S7 Ep. 3 - Ideas for 2025 PGF Builders W/ Juan Benet

Transcript

welcome to the green pill. network podcast if you're just joining us we're building a coordination a network Society of thousands of hackers dreamers and doers using crypto to bring positive sum digital systems to the world this podcast features the people who are doing it and we publish a season of 5 to 10 episodes every couple months this season is about how to Dow uh myself and my co-author punar have released a book entitled how to Dow that is all about the why the what and the how of dawing how do you start a dow how do you join a DA how do you think about Dows in a way that's uh native to Dows and not just from the old world and old organizations if you're interested in picking up the book you can get it at howto dow. XYZ and for the next couple months we're

going to be dropping how to Dow Alpha on the green pill podcast um if you want to learn more about the green pill podcast you can visit our website at Green pill. Network where you can download the green pill book for free join the Discord and become a member of your local green pill chapter all right today's episode is ideas for 2025 public goods funding Builders with Juan Benet we are on a tear interviewing vitalic buterin interviewing Ai 16z and next up I've got Wan D the founder of protocol Labs inventor of ipfs and filecoin which has been driving innovation in decentralized web Technologies and building more robust foundations for our digital future through ipfs filecoin and the

other stuff that is being built in the protocol Labs ecosystem and joining the discussion as well is sel Reen who is a grants Innovation lead specialist at gitcoin and today we are talking about how to think about building public goods in 2025 so in this episode we cover a real breadth of things with laan we talk about where we are at the end of 2024 what's what's the lay of the land what are the experiments that are that are being done where are we starting to scale and where is more experimentation happening we talk about how to speed things up when it comes to public goods funding Legos and the composable modular funding structures that are going to allow us to build a lot faster better

faster cheaper than we ever have before and then we get a little bit meta we talk about how are we actually going to fund the builders who are building public goods funding mechanisms and from there we get into all the different mechanisms credit assignment retro public goods funding deep funding drips Ai and allocation hypers certs um one is just um I I think that the number one thing I admire about Juan is just his curiosity he knows about so many different things and um I think that he's just kind of self-taught on all these different levels and um protocol Labs has the capital and the vision and the ambition to do something really cool in this space and always enjoyed talking to Juan about public goods funding

Capital allocation and this episode I think is a great snapshot of what builders in 2025 should be f focused on so I think you're going to enjoy this episode with myself Juan and Sage all the next voice will you hear will be myself and Sage all introducing Juan enjoy hi ju how are you doing very well uh thanks for having me very excited to talk to you about public goods funding in 2025 and what ideas Builder should be focused on in 2025 um I'm wondering if you could start us off by giving us sort of a lay of the land uh why does this matter what are you excited about and you know if you could just give us a

breath of of the ideas that you're thinking about when it comes to ecosystem building and public goods funding as we cross into 2025 I personally think that the ability to rewrite how funding structures and economies work is one of the most exciting things in the crypto space uh it's the thing that uh it's not the reason that initially brought me to crypto but it's the the kind of like why I stayed longer term sort of thing where I think the leverage that we have as a community and being able to make new new experiments with different kind of funding mechanisms and new kinds of economic structures and incentive structures is incredibly powerful and could help rewrite how the Future Works and how future organizations Nations

communities like all work and I'm especially keenly interested in helping accelerate R&D uh and sort of Science and Technology translation and so on and that's where improving mechanisms here here could lead us to a much better landscape where we can um do more science faster and we can extract the fruit of science uh in terms of technology and products faster you know so many good ideas like don't you know we've come up with a good conceptual results but we haven't been able to translate that into better kind of improvements for the world and so I think like you know very concretely where we are today is that the whole Community you know compared to maybe like 2020 or or so we''ve done a really good job exploring a lot of the IDE space like we've over the last few years

have tried a lot of different kinds of things everything from you know initial experiments with credit assignment and dependency networks to then experiments with better crant funding mechanisms and better ways to collectively vote on and assign funding um better mechanisms to incentivize the dis the the funding of public this you know things like what quadratic funding um merge both of those ideas and and um and so on and uh We've also explored a lot of like more you know kind of like next Generation type ideas things like hypers Sears things like retrospective funding and so on and we've done a lot of really good experiments that have illuminated some of the things that seem to be working

some of the things that don't work and and in general we've been able to like do a lot of like good point experiments we've also been able to scale funding quite a bit so today you know the crypto World Flows at a very large amount of capital to a lot of the production of goods both in um in in like these new kind of mechanisms and also in like you know kind of standardized grant program type things we're talking about like first of all like tens of millions of funding per year like close to like a 100 in a lots of Grant programs around the ecosystem and we're also talking about probably the largest scale retr funding funding structure which is like the block reward powering all the deeper networks I kind of classify that mechanism and that structure within this

the same kind of public goods funding uh model and that is you know like billions per year uh which is like really really massive and that that's kind of like the potential here like we can get get to a model where we are allocating billions and tens of billions per year um uh through these these kinds of mechanisms and and so I think kind of where we are currently is that there's through a lot of these experiments we've found a set of ideas that that work those illuminate maybe what we should be doing next like they they show a set of possible next experiments there's a lot of capital allocators now um taking these kind of mechanisms much more seriously there's a lot of dows and treasuries that whose charge is to improve the networks that they that they

are associated with and they need better and better mechanisms to to try and and do that better ways of like deciding what to allocate better ways of collecting of uh sourcing funding and so on you know in the podcast that you did with the you kind of went into a little bit of a division between what is how do you Source the funding and the second how do you allocate it um I think the good news here is the protocols over the last years have alloc um been able to source and allocate a very large amount of funding towards public goods but we don't yet have very good mechanisms on how do you granularly distribute that and allocated to smaller projects that are very very impactful so that's kind of where I would focus a lot of attention maybe on the sourcing of funding side I would focus attention on

this idea that in general new networks that get built should be allocating some large percentage of their emissions or their token Supply to public goods funding so I usually kind of recommend you know at least 10% but you know Clos to to 20 or 25% if you can uh swing it over the long term and you want that allocation of of uh funding to be F kind of like this cone of impact closer to the things that matter to your community but then extending to like the deeper and deeper and longer term dependencies so you know maybe initially you start by funding things that are you know very close and immediate to to your your project so so things like you know Goods that you need right now like software libraries that that develop the product or something like that or or or some

other kind of kind of project and and then the deeper dependencies include some of the scientific research that needs to be done or some of the support for a lot of the people in the community that that are expanding the scope of the of the network and so that I think from the sourc planing like that's kind of we're in a really good place and there's of course more to do but but we're doing well on the experimental side of how to we allocate that funding um again we have a lot of like really good experiments but we are seeing maybe we have a good gradient of like what next experiments to do but my sense is like we're not able to experiment fast enough so I think from my sense of kind of where we are I think we're reaching a point where there are many more ideas

and potential projects to do than than our capacity to like actually do them and we need more people and more experim like we need to increase the rate of experimentation across the across the ecosystem and that could happen in a few ways one is by dropping the cost of of experimentation by building like more modular and composable tooling that enables people to like experiment faster second is by coordinating better across the ecosystem of like making it cost a lot less time to talk to people about these experiments and kind of convince them to try things and whatnot so like um doing some amount of the communications coordination work and third which I think is probably the most important is we need a a funding structure for pgf Builders so I think like one of the the issues is that pgf

mechanisms in general don't take tend to have a strong haven't yet had a had a strong take rate in a sense and so there isn't that much of a Financial sustainable element in pgf building and because you know when whenever you're in pgf land people tend to think oh like we're just doing you know sort of like good charity work and that's not that's not the point at all it's like we're we're creating valuable allocation mechanisms that net back a lot of um utility to the individual networks that are that are producing these goods and you need some kind of financial structure there that supports the R&B that goes into the pgf funding itself and so that part I think it like requires a lot of work next uh where we

should be creating kind of like some conventions or standards of like what percentage overhead is like the right amount of percentage overhead for developing the pgf mechanisms because that's what's going to enable hundreds or thousands of Builders to go out there and develop new kinds of mechanisms and um try out new things sample the space do more experiments and figure out what works today a lot of the pgf builders are basically raid limited by not being able to um kind of like have enough Capital to pay a lot more people to do this kind of experimentation like when you compare it to say the number of people working on scalability Technologies it's kind of crazy like how many additional you know L2 L3 scalability Frameworks do we need like you know it's a super saturated space and that's there because there is

a very large amount of funding going towards those kinds of solutions like that the world kind of has priced those well and and reor those and whatnot and that generates a lot of attention towards those and in reality this other stuff um this pgf allocation stuff might end up being dramatically more impactful to these nwor and communities um but currently doesn't get this kind of like attention allocation because there isn't currently like a super clear and viable business model and path for pgf Builders and so I think one of the core things that we need to do as a community is like refine those and set some good standards and um very clearly show that because this is solving a huge key problem for the space it needs to get what rewarded and that'll kind of help generate you know this whole like Builder funnel for bjf Builders where

where like you know it' be amazing to be able to see you know every single hackathon in the community having prizes for new can of pgf allocation mechanisms and making it viable and possible for new Builders into of pgf mechanisms to take those and turn them into some kind of business for themselves and for others where like if they can kind of create an important Improvement and it can help improve a lot of networks they should be you know created rewarded for doing so and so I think that that kind of like problem if we solve that problem as a community first then we're going to enable like a ton of other um pgf Builders um around the space and yeah and then sort of like beyond that I think that there's a um a lot of interesting solutions that have been or a lot of interesting ideas that have

been explored in the last year or two um and you know maybe like I run to like interesting uh next steps of take uh you sort of across the board um and so maybe you can like dive into into a few of those yeah let's do it first I want to recap what you said and then we'll dive into a couple of those experiments so a a couple of the themes that I heard in your answer one which was I think conveyed a real breadth of the lay of the land in 2025 is that there's starting to be more and more experiments that are done uh but I think that there's an opportunity to speed up things in 2025 increasing the number of Builders building pgf mechanisms balic said in our episode with him that 2025 could be a year should be a year in

which funding infrastructure is elevated to a first class status alongside things like Dows and defi as something that we worry about in the space and so how do we how do we manifest that well we should fund the builders that are building funding infrastructure we need a structure for that and then also there's a lot of power I think in these composable modular funding structures where you have like take deep funding for example you know it's got Oso doing data getcoin doing weaving pairwise for selection and then AI agent like the the specialization I think and the pluralism of those things working together create is just that composable modular funding infrastructure and you can go way faster with the with those structures those are things I heard in ju's answer SEL I want

to do you want to play anything back that you heard and also next question uh I think should be for you SEL I actually love the fact that you emphasize always on R&D for the mechanisms right that is needed so this is something that I would love to know from you like how do we balance the need for like rigorous research and then also the urgency of implementing it fast right and testing it out in order to kind of build high quality successful things that you know end up improving the status quo you have to do a lot of product iterations and then you end up needing to do a bunch of research to advance the the ideas and the science behind it so a lot of these kind of funding mechanisms because they're economic Innovations they tend to have

very important game theoretic type uh ideas underneath them so you know everything from like Assurance contracts which is like the basis of how Kickstarter and things like that work or you know dominant insurance contracts which has like an improvement to that that haven't been you know widely deployed to things like prediction markets to you know hypers to like uh you know walk rewards like all these mechanisms have strong game theoretic elements underneath the hood and that science and that that idea space needs to be very robust and needs to be well researched and at the same time when you look at say academic Game Theory research there's piles and piles and piles of papers of inventing lots of kind of kinds of mechanisms and and doing a lot of study of mechanisms but few of that few of those things is

actually translating into wide use and where there is wi use is you know where people are building products and they playing the map to users and and so on and so you have a ton of builders that are like actually trying to solve problems for themselves right away and sometimes they come up and with the with really good new novel ideas um but sometimes the ideas are not as well tested or as rigorous and so on and and they could really like use a bit of literature review of finding out what other ideas have existed in the past this is a fundamental issue in kind of building products and doing science ific research where we need to kind of bridge those two spans and you know sometimes there one person that can like do both the product duration and have like a really fast Cadence to like deliver the

product and deliver the good and you know get get you know solve the problem and they can simultaneously do the Deep research required but more often than not it's actually teams of people that that do both of these together and you want like a wider community that is able to do you know both of these things at the same time um our practice in this broader community of publishing what we're doing talking about it discussing it talking about like the problems the next ideas and so on very widely is very good to stimulate that kind of discussion um I do think that we need a lot more strong kind of like research background folks in the space I think we have a like a really great supply of Builders but less kind of like deep analysis deep research and this is where

I think um you I think fun in common since a community started with the goal of like bringing um a lot of these Community groups together and has you know eded uh to some extent I think we need an even like stronger version of that like I wonder if there's like a a need for both like a I don't know an academic format style conference where we like talk about these kinds of you know people like actually publish papers and like this sort of stuff and like you you like nerd snipe a lot of the deep uh game the theory researchers to help you know prove things and then on the flip side creating a valve for a lot of the people that have spent their time researching these things in theory to then actually have positive Global impact in terms of like getting their

ideas out into the world and used by millions of people or like you know benefiting hundreds of millions of people um like that would be kind of like a really great setup again I think it it also comes down to like funding mechanisms like having the ability to form these teams with funding so that they can like go and productize the solutions is kind of required meaning you know in the early days of crypto sorry cryptocurrency world the the cryptography Community had come up with tons of great innovations that were stuck in Academia and had not yet been translated and it was really the the cryptocurrency world that emerged with large piles of you know magic internet money that suddenly was able to fund a huge amount of that cryptography

research and get it out into the world like zero knowledge is one perfect example of this where huge amounts of research had been done in zero knowledge and advancing the field and all of it it was stuck because there was no like broad application the blockchain cryptocurrency world needed those applications and had the money to to to fund the work and then you had like all these kind of hybrid academic and technologist and product Builder teams that were able to kind of push the frontier forward and do it right and so I think we need some kind of similar type of situation here with these kind of better funding Primitives where um we kind of need the big funding results to be available to those so that those teams might form and I think it starts with like the the Dow treasuries that are trying to do this funding allocation

is like they they need to be factoring in some significant overhead in in uh in their deployment of capital to go to the groups that are doing the R&D of improving the the mechanism allocation right at the end of the day like um allocating the the the funding much more effectively like what to produce higher scale of impact like you know if if you come up with a mechanism that like doubles the impact of the of the allocation money like that that's an amazing Improvement that you should be willing to pay like a very large fraction of the of the of the funding for and so if if we I don't know normalize something some like takeway between 5 and 20% of the allocation going to the people that are helping produce that impact um that could be like a massive result and then after

that then we can like prioritize optimization of cost reduction to like make it cheaper and cheaper and cheaper to be able to do this kind of allocation and and there I'm like pretty excited about um AI systems to be able to to greatly accelerate and and drop the cost in in a lot of this um kind of allocation work so many of the pgf mechanisms that we have today are very labor intensive and Communications intensive because they require um externalizing a lot of preference um you know part of the reason why markets work so well is that they implicitly they they don't force people to externalize their preferences by like verbally you you do it in like the action of buying and like the actual transaction and like that's a super super powerful mechanism and so we need some kind of like similarly cheap type

of structure potentially using markets or prediction markets to to be able to do a lot of this pgf um type of uh funding without requiring so much conversation because the conversation can be very very expensive the other option is to just throw LMS at it like we now have like ation machines that can like you know ingest vast quantities of of text and information and um be kind of like aligned to a set of preference value values and then you can kind of um stick them on on on on doing this kind of thing so I would love to see so some of the experiments I would love to see in 2025 include um try and use LMS for a lot of these kinds of mechanisms so try to use llms in retr pgf rounds or in Grant rounds or um in prediction markets or the Deep funding um stuff that uh

Deon ballic and others are are are working on and in usage all um and so I think like those kinds of experim I think could be could be really good at like dropping the cost uh but I think like first and foremost we should like very strongly normalize the the kind of like allocating some amount of funding to the builders themselves because that's what's going to drive the optimization here and like enable us to sample so many new ways of of uh of doing this allocation yeah I love that that's definitely one of the goals that I have to you know how to bring the right people together because I feel the mechanism space needs that kind of an exploration which is a bit academic but at the same time you know this fast uh implementation right like for example we had one session on Chaple values and we

figured that the weight of the graphs can actually be fig you know identified using shapley values right we need phds to do that so there are solutions that exist in academics but that you know Gap needs to be bridged uh in a way which is like which has that sense of urgency also so yeah yeah let's maybe dive into some of the kind of interesting experiments that we could um we could see in 2025 so um you know one of the things that that we've seen Advance a lot in 2024 is the optimism retr pgf type programs so um optimism uh pushed the retr funding idea into this like uh round based structure um that has been really successful um they've been deploying I think close to $100 million

do or more already through it um and that has been quite successful in like funding a lot of critical dependencies for both the optimism community and the broader ethereum community and has gone to like a bunch of dependency projects and that program inspired the fat R pgf program and the lipop r pgf program and there's you know these kind of like other uh retr pgf programs that have been kind of following that model the this kind of program is also seeing a lot of experimentation um I think the uh uh open Observer open sour OBS server folks and and and other folks are like um refining these programs to try and and again drop their cost and increase their impact so um they experimenting with experimented this year in uh doing

like metrics Focus rounds or rounds where like you get experts to indicate metrics or to weigh metrics and to kind of like help tune an algorithm and I think those are like very very good experiments that can help um again decrease the cost of doing the allocation but also ideally most importantly um increase the impact because you really like all these mechanisms should be first measured on impact and second on cost and so I think those those are like pretty good ideas and I think we should see those ideas percolate to the other similar kind of programs uh in 2025 so it would be great to see some more experimentation in this direction in um in Phil pgf andp retro pgf um separate from these uh I would love to see more retr funding experiments that just don't have humans in the loop at all or or like where you

you like greatly minimize them or they're faster Cadence like one of my uh gripes with a current way that these programs worked is that the cycle is just very very slow it's like you know a three to six months cycle each time and that doesn't allow a lot of faster optimization and so getting into a mode where the Cycles can be much faster you know closer to like block rewards you know part of the reason why block rewards work so well is that they happen every single block right like uh the the blockchain like just rewards on an ongoing basis um and like that's fantastic I felt like I could add value by pointing out that I did an episode with AI 16z uh in which they talked about using AI to bring the the quarterly rounds down to weekly or even

like like you you have that amount of intelligence processing then you can provide instant feedback loop and the difference between uh what retr funding is all about a faith-based agreement that if I do something good for this ecosystem I will get paid eventually and it has to be faith-based only until you get the money and if you have that feedback loop on on like an hourly or daily basis I mean I'm just saying making your argument back to you at this point but like it's interesting how like the big brains have all independently conversed on this on this direction yes yeah yeah I think there's like a um everyone is like once it once should be faster um is pointing out that AI can be like a really good way of of doing a lot of the intelligence work and and and yeah I think like you know these kinds

of mechanisms the reason they work is because you have like this public reward that everyone can see that everyone you know is committed you know it's a it's a an advanced Market commitment in deploying this funding towards this kind of good and that causes a lot of people to do that work um it's just that the cycle distance is is so long and especially if you start changing a lot of the rules and do a lot of experiments which you want to do to to try and find the better structure but that also like changes the the the game a bit on people and so it's hard to like stabilize the mechanism and whatnot like um one of the things I love about the block reward structure is that it's just so clear and direct and so fast that you know you can have these like large protocol emissions

associated with um this incentive and then in a matter of like a few months to a year or even you know a few years you have these enormous scale infrastructure networks being developed um and that drops out of like the you know consistency of the of the mechanism and the frequency at which it adjusts right like so it adjust you on a per block basis um but yeah I think even like adjusting daily would be good enough I think weekly probably also good enough I think once it starts getting into monthly and quarterly it feels uh very slow now of course you know very naturally like if you if you need the input of tens of people like yes you're going to end up in like a monthlong process and so yeah I think uh I'm very sad about like using AI models or using

more metrics type things and so on I think like um broad set of mechanisms um which you know a set of those like kind of kind of described as impact evaluators that it's not a great name for them um but like I think that this whole class of mechanisms um should get a lot of experimentation in 2025 so I would point a lot of Builders towards these and one of the things that I think you know talking about another mechanism structure um one of the things that can enable a lot of this experimentation is going towards much more composable modular funding routing and so one of the things that I've been pushing for a few years now is like this model of um doing these uh networks of funding allocators where you try and get these mechanisms to be as small and

self-contained as possible like very Unix philosophy type where you know one small thing doing doing it really really well and then you kind of wire up the mechanisms and when you look at like private capital and the successes of VC and the successes of of many other kind of financial industry systems it all drops out of like this very clear composable modular structure that is achieved through you know like the the idea of a fund like when you can create a fund you can like raise money on one side and then deploy money on the other side and then if you're successful then you like return more money and if you're unsuccessful then like the fund loses money but like that composability is super applicable and it's what gives rise to like this massive scale uh Capital system and so getting to like

that same kind of like modular structure in pgf where we can have like these kind of like money routers in a sense being able to race money on one side and then allocate it on the other side then have like a strong sense of like metrics and impact that can allow the funders to decide whether to increase the funding or decrease the funding the next round will help a lot and one part is to kind of disseminate and socialize this idea and get more people to do it the second part is to actually this build composable Primitives with software and this is one of the reasons I'm super excited about Alo and um all the stuff that you're doing Kevin and and and others because Alo is like a like the right structure to be able to comp compose all of these kinds of mechanisms

into this this kind of network where you can very cleanly and clearly see this these are you know the set of mechanisms that exist here's how they wire up together here's what each one of these mechanisms is supposed to do um here are the participants of this mechanism and like getting all of that clear metadata to be composable and onchain and like easy to ride through will be super super super valuable in in getting all of this experimentation to speed up right so I think like one of the things I'm like really excited for 2025 is like enabling a lot of pgf Builders to spend way less time on all the root like operations side of running one of these programs which is kind of like oh great like how do you go and raise money from people and how do you like you know once you

have figured out your allocations how do you distribute it out to people like I've just seen so many people spend an enormous amount of time having to deal with like oh you have to like give this thing to this person you have to kyc that person it's just all this Builder energy is going into like solving Road problems that shouldn't be there and if we can like build some software primitive some infrastructure that enables pgf just be like Oh I'm just going to Smart uh like take a smart contract and deploy it and like that's my P my pgf mechanism and like you're good to go and you can like get at it to the network and automatically you you can have something kind of like defi Lama but for bgf mechanisms where you can see like here's what the allocations are are happening here's what they're

doing here's like the impact metrics like that that level of kind of both financialization and like instrumentation will be fantastic and like once once we can do that and make and drop the cost of innovate innovating here um we're going to be able to get like hundreds of Builders to try out new mechanisms um and so on so I'm pretty excited about like that composability composability and modularity in the design space s next question for you yeah um one of the things I would love to understand is do you think it's possible to align the principles of public goods with the financial dynamics of startup ecosystem like such as Venture like returns like what would that look like yeah it's a good question like Venture style returns means like a a 100 or a thousand X return on the

initial investment and and that's sort of needed in order to fund these massive scale product build out right like the amount of produ like a people that you need to build something like Airbnb or you know coinbase or something like that it's just like you need hundreds of thousands of people working on it and so that's why like you end up with these like institutions that require enormous amount of amounts of capital and therefore need to return massive scales of capital I don't think we need to go that high I personally think that like getting like a five two to 5x return is like really good and if you can do that in a pretty fast space uh you could like have a lot of success so I think we can

end up with a model and system that is like VC like in that there is some initial investment and there's some return later but the numbers can be much smaller because you don't need hundreds of thousands of people to create new mechanisms you need like somewhere between one and 10 and and that alone could like generate a lot of experimentation so I would say we should normalize ad a that you you could go on like a like a venture in a sense in terms of you're going to come up with a new idea you're going to potentially raise a little bit of funding initially to be able to do the R&D work for a new mechanism and then you're gonna get a very strong reward by being able to create a business on the other side of like making sure that the groups that use your mechanism can route some reward

to you but I think that return you know could be anywhere from like 1.5 to 10x like it doesn't have to be 100 or 1,000x like if it is awesome like if you created something so valuable and so great that like the entire world is like fine giving you a 100s return multiple on your initial investment because it's like that valuable for the world like great awesome like more power to you um hopefully allocate that wisely as well and like route your funding to to other projects that really matter but what's most most likely is that like you know a much smaller return multiple can be like really valuable here we just have to normalize that transaction and that structure I think like in general people tend to not like like evolving markets is one of the thing that things that happens uh kind of organically and like

one group starts doing something and it shows that a strong result and then everybody like copies it and starts doing it as well and so I think here we just need a few groups that do this that like just commit to like you know properly rewarding the folks that do these kinds of Innovations and if it works really well like a lot of other parties are going to copy and and and scale it and so I think it's is something that like I'm already you know in the set of pgf mechanisms that I'm working on WE explicitly account the Opex cost of running the mechanism and doing the work involved in it and we like quantify like today like the overhead is massive like some of these like have an overhead of like 40% which is crazy right like like you ideally want that over overhead to get closer to

five to 10% but that's fine in like the early stages of R&D of any mechanism and that's and and that's good and so like we're fine with that kind of like initial investment and and I think like we and so that that you know we're normalizing already like factoring in the cost of doing the work of of doing allocation I think The Next Step Beyond that is like factoring the idea of like creating like a like a mini company or something that is doing this development and then get some reward in the long term and I think this could also happen in smart contracts right like you you could have like a very simple smart contract that like represents that little company and has like some splits associated with that in terms of like the people that contribute to it um the good news here is that

because we're doing all this like credit assignment work uh we get to like kind of like dog food our own stuff and so like we could create like these very simple kind of mechanisms on how um how to reward the people that like contributed to the creation of this maybe this something that Al could also work on is like have a clear community of builders that are pgf builders that are working on this stuff that uh then we kind of like make visible in the network like not only do you see like the network of allocators but you also see the network of Builders building the mechanisms and then maybe we can like drop the cost of innovation here by enabling these like groups or like teams to form to them that then can like get rewards together or something like that I think that can like help normalize the

whole thing there's kind of like a you know speaking of the modular structure uh to me the primary modular things are these are the mechanisms these are the builders these are the experiments that they're that they're working on these are the pilots that came out of those experiments these are the results and if you sort of think about those as each like data objects that each have their own sort of relational structure with each other you're kind of starting to see I it's just Legos of different types that are being assembled here not to like abstract the beautiful human collaboration that we're doing in into this like in a reduction way but I really want to like Focus here on the composability of these things and I think that there's a real opportunity

to map like oh this mechanism is good at generating function and F funding and this mechanism is good at allocate Distributing the funding and like even then you can string mechanisms together and um you know it's kind of like with with with money Legos like um you can do stuff way faster than you ever could because of the Legos that were available and I imagine like I I truly believe my heart of hearts because of this the the how open source all these things are that three hackers can walk into a hackathon and build a public goods funding mechanism just over their weekend better and faster and cheaper than a government that has a thousand software Engineers um and a huge budget 10 years ago because of the open source

Lego um and and and it even it gets faster because once you build the funding infrastructure to fund the funders then you have funding that you can drop on top of them yes and so just accelerates and I think that that's what makes me really excited about um about this is just the accelerating pace of uh of of innovation in the space and um you know we can only re gonna be a completely agree yeah completely agree that like you know a getting to an experimental model where a small group can do that R&D be successful and so on like I think will be will like supercharge this entire entire ecosystem um I think the yeah the money Lego like the um funding allocation money Legos thing I think is is is like an idea whose time has come like a lot of us

have been circling around this for for a while and I think like now we have like the building blocks in place um I personally think also that uh finding like the the large list of things that current pgf Builders have to do to S up a new mechanism and just explicitly going after cost reductions there will be very valuable right so like oh man everybody has to like incorporate some kind of entity in order to have like the legal structure to be able to get funding from a certain party to be able to Route it and then on the other side you need to be in order to be able to allocate the funding to a a party you need to like do some kyc or whatever like all of that is like stuff that you could solve once like you could have like one system that like solves that

problem once and then you let you know uh it becomes like really easy for like people to kind of innovate on top and I think like those kinds of cost reductions can help a lot here like just the friction to do doing this is is is hard um think of like the kind of innovation that was unlocked by stripe Atlas where you know strip let you incorporate a company with a few clicks and like boom like that enables just tons of people to create corporations and so or or like you know the ability to have a banking account by just shipping a um multisig that you know um or or like ship an organization by shipping it Dow right like that that's like that's the kind of the Innovation rate that we need here and so far the pgf builders have had to deal with like

super Draconian cost structures and like we have to kind of totally change that maybe another idea that I maybe I want to highlight like two two more ideas that I think are super valuable for 2025 one is uh we're seeing a Resurgence of credit assignment uh in general uh which is really really good like a lot of these ideas um you know that first started gitcoin and Source grid and other projects were like hey you have a lot of information available to us in terms of like the dependency graph of projects and who worked on those projects and so on how those things can combine to give you like a better set of funding mechanisms like that that got resurfaced by drips recently and like drips has been pretty successful in like creating like the dependency graph and

like the drip splits and whatnot but drips itself did did not like kind of give you a lot of power as a funer to like routed in a certain direction deep into the tree and that's where I think like deep funding now added on top of the drips dependency graph can be like super super interesting and helpful where like because you can see that dependency on a large cone of of um of of interests uh for for a particular project or Community you then get to like run a few different models on how you would do that location across that very large depend tree and then as a community decide where you want to allocate your funding deep into the tree like that that is I think a a very good unlock for um for a lot of these kind of credit assignment type of ideas and I I

would even return to say like it may be really valuable for a lot of Builders interested in the stuff to like go back and read some of the stuff from like the early getcoin days and early Source grid days on a ton of the ideas that came into these kind of credit assignment things some of the things that I haven't seen yet are like um we need to build very simple onchain composable credit distributions I think drips list is a close version of that but I think it's not it's not even modular and composable enough I would love to see this like kind of just in a in a smart contract where you can kind of like create a node in the graph here by just like submitting a smart contract into the network and you now have an onchain primitive like I would love to have that

um and once you have that as simple as that then you build like one website that just Aggregates all of these like they can find them and and you can just see kind of like how the neuron Network is sort of like wired up and and that's if we build that infrastructure then you can build things like deep funding and other things on top much more easily because you can kind of like externalize this huge dependency graph and at least some idea on what the credit assignment has been so far and then you can just keep adding to that by adding more nodes to that graph and I think I think something like that could be could be super valuable um I think another set of ideas that I think could be very powerful in 2025 is is um uh is hypers so I think hypers is probably the

one of the mechanisms that I'm most excited about um I think it has an extreme level of power it's one of those things that's like as significant as maybe the invention of like the stock certificate which you know kind of powered like the last few hundred years of of of uh you know Industrial Revolution and optimization and product building um and so I think like the the you know the impact certificate and and kind of like hypers soures and implementation of that on on chain is that level of powerful and we just need a lot more Builders trying experiments and trying to build different things with hypers certs and so this is the thing where like I think the community needs to be able to like have a fast pace of of innovation there and I think like right now is hypers Source ecosystem is kind of like rate limited

by by by the same things that we were talking about earlier like there's just not enough people trying to build it because there's not enough um kind of funding flowing to these ideas so if maybe we create the infrastructure and normalize the idea that like by building better allocation mechanisms you can get a strong return and that helps bring a lot more Builders then a lot more Builders can go to things like hypers certs and credit assignment and you know drips and deep funding and so on I think you know in hypers specifically it was like this really cool idea around using hypers search to do sponsorships so like think about how many sponsorships exist today for events like you could do all of the event sponsorship through through hypers Sears the other thing is you could do you could use hypers Sears to fund long-term open source projects so

one of the things that we're trying out in the thep ecosystem we haven't like finished this this kind of like ongoing work but um we're kind of like going back and and looking at all of the history of contribution on a per year basis and then we're going to issue hypers Sears per year for the project and so you'll have a web page where you can go and you can see like the kind of like contribution board for both the people working on the project and the people funding the project yearly and so we'll have like a yearly history for you know everybody who's been contributing to lip and that gives us like a you know these kind of like mechanisms on chain to play with um for this kind of structure and I think like that could be a a pretty neat way of using the hypers

source infrastructure to fund any open source project and and kind of like it mixes with the fundraising and it mixes with the um with the uh um kind of like deployment deployment of of funding and so on yeah I think there's an amazing breadth of things that you mentioned there onean and looking forward to to articulate those to the community and Route Builders towards the the best mechanisms as we're bringing the episode to a close I've got to start uh bringing us towards a coherence towards the end and I'm going to ask my my final questions if we are maximally successful in 10 years what does the world look like how has the world changed as a result of of of this work being successful and look 10 years is a long time you know 10 years ago uh people didn't think AGI was going to be a thing

and 10 years ago people didn't think crypto and the financial system that we built was going to be a thing and so on like in in a very quick span of time we've been able to show that both of those things are happening in in a fast scale I think 10 years from now I think crypto could probably grow one maybe two orderers of magnitude in value so this would be like going from like around three trillion to 30 maybe 300 trillion like that's maybe where I would put it um I don't think it'll grow more than that in that time frame but it could it could and so that starts putting crypto on the scale of like larger nation states in terms of our capability of funding things and so by then if we can solve a bunch of the these kind of like hard funding

allocation questions we could have these networks funding at the scale of nation states like we could have these networks funding science and R&D in kind of like similar quantities as like I don't know NSF and ah in the US or like the equivalent agencies in in the EU and in Japan and in China and all these places and so I think like the level of power and potential here is enormous like if we if we get these mechanisms right then as crypto itself scales the the scale of capital allocated through them is just going to grow tremendously so I think 10 years out from now we could be fun like having this incredibly fertile grant for Innovation powered by these mechanisms you a much more um the centralized

efficient more impactful way and in 10 years our experiments are likely going to start refactoring existing infrastructure right like when you go look under the hood for a lot of the existing mechanisms that people use to do this kind of allocation in the traditional World those systems can see a lot of improvement and I think within 10 years we'll likely see a lot of these canel mechanisms percolating into into those models and mechanisms so I think I wouldn't be surprised if in 10 years from now we we start seeing like National agencies AC around the world starting to use some of the mechanisms that we're describing here to do their funding allocations um one one idea that I think has an enormous amount of power that could be this kind of like you know very fast rewriting of of how the world

works is prediction markets I think like once you start mixing prediction markets with funding flows you can very quickly show that you this kind of like small network of experts can outperform like you know some um more closed off uh network of experts that kind of has has like a like a specific kind of mate to to do the work or whatever and um there's been a lot of papers that show that and so I think like we could be doing dramatically more impactful and effective allocation of the world's resources right like we we could be this could be hitting like the hundreds of billions of dollars per year sort of scale of funding allocation and and so I think like that's the that's the space potential uh in terms of the quantity now how that gets used that would be

like um accelerating how science gets done accelerating a translation of science into into Technologies and products that benefit the world you could then start you know maybe building impact markets not just for um green and carbon offset type things but also for like reforestation and for the ocean cleanup or for many other kind of like broad public good type issues you could also see these kind of allocations starting to also inspire better models for private funding where like maybe we can like build much more uh public and open and less kind of like World Garden type VC funds that end up working much more in concert with the networks and economies that that depend on them um we could see we could see just like this proliferation of much

more regenerative and um uh and like positive sum and incentive aligned models like percolate through through the the world to kind of like help us achieve like this kind of Regen ppan type of vision that I think a lot of us are are super excited about yeah when you talk about the scale of this thing surpassing nation states uh and I think about how much nation states fund and how much more effective onchain Capital allocation structures could be just makes me so bullish I'm wondering if is there anything that we didn't ask that you want to say uh ju I'll go to your F you first and then stage all would love to hear your closing remarks yes and maybe if I could leave all the listeners with like a very small concrete set of things remember is like one experiment a

lot in 2025 like if you're not a bgf builder now like consider like come join us two I would love to see all the hackathons out there starting to add pgf prices uh for every single hackathon just like enable a lot more Builders to do this three I think we should normalize um the funding streams to make sure that like pgf Builders can get uh a significant return because that's what's going to cause a lot of more participants to come through for I would love any new network that appears to dedicate at least somewhere between 10 to 20% of their supply uh to pgf building of course waited to the things that matter to your community first but to be routed through these mechanisms and then five I would love for us to like you know get ready like the the Bull Run is happening in crypto and

crypto is very cyclical right like we we've been you it's possible that like that will disappear at some point but I just don't think we're there yet and so I think like we're likely to see a bull market this year in 2025 and then you know some winter after that and so for all the Builders out there right now leverage this time period to do very strong fundraising so that you can weather the winter right after that and so you know five like as a as a biller like make sure that you take into account like the cality of crypto and like you you lever the BL run um into the into the winter thanks ju amazing uh SEL any last words uh is there anything that we didn't ask that you want to say as we close out yeah um one thing I definitely want to ask is that I see you

as a leader who is so invested in Innovation where do you draw your inspiration from is it books is it people experiences and related to that like if you could give two book recommendations or movie recommendations what would that be since people will be watching this right before going into holidays uh yeah totally uh a great question so I get I draw my inspiration from you know everything that I come in contact with like books media people um conversations um and so on I would maybe highlight there's a lot of like things that have been deeply inspiring to me but I'll highlight a few that I think are extremely influential for me like one is C against Cosmos um I watched that at a very early age and they was like deeply deeply inspiring in in my

orientation towards science and discovering reality and then and also having like this very kind orientation towards it it's like a it's not like a like a I I think the world is just so conflict oriented today and Carl just gave us like this extremely thoughtful generous kind way of seeing reality and truth and and and science um so I think I highly recommend watching the old you know 1980s Cosmos um it can feel dated now but like just kind of like have to get get into it and I think like it's just an amazing piece of work the second one that I'll highlight is David D's the beginning of infinity and that's a phenomenal book that kind of brings together different models of of reality

through different theories of physics and also develops like a theory of knowledge and how um Humanity's curiosity and our development of more science and more technology boots us into like this potential like great set of possible Futures and so I think that is like also deeply deeply deeply inspiring I think I'll also highlight Mark Miller and Eric Drexler who um uh are you know helped build the crypto Community like very early on in like the 70s and 80s uh 90s and um and maybe highlight the the Petopia type of uh ideas that they they fered um they it hasn't been as like widely talked about

but I think um you can find their talks online and so on I think those are like super super inspiring and then maybe the last group I'll highlight is foresight Institute um which you know Alison dman and a few other people are are pushing um is super deeply inspiring in terms of Illuminating where the technologies that we have today and are experimenting with today are going to be lead us in the next 5 10 20 years and they are phenomenal in terms of highlighting where all of this stuff could go lots of different kinds of fields um and they gather incredibly inspiring people and give to give talks like online and all the videos are out there that you can find and watch so I think that Community is just deeply deeply inspiring to be to be with Wow SEL you want to ask one one

last question I keep on saying this is the last question but SEL was DM me and she was like I have one question I want to ask one really quick yeah because I I love knowing more about people so I just wanted to ask like what's the one thing totally unrelated to work or web 3 that you absolutely absolutely want to take off your bucket list in 2025 oh uh in 2025 um I'm really lucky that I get to like as soon as I sort of like want to do something I kind of like maneuver to to find a a way of doing it um I think like I really love nature and I really love exploring nature and I've gotten to be really lucky in um getting to see a number of volcanic eruptions uh in person uh over the last few years and so

I I really hope that we get to see another one this year and um yeah just uh honestly spending a lot more time with my partner and we're like fostering cast right now and and like the kittens are like incredibly uh cute and what on so getting more time to do that I think probably like more time to um just enjoy life I think would be would be great uh but it's also like hard because there's so many exciting projects to to do so I don't yeah I think we were on a call last week together and you had cats crawling all over you and I'm I'm sorry that we neglected the podcast listeners uh of that but um Juan you remain an inspiration to us all thank you for dropping all these knowledge bombs on this episode I hopefully this guides web rebuilders in 2025 and thanks again for

coming on the green pill podcast SEL thanks for hosting we just heard from Juan Benet the inventor of filecoin and ipfs and the founder of protocol Labs I am excited about this episode because I think Juan laid out a true breath a menu an alart menu that you Builder can listen to in 2025 and figure out where you want to route your resources your attention your capital in 2025 and so I hope that this episode was an interesting episode for you but mostly I hope to in three or four years meet a builder that says um this series of episodes that you did with vitalic and Juan and AI 16z led me to my next venture so that's what we're trying to do is we're trying to Route the builders

the highest leverage and most impactful problems thanks again for listening to the green pillow. network podcast again this podcast features the people who are building a coordination a network Society of thousands of hackers dreamers and doers focused on using crypto to bring positive sum digital systems to the world like Juan Ben like SEL like protocol labs and like all of the builders that draw inspiration from them if you value our work you can collect this episode go to pods. Media green pill to find all episodes and to start collecting while you're at it you may want to pick up a copy of the how to Dow book at how toow doxyz this is the Dow season of the green pill podcast and that book how toow is designed to be your Guide to the future of dows and uh

you support Dow Movement by helping us climb that bestsellers list and picking up the book so um if you've been a longtime listener of the podcast if you gotten value from it please check out the how to Da book at howto dow. XYZ and while you're here might as well make an offering to our algorithmic overlords Please Subscribe rate and review please tell our AI Tech overlords that you like the green pill podcast when you gratify us with your likes and sharing this episode with your friends that helps us spread the green pill message the regen message the public goods funding and web three message to the world and I really appreciate your subscription to the the network and uh you you're supporting uh our mission over here at green pill in getcoin and I uh I'll I think this is a

fun episode I will see you on the next one peace [Music]