Game Economist Cast

E24: Tragedy of the UGC Commons & Gold Bugs

Phillip Black

Send us a text

Eric reminds us that The Wizard of Oz is a heated monetarism debate, while Chris goes six layers deep on enforcing royalties. Phil thinks Warzone Mobile is a good executive Powerpoint math muddled with poor execution, but good economics secured him Taylor Swift tickets. 

Speaker 1:

Yeah, the Wizard of Oz. You guys know about this. You guys probably know. No, the Wizard of Oz was originally written as like a pro minting silver minting propaganda piece. Oz O-Z ounce was a reference to the US mint and in the movie it was Technicolor, so they made her shoes ruby. But the whole thing, the yellow brick road, is like oh, everyone says, follow the gold standard, follow the gold standard. You go all the way to washington. The wizard at washington who says follow the gold standard. But he's full of shit. He's just a guy behind a screen and the real way to go home was her silver slippers, which she had to tap. The solution was in their silver slippers all along. We just need a mint silver and a flay away all of our farmers debt every year I learned something else about that movie don't.

Speaker 3:

I don't know if I rank this above or below the thai story, the Thai restaurant story that's still living and breathing.

Speaker 2:

That's important to this day. I got Thai food the other day, eric and I was like this is soulless.

Speaker 1:

You'll never stop thinking about it.

Speaker 3:

Did you ask the owner where she got her subsidies?

Speaker 2:

I didn't, but I could taste the subsidies in the food.

Speaker 3:

Let's start with utility. I don't understand what it even means.

Speaker 2:

Everybody has some kind of utils in their head that they're calibrating, there's hardly anything that hasn't been used for money.

Speaker 1:

In fact, there may be a fundamental problem in modeling.

Speaker 3:

Episode 25, coming in hot, you can taste the subsidies in the food. We're back in town after gdc recording. I think we feel more energized than we ever have. We have an exciting lineup of guests this year. I've been teasing. We have one probably in the next episode and then two or three more lined up down the line in the next couple months. Hi, I'm phil from game economies consulting.

Speaker 3:

I'm joined today by I am er Eric from Superlayer, and I'm Chris from Star Atlas, and we have three wonderful topics to talk about today. Chris, what are you gonna be talking about?

Speaker 2:

I'm going to be talking about ERC-721c and some of the other token standards that Limit Break released about a year ago today, just a topic. That is something that's quite important in Web3 and gets at the heart of the whole entire purpose, the whole entire mission, and it's something we don't talk about a lot, and eric, what are you talking about?

Speaker 1:

I've been looking at the fortnight creative ecosystem epics, big ugc plans and how they do revenue attribution.

Speaker 3:

It's more like a command economy than a free market and the trade-offs both of you have stub stacks which we're going to link out to in the show notes you guys are trying to publish weekly. There was some sort of bet. I heard about that.

Speaker 2:

Someone is losing one soul. I think both of us failed. Eric failed last week, I failed this week. So we can you do a smart contract for this one could.

Speaker 1:

You would need a they call an oracle some way for the smart contract to verify that we have published on sub stack this would be a good suback article how I wrote a.

Speaker 3:

How I wrote a smart contract about a bet on substack I'm not technical enough to do that.

Speaker 2:

I was gonna say it requires a little more coding skill that's part of the question.

Speaker 3:

You got chat gdp on your side oh, chat gpt.

Speaker 2:

Don't get me fucking started on chat gpt, but before we begin, let's talk about what we've been playing.

Speaker 1:

oh, chris, I think it sounds like chris has lot on his mind A lot of board games these days.

Speaker 2:

No, no board games. Ever since we got back from GDC I have been playing board game-like video games. I got recommended two different games this last GDC Well, one was before GDC and then one was at GDC that are very similar Not super similar, but kind of a. There's obviously not DLC, but there's microtransactions and DLC in the game, but it really feels like a polished, complete product, no ads. It's a really good game. And I also started playing Square Valley, which was recommended to us by John Cross.

Speaker 2:

Now that is a premium game, so $2.99, buy the game and you have all the features. And that's a story game and that's a tile place or point salad salad game. So you put your farm next to your sheep and you get bonus points for combining a sheep in a barn or whatever it might be. So you have these scoring matrices that you know you want to optimize. So this tree gets plus one point for every single tree that it's connected to, so you get a big forest. This is like really classic board game designs where you have this I call I said point salad where you add a bunch of points together and there's different bonuses and scoring conditions and whoever has the highest points at the end wins. In this case it's a solo game. You just play to try and get the to meet the criteria for that match, for that game, that specific map.

Speaker 2:

So two similar games, very different monetization strategies. So polytopia uses, like I, they have in-game transactions. You can buy different clans and I think you can buy different maps. I actually was trying to see how to spend money in that game and the only thing I could really figure out is how to buy new clan guys like new armies, and I think they're like maybe I don't remember the price I want to say between two and five dollars. I don't know what the ltv is for users in polytopia, but for such a polished game they can't be looking at more than ten dollars ltv or sorry, like ten dollars max ltv for a player. Once you bought the six clans there are to buy, you're like out of content.

Speaker 2:

As far as I can tell, and I've always been a big advocate for if you're going to have a free to play game, you shouldn't make it super hard for somebody to figure out how to spend money. I actually even think Brawl Stars was a little bit hard for me to spend money. It wasn't like obvious to me where I should go, what I should do. I think that was one of the criticisms I gave was like I was nervous to spend money. Even if I wanted to spend money, I wasn't sure what to spend it on. Only criticism to Polytopia is just they've got a really beautiful, polished product, very much the supercell model.

Speaker 2:

Let's give me some ways to spend some money in that game, because with a 4X game you can imagine a million different ways to monetize, to incorporate different mechanics. I'm not a free-to-play expert, nor am I a mobile person, but I imagine that they have some improvements. They could make Square Valley premium game $3. Not really much more to say there. I think they could probably get away with it and I don't know if they do this. I haven't actually dug too deep into it, I've just played the first 10 levels, but they could probably get away with once you get through the core game. Add-ons, special features like the dark forest for $1, something like this where you add in a map or you add in a storyline. But both really good games and I've been playing them on ipad. Recently got an ipad super pumped about it.

Speaker 3:

it is the ultimate mobile gaming machine you and super circa 2018, still believe in ipads the future oh my god, it's amazing.

Speaker 2:

It's amazing. I've been playing chess on it beautiful. It's amazing because I don't like to use my computer after work, because it reminds me of work, so I I need, like a big screen that I can poke around on. That's not my. I have an iphone 7, so, like this thing is an ancient piece of technology that has a tiny little screen that I can barely see yeah, I think tablets are great for mobile games, but I don't know, I never really caught on don't little kids no kids are using.

Speaker 2:

I always see kids with tablets, like all of my nieces and nephews have tablets.

Speaker 3:

I think it's because they don't have the dexterity to be able to navigate a very small screen, though I think they probably have more dexterity on a tablet.

Speaker 1:

Also, tablets are cheaper. The base iPad is pretty cheap.

Speaker 2:

It's $300, some oh yeah, I got mine for $250.

Speaker 3:

Because it doesn't even tie to a phone. You don't need a cellular plan.

Speaker 1:

Don't have to worry about your kid calling 911 by accident. Does that happen to you, Eric? Yeah, no, they mash buttons. They see a thing and they just mash buttons. A little operator, yeah.

Speaker 2:

You don't think that the kids are going to want this. We grew up with controllers in our hands when we were kids and we still use controllers. You don't think those kids are going to grow up and be like I had an?

Speaker 3:

iPad. The dual analog sticks, the virtual analog sticks, that's been my big thesis.

Speaker 1:

These cracked out 10-year-olds playing Fortnite on their iPad. They're super fast, they're doing all this stuff.

Speaker 2:

I don't know how they do it. I'm terrible. I tried to play Vampire Survivors on the iPad and I died.

Speaker 1:

You broke your Vampire Survivors, cherry, five years late. How is it?

Speaker 3:

I know it's very interesting to play next to Vampire Survivors. They live-serviced all of those mechanics. If you care about money, eric, what have you been playing?

Speaker 1:

I've been playing a bunch of Duelists of Eden. I read an article about it. It's like a niche indie fighting game.

Speaker 3:

Surprise.

Speaker 1:

It's the two intersections of the games I like, which is indie games and fighting games. But it's really cool, interesting. So if you've played Mega man Battle Network, it's heavily inspired by that. So it's a fighting game like in terms of experience, but it's deck-based, Like you build a deck of 8 to 10 cards, which is your attacks, and that means there's a lot of and you only have two cards in your hand at a time. So there's a lot of thinking on the fly of when a fighting game you might be like, oh, medium punch into shoryuken is a combo, so I practice this specific combo.

Speaker 1:

But in this game the cards you have available aren't always the same. So you have to know oh, this card combos into this other card, but I don't have card B right now, but I have card C, and how can I string these together? And similarly, you have to watch what your opponent has. Oh, my opponent has a bunch of things that threaten the front row but not the back row, so I should move to the back. And then they use a couple cards. Oh, now they have moves that target this zone, so I have to change where I'm positioning. It's tough because there's a lot of card literacy. You have to read what cards you have and your opponent has and adjust your game plan.

Speaker 3:

Is it real time?

Speaker 1:

Yeah, the thing about the game is like it's probably around 50 concurrent users right now. It launched at 250 and now it's a 50, which, with any multiplayer game, you know, once the player base drops below a certain level, the game sort of starts to die, and I wrote a little article on this. Check out my slip stack link in the description. But yeah, they, once it hits queue death, once the population runs out, the game just dies, right, and so I think games who are in this situation need to aggressively consolidate queues and improve matchmaking times in order to keep the game alive as long as they can, and I think the devs done a good job with that. How does this compare against Clash Royale? So the card and mana system is very similar to Clash Royale. You've got a hand of cards, you've got a mana that refills over time and it costs monoth cast spells, but the moment to moment gameplay is very much like a twitch action game as opposed to like more of a slow strategy game yeah, jesus, this thing looks intense.

Speaker 2:

Man, that's great opposite of the type of game I like, especially competitive and fast are like the opposite of me not competitive and I'm not fast. Yeah, it definitely had a pretty steep learning curve.

Speaker 1:

At first you, because not only you're trying to play a fast fighting game, but you also don't know what any of the fucking cards do, so you're just mashing.

Speaker 2:

You don't know what each button does, so it's got to be one of the only card games that's real time, right. I've heard of board games that are real-time card games, but typically the card games I go, you go, I go, you go. How does the card replenishment work?

Speaker 1:

do you like just?

Speaker 2:

every time you play a card, you get a new card.

Speaker 1:

That's right. You've got a deck of cards. As soon as you play a card, you get to draw a new card, and when you use all the cards in your deck or you manually shuffle your deck, it takes three seconds and then you reload your hand.

Speaker 2:

When your opponent's shuffling.

Speaker 1:

It's like a big vulnerable moment because they have no powerful attacks. Oh my god.

Speaker 2:

It's very meta. Oh, my opponent's shuffling his deck. I'm gonna attack him with fireball in magic gathering fucking bullshit. He's like fetching his land. He's trying to like shuffle his deck real fast. So why do you think that this game died?

Speaker 1:

or I guess it was small to begin with yeah, I was gonna say what was the da was a thousand or something yeah, I want to say their peak ccu was like 500 on launch day but it very quickly was in the low hundreds and currently below 100.

Speaker 2:

Do you find that, like people in the fighting community grab just are? There are certain genres. It's like any game that comes out within that genre, everybody plays that game. Who's obsessed with it? Do you find, like I try to play most ccgs, especially digital ones? Do you find that's true for fighting games? Or are they more set in their ways, like I'm gonna play street fighter and that's and that's all I'm going?

Speaker 1:

to play Depends on the people. It's definitely a lot more dynamic than other live service things like MOBAs. People just play League or Dota and that's it forever. So there's some people who are like they only like Street Fighter games. But there's a decent amount of crossover where a Street Fighter player will try Tekken if the new Tekken game is hot and all the people they know are playing it. But there very much is a what's the word? Like a shelling point, like critical mass. There's a thing where, like whatever a lot of people are playing is the thing you want to be playing.

Speaker 2:

Network effects yeah. Bandwagon effects yeah, because the cascade effect.

Speaker 1:

Your enjoyment of the game depends on how many other people are playing it. Right. It's not you just consuming it in isolation.

Speaker 3:

That's network effects so really subtle positive externalities right by that. Speaking of things people won't buy, I've been playing call of duty warzone on mobile. This has been activision's attempt to bring the call of duty franchise to mobile, and if you're confused because there already is a call of duty mobile, then you should be because that was released a couple years ago by Activision and Timmy, which is a studio from Tencent which has done gangbusters in terms of revenue. It's crossed over a billion dollars since it was released. They finally got monetization to a good place, but it's one in the east and in the west.

Speaker 3:

I always thought the revolution of that game was that it brought virtual dual analog sticks to a mature audience on mobile, which is something that I think was very hard for gamers to wrap their heads around, especially mature gamers Like this is where mobile started. It was just porting HD games onto mobile and slapping some virtual analog sticks, but that was always. The issue is that it's hard to use virtual analog sticks on mobile Because when you put both of your thumbs on the screen, you don't have an additional finger that's free, in a natural position, to be able to fire. And when you look at players who play at a high level on mobile devices for first person and third person shooters. They use a claw like motion with their hands, and so you have your index finger, which is positioned in a way to press the fire button, which is extremely awkward to hold, especially if you're holding the phone with your hands itself rather than laying it down on another object, and so I thought there was going to be a big breakthrough moment, but nothing's really broken through after Call of Duty Mobile, but still it's made a shit ton of money, still successful. So why not do Warzone and then try to be a really bad faith partner to Tencent, which I'm sure they have a revenue share agreement?

Speaker 3:

And this is like, ultimately, the dream that Bobby Kotick, who was the CEO of Activision, had. I think it was the same dream that we shared, or at least the executives that EA shared, which is that you have one engine, which is rumored to be the case with Call of Duty mobile. They have an Infinity War engine. That is for both the HD version and the mobile version, and it'll be zero marginal cost, because the thing that's super interesting about this game is that the content that's on the HD version is also on the mobile version and there's account linking so you can bring over all the purchased content you've had beforehand. This is something that Activision has been running in the Warzone series itself. So when you buy the next Call of Duty, a lot of your cosmetics are going to be ported to the next version. You don't need to purchase things again.

Speaker 3:

And so you can imagine how tempting this is on a slide. How tempting it would be if a game economist were to argue that hey look, it's zero marginal cost, we'll just build up the SKU. When we have one piece of content, we can amortize those costs over a much wider swath of players. So let's say, the mobile SKU adds five million marginal players. Okay, now that one piece of content is going to monetize across that wider swath of players.

Speaker 3:

It makes so much sense on a spreadsheet, but the execution is horrendous. It's so bad. The phone heats up like no other. The engine clearly is not in working condition. Not only that, the focus is on Warzone, which is the Battle Royale mode, and the draw distances are really long. In Battle Royale maps, player counts are very high. They didn't do a ton of adjustment to the player counts.

Speaker 3:

It's still a large number of players and the big revolution of call duty mobile, or at least part of the revolution, is that they popularized autofire. So, rather than needing that additional finger that you would have to make with that c-shape, as long as you just align the crosshairs, it fires. Well, that was the original mobile, that was the original. That was the original mobile version. Is that they have the the auto fire feature, and that's still here, don't get me wrong. But when the draw distances are so far away, auto lock or auto fire doesn't trigger as often, and even it does, it can be slow, and so the experience just feels. Now they do have traditional call of duty modes. You actually had to unlock them a little bit. You have to level up quite a bit to be able to get to the other team deathmatch modes, which kind of reduces this problem because you're playing on much smaller, more traditional Call of Duty maps.

Speaker 3:

But nonetheless, like they still don't have the quantity of China. China just makes a fuck ton of content, which is part of the joy of playing the game, and they have loot boxes. They have a ton of fun monetization mechanics, and I've argued for a long time like the monetization mechanics increase engagement to. It's not a price that you pay to generate revenue. It's something that subsidizes retention. I've churned. I think it's over for this game. Numbers don't look great and I play a fuck ton of call duty, so I'm just really sad about this. I thought there's going to be another big moment for dual skews and this whole cost amortization across both skews, and I don't think it is. I have a question about hardware.

Speaker 2:

What about? Are there any ipad or or mobile like controllers?

Speaker 1:

that allow you to more easily.

Speaker 3:

I already have one now, you see that thing? Oh really I have a razor oh, that's right.

Speaker 2:

For some reason, I thought that was like an additional controller for your steam deck or something like that no, it makes the phone into a steam deck.

Speaker 1:

It's crazy that's sick.

Speaker 3:

Uh, for my ipad it doesn't work that well damn it doesn't work that well. And the other thing they do is they put you in different matchmaking pools and the problem is exactly as Eric has written about before the pools are too small to be able to support stable matchmaking.

Speaker 2:

How do they know that you're using the tool? Doesn't it just touch? Does it plug in or does it? I figured it would. Just it would basically side of the and tap the screen with little analog pads.

Speaker 3:

There. There is a controller that does that's physical rather than having any digital connection. Yeah, get around that damn matchmaking.

Speaker 2:

But I would pay money for just two joysticks because I want to play vampire survivors on my ipad but I just can't do it. My thumb ends up in the middle of the screen and I'm like a toddler, like I understand your pain so they segregate the controller versus touchscreen matchmaking.

Speaker 3:

They do. And just to be clear, there also is no crossplay between the HD version of Warzone and the mobile version. But even on mobile itself they'll segment those two audiences.

Speaker 1:

They should just combine them into one pool and just adjust MMR accordingly. I completely agree with you. That's what Fortnite does.

Speaker 3:

I actually don't know if they're doing this on Warzone, because I actually couldn't get my controller to work, but I know they do that on Call of Duty Mobile. So is Call of Duty Mobile still a pretty high DAU? Oh, it's 16 million last time I checked. It's fucking insane.

Speaker 2:

Okay. So why didn't they just come out with Call of Duty Mobile 2, instead of trying to port Warzone over to mobile?

Speaker 1:

Because they don't want to give the rev share trying to bring it in-house.

Speaker 3:

Yep, yep, and I'm sure the P&L looks really complicated, I'm sure the ownership of the code and a lot of those assets is very complicated too, which is crazy, right, because they're bidding against keywords, against one another, they're bidding against marketing. So if you have one player who's interested in Call of Duty, who's going to drive the highest bid, probably the other freaking game. So you're artificially increasing your own bits and user traffic.

Speaker 2:

So it's just a bad decision to publish with Tencent in the first place.

Speaker 3:

I just think this is really dumb. I think this is a whole bad faith negotiation. I think it's just really complicated. The problem is they partnered with Timmy and I think for a lot of studios you've got to own the means of production, or at least you've got to have a pathway to own the means of production.

Speaker 2:

And even though Timmy has made a lot of money for Activision. The fact that you can't internalize any of that revenue, I think is a big mess. That's the advantage Nintendo has they have complete and utter and they've always had complete and utter control over the production of their assets. It's a decision that's completely antithetical to the Nintendo philosophy.

Speaker 1:

I have a comment on this. Aaa Western studio tries to port to mobile with a Chinese developer. So we've seen this pattern a bunch of times, you know, with League of Legends, wild Rift, you've seen it with Diablo Immortal, you've seen it with Apex Legends mobile. Wait, wild Rift was China. I think it was also Timmy, really.

Speaker 3:

Yeah, are you serious?

Speaker 1:

I don't know if exactly. I think it was Timmy. It was definitely a Chinese mobile developer. And the story you hear with all of these first is that they tend to be very successful but on the other hand, all the Western devs hate working with the Chinese mobile developers. There's basically like the recurring story of they don't listen to what we say, we like, we put, try to put these guardrails in place and they just do their own thing. Or maybe they're unhappy about the rev share or something and they ax these agreements and they're like oh, this studio is probably Chinese studios, actually probably stealing our code and using it in their other games, and so they try to do it in-house and they fail to do it in-house.

Speaker 3:

The recurring pattern, not. I think this is all in-house, I think Wilder's all in-house.

Speaker 1:

No, whoa they had a set of North American, la based game designers and the development team was in China. But they kept throwing thing information over the wall and it not getting. And then they got so fed up at some point they cut the Riot design team and were just like fuck it, timmy, you guys just run with this.

Speaker 3:

Okay, that's new information. I didn't know about that, which is ironic because that was the whole drama to begin with, that Tencent developed Arena Valor because Riot thought they were too good for mobile.

Speaker 1:

Yeah, exactly. And then they were like oh actually, let's make a League of Legends branded one. And then yeah, oh shit Anyway.

Speaker 1:

Yeah, this is recurring and, like the Asian developers are way better. Like you said, the COD mobile controls like so much better than the Western developed ones in Warzone, right, and we see the same pattern as well with Kings of Glory. The Asian developed MOBAs have all sorts of auto-fire, auto-aim like smart pathing features that make it work much better with the shitty mobile controls, and these PC developers don't know how to develop all these quality of life features.

Speaker 2:

So I want to start this conversation, or this part of the cast, with a few caveats. The first one is I work at a studio that works exclusively with Solana. I try to follow Ethereum stuff, polygon stuff I've obviously used the name in Polygon Polygon more so than Ethereum, and we think it's important to cover. It also helps to emphasize the importance, or, I guess, the secret sauce, to Web3. The importance, or, I guess, the secret sauce, to Web3. Why do Web3 people, why are they so obsessed with these like NFTs, these tokens, this blockchain, why are they so interested in this stuff? So this will more so be, I think, a conversation that will help people who don't know a lot about the topic or the space understand more.

Speaker 2:

For certain people, this is probably just going to upset them because we're going to butcher some stuff, but I still think it'll be a valuable discussion. So Limit Break is a studio behind DigiDigaku, at the very cutting edge of Web3 Game Studio, and they're more focused on the ownership the exclusive ownership of an item than some of these other games that are more focused on earning or play to earn. Those are the classic Axie examples. I would argue that Star Atlas is certainly in that camp where you can come in. You can buy some ships and you can fly around and you can try and make the in-game currency which has a floating price in the real marketplace.

Speaker 2:

So what Limit Break had to do in order to succeed in their monetization strategy, which is almost a pure royalties and secondary sales method? So they're going to get secondary transaction revenue through the secondary sale of their NFTs through a royalty. So maybe this is 5%, maybe this is 15%, maybe this is 100%. If I create the NFT and somebody exchanges it on a marketplace, I'm going to get some portion of that sale price. This is essentially a tax.

Speaker 2:

The other, at least in the early days, my understanding of their monetization strategy was the ability to sell items into the secondary marketplace as a primary. So let's say that there's a demand for a specific type of item and it's not being filled, or there's sufficient demand for a specific type of item and it's not being filled, or there's sufficient demand. You can, as the publisher, you can step in and say hey, there's there's 10 000 people that are willing to spend 100 on this nft or this item that's needed in the game. I'm going to sell 10 000 of those. And it's this opportunistic secondary selling behavior at least I believe that was digi dug out in the early days I don't know, and I think that's how most fungible token projects work.

Speaker 1:

They drum up some existing circulating supply and some live market and they sell into that to move cash in. Yeah.

Speaker 2:

We do something similar at Star Atlas. Right, we have primary ships on the market and then players can list secondary ships on the market. So if you buy a ship, you can go and you can sell it on the marketplace. We've actually had an interesting phenomenon occur recently where the secondary price of the ship event finally or sorry, I should say the primary price of the ship, the price we sell it for finally fell below the secondary price. So if a ship is sold out, we don't really care what the secondary price is. If a ship is not sold out, we're almost always going to have our sales cannibalized by secondary sales. Right, If we're selling a ship for $100 and somebody else is I'm going to sell for $85, then we're not going to have any sales go through. But we could, for example, discount a ship in order to sell more volume. So let's say that it's 85. We don't do this. We don't opportunistically discount sales in Star Atlas. But that's more so a philosophical decision as opposed to a business decision. So we choose not to opportunistically sell into the secondary market and what that would mean for us would be we've got these styles of ships sitting on the marketplace at $100. The current secondary market price. The actual market price, I should say, is $85. Let's sell this for $50 off. We don't do that, but some people do and that's a very reasonable monetization strategy, or monetization or revenue strategy. That's limit break.

Speaker 2:

In order to do this, in order to have these enforced fees, these enforced revenues for the company, they needed to generate new protocols, new blockchain protocols. The way that the blockchain works, especially chains like Ethereum and Polygon, is a subchain or a layer two of Ethereum is everything is written in what's called smart contracts. Smart contracts are just code that are stored on a blockchain and those pieces of code can be called by a user and, as long as the user satisfies a certain number of criteria defined by the piece of code, it will run and it'll send and it'll execute a transaction on the blockchain. So the way that these programs are written up is they're a series of a bunch of different instructions and each instruction does a very small thing. So you can almost think of this as like programming when you're writing Python code. Python is basically a bunch of big programs that are composed of a ton of tiny instructions and you could think of an instruction as what's the basic programming language. It's almost like machine code not quite machine code, it's like assembly, it's like push, pull. These like very basic computational operations add, subtract that's what an instruction is.

Speaker 2:

So, for example, in Star Atlas, we have a bunch of programs and each of those programs is composed of hundreds, if not thousands, of instructions. So when a player, for example, earns XP, that is going to be a part of an XP program, but that program is going to execute a bunch of different instructions. So let's say the player gets XP, their XP account number is going to go up and we're going to call a specific instruction that's going to push that number up. We're going to call another instruction that's going to say did they level up? And it's going to level them up. There's a bunch of different instructions associated with a bunch of different programs. That's how blockchain architecture works.

Speaker 2:

Now, I just talked a lot about Star Atlas and that's really how Solana works. We have tons of blockchain programs that allow players to interact with one another in a specific way, in a constrained way that we want them to be able to interact with one another. You can call blockchain code from anywhere. As long as you satisfy the requirements that program says you need in order to interact with it. You can interact with it. For example, you don't need to use the UI. You can interact with a lot of the Star Atlas programs without ever playing the game, without ever getting into the actual Sage game. You don't have to go to the website. You can interact with it directly by calling those programs, as long as your wallet, your account, has the associated values, in order to send through some sort of execution, or sorry, in order to execute some sort of instruction. You will satisfy that, Okay, that's how this smart contract, blockchain architecture stuff works. Now, that's all about programs and that's all about units of account.

Speaker 2:

This number goes up, this number goes down, and it's mostly talking about accounts, and it's mostly talking about semi-fungible or fully fungible tokens. Semi-fungible token is just something that has an integer value, so, like fuel in our game, is a semi-fungible token. It's basically a commodity. There's millions of them and they're all identical. An NFT is completely different or it's very different. It's non-fungible. It's each and every single one is unique. Now, the way that we do this in Star Atlas, or at least the way we've done it so far, is basically by having a bunch of on-chain programs that are almost pretending that everything is a non-fungible token, even though they're actually semi-fungible tokens. Now we token, even though they're actually semi-fungible tokens. Now we are going to move to a non-fungible token model into the future.

Speaker 2:

But this is what Limit Break has been doing since the beginning. So they have a bunch of NFTs that represent the characters in their game and they need a way for players to be able to trade those unique characters in a marketplace where they can still make money. So, let's say, they have a marketplace that's called Limit Break Com and you go there and you're able to trade. That marketplace is going to be using a blockchain program called Limit Break Marketplace Sell, Limit Break Marketplace Buy. It's going to have a whole bunch of instructions and players are going to go through to that website, they're going to push through instructions through that UI and there's going to be within those instructions, within those programs, there's going to be some sort of blockchain code that sends money to Limit Break whenever a transaction goes through. So they're going to get their 15% or whatever it is.

Speaker 2:

However, what happens when a player says I don't want to go to LimitBreakcom, I want to go to OpenSea, or I want to go to one of these other marketplaces, or I know Joe's got this nice, shiny NFT. I'm going to go send Joe a thousand bucks and Joe's going to send me the NFT and we don't have to worry about that. We can just do OTC over the counter, just peer to peer. There's no middle person. So with these external marketplaces and this OTC, there ended up being a problem for Limit Break in that they weren't able to properly monetize because people were just getting around. Why would you pay 15% if you don't have to? I'll just go to OpenSea or some other marketplace where there's a lower fee and avoid the thing altogether.

Speaker 2:

Now you might say these different protocols OpenSea, for some reason, that's the only one that came coming to my mind right now but these different NFT marketplaces that allow players to exchange different NFTs for one for a currency, whether it's a cryptocurrency or it's a stablecoin. The problem is that there was no agreement between those different programs, those different websites, on how these things should be handled. So should we charge 15%? What kind of permissions Should users be able to send it to anybody? Should it be a whitelist? So how do we handle this type of trading was completely up for debate, for debate.

Speaker 2:

So what Limit Break did was they said okay, rather than try and clog the many holes that are in this boat, we're just going to build a new boat. We're just going to come up with a new token standard. So they came up with two new token standards, and a token standard is basically just a series of blockchain instructions and code, like I was saying and apologies if there's any super expert blockchain people listening, but they're probably going to be insulted by my butchering of this topic, but I think it's sufficient for the layperson. So what they did was they said okay, here's a list of code and this is how the current Ethereum polygon, Ethereum-based blockchain NFTs work. And what they said was let's rewrite it so that there's additional features and restrictions, importantly, restrictions on these tokens so that they can or cannot travel over the chain. So basically, what they're saying is, rather than trying to adjust every single program that's out there to try and make sure that we're getting our royalty and even if we do that, players can still always go OTC let's create a token, a piece of blockchain code that abides by the rules we want to enforce. So basically, there's this.

Speaker 2:

So they created two new token standards ERC-721-C and ERC-1155. I think that's yeah, 1155-C, which are just new versions of ERC-721 and ERC-5. 721 and 1155 are basically NFT or SFT token standards. 721 is purely non-fungible tokens. 1155, which is a more kind of general, one can do semi-fungible and NFTs so they can actually break it down into they have decimal points, basically for the 1155.

Speaker 2:

The problem with these tokens is they didn't have really good royalties or restrictions on who could trade what. So what they did was they came out with, so what Limit Break did was they said okay, let's adjust this, let's build on this architecture that's already built out there. That's very important. All blockchain code is public. All blockchain code is open source. That's the whole entire thing. So what they said was okay, we can't enforce these royalties with the current protocol. We're going to come out with new protocol, we're going to update them, and essentially what they did was the RC721C and 1155C built in security layers and a royalty fee structure.

Speaker 2:

So if that item transfers, there's, I think, six or seven different levels of security. Now these levels of security have to do with who can send a token, who can receive the token and who has to pay and who gets paid when that token gets transferred. So level zero is going to be anybody can send anything and there's no fee or there's like a fee that's optional. Level seven let me actually say level six is only specific people can send, only specific people can receive, and there are fees that are enforced when the token is being transferred from one player to another, and they have a bunch of really cool things that you can do in order to make sure that the revenue, that's the royalty, is being split between multiple parties. You can transfer the royalty to a different person, so limit break could potentially transfer the royalty rights from themselves to somebody else. So that's what these tokens did.

Speaker 2:

Now, why is it important? It's important because in the real world, we have these physical laws and these physical constraints that allow us to pretty easily enforce these different rules. In blockchain, it's the wild west. Everything is code, everything is changeable and anybody can call anything at any time as long as they have the correct permissions. So without the correct restrictions on a specific piece of code, you can end up with people bad actors coming in and adjusting things when they necessarily shouldn't be able to adjust things or change things or put transactions through. So you have to be very careful when you're building this stuff out.

Speaker 3:

So here's what I don't understand. So the whole point of the standard is to prevent the situation that you described earlier, where I want to buy something for $100, but the smart contract says that if I pay you $100, 5% of that is going to go to Eric, and ultimately I don't want Eric to get that fee. What I'd rather have is us take the fee that Eric would get and we can split it between us. Us take the fee that Eric would get and we can split it between us, so I will pay a little bit less and you'll get a little bit more revenue. We're basically going to eat that margin, and so what I'll do is I will give you the hundred bucks in a separate transaction. I'll just give it to you, you don't have to give me anything back. And then you'll give me the NFT in a separate transaction. I don't have to give you anything back, and so those look like different transactions, even though beforehand we had a royalty attachment, because you gave me the NFT and I didn't give you anything in exchange.

Speaker 3:

In one of those transactions. The royalty triggers nothing, right? It's 5% times zero, and zero is the transaction fee. How does the standard prevent that from happening and let people enforce royalties?

Speaker 2:

So I don't know, I haven't literally read through all of the blockchain code, but basically it does it through those security levels I was telling you about. If the item transfers hands, so it matters not what the person, I don't know how they necessarily calculate like that amount, but you can essentially put some sort of fee that's associated with transferring that thing. So it's not necessarily a percentage of the sale price. Oh, interesting. Okay, it's just some fee.

Speaker 3:

So it's not necessarily a percentage. Oh interesting. Okay, it's just some fee, so it's a fixed you can almost say a fixed fee. If this gets transferred, this guy gets $5, no matter what the fee was.

Speaker 2:

Yeah, so I mean you could also see it as like a price floor. So that's my understanding of how it works.

Speaker 3:

Not only that, the limiting is also very interesting too, because when you go onto a platform like eBay, you can still do exactly what I described on eBay and avoid the fees, but you need to have a level of trust, and so if you create a whitelist and say you can only trade on these platforms, you can get to an eBay based place even though that's a possibility at the margin makes this stuff much, much, much harder to execute against.

Speaker 3:

The thing I love about this is like the whole promise of Ethereum was that it's a coding language and you can do anything you can with coding. That you could with Ethereum, and I've seen so few people take advantage of that, because, to me, what this is, it's a move to closed systems. It's a move to more closed shit, which to me, is just a sign of maturity. This is actually, to me, this is, as a Web3 skeptic, this is me believing more in the system. Now that things are starting to close up and a lot of the Ben and Jerry's libertarianism is starting to wash away, this is good. This is more control of things. We need to run effective businesses.

Speaker 2:

Yeah, the problem is the whole industry is run amok with this consumer mentality that the consumer is now the one who should get everything. So there's a misunderstanding of the consumer and producer surplus and there's this belief, especially by consumers. I know firsthand that they deserve all of the surplus. There should be nothing left over for the company. The only purpose of the company is to kickstart this thing and create a bunch of open source code which, who knows, maybe there's some sort of revolution there that I'm just not able to see, but I do think it's very damaging and it's anti-capitalistic and it just there's. What motivation do I have for creating something if I know it's always just going to go to the players? So I enforce a royalty. I enforce a royalty. If the players get pissed about that. What do you want me to do? How do I justify ever building this thing? I'm never going to build this thing, so I agree with you.

Speaker 1:

One still so on starless, you guys had some issues with users going off platform to trade and dodge fees. Would you guys consider implementing something like this?

Speaker 2:

so obviously we're on Solana, so from a practical standpoint, I don't think that we can use these. We can't use this exact architecture, obviously. So the architecture on Ethereum is quite a bit different than that on Solana, but the other thing is there are it's a different programming language. So I think ETH uses Solidity, solana uses something else.

Speaker 1:

Solana uses Rust, else. I think okay, yeah that's what it is.

Speaker 2:

I know that there are t protocols. I don't know if there are quite erc 721 protocols, and that's could partially be because of the difference in the way that solana is used. Take a take star atlas, for example. We actually don't have any nfts, at least not yet. We're trying to transition into an nft world and this emphasizes the importance of the ownership, not necessarily the importance. We actually don't have any NFTs, at least not yet. We're trying to transition into an NFT world and this emphasizes the importance of the ownership, not necessarily the importance of the thing. The thing. They can interact with the game the same way they would if it was an NFT. It's just not technically an NFT. But yeah, I know that there are protocols out there. There's things called compressed NFTs on Solana that perhaps have that kind of ability to enforce royalties.

Speaker 2:

I haven't actually done a lot of research. I know more about the Ethereum standards than I do about the Solana standards, but that's just because Limit Break has really been pushing the edge in terms of this and, frankly, there's not a ton of games on Solana. That is the reality. A lot of games are on Avalanche. A lot of games are on Polygon. It's another popular one that.

Speaker 1:

I'm not Forget about whether it's technically feasible. Assume it's technically feasible, would you want to implement this? That's a good question.

Speaker 2:

So the problem is, I guess they do have the 1155 standard. So, as long as we could, if we could implement it on both SFT, Sumbie Fungible Token, so that's like a commodity and an NFT, I would definitely see a use case. I would be surprised if we wouldn't want to use something like this.

Speaker 3:

So what prevents every other blockchain from just copying the standard instantly? Do they interpret code in a different way? Do they use the same language in smart contracts?

Speaker 2:

So different. Okay, so, solana, this is definitely going outside of my scope, but I can talk about this a bit.

Speaker 1:

Oh, okay, so Ethereum uses what they call solidity, which is basically JavaScript and everything. If you may have heard EVM, ethereum Virtual Machine. Everything that uses EVM is using Solidity, so basically you can copy paste the code pretty much, and so all Layer 2s and a lot of Ethereum forks, like Polygon and Arbitrum and stuff, are using something very close to what Ethereum does. Some of them have slightly different things, like I know all the zero knowledge proof layer twos, optimistic rollups. There's some stuff that's slightly different, but it's pretty easy to copy. Solana uses Rust, which is a different language it's actually pretty popular among game developers which is strongly typed and supposed to be a lot more efficient but a little harder to work with than Solidity, and so it is harder to copy stuff from Ethereum to Solana, but Ethereum to pretty much most blockchains is pretty easy to copy, so we should see standards like these emerge in Layer 2s.

Speaker 3:

and what other weird chains.

Speaker 1:

Yeah, so it's already on Polygon, because it's just a straight copy paste.

Speaker 2:

Now there is also the difference in how the blockchains handle memory. I know, like 1155 uses less memory than the 721. And again, don't even ask me like why that is. That's like far too machine code for me, but I know Solana is fundamentally different. The way that they process transactions is fundamentally different from Ethereum. Even now that Ethereum is proof of stake, they're still very different, because Solana is proof of history, I think so proof of history just the way that these things are transacted.

Speaker 2:

It requires a completely different protocol. You would have to rebuild it from the ground up. It's not to say it can't be done efficiently. That said, if anybody's been using Solana over the last few weeks or months, they know that the whole entire thing is an absolute dumpster fire. Transactions are failing left and right. It's extremely difficult to get anything to happen, and it's because of the way that people build blockchain programs. When you're sending through a transaction, we'll have three or four instructions built into it. So you have this 256 bit instruction or transaction that has a lot of different things that need to happen in order for it to succeed, and then you batch transactions together. I don't know if they do this on Ethereum, but batch transactions together, and so then, when you send one super transaction through, you've got five sub transactions within that. If one of those fails, the whole thing it's. I don't know if it's feasible, phil, I'm sure it can be done, but I think this is a huge success for ethereum, in my opinion, and polygon cool congrats to the lemon break team.

Speaker 2:

Breaking the ceiling, breaking the glass I know I went a little bit long, but I want to make one little comment on like marketplaces, about safety. So we ended up we have an otc community or an otc operation. In star atlas we know the players exchange potentially millions of dollars worth of resources over the span of probably every quarter millions of dollars are transacted outside of our ecosystem. So there's a lot of potential secondary revenue that we're missing out on. It's not a lot right now, but at scale it would be a lot Now. We also ended up with someone who reached out to us who was really pissed because they got scammed out of a $25,000 ship. They ended up sending the money and they didn't end up getting the ship back, or I can't remember if it was. They sent the ship and they didn't get the money whatever.

Speaker 2:

So there's a lot of risk operating an OTC and it can really only be done with these very elite members of a community, and if you don't have that elite status, you're not going to do business with somebody whose account is five days old.

Speaker 2:

So that's one thing I've actually harped on this in an article on my sub stack, which I also have a sub stack, but basically one of the things you can do in order to encourage secondary market revenue is to make a good marketplace, make an efficient marketplace this is rich coming from me, because our marketplace is not very good at Star Atlas. We all know this. We're working on improving it. But make it a safe marketplace, make it an efficient marketplace, make it a very cool kind of slick experience and that is going to get you people transacting in your marketplace. Because basically, what you're doing is you're saying look, the fee is the premium and the fee that you would pay in our marketplace is lower than the premium you have to pay if you go outside of our marketplace, because you are exposed to risk and you're exposed to a bunch of hassles. You wouldn't have to if you were just in our marketplace.

Speaker 1:

Let's talk about Fortnite Hit us up Eric Fortnite Creative I'm sure you've heard about this UEFN, which is Unreal Engine for Fortnite. It goes by a lot of names. Fortnite's trying to make their play to become the user-generated content platform where people make modes in Fortnite that hopefully someday become their own big spinoff game. So if you play Fortnite imagine most of the listeners of this cast aren't of that demographic but if you play Fortnite, you go in and there's a whole giant menu of modes. Right, it looks like YouTube or you or netflix. You go down, you just scroll past modes and modes and broadly I would say there's first party mode. So this is stuff like rocket racing, where the rocket league guys made a racing game, and lego fortnite, which is lego, for it's like a survival crafting game, lego themed. These first party games I don't really count in this category. They're not reallyated. So we're going to be talking about the other host of maps, which is all sorts of random modes like box battle and only up mega ramp and half the titles are in all caps to get attention. And talking to a couple of Fortnite players, there's a mix of maps in there. There's the what have you told me? There's the super sweaty maps, which are basically high intensity combat training for Fortnite players who want to play Battle Royale and want to train their aiming or train their building mechanics to get better at the Battle Royale. And then there's these very arcadey casual games where you know these like 12 year old kids hang out, they just treat it like their group voice chat, mess around in Fortnite together and play these different modes very similar to Roblox, and it hasn't really evolved past that. There's a lot of people using it, but I think the dream for Fortnite is for it to spawn some innovative new mode that becomes a new hotbed of attention that brings a whole host of more players into Fortnite.

Speaker 1:

So Fortnite, to incentivize people to make these games, they have some kind of revenue attribution, revenue share program. And I think it's worth looking at a couple other creator platforms, everything from Roblox to YouTube to Netflix, to see how they monetize and how they share their revenue. And broadly I would say there's the free market, direct attribution approach where each piece of content generates revenue. In YouTube, it's the ad views on the ad On Steam. It's literally people purchasing the game on Steam. And then there's more indirect methods where the platform uses some attribution formula views on the ad on steam it's, you know, literally people purchasing the game on steam. And then there's more indirect methods where the platform uses some attribution formula to attribute revenue.

Speaker 1:

So spotify and netflix do this where if you have a show on netflix, you don't get your show, it doesn't make a specific number of dollars. Netflix has a pool of money that they allocate to each show based on something. Spotify, I think, did this based on listen time. The more people listened to the song, the more you would get paid proportionally. And these two methods have their pros and cons. So on the more free market side, you've got app store congestion. If you've ever tried to browse a mobile app store, it's absolutely horrible, it's just completely crowded. And you've got some issues where I think with Roblox, they basically what ended up happening on Roblox is they adapted a whole bunch of classic free-to-play mobile monetization strategies like loot boxes and pay-per-progression and stuff like that. And I know Fortnite specifically wants to avoid that. They see it as a brand risk where they don't want this to be this like hyper-monetized thing where, like kids are getting tricked into purchasing loot boxes. They're trying to avoid that brand risk.

Speaker 1:

And so fortnite went with the platform heuristic, algorithmic attribution right, similar to spotify, which has its own downsides. So spotify famously had that playtime based attribution. And there is this, a band that asked all their fans to listen to an album on repeat while they slept. The album was pure silence and they racked up like tens of thousands of dollars of revenue attribution before they got canceled up. And yeah, whenever you have a centrally run revenue attribution, you can think of it like a command economy as opposed to a free market economy. There's risk that the way you're pricing things and allocating revenue creates perverse incentives for people to listen to silent albums on repeat, and fortnite ran into this exact problem.

Speaker 1:

Right, they've changed their revenue attribution quite a bit over the years.

Speaker 1:

Initially, there was this system where you would sponsor a creator I don't know exactly how I think you like picked a favorite creator and then five percent of your revenue, any skin purchases you made would go to that creator.

Speaker 1:

But then there was a bunch of tricks these games would do to try to get you to pick them instead of someone else, where they would lock you in a room, basically, and basically make it seem like you had to register that creator's code in order to get a power up or to leave the room and access the game or something like that.

Speaker 1:

So then Fortnite moved towards an engagement-based attribution model, again similar to Spotify, where it was based on playtime and DAO and retention whatever engagement they had and these games would try to optimize that as well. Some games you would have to be logged in for 20 minutes and then you would unlock a power-up, so it's basically you must idle in the game as a way to boost their engagement stats and as well as I think I'm pretty sure the revenue attribution ended up over-attributing to the custom modes and not enough to Battle Royale, because they soon afterwards changed it again to do engagement-based. But there's also some kind of revenue-based as well, where if a player buys a skin and then plays your mode immediately afterwards, you get more revenue than if a player never spends and never plays your mode. Phil's pointing at his tongue.

Speaker 3:

It's just so gross. It's such a disgusting way to think about causal effect. It's definitely not perfect, but I don't know. Do you have a better solution?

Speaker 3:

The thing that really troubles me about the Fortnite program right now is that it's cosmetics based. So, unlike the Spotify example, the Spotify is directly incentivizing the activity that they want. It's about plays, but Fortnite is about cosmetics, and so if you sell a bunch more cosmetics because Fortnite, let's say, had a tie in with Dragon Ball Z, which they did, that means that they're going to pay out an increasing amount to a bunch of the experiences that are on Fortnite, but none of those experiences actually helped increase revenue. The things are completely disentangled from one another, and so it's to me a misattribution of revenue. And even if I went to a playground, what about that experience made me purchase a cosmetic? I can't think of any experience in Fortnite where playing a particular playground then made me want to go purchase a cosmetic because of that experience in the playground. Even at the margin, I can't even think about a single experience that made me more likely to purchase. So to me it just feels like they're flipping a coin on who gets attributed and who doesn't.

Speaker 2:

What about? You say that it's about purchasing cosmetics, which it is, but it's first. There's a step between when your cash spins on the cosmetic and it's your cash spinning on V-Bucks. Is that not the currency here? Couldn't they, wouldn't it, would it be insane for them to just charge vbucks to access certain game modes and then user creators can, they can increase their prices if they would like.

Speaker 3:

They could go with the standard actually has this. They implemented this. It's just not very popular because it's friction, defeats the whole purpose of having a free-to-play game, which is why roblox has true games within a games, where a developer can implement a purchase inside of their game and then that solves the revenue attribution problem. Because they made a purchase inside your game, we're going to attribute it to you.

Speaker 2:

Yeah, no, that's a good point, yeah.

Speaker 1:

So I think Phil's causality point is really the heart of the issue. Right Is that they're doing this heuristic. But really what they want to know is how much incremental revenue did this mode drive? How many purchases of skins did box battle 2b2 actually create? Right, and in theory you want to attribute marginal revenue to marginal product. What games are actually driving revenue? But they don't know that and their guess at that is average. They're like oh, on average, two percent of the time is spent in this mode, so we should give two percent of our revenue to this mode, which was probably imperfect. I think a really strong example of this is fortnite. Battle royale. The main game is probably the biggest driver of people coming in, playing the game, purchasing it. There's all these spin-off games in the creative mode which are battle royale with a slight twist, battle royale in a tiny map, battle royale where you have a tragedy of the commons problem Interesting Exactly.

Speaker 1:

Yeah, so they're taking all the IP and all the valuable stuff that Battle Royale produced, making a slight spin on it and absorbing some of the engagement time and therefore a share of the revenue, which is probably why Fortnite changed the revenue algorithm.

Speaker 3:

It's like the voter median theory, in some ways. Right Like it's. You have a beach where you put the hot dog stand. Where do they go? They go right next to each other. Right, Because you maximize the distance to the left and the distance to the right for all those customers.

Speaker 1:

Exactly, whereas really they want to be maximizing social welfare and for the platform as a whole. And in the hoteling beach example, you don't want the two lemonade stands right next to each other in the middle, you want them spread out to cover as much of the beach as possible. And so I think with any command economy, any sort of centrally run attribution model, you're going to run into these kinds of perverse incentives. And to your point, with Spotify, it's a lot simpler. Or Netflix, where maybe I didn't sign up to Spotify to listen to Taylor Swift, right, she's not.

Speaker 3:

Taylor Swift's on Spotify. She came back, she is now.

Speaker 1:

Oh, she came back.

Speaker 3:

Oh, I know because I'm about to buy 500 tickets. You're a swifty uh, so buddhika is, and I was trying to get them for her birthday and I couldn't get them because they didn't do secondhand sales over ticket master. But I've been on reddit and, like my reddit post looking for tickets became really popular. So now people are like posting hey, I have tickets available, so I dm them immediately.

Speaker 3:

We found a quick aside. We found a woman in vancouver and I was like holy shit, this one's probably going to fall through too. So I actually offered to pay her money. I said can I put a deposit in? Because I wanted to have a commitment device and also to show that I'm serious. So a signaling device and to also a commitment device. She said yes. So we and she was canadian, so put it as a canadian bank account we wired her 150 bucks. That's my deposit, and so when the ticket transfer, we'll do the transaction off platform. We'll do the transaction. We'll do the exact thing. Erc. The exact thing that ERC standard was trying to avoid is we'll do the transfer on Ticketmaster and I'll send her the cash. We avoid all the Ticketmaster fees.

Speaker 1:

Wow, good job. I love to hear it. I love the dodge form fees.

Speaker 3:

Well, and I've been listening to her music, music, so to that point.

Speaker 1:

You might sign up to spotify to listen to taylor swift and you might have a bunch of play play time on a bunch of random indie artists. But, like the, the thing motivating your sale was the, the taylor swift. The algorithm isn't. The playtime based algorithm probably isn't too far off for spotify, but with games like you said, it's cosmetics driven. A game might be super engaging but not drive skin purchases at all. A game might be very successful at driving skin purchases. Imagine like a fashion show game. The engagement time might not be very good. So the incentive alignment probably is a lot harder in Fortnite because there's many more layers to the journey from engagement to monetization compared to Netflix. So yeah, I think they're going to keep iterating on this algorithm.

Speaker 3:

I think they're still going to flail around for a while. So here's the thing I'm interested about Eric. So think they're still gonna flail around. So here's the thing I'm interested about eric. So one of the things you mentioned about the algorithm is there's this immediacy thing. They added to the algorithm. Where I went in, I bought vbucks, I played this experience and then I made a purchase of the vbucks. In this experience I get almost like a special attribution boost. What sort of experiences do you create to maximize that incentive? How do you min max?

Speaker 1:

that at a base level, I would say for the creator or for the creator yeah, I don't know. I think you want a experience that highlights your character avatar right.

Speaker 1:

So, for example, you can imagine a tetris inside of fortnite, but you don't see your character at all, so it's not at all motivating to purchase a skin for that, as opposed to a mode that really puts your character front and center right, all the battle royale likes I mentioned, maybe a fashion show kind of game. But yeah, I think that's. The point is that some games might be super engaging but not drive revenue at all.

Speaker 2:

I still think they could just charge V bucks. People earn V bucks during the season in Fortnite playing Lego Fortnite.

Speaker 1:

I think it's the same problem as the app store. Right, you put $3 on an app, you get way fewer installs than if you make it free to play, because for some reason, the people are on the app store expecting free to play.

Speaker 3:

I actually think you're right, Eric, that they should implement this in the short run because it's easy to do technically and it solves the attribution problem before they get to individual playgrounds allowing MTX. It's worth the experiment because it's so cheap to implement. Implementing MTX in experiences I'm sure is on the roadmap, but experiences, I'm sure is on the roadmap. But for listeners who don't know, they recently added the ability to have a saved game. That's the state Fortnite creativism Like. Now you can log into the game and you have a save state. They didn't even have that shit yet. It took Roblox like 10 years to get where they are today Even more than that but took 10 years of just nothingness. And if you look at their charts it's like a rocket ship after the 10 years, but for 10 years it was nothing too. It's tough shit.

Speaker 1:

Creator is very limited right now and even though first party games like rocket racing and lego fortnite, those aren't actually built in fortnite creative or uefn. They're built in unreal 5 like the more hardcore fully fleshed out engine and it creates this problem I think you guys have talked about in the past. We're like if you're a roblox developer, you're developing on roblox, but what's the transition in or out? And as a developer you're like I could pick Roblox or I could pick Steam and I have to commit early on. And I know that one side has platform control and has all these issues and it's a big risk if you're going to throw millions of dollars into a development.

Speaker 1:

So what Epic is trying to do is make it a lot easier to switch between Fortnite Creative and Unreal and I think on the roadmap they want UEFN to basically be at parity with Unreal Engine 6. And they want people to basically be able to move the game in and out very easily. So as a developer you can flex back and forth and you don't have to commit before you even start development. Which platform you're going to lock yourself in. I think they've got a long way to go, just based on the current tooling, but that's their future vision.

Speaker 2:

So they have EGS and that's like their secret sauce.

Speaker 1:

I think it's part of the value prop in terms of off-ramp you could off-ramp into anything, off-ramp into Steam, off-ramp into your own thing right, the engine is super open. There's a lot of stories of Roblox developers getting big and then leaving Roblox because they're like. My ceiling here is a lot lower. Yeah, so, speaking of all that, optimization gaming, the algorithm, discovery algorithms are super important on platforms like this, youtube, famously, every time they tweak their discovery algorithm, different types of content gets first surfaced right, first, super short. Obviously, youtubers love making content complaining about YouTube. When the algorithm favors short viral videos, like number of views, you get like those short punchy videos. When it favors play time, you get these super long, 10 minute let's play videos right. And similarly, fortnite's discovery algorithm is very crude.

Speaker 1:

At the moment, I'm pretty sure there's almost no personalization, and likes and favorites are how you get to the top of the front page, and if you get up there, obviously that drives more, more play, and so a lot of creators have been finding tricks to try and increase their favorite and like count, and so one one recent tactic that's been employed is in the game there'll be like a little lightning bolt item which looks like a powerful weapon you can pick up and when you interact with it, a little pop-up appears. It says in favorite to support us, which implies potentially that if you like and favorite, you'll get this lightning bolt. So people will go and favorite, you don't actually get.

Speaker 3:

Oh, you don't the games.

Speaker 1:

You don't actually get it. Yeah, because the game doesn't know who liked or favorited, but it makes you think you will. And then these games their favorite count shoots up as soon as they added this feature and it's been proliferating. First it was like one or two modes doing it, but now can I never get the lightning bolt? No, it's not a functional thing in the game. In base fortnite the lightning bolt is a powerful weapon, so in the game you mode you might think, oh, this looks like that lightning bolt oh, that's awful.

Speaker 3:

That really is a dark pattern.

Speaker 1:

I didn't think it was that bad when you explained initially, but no, that's a dark pattern yeah, and on one hand, it's like you want to boost, like everyone's trying to figure out how to boost their content. But, yeah, if it's creating these ear flicks and confusing patterns, and like the games that are manipulating people are getting promoted above ones that are trying to be honest, so you might argue this is hurting the platform one of my favorite parts of the article eric was like the uniformity thing where, oh, if everybody's doing it, then nobody's doing it.

Speaker 3:

However, it disproportionately impacts potentially disproportionately impacts types of creative and it's bad equilibrium to your point like, if everyone's doing it, no one's doing it, but it's still a bad equilibrium, yeah and it's because the incentives are all fucked up.

Speaker 2:

I liked that part. I liked another part too.

Speaker 3:

I'm trying to look for it can you publish a piece on ugc web 3 for your sub stack, please, please. Where is it? I want my composability.

Speaker 2:

I don't know of anybody who's like actually doing UGC. Everybody talks about UGC and it's we're not doing UGC, I can promise you that. I need to look into. There's a game called Mix Mob or something like that. They were all about that, UGC.

Speaker 1:

And they're actually a game you can play. Yeah, some of the modes started doing this trick to boost their like and favorite count and other ones quickly caught on. The first one to implement it shot up to number one, but now they've gone back down. And the thing is, once every single mode is doing this, every single game will have their favorite count boosted. And I think this is very similar to what you do in mobile apps apps, where if you've used a mobile app, you'll probably at some point get a prompt oh, did you like the app, please rate us five stars. Or if you've gone to a yelp restaurant restaurant, that say, hey, please give us a good review on yelp. This kind of review farming behavior is pretty common on other platforms and I think at first one game might be able to boost their stats, but once everyone starts doing it, it flattens out.

Speaker 1:

And the point of this favorite ranking for the platform is a signal to say what games are good or bad. And the question of is this bad for the platform is boils down to is this distorting that quality signal right? Is this making bad games get favorited more and good games get favorited less than they should be? And if they are all implementing it and the effectiveness of it is uniform, that is, let's say, all game types are able to use this tactic to the same degree. Then I would say no, it's not actually causing any major problems, Just a minor ear flick dark pattern.

Speaker 1:

But you can learn to ignore it. However, I think there is a real risk if certain games are able to employ it more effectively. It more effectively. So you could imagine if the game knew that when you favorited it actually gave you a power up, those games would be able to abuse this much more than games that didn't have power ups to give you for favoriting. Alternatively, you might say that a quick, casual game maybe can farm favorites from a wider base than a more hardcore, deep engagement, narrow player base game that can't farm as many favorites from its audience.

Speaker 1:

So yeah, I think it really, you know, time will tell, but I think it seems to me like this is a minor trick, that, like in the app stores, some apps did it and were able to boost their ratings. Then all the other apps are asking their users to rate them five stars and we'll see everything settle back into equilibrium here, I agree with this analysis.

Speaker 2:

Yeah I liked it. I liked it a lot like subscribe. I do think you can subscribe to eric so engagement, farm the shit out of it.

Speaker 3:

Dark pattern, it leave us a five-star review what can we send you? That's what can we send. What can we send?

Speaker 2:

a listener leaves a five-star review I'll send them a picture of a lightning bolt don't, eric, don't you have an onion coin? We can send them oh yeah, yeah no onion coin.

Speaker 1:

You you should buy onion coins. Hot new coin on Solana, currently priced at $0.0420.

Speaker 2:

This is why this blockchain is fucking broken.

Speaker 3:

This is interesting. Why are Web3 games not encouraging more wash trading for in-game rewards?

Speaker 1:

Oh they do so.

Speaker 3:

They engagement, farm themselves what do you mean wash trading? Wash trading is very common in crypto projects so so wash trading for listeners who don't know is you basically just trade with your friend. They trade you back and that marketplace transaction increases the volume of activity and that can move you up some leaderboards which people look at. So do a lot of it, oh this is a.

Speaker 2:

This is this whole podcast is all about perverse incentives. It's just a whole podcast about fucked up and there we have our episode title seriously, though, now if you've got a leaderboard, who's the metric is a gameable metric. Now, the nice thing about having a marketplace and this is one of the I was going back to the marketplace incentivizing people to use the marketplace, give them benefits of using it but yeah volume.

Speaker 2:

if there's a, if there's a royalty fee, then that's not free, and that's why I'm a huge advocate for costs in Web3. Costs are essential. Nothing should be free in Web3, but that's just my personal opinion.

Speaker 1:

Ironically, the guy in the super low-cost chain.

Speaker 3:

Web3? More like Web3.

Speaker 1:

No, but I'm with you. The amount of wash trading on DEXs decentralized exchanges, which usually have a transaction fee, is a lot less than the amount of wash trading on centralized exchanges, binance or Coinbase, where the people doing the wash trading can get special accounts with zero fees and they can basically just fake the stats. So if you're ever checking stats on a site like CoinGecko, if it's a centralized exchange, chances are those numbers are fake. If it's on a decentralized exchange like Uniswap, chances are it's real.

Speaker 3:

We are off to see the eclipse. See some stars. Catch you next week. Got my glasses. The Amon Economist cast episode Perverse.

Speaker 1:

Incentives Silver Edition the Silver.

Speaker 2:

Wizard of Oz. See you guys.

Speaker 1:

We should teach this to our children. Economics is major. Everyone has to major in economics. Number to our children Economics is major, everyone has to major in economics. Number one for personal survival.

Speaker 3:

Economics is major.

People on this episode