really powerful for games because once you have an incomplete information game, then information itself becomes thefirst-class resource, and combined with the economic mechanics of crypto, you can get all sorts of dimensionality that you wouldn't otherwise.

Sina [00:18:34]: Interesting. It also makes me think of like with the reveal marketplace, there could be an incentive to just go explore the universe. Right?

Gubsheep [00:18:42]: Exactly. Yeah.

Sina [00:18:43]: And so there will be players who are not even maximizing their, this is a pretty interesting idea that they're not even maximizing the score that the game is telling them to maximize. They're just looking at these reveal requests and seeing that, oh, like I could go in this corner of the universe and just map it out and that would be valuable.

Gubsheep [00:19:06]: Right, right. Yeah. There are definitely different ways to play. And I think that like you know some players become specialists in like in discord. Some players are like selling map data and they're basically getting there, like mining rig set up. So that way they can sell off chunks of a map for like five XI each or something. Other players are like artifact hunters. They're just sort of, rather than trying to build a big empire, they're just like jumping around the universe and like finding and harvesting like artifact NFTs. Some players are really into the coding aspect of the game. So, they're building plugins and sharing those and encouraging others, and educating others on how to get involved as well. And then of course there's always going to be hardcore competitive players as well, who are wanting to engage in PVP going for maximal score writing scripts specifically to help them achieve those goals.

Sina [00:19:58]: Yeah. One thing I remember you saying is that you are trying to build a game that's fun to play, right. That it's good for the sake. Like it's enjoyable to play and especially the kind of economic mechanisms that emerge out of it is secondary for the time being. And now we're talking about the different archetypes of players that are emerging, and it feels like there's just a whole emergent, like property to the way this game is growing. And I'm curious in terms of like your focus, what do you see as the key pieces of this to focus on? Like, what is the highest leverage things that you are focusing on so that this thing can continue like organically growing?

Gubsheep [00:20:50]: Yeah, I think well first to respond to one thing that you mentioned about economic mechanics even though we are seeing like these flywheels happening, were stuff in the game is starting to become valuable to some players and that incentivizes players to like spend money on like AWS mining or things like that. The latter part of that loop is something that we were seeing even before anything in this game was worth anything. You know, players were just of their own volition. They wanted to have knowledge more of the universe. So, they, you know, spend money mining out like some large quadrant or some like very large region. And that to us was a great sign because that meant that this game isn't just something that people are playing due to an extrinsic motivation, which is like you've got some like a financial mechanic or something, but rather there's something like intrinsic to this universe that's compelling to people.

And so as much as possible, we want to lean into things that enhance or add complexity to the intrinsic motivators that make people want to play the game. So specifically, things that we're like not focusing on are going to be things like promoting or trying to hype up the value of Dark Forest NFTs. You know, if players want to build a marketplace or transact NFTs, they're more than welcome to. But we think that at the end of the day, the price of a prize is something of an extrinsic motivator. So that's less of a focus on the other hand. Things like building out tools for third-party developers allow us to tap into a deeper intrinsic motivator that gets a lot of players excited about the game. So, you know, a lot of people who play the game enjoy writing plugins or trying to modify the client.

So making it easier for them to do so allows us to build that community out. And you know, there's always stuff around like making the game itself easier and more fun to play and more accessible too. So, things like adding new mechanics that can have multiplicative combinations with other mechanics in the game are always something that we try to do. Every version that we come out with just broadly improving, like is like quality-of-life improvements on the game in general, with performance and graphics and things like that. So, all the things don’t sound like groundbreaking or anything because it isn't, it's just like what is involved in building a good game. But at the end of the day, I do think that we have to build a good game rather than a profitable financial application for this kind of thing to be long-term viable.

Sina [00:23:26]: Yeah. When you say game mechanics that together unlock new areas what is an example of that?

Gubsheep [00:23:36]: Yeah, so like this last version we introduced so for those who aren't familiar there are items that you can discover on certain Dark Forest planets that are called artifacts. And these artifacts basically grant powerups and stat boosts to different planets that you own that you might choose to deposit them on. So, for example, if I discover a spaceship artifact on one planet, then I can activate that artifact, I can move that artifact around my empire and then activate it on another planet and that planet will get like a range boost. So, last version we introduced artifacts with status effects that are more than just passive stat boosts. So, we introduced wormholes that allow you to basically connect to faraway planets and speed up any moves that go between them. We introduced the Photoid Cannon, which allows you to fire off a single, very powerful attack from a planet.

And it’s consumable. So, this is like a one-time use thing. We introduced a planetary shield, which allows you to give a one-time, very strong defensive bonus to a planet of your choice. And you know, like various of these sorts of like active status effect artifacts. And what we saw is that like players were using these status effect artifacts, these active artifacts in combination with each other in interesting ways to basically do things like, you know, pull off a heist where like you would fire off a move with a photo toy cannon from planet A to planet B and that move would be carrying a wormhole. And then you would use that wormhole...

Sina [00:25:20]: Oh my God.

Gubsheep [00:25:22]: And connect it back to your original planet. And then you would send a black domain along the wormhole, and then you would like destroy that planet and then transfer its assets to like yourself or something. So there’s like this there's this whole like combo system that was sort of evolving towards the end of the last couple of rounds where people were figuring out how to use these artifacts in combination with the other natural mechanics of the game to basically come up with strategies that like us, the developers hadn't envisioned because we simply couldn't like the total number of combinations for how you can put these things together is too much for us to be able to predict and advance how people are going to use them.

Sina [00:26:02]: Yeah. Well, it sounds like a big part of how this game is evolving is the community around it. Right.

Gubsheep [00:26:10]: Definitely.

Sina [00:26:10]: Both on the side of like users doing these weird combos to like developers building plugins. And like, to your point of finding intrinsic motivators for people. What do you think is drawing people to this game? Why do developers go and spend like a week not doing anything else but like playing a Dark Forest around what's happening there?

Gubsheep [00:26:40]: Yeah. So, for one thing, I think that we've got a really, we've got a very active developer community of just like a lot of really like friendly and also strong developers. So, you know, I know that Blaine and Jacob who are two of our most prolific third-party developers have done an amazing job of educating new Devs and giving feedback on different people's plugins and stuff like that. And there's an effect that I think starts to happen where it's like, you know, if you're joining this community and you're seeing all these other people like create and share all this cool stuff, you feel that you might want to participate as well. And you'll get like some like street cred for coming out with like a really awesome or creative plug-in idea, or really like a neat piece of technology or infrastructure that other people can use.

So yeah, I think the feeling of community is very important for a lot of people. There's a little bit perhaps of a competitive aspect to like who can create the coolest stuff as well. I think like another thing that's really powerful is that Dark Force lives under this larger umbrella of the Ethereum slash crypto movement. And especially because it is in a lot of ways, the first game to use a lot of new technologies that are becoming available to application developers. There's this feeling of you getting to be at the forefront. And you get to also bounce around ideas with other people who are also at the forefront and also part of the same movement as yourself. The thing that's really inspiring to me is just jumping in our discord and chatting with a bunch of other players. A lot of them are also just even familiar names on crypto Twitter or something like that. And it's basically just like another like channel for interacting with this broader crypto community or broader like ZK development community.

Sina [00:28:34]: Yeah. It makes me think of the saying that building hard things is sometimes easier than building easy things. Because you get everyone excited and it's just inspiring to be at the frontier and doing things that haven't been done before.

Gubsheep [00:28:55]: Yeah, for sure. I think that like it's much easier to get people excited about something that's like, not just another, like web two web app or something like that. Like, we've all seen a million of those at this point. When there's something new going on like it's going to drive a certain kind of person who is naturally curious and eager to innovate. And when you get a bunch of those people together, then I think that like really interesting things start to happen.

Sina [00:29:25]: Yeah. And maybe it would be cool to talk about how Blaine and Jacob found this game. Oh yeah. And I feel like there's almost this like an evolutionary history of like how various interesting people have found our forest and then been a part of it going to its next stage.

Gubsheep [00:29:43]: Blaine and Jacob discovered the game very early on. I think either in version 0.3 or version 0.4 both of them actually won V 0.4. And the way they did that was they were two of the first players to realize that the game is highly programmable and that you can write powerful scripts and automation and client modifications and stuff to achieve your ends. So yeah, they took first and second in version V 0.4, and afterward, we were really curious how they had won because we had seen that like V 0.4 was the first version where I think actual programming was heavily used to give players an advantage. So, we got in touch with them, and we asked them to sort of break down their strategies and how they'd gone about growing their empire and like winning the game. And we found that they'd just been doing like all this crazy stuff, like hacking into our code that we like, it was just mind-blowing to us.

Sina [00:30:54]: How did they even find out about Dark Forest?

Gubsheep [00:30:57]: I think that someone must have tweeted out about it back in like mid-late 2020. I know Vitalik shared the game at some point on Twitter in early fall, I think. So, a number of players found out through that, and then we also had this hacky little system going where in order to verify your Twitter, you would have to like send out a tweet with a signature.

Sina [00:31:20]: That's a good move.

Gubsheep [00:31:21]: Yeah, definitely. That helped get a lot of people interested in like, huh, like, why are all of these, like, you know, crypto Twitter accounts tweeting out like these like bite strings that are tagged with like #darkfacts.

Sina [00:31:33]: Maximally mysterious tweets.

Gubsheep [00:31:34]: Right. Right. And now I think that this is a pattern that we're seeing more and more I know that like to verify with like mirror, for example, you also have to do this. I think actually, like one of the first examples of where I saw this was like key base like a couple of years ago, if you wanted to verify your key base account, you had to like tweet out and [laugh], that was pretty cool to see. They found the game, I think through Twitter one version V 0.4, we got in touch with them and tried to figure out what the heck they were doing to play this game so effectively. And then since then, you know, like we started sharing our vision with them of this like very programmable interoperable you know, blockchain-based game. And they were super excited about that. So ever since then, we've been going back a lot back and forth getting feedback from them on third-party tooling to help out with other third-party developers. And getting ideas from them as well about like how can we make the game more friendly both to players and programmers.

Sina [00:32:41]: Yeah. Yeah. And having we're recording this after a mini-con that we hosted in Seattle and it was the first time I was meeting Blaine and Jacob and was just yeah blown away by how excited they are about the different directions that this game can go in and are like, just coming up with all these like crazy ideas that I wouldn't have thought of and also how good they are as engineers they're like really solid engineers who are just working on this game because it hooked them.

Gubsheep [00:33:16]: Right. And it really does speak to the power. I think, of having like an open ecosystem and an open community like a lot of the stuff that they're building out is stuff that like our quote-unquote like core team definitely doesn't have either like the experience or the perspective of the knowledge to have considered and we see that also with players like I know DF DAO was doing a ton of interesting stuff last round as well with like, they assembled a team that operated like a Multisig wallet and had like various different [33:46 inaudible].

Sina [00:33:46]: What is DF DAO?

Gubsheep [00:33:47]: DF DAO was a like team of players who in the last two rounds they scored really, really high and their strategy was essentially realizing that a single account doesn't necessarily have to map to a single person. So, you know once you have this like realization, I think it opens up like a really wide range of strategies. You can assemble these like teams or DAOs that have people in different time zones. So that way you can always be active.

Sina [00:34:21]: A player that's awake 24/7.

Gubsheep [00:34:23]: Exactly. So, you know, that you're like never going to have, like a lot of players have the problem where you go to sleep and then you wake up eight hours later and like someone, 12 time zones away from you has just like taken like the Eastern portion of your empire or something like that. So then beyond that, you can have certain members of the DAO who are specializing in negotiation or in writing automation. You can have players who are, you know, feeding like sub empires into the main empire by likeconquering boundaries and then transferring them into the main Empire's account.

Sina [00:35:00]: Wow. There could be, yeah. There could be full incentive systems. Like they don't even need to be cohesive units that are playing these things. Definitely could be just self-interested, like parties that are running different portions of this empire.

Gubsheep [00:35:15]: Exactly.

Sina [00:35:15]: Or they could even be different players. Right.

Gubsheep [00:35:18]: They're all bound by smart contract someday and yeah. So that' stuff's really cool. And it's another example of a thing that like our like core team definitely would not have thought about on our own because we're not experts in DAOs and we come at this more from the angle of like oh we want to build a cool video game. What kind of cool video games have we played in the past? So that's been really cool to see. And yeah, I think like all of these things are like, we get to harness the whole power of the whole creative energy of an entire community. When you do try to make the game open from day one.

Sina [00:35:54]: Yeah. I think maybe this is a good segue to talking about the future of this game and you are really leaning into this direction of openness helping an ecosystem create be built around this and even interoperating with other games and like, what are some of the things you're focusing on there?

Gubsheep [00:36:17]: Yeah. So, I think that one thing that's really exciting about crypto native games is that they can permissionless inter-operate with each other. So, for example, you know, we're in contact with a team that is building a new game ZK dungeon, which is sort of like a dwarf fortress or dungeon keeper-like game, but on the blockchain and using ZK tech. And the exciting thing about that is that without having to, well without, with the big asterisks, because you know, there's always the need about developer tooling and whatever else. But you know, technically without being in communication with us, a team like the ZK dungeon team could just decide that they're going to look into the Dark Forest artifacts contract and like create mechanics like, oh, you know, if you drop your Dark Forest pyramid into your ZK dungeon cave, then that like powers up all your units and your dungeon or something like that. And that makes the Dark Forest pyramid something more than just like an item in a specific game that's only been granted value by virtue of being in that one universe. Like now the dark forest pyramid is something that you can kind of like pick up and walk around with throughout an entire ecosystem of crypto worlds. So yeah, interoperability and having multiple games that communicate at the interface level seems like a really, really powerful thing.

Sina [00:37:47]: How do you think that would impact like the strategies people use within one game, because now there's like external incentives, right.

Gubsheep [00:37:57]: Definitely.

Sina [00:37:58]: Out of the system there's like other reasons why people would do certain things.

Gubsheep [00:38:03]: Right. I mean, you know, and I can only speculate on this because we're still so early. But one thing that I think about here is that some people are going to be better at or are going to care more about their identity or reputation in like one universe versus another, or their ranking in one world versus another world. So perhaps we would see like natural, like specialization, you know, some people would be like, oh yeah, I'm the person who likes farms like dark forest pyramids or something like that. Because I'm like really good at Dark Forest and then there is someone else who needs the status effective of Dark Forest pyramid in a game like ZK dungeon. And then there's a natural sort of like a positive-sum relationship that can form between those two games.

Sina [00:38:43]: It trades through like allowing people to lean into their competitive advantages and specialize. And like trade between different crafts that they developed.

Gubsheep [00:38:55]: And specialization and competitive advantage are exactly what makes like economies interesting. Or like market-based economies interesting. So hopefully, you know, you might be able to see some emergent dynamics like that evolve. But yeah, like we're really excited to try to see that world through and to that end, this project to us is not just about building Dark Forest, this one single game it's about trying to figure out with all of these other stakeholders and voices in the room, like Blaine and Jacob building project [39:29 inaudible] and DF DAO experimenting with what the future of like guilds might look like in games or all of our other third-party developers or the folks working on ZK dungeon, how do we create an open ecosystem where these things can all be connected together? So instead of just trying to like, you know, make our game really good and like shut out the competition or something, I think it becomes more like how we foster a community?

Sina [00:39:58]: Yeah. That's super exciting. It's the definition of like positive some thinking that we're growing this together rather than we're competing on this like a narrow thing that exists today.

Gubsheep [00:40:10]: Definitely. Right. I think that likeit's so much more powerful to have, let's say a handful of games that all interoperate with each other rather than a bunch of like siloed individual experiences. I mean, that's what crypto's all about.

Sina [00:40:24]: Yeah. Have you been inspired in your thinking in this way, by any particular other like ecosystems or things that have been built in the past? What's played a role in like shaping your worldview here?

Gubsheep [00:40:40]: I think that the biggest inspiration is definitely the Ethereum ecosystem itself I think just looking at how, you know, even following like, you know, the way that the Ethereum foundation has gone about decentralizing, Ethereum, like not just the technology, but also the community. You've got this ecosystem of like research organizations, nonprofits companies’and individual developers that all sort of like work together in this harmonious and positive some way. And it's, yeah it is sort of unconventional because like the first thing that I think a lot of people might think about when you're trying to like build something successful is like, oh how do I build my like company and make sure that it's got like hard MOS that allows it to, you know, uniquely succeed and like fend off competitors or something. And I think that there's something of a paradigm shift that has to happen when you're working on a technology landscape whose affordances naturally lend themselves to explicit positive games. So, thinking about now, like how do we actually like to foster healthy competition and in turn channel that competitive energy into cooperation in different ways as well. So yeah, the Ethereum ecosystem I think is really unique in how it's gone about evolving. And hopefully, that can provide a lot of lessons for us if we're really wanting to build this kind of ecosystem of crypto native worlds as well.

Sina [00:42:15]: Yeah, totally. Do you like we were talking about how this could evolve and like there's so much here, but do you see any fundamental bottlenecks or any building blocks that are needed that aren't actually like solved that are standing in the way? Or is it just a matter of doing the thing?

Gubsheep [00:42:38]: I mean a big question almost like an elephant in the room for all of this stuff is the question of scalability. Right now, I know that like moves on dark forest can take upwards of like a million gas because there's a lot of like ZK proof verification is very expensive on-chain still. But even beyond that dark forest is just like a complex game. It's, you know, more than just like a contract that's tracking ownership of tokens, it's like doing a bunch of like fixed-point math to calculate how far your ships went and like how much energy has decayed on the voyage from planet A to planet B. So,scalability I think is going to be an open question for a while. And I know that you know, looking at some of the ZK dungeon contracts calls a lot of those are even more expensive, like multimillions of gas per transaction.

So for us, what this means is we are just very actively tracking the work on scalability, both at like the protocol level and in terms of L twos as well. It's exciting to see a bunch of optimistic rollups move closer and closer to production and also to see the progress on ETH 2 which really feels like it's moving at like a pace right now. Obviously, also I think that side chains have played a very important role for us so far. XDAI has been just, you know, a really phenomenal environment for us to be running dark forest on and working with and talking with the XDai team. We're excited to see how they can contribute and how side chains can contribute to like the larger vision of a scalable Ethereum ecosystem. Yeah, I think that's probably the biggest question that I have a lot of uncertainty around, but I also know that this is not just a question that affects us.

This is sort of like an existential question for the whole ecosystem. So yeah, like in my head maybe this is also where some of the conversations around like public goods and infrastructure come in because it's like, if so many things at the application level depend on scalability work, then what is the responsibility of application developers and entrepreneurs in the space to help support that kind of stuff as well. And that's why I thinkthese organizations like the Ethereum Foundation are so powerful and so important for the ecosystem.

Sina [00:45:02]: Yeah, totally. Well, I think layer twos are very valuable things to work on. I don't think they're going to lack incentives themselves.

Gubsheep [00:45:15]: Yeah, for sure. And I think also like another interesting spectrum that we might see is likethere's going to be all sorts of like trade-offs that are necessary depending on like the security guarantees that your application requires versus like how accessible you need to be. If you're running a game or something, you don't need the same security guarantees that an optimistic role of like optimism or arbitrary built on the Ethereum main chain will provide. Maybe likethere are also questions around like how strongly do you need data available for your application as wellas far into the future?So, I don't know ifthere are futures even where you could imagine that asyou have like applications, specific chains spun up, or something like that. Or you have like execution environments where you have things like interoperability and likecryptographic mechanisms that ensure that like validators of the system cannot forge transactions, but you lose censorship resistance and data availability, and you just commit like hashes back to the main chain. And you make the data available on like IPFS or some hosted service.

Sina [00:46:29]: Totally. like a Validium model.

Gubsheep [00:46:31]: Right. And then provide like Merkel proofs. So, if you need to prove that you can like withdraw some NFT from this like game that you played.

Sina [00:46:38]: No idea right now. But I know that like that is application-specific rollups or you know a world of many rollups is explicitly a part of arbitrator's vision. Right. They're not seeing it as there's going to be one canonical roll-up. And even Reddit is building its own application-specific roll-up. And I mean, in terms of like needs driving innovation, you know, games are really at the forefront of things like where you're just making a ton of moves. Like there's the game logic is quite complicated this and other, like let's say like Reddit launching where they have lots of users, these are going to put the pressure on the infrastructure to like to improve to meet their needs.

Gubsheep [00:47:32]: Yeah, I hope so. I think thatlike if you know, we live in a world where there aren't applications that are pushing the bounds on demand then perhaps there aren't those naturally grounded feedback loops that are telling protocol-level designers what they need to be optimizing around. So, another hope for ZK games like Dark Forest or ZK dungeon, or other things, is that they can provide some feedback to people who are thinking about like scalability and stuff on what needs to be supported or what there's demand for support for in the far future.

Sina [00:48:14]: This is something I've thought about for a while. And like Dark Forest is a very good example of it, where I feel like applications are like fundamentally a lever for moving the space forward because one, they build useful things with the technology like that's their prime motivation is to like do interesting things. And two, like that interesting thing ends up attracting talented people to work on it. So, it's, you know each kind of interesting application is a node that's growing the amount of talent that's in the space, because like they're just attracting people in.

Gubsheep [00:48:50]: Right. And there's a feedback loop that you, that gets built up.

Sina [00:49:00]: And then, yeah, the last point is that they kind of have needs from the infrastructure and over time, like have to start solving their own problems. And to me, like web two examples of this is, you know, Google building Kubernetes or Google building the Go programming language.

Gubsheep [00:49:20]: Right. And the exciting thing to me, though about the way that this could play out in web three is that because web three is fundamentally a more open paradigm, you could see that the infrastructure that gets built is shared between different applications and different teams. And you have like open standards evolving rather than individual companies with like IP considerations or considerations about like a moat or whatever else, having to reinvent the wheel every single time they need like a scalable, like whatever, whatever store. So yeah. I think that's going to be really powerful.

Sina [00:49:52]: Totally. Well, good stuff, man. We've been talking for some time.

Gubsheep [00:49:58]: Yeah. We can call it here.

Sina [00:50:01]: Okay. cool.

Gubsheep [00:50:02]: Thanks so much, Sina.