 Cool. Welcome everyone. I think this is one of the first-ever workshops at DEF CON, featuring a really new exciting space that has evolved over the past, really only over the past year, called DSI. And DSI is short for Decentralized Signs. And so the goal today is really, the title of the talk is my contract in Petri dishes, essentially merging these two fields together. And then with a goal to who everyone is here today, to kind of look at beginning to create a shared technical infrastructure roadmap for what is actually required to make DSI a success. Cool. So we saw some people coming in. Maybe a few words about myself. Hi everyone. My name is Paul Kohas. I'm the CEO and co-founder of a company and ecosystem protocol called Molecule. And I had the amazing opportunity last year to work with a great group of people to launch one of the first DSI centric research organizations called VitaDem. Maybe just start with that. Maybe a show of hands. Who's heard of VitaDem? Okay. So, almost pretty much everyone. It was a little bit hard for me to gauge the audience of the talk today, just because there's obviously a lot of developers at DEF CON. It's a developer centric conference. Cool. Maybe I can share a few words just about myself. I actually don't have a background in science, but rather in economics traditionally. However, science and specifically like biotech has been a huge personal hobby of mine since my teenagers. I spend a lot of time in online biohacking communities as a teenager. And some of you may be familiar with these. There's all of these different subreddits focused on tropics, for example, or biohacking. And so as a teenager, I spent kind of time in these online communities that were, for example, diabetics. And then we're looking at alternative ways of accessing insulin. As many of you probably know, insulin prices in the United States have skyrocketed over the past decade. And so these communities were looking at ways of producing their own insulin kind of in a garage, which sounds super risky, but actually often people are our current healthcare system really like underserved certain patient population. Similar cases in, for example, in different HIV communities that were kind of collaborating online to develop their own open source gene therapies. Or in other cases, you have communities of psychedelic researchers and enthusiasts that are kind of exploring hundreds of different novel psychedelic compounds. So what happened in these communities, someone would say, Hey, here's an interesting chemical structure. I think it could do this, this, and this in the human body has someone tested this. And then someone else would say, Yeah, I tried it. It did this and this and this. And then essentially people, some of you might know this as trip reports. It's essentially a form of like very anecdotal data collection. But I found this form of like open source drug development really, really fascinating. And at the same time, then I began looking at like the larger macroeconomic landscape of farmer. And because I started asking myself, well, why are these people online in the first place actually, in many cases, engaging in very dangerous experimental behavior on with their own bodies? And it's often out of sheer necessity, because either through lack of access, or because, because certain medications and therapeutics are fundamentally overpriced. And then I went on and studied studied economics. And during my time studying economics, I got an interest in trading biotech stocks. And so a biotech stock typically trades around kind of the success of one of the core assets that is being developed. Okay. One of the core assets that are being developed. And so if positive data emerges about one of these assets, typically the price goes up. And if negative data emerges, the price goes down. And so within biotech companies, this data is extremely well guarded. It's typically only released like every two or three months. Typically only released every two or three months. And when the data is released positive or negative, the stock typically doesn't have 5x or it does minus 80% in a day, which is actually really inefficient if we think about how markets work. It's like this fundamental information asymmetry. But furthermore, most biotech companies actually only have an incentive to share positive data. So a biotech company could be generating 100 studies on a specific compound, and two of them are positive. And let's say eight of them were inconclusive and 90 of them were negative. They would only show the two that are positive. Because you typically wouldn't want other people to know that what you're developing actually isn't working. But in the context of science, this is extremely inefficient. So in science today there's this problem called the reproducibility crisis that is actually quite hard to reproduce other people's data sets that have been generated. So as I began looking into these companies, it reminded me of my experience as a teenager in being in these much more open online shared communities, where people just kind of live shared data in real time. And I thought to myself, hey, there's actually, there's something there on the one side, we have a pharmaceutical system that's incredibly inefficient, and on the other side, we have kind of these online open source communities of patients, researchers gathering that could potentially be much more inefficient. And a lot of this reminded me of the way that open source software is done, which is I think a really big topic in the crypto space and in the Ethereum space. And so in open source software, you have to remember that really up until the 80s and 90s, open source was not a thing. And actually famously Microsoft called open source software a cancer. Many, many years actually actively tried to sue the Linux Foundation. And pharma today, or like biotech, and actually much of science still works like open source in the 80s and 90s. You have to imagine like pharma companies are like giant IBMs that are trying to develop everything close source and effectively buying a drug from a pharmacy that is kind of licensed through a very, very long complex close source process is similar to like in the 90s buying a Windows CD and then going home and providing a new CD key. I don't know how many of you remember doing that, but and today that would be unthinkable, right? Unthinkable to still go into a physical store, buy a CD that is protected, and then enter that CD key. But that's essentially where much of the scientific system is still in today. Still not? Look at that. And so as I began getting more and more into this, I kind of realized that there's a fundamental opportunity here to create a much more open and shared system through the use of Web3 technologies. Because the fundamental difference often in science and in software is that science is in meat space. It exists in the real world. And so you can't just as easily as easily replicated as you can in other industries. So here today we're going to look at what are the problems definition and goals of the D-Sci community and space as a whole. We're going to look a little bit about what we've accomplished so far specifically by going into one specific use case. We want to look at where the whole space is going and the different Lego blocks that are emerging. And then essentially quickly move into workshop sessions and different breakout groups to explore specific building blocks that have emerged in the D-Sci space. So D-Sci has actually only existed for around one year. It's a very, very new phenomenon. And the goal for this workshop today is also to essentially help define a future roadmap of where the space is going and how we can build technical solutions to essentially enable the next generation of D-Sci developers to build permissionless open systems that can really serve future generations. I asked briefly before who knows about Vida, but it would be really interesting actually to know who maybe in the audience actually has a scientific background. Okay, it's actually much more people than expected. Can I maybe ask you what your background is? Social scientist. Okay, cool. And there was a hand up over there. Dentistry. Cool. Anyone else? Yeah? Medicine. Okay, perfect. And there's this, yeah. Can you see that again? Oh, pre-clinical research. Okay, that's our job jam. Also, doctor. Used to be a doctor. Now a D-Sci dev. Kevin? Anatomy and cell biology. Cool. Awesome. What a great crowd. Just keep haunting me. So even though the space has only existed for a year, there's actually, we've seen almost a Cambrian explosion of D-Sci projects. And there's a lot of tinkering and like ongoing development. Big shout out to a team called Altraware Bio. If you want to follow them for like updates around the D-Sci space, this overview was actually created in, I think in May. So I think by now already, there's a whole range of new projects, DAOs and systems that have emerged that are actually not on here. But just to see how quickly the space is moving. And as any kind of new industry emerges, one of the most important things is to avoid redundancy. And if you, maybe some of you remember how the whole D-Fi space and D-Fi community came into being, it's really through building open interoperable Lego blocks that we can all build on together. And so the biggest thing that you actually want to avoid in the early development of a space like this is, yeah, redundancy. And you want to enable permissionless interoperability. Permissionless, for example, means not introducing friction in the use of services. Permissionless means building in an open source way. Permissionless means creating events like this for communities to come together and share, rather than building in silos. Because now that we, like, in building out the D-Sci space, we should really ensure that we're not kind of making the same mistakes as the industries and people that came before us. And it's tempting to do so. And it's tempting to do so because actually creating friction between systems is the way that most industries actually capture value today. It's about introducing middlemen. And so I represent an organization called Molecule, and this is also something that's really important to us. Like, how can we enable open permissionless systems? How can we enable other people to build together, rather than introducing frictions in systems? Yes, cool. So we have a big opportunity here today to define the roadmap for future builders and create an open collaborative technical infrastructure in the same way that Ethereum enabled an open collaborative infrastructure. And so D-Sci will be most successful, in my view, if really enabled as these interoperable permissionless level blocks that we can all put together and build on. So now the question for today's workshop is, what Lego blocks do we need? For which types of D-Sci application? Pretty simple. And if we manage to build those Lego blocks in the right way, then I already think in two years, we won't, in two years, we won't have, we won't have 80 organizations, but we'll probably have like 2000. This is essentially what happened in the Web3 space when DeFi came around. In probably in 2018, 19, I remember there being like maybe two, three, four, five DeFi applications, they were extremely early. But if some of you, for example, remember MakerDAO, who then launched, or not remember, but no MakerDAO, who launched CDPs and DAI. So DAI essentially was a building block that enabled the emergence of a huge amount of DeFi applications. And that's one way to think. So science today is full of problems. Funding in science is incredibly difficult. It's highly competitive. It's very asymmetrically distributed. Often only goes to the best universities. It goes to scientists that have a lot of academic pedigree. It's very political to get funded. And there's a really interesting insight that actually most scientists today would change their topics or their field if, if funding was not a concern. So almost like scientists today have to be extremely opportunistic. They can't often actually work on what they want to work on. They can't actually work on their thesis. They often have to just chase the money. So one of the first researchers that we worked with at Molecule, and also one of the, the first researchers to get funded through VitaDAO, gave us a famous quote, which was that he spends 80% of his time fundraising and 20% of the time doing the research. And so fundraising in this context actually means getting grants from different, different institutions. Then the second thing is replication. I gave this example earlier in with biotech companies, but much of science today is not reproducible. So someone will say, here's a study. I got the following outputs. And then you'll try to run the same experiments and not get the same outputs. So how do we get there? The, the further thing is competition. So science today has become hyper competitive and that really creates perverse incentive. Often you have, you have competition for the same grants and then there's an issue with essentially like scooping each other's work. So it's also, we don't have a system yet today where there's a lot of scientific integrity actually between different scientists. And then a last plum is that science today is still really inaccessible. Much of important scientific literature lives really behind paywalls. It's not, it's not open source. And science is often to many people still not very accessible. If we think that actually before any of, before the whole NFT art craze, I never felt that art was very accessible to me. It's like quite a complex actually process to go into a gallery to figure out what you like to, like art wasn't very accessible for NFTs. But so I think Web3 actually has the potential to make science much more accessible to the, to the broader population. For example, if you have a specific disease and you're really interested in maybe funding therapeutics in an area or actually getting much deeper involved in it, it's not, it's not that easy for you today. And a lot of these problems that we actually face in the scientific community are the same problems that we have in other industries today. They stem from centralizing authorities. So funding, for example, is largely distributed through governments or it's largely distributed through very powerful, large private corporations. And decentralizing authorities essentially affect each part of the scientific value flow. Another example is the publishing houses that essentially control much of how what is actually published. So if a publisher doesn't want to, doesn't want to publish your research, it may just mean that it actually never, never gets disseminated to, to a much broader audience. And so this is what has to, what we've, well, we have to start changing, move away from the centralizing authorities that actually govern access to science and govern truth and move into a much more open scientific system. So the question is, and maybe one, one addition here, how many of you are familiar with the open science movement? Does anyone want to give a definition of open science? Yeah. Similar to most of the things that you've heard so far, right? They're looking to, for ways to lower barriers. I remember talking to the center for open science who was working on a replication challenge. And basically trying to find a lot of, I guess you would say, web to ways of addressing these problems. And why do you think open science today, do you think it has succeeded? I, it doesn't seem like it's gone as fast as you would have hoped that it would go. And I think that they don't really have that much fun. Yeah. Yeah. So funding, huge problem. And funding is often based on incentives. What are the incentives for open science today? Goodwill. Yeah. Let's build a better planet. Like yay. Unfortunately, like, like goodwill doesn't pay. And unfortunately, we live in a very capitalistic and profit-oriented system. It's actually the most interesting application of web three. I think today it has really been creating incentive machines, incentive machines that fundamentally alter people's behavior. And so I think a fundamental thesis for us was that the reason open science has not succeeded is because, because there's no incentives around it. Um, and taking this example that we had earlier with a biotech company that actually only has incentives to publish positive data, but not the negative. If we now had open public markets, as we do in the crypto space, for example, if I, you can make an analogy that if I find a bug in a crypto network that's like open source research. So actually crypto networks are relatively open source. So if I discover a bug, that is actually now valuable negative information. So I could be like, this crypto network is flawed. And now I have two options. I could, I could hack it and stand to benefit from that. And through that, actually, I'm revealing that this code is fundamentally flawed, which is actually valuable to society. It's like, this was actually flawed. Or I could, I could also just ping the team and tell them, Hey guys, your code is buggy, you should fix this ASAP. Now imagine if we had the same thing for medicine. If, if there was an incentive to now create both positive and negative data about an asset, I could be like, Hey, there's a drug here that's in development. There's a lot of kind of, there's a lot of hype around it. But actually I as a scientist have a thesis that this drug might actually be toxic and humans in this, in this use case. And now I have an incentive to release that data or actually to create it. And so all of these incentives are actually for open science are missing today. And that's what we're going to build together. So what if science is decentralized? Maybe first let's ask the question of why is decentralization so important in decentralized system users can participate in a trustless system. So you don't need to trust a middleman. I don't need to trust a publishing house, for example, I can trust other participants in the system. It lowers the risk of systemic failures. I don't know if you have some of you recently heard about the big Alzheimer's trial or Benji, can you maybe quickly spoke to anyone from the audience? Yeah. I mean, there's been this whole issue around like the amyloid data process in Alzheimer's and there's been a lot of career academics who have built their reputation in their laboratories and their funding around Alzheimer's and their particular thesis. And I think they managed to through political force lobbying through the power of the Alzheimer's institution and amyloid data really through a drug through the approval process that just goes before. Yeah. And it got extremely far just based on a fundamentally flawed hypothesis. And that fundamentally flawed hypothesis was promoted by centralizing authorities. Yeah. And so actually many drugs, I think not just one, I mean, many drugs actually started being developed just based on that thesis. Yeah. Another one is just like avoiding censorship resistance. Avoiding censorship resistance fosters a much more open culture. And the last point is really enabling global collaboration. So what is the why, what and how of design? So why are we doing this? We're doing this to build an open science movement and to make science more collaborative and make it accessible and open to everyone. Like if you're a promising scientist in Nairobi, you should have the same opportunities and chances as a promising scientist at Harvard. Today, that is not the case. In the same way that if you're a developer on Ethereum, I don't really care if you're in Nairobi or in Cape town or in Boston does make a difference. The same thing should apply for science. What are we doing? We're building a global open alternative to the current scientific system that anyone can participate in. So equal access to anyone, equal opportunity. If you have a promising, actually, Vincent, I love that, that example that you typically give around, around getting funded by a Dow as a young researcher, you could be in a 16 year old whiz kid sitting somewhere having a great idea looking for funding in the current scientific system, you'd never get funding. You have zero credibility, zero trust to actually receive funding from a reputable institution. But with these Dow's that are emerging, like age doesn't really matter anymore. What matters are your ideas, your integrity and the data that you can produce. And how are we doing this? So Web 3 fundamentally has a technology that enables scientists to raise funding, to run experiments, to share data and distribute their insights much more openly. So what decide verticals have emerged? Because the scientific field is extremely broad. So the area that these sites are targeting and attempting to disrupt, one is funding of data, IP and impact, which is a huge field. So what we're essentially trying to enable is faster and more democratic funding mechanisms that enable communities to form and govern impact or novel forms of IP. Another big vertical is publishing. So having transparent open access publishing with aligned incentives for all participants. Having incentivized peer review systems, for example. The third one is Dow's and research governance over these research assets. So Dow's present in a highly new way of organizing researcher or patient involvement and for example clinical trials. This has never been possible before. If you're a patient today, you have no absolutely no choice what lands, what comes onto the market. You have no choice over the price. It might be a life-saving medicine that you need every day to survive. You have no influence on the price, on the access, whether it is actually the web medication. Yeah. And neither do researchers. So we have this very long drug development process, but we've removed the core stakeholders from that process. And another one is identity and reputation systems for academic credentialing. So how do we actually trust scientists in these different systems? Maybe I want to ask into the audience. So these are kind of these side verticals that we've seen emerge across the field. Do you feel that anything is actually missing there? Yeah? Yeah. How would you just, like, in what use case? Yeah. Really good point. Yeah. Yeah. Yeah. So what we're going to do, like, what are you? Yes? Addressing, like, the replication or reproducibility of data? That's actually a good point. Well, in publishing, actually. But we should add it. That's actually, let's note that as a point for the workshop later. It would actually be awesome to, like, not just consider these areas, but really map out holistically what all the different verticals are. I want to go a little bit into just one of the verticals, which we're working on specifically, but we also don't have to spend too much time on it. There's a lot of other talks that I've given in the past where I can go into this. But of all of these areas, where we're currently seeing the most traction is funding in IP. Because it's, like, it's relatively simple today, I think, to facilitate funding use cases through Web 3 technology. I think it's still much harder, for example, to facilitate identity or reputation systems through Web 3. So what are now the subverticals in funding in IP? So we have retroactive public goods funding. Some of you may have participated or even benefited from the recent DSI Gitcoin route. So that was a great example of that. The other one is quadratic funding, having a fairer, more democratic weighted balance actually allocate funding to projects. The third one is having DAOs and tokenized incentive structures, such as Vita. And the fourth one are IP NFTs, which are a new form of on-chain native IP and data ownership. So actually for each of these verticals as well, something that we could do later, we could start defining all of the different sub areas in that vertical. I want to quickly maybe ask a question into the audience. So do you want me to go deep into one of these verticals now? Or should we actually leave it much more open? Go deep. Okay. Let's go deep. My goal, like, yeah, the only thing I want to avoid is that we prime it too much about this one use case, because there's so many other use cases. But what is interesting about looking at funding in IP and specifically what we as molecule have developed, which is the IP NFTs, is that it's already very applicable in this theoretic. But so if we look at one of the like the fundamental problem spaces of how does innovation and biotech research actually work today. So you have innovation emerging at the fringes of an organization. And then in most organizations, innovation is driven through a funnel. And then essentially you curate ideas as they emerge in systems. And eventually something goes into a market. So you have idea generation, you define the project, and it moves through this funnel. And you can actually map the same process to the entire drug development pipeline. So this process today can take like industry, industry averages to take up to 10 years to actually develop a new drug and to drive it through all these cycles. But now one of the biggest problems with that is that like, cancer is a global phenomenon, yet every company developing cancer therapeutics is doing it in a silo by themselves. And you could ask, wait, wouldn't it be much more efficient if this was done as an open and public funnel? Imagine instead of every company having this funnel by themselves. And like, we're essentially investing so many resources on a global scale to try to achieve the same thing. Wouldn't it be much more efficient if we had a giant open funnel? And this is essentially what we could build with referee. And you can imagine each of these little dots are now individual research projects or individual assets driving through this funnel. And why is it, why is it that way? The reason actually behind that everyone is doing their own thing is because we have something that called the patent system that fundamentally uses IP. But patents are really legacy legal physical assets. Patents are literally often still in most organizations boxes of papers, which if you think we're in 2022 is absolutely mind boggling. It's somewhat ridiculous that we're using an extremely bureaucratic and outdated IP system that makes IP very hard to enact. It's often much too expensive to claim IP in the early stages. It's really hard to get IP out of the university. The patent filing system takes ages. And so what happened actually often that a lot of IP or a lot of promising projects that are in here never see the light of day. And it's not because of, because they're actually not interesting projects. It's actually because the entire innovation system that we built around them is really inefficient. And so there's something called the value of death typically happens here. So actually most of the world's innovation never even enters like enters further stages, which is if we think about scientific progress is really a shame. It's like there's so much good science that never sees the light of day. And if we think about data and software, data and software today is completely virtual. But if you think back 40 years ago companies literally had still had like entire floors with filing cabinets, pulling hundreds of people to sort through those data sets. This is an example I really like using. This screenshot was taken about a week ago. This is the current state of the art of the US PTO patent search system. That's literally how it looks in 2022. And the only other, the only alternatives to this system are paying really expensive like legal software to essentially search and discover patents. So if we, if I was someone who was like, Hey, let me look for interesting cancer research that's been patented. And I'm like trying to find something in here. Like this is, this is what I'm supposed to use from the government. Of course, there's better systems than this, but they're typically proprietary. I'd have to pay for them. Yeah. And so this, this is what we want to get away from. Yeah. Yeah. What do you think? I'm not sure. It looks like a lot of data is pretty profitable. I mean, I think the first thing, the first thing I would say is it's extremely static. It's impossible for me to actually contact the researcher or the Institute that's working on it. It's hard to, for me to see what's done or if anything has been successful around it. Yeah. These are the work slides that are on the shelf. No. Depreciation here is a very low resolution, like high volume database that the government probably doesn't have a lot of people building like interfaces to go over this. So I think this is more the problem about just like fundamental transparency. Like uniquely extraction. Reason or meaning from a lot of this data. I think also like the scientific data that underlies the IP is also missing from here. Yeah. Usually they'll have some like black and white, like black and white graphs who never like audit under like data understand how experiments are run. I also, I have no idea if someone's actively working on this. I also, yeah, it's, and actually if someone has patent of it, it almost disincentivizes me to work on it because like fundamentally, for example, if, if let's say I'm a researcher at a university and I look at a new interesting molecule that I've kind of come across, and I search whether it's been patented and I read that it's been patented. Like I don't have an incentive to now work on it because I could literally get sued for, for engaging in that, in that, in that research. So IP monopolies really kill innovation at large and they also lead to much higher drug prices. So what we have a system today that rather than the best science winning, we really have revenue drivers that dominate, that dominate medicine. So at molecule, we've really spent almost the past four years asking ourselves what could a better system look like for this and trying to build towards an open market for IP. So kind of trying to reimagine the system and making it much more open and transparent. And one fundamental thesis that we have is that most of the world's potential scientific talent remains untapped. And so for example, if we think through what happened in, in the NFT market, the NFT market for, for artists and now for music, fundamentally has changed how we interact with art and fundamentally has enabled a different creator economy for artists. And so what if we enable a creator economy for scientists to really help, helping the best innovation across the world rise to the top. So we are building towards an open transparent marketplace for research funding that is underpinned by a new, a new DSIDE primitive. And now the question is really how do we bring legal IP and data into web three? And one kind of DSIDE Lego block that, that we started building on is essentially a new, a concept called an IP NFT. So what is an IP NFT? IP NFT is first, it's a legal contract in a real world license that is tied to research or data. So it's a legal contract that now takes IP out of an organization or out of university and attaches that to an NFT. The second is it has a storage layer. So you have a decentralized permanent data storage with public and private data repos. So almost not imagine this NFT is like simple, simple story saying it is like, has like a decentralized Google Drive attached to it where the owner can now provide read and write permissions to anyone that wants that. It's transactable in the same way that an NFT is transactable. It's discoverable through public metadata around that NFT. So I can now publicly on chain start searching through the metadata of these assets. And then fundamentally, and I think this is the most exciting part, the IP now becomes programmable and composable. That means you have governance, you have fractalization. If you're interested in actually IP NFT, fractalization, we have one of our, someone from our legal team here, we've developed a new framework that's called the friends framework. Because fundamentally what you need to avoid is making any of these assets like securities, because then they don't become transactable in web three and they don't become Lego blocks, at least for now. You can have programmatic royalties or pay for success models. We also have someone here from the crowdfunding cures team, which pay for success models represent a really new fascinating way to essentially do repurposing of drugs. What does this kind of look like from a kind of from an on chain perspective? So you have a real legal contract that's mapped onto a smart contract that then maps to metadata that is sort of our weave, and then you have encrypted and private patent data that now fundamentally protects the IP. And this latter part you can open up, you can open source it, or you can keep it private. One big thing that we read is it's actually really hard to do open science in a field like biotech, because we began working with an organization called the open source pharma foundation. This was in 2019. And essentially what they do is they create public GitHub repos for open source drug development that they're doing. So they were researching malaria, open source isomyzatoma, essentially different tropical diseases that it really underserved. And so what they realized though in doing that, as soon as you publish, for example, the structure of a new molecule in a GitHub repo, you open source it, and you say this could be used as a malaria drug, it fundamentally forever becomes unpatentable. And now they were working with the development in a Gates Foundation, for example, and then realized throughout doing that, that the Gates Foundation essentially said, we'll never be able to fund any of your malaria drugs, even if they work, because they become unpatentable. And because the problem is, if you don't have any patents around drugs, you can't essentially get the asset through late stage clinical development. No one is going to come and pay, let's say the $100 million plus that you need for like a stage two or stage three clinical trial, which is like, which is a big shame. It's just how the system works today. It's fundamentally impossible to do open source drug development. However, I think as DAOs and communities, for example, get much larger, you can actually move more into impact driven development. So maybe, so to close the session, IPNFTs, in our view, are really a first composable Web 3 Lego block. So they can be transacted like NFTs and applied for funding. DAOs can now build portfolios of research of IPNFTs. So we have this new phenomenon a bio DAOs that have been emerging, which are DAOs that are now focused on different therapeutic areas. So since we did now emerged, we now have a DAO focusing on neuroscience. There's a DAO focusing on psychedelic research. There's a DAO focusing on women's reproductive health. There's a DAO focusing on hair loss. There's another DAO focusing on synthetic biology. And all of these DAOs can now use the IPNFT as a basic building block to build an on-chain portfolio of IP. NFTs can also be fractionalized and be inserted as these Lego blocks. You could imagine an IPNFT being fractionalized, and then those fractions going into different on-chain liquidity pools. They're fascinating new mechanisms where the fees that are now being generated through these liquidity pools can actually be used to fund the research as well. So everything that has been built in DeFi for the past two years, three years, can now also be cross-applied to DeFi. For example, also data access can now be granted by multi-signature wallets, and we're really only beginning to scratch the surface of what is possible here. One of our core goals in molecule for the next two months, three months, is actually fully open sourcing this first version of the IPNFT protocol to enable anyone to start playing around with it and essentially just tinkering around with it, because we don't actually know yet how many applications this can be used for. So what are the basic DeFi Web 3 Lego blocks? So I think we're looking at transaction layers, we're looking at data storage layers, we're looking at compute and execution layers, and then finally we have identity layers or reputation layers. The IPNFT, for example, combines what I would say is a transaction layer and a data storage layer, but it doesn't actually touch any of the other layers. And so now we're slowly moving into the workshop part. I think Tyler just left. So the goals for this workshop today is essentially now to look at both what Lego blocks do we have, and to look at how they match with different Deci verticals. So we have different Lego blocks that are merging Web 3, how do we map them into these Deci verticals to solve fundamental problems in Web 3. So the goal of this workshop will be to really explore how we can, how decentralized science can improve the process of funding doing and disseminating science through the use of decentralized systems. So now we're going to look at each of these different elements. And yeah, since you start watching them. Vincent? Yeah. So the fault of the workshop will be that we now have a breakout session for about 40 minutes where we can gather around these different topics. We also have multiple members from leading organizations from these different areas in the room with us today. We can then do like breakout pitches where people from that have not formed through these different teams can essentially pitch their ideas. And then we can all discuss them together. Vincent? Yeah. So maybe like basically we'll collect all the outputs and nodes kind of from each table in a shared Google doc, which also has some of the like links and instructions so you can scan this link. I think it's also under bit.ly slash Deci Booker. Yeah. Tyler, do you also want to run, explain the workshop now, do you want to run through it again? Sure. Yeah, so is there a mic or is there a mic? Yeah. Yeah. So I think the thought was basically there's these four high level verticals that we sort of identified, but obviously this can be opened up much more broadly depending on if someone has an individual like interest that they want to tackle that isn't met by the list that we have here. But so we could look at things in the context of funding an IP, which could be things such as like basic science, for example, grant funding, translational research, productizing medicines all the way. I'm taking a very less science approach, but we could also look at how this relates to social sciences, for example, data and reproducibility, which we can think about in the context of like how scientists work today, how data is captured, what are some of the incentive structures that exist for problems around how data is actually reproduced. We have this sort of negative data and sort of like positive and negative data problems with regards to reproducibility and also incentive structures, things in the context of publishing, like how to make incentivize open access, the culture of science and how it functions to largely reward scientists that only publish in very high impact journals and the effect that has things like identity and reputation. So potentially exploring how things like impact factor, for example, create issues in the current scientific system and rethinking how a lot of these things would look in a sort of DSi-native, web-free context. And so within each of these, I think there's obviously a lot of ideas that we could tackle. I think functionally it would make most sense to sort of organize around either the use case that we want to design, which could be like creating a fast ground mechanism for basic science research funding using Ethereum or creating a reputation system for DSi-dallas, as examples, or we could also look at just a particular problem space. So I think there's quite a bit of flexibility, but I think the idea would be to organize in individual tables that are looking at a specific problem that we sort of agree on and then explore that, map it out a bit. And I think, yeah, I don't know if you want to open, you could open the doc that is on here just so we could maybe go through. The doc has the same stuff that's in here. Yeah, just a little bit more. So I think there's a couple of different things. I think it's also, depending on... Todd, do you want to take this? Yeah, so we've prepared a doc that we're going to share a QR code for, although I think we might have been rugged on the internet front, so it might be a bit difficult. You can't open the doc here, no worries. Yeah, so you can scan this with your phone and we've basically prepared a document that talks a little bit about some of the goals of the workshop. I think there's a couple of different ways that we can approach this. The first is really just to facilitate robust discussions that are currently looking at what is the status quo in the sciences today? So if we look at something like funding, for example, or something like publishing, what is functioning really well in the current system? What is not functioning so well in the current system? And basically identifying this sort of where we're succeeding and where we're not. And then begin to think about what aspects of the current system we want to migrate into this novel, let's say, DSI focused architecture, what we think needs to be upgraded and what can go. And then to actually begin thinking through, maybe by a show of hands, how many developers are in the room or technically oriented folks, let's say, you don't need to necessarily be a developer. So yeah, maybe an interesting way to organize if we're capable of doing this would be to, in a perfect world, this might be challenging, but we would have like one developer per table, one UI, UX person per table, one scientist per table, one maybe a social scientist per table and try to create like little working groups that are basically trying to prototype what possible infrastructure could look like in the context of DSI to solve some of these problems. Does that make sense? Do we want to maybe start by like, we can also throw up a couple prompts or like use cases that we want to tackle people could raise their hands and we could also just walk around and try to organize, self-organize into different topics. That's extremely difficult to read. But yeah, totally, there's a lot of flexibility, I think, from this point in how we actually, how we actually run this. But I think that the prime takeaway or the main thing that we should be working towards is basically just trying to create a clear roadmap to solving some of the problems that we've identified together in these verticals. Does that make sense? Any questions? Cool. Would it make sense to start by actually organizing a bit based on expertise so like devs can get together, product people can get together, UI, UX people, scientists can get together and then we can actually from that to like organize into actual groups and tables for specific problems or? Or I would consider organizing by vertical though. Yeah, we can also just create a bunch of verticals for example and then yeah ideally we just want even distribution of people with particular expertise in each area if that's possible. Yeah and then probably for verticals. Otherwise you one devs to only be together. I think that would just be the first step and then totally work. Oh yeah, we can do that too. Are there any major like, does anyone have a prompt or an idea or a specific like, that's just the idea that they would like to work on and we can get a kind of thinking about that. I think every table is like built in public and then also then you can put it on which top table. So do you want to maybe take five or ten minutes to brainstorm some more ideas and then begin organizing around this? Yeah, so if you for example have a specific problem in mind you would like to work on, maybe write it down on a piece of paper and grab a table and yeah just begin organizing. Maybe maybe maybe we start like this. Like what are some of the like should we start with like problem spaces like again like a vertical like publishing or like funding and then someone can share their idea and then people can gather around that idea. Okay, what idea do you want to work on, Tyler? I would love to think about a novel reputation system specifically for DSI. So looking at the current faults with impact factor and basically how we deal with academic credentials and think about what the V2 of this looks like for DSI. So if you want to think about this problem you're more than welcome to come join me at this table to tackle it. Okay, reputation. I have an idea. Cool. I'm curious to look at publishing from a hypertext perspective so like because I see how referencing is super old school like you're publishing PDFs it's super hard to look at where research came from so I'm interested in exploring like the web-native hypertext system and migrating that to how research could look like in an interactive web environment. Cool, publishing at a high level. My name is Michael. I'm interested in like the scientific process so like the data collection, the data storage, any sort of computation that happens around it in order to like hypothesis generation. Cool. Okay, how would you, what would the, what about talking about? Data collection. Data collection, awesome. What? Data collection and scientific process around data connection. For me it's legal protection for fundraising around IPNFTs, fundraising around the science and the legal structures that people want to employ for that. So if you want to do legal, over here. Jesse's a lawyer. What? Not your lawyer. I was interested in identity and reputation in general but I think that's more likely what you were speaking about. Okay, awesome. Cool. Perfect. I would do one on hyper certificates and kind of impact markets and like retroactive public goods. Dope. We have maybe one for one more, I guess. Decentralized clinical trials. Oh, that's awesome. Cool. Okay, very complex. Decentralized clinical trials. Anyone else? Does anyone have a, does anyone have, everyone have a topic that they're like interested in doing or helping with? Cool, I assume so. Cool. Time is slowly up. I want to slowly start bringing all of you guys back. What? No, why will it be? Call my attention. I don't know. I was just, yeah, I was, did the pick get you? Tobacco? Now I got, now I got your attention. Okay. Cool. Hey everyone. So I want to slowly start bringing us back because it would be awesome if each of your groups can now kind of present the specific vertical that you explored and some of the solutions, some of the Lego blocks that maybe emerged around that. So maybe if someone from your team can come up or do you want to stay in your circle? What do you prefer? Stay in the circle? Okay, but you have to stand up while you're presenting. Okay. Who wants, and so maybe take two to three minutes to walk through the findings of your team and maybe the solutions that you discussed. And then it will be really awesome if someone from your team can actually capture it. And if you haven't already just write it and put it into this document, if it is already in the document, I'm actually just going to scroll through and check. If anyone can see, I think I misplaced the mic somewhere or maybe the support people took it, but it should be somewhere. And yeah, so just to describe kind of that main outputs that you're also going to present into this doc so that we can share it with the world and the community off to this. That would be really awesome. Okay. Who wants to go first? Cool, Tyler, floor is yours. You have to stand up. So we were looking at a scientific reputation system and looking first at how our current like scientific reputation system functions. So we first map the current system out into two verticals. One would be formal credentials and other would be sort of informal reputation. And formal credentials are things that are verifiable, measurable, and somewhat quantifiable. These are things like academic affiliation, degree, impact factor, h score, pedigree, citations, for example, and funding sources. And then there's more sort of informal, more let's say qualitative reputation things. This is like social media and press, collaborators and peers who you know, political affiliations, hype, your ability to influence others. And we mapped out some pros and cons in each of these. If you want to like look at these in detail, you're more than welcome to. They'll be in the doc. But just as like a high level example without going into excruciating depth on each of these, things like academic affiliation can demonstrate or attract quality. They could signal trustworthiness. They can be associated with specific areas of expertise. They could act as sort of shelling points. But on the other side, they can create inequality. They can, for example, foster sort of like certain reinforcement patterns, create like legacy students. There's a bunch of problems and sort of fostering. Sorry, this doc is moving in real time as I'm reading it. I'll just talk through it rather. I think I can do it from memory. There's there's a bunch of things in the context of this current, let's say verifiable metrics that still need work. So impact factor is something that I think can be quite positive in terms of it can demonstrate readership and citations, but it can also be heavily gamified and is used to largely sort of reinforce certain popular ideas. Many Nobel laureates, for example, don't publish in high impact journals. Their ideas are initially rejected. So we were basically trying to look at in the context of all of these different verticals, whether they be like verifiable credentials or sort of informal credentials, what are the sort of things that we want to bring over into into DSI in the context of this, let's say, new reputation system. And so we did a little bit of work thinking about things like credentials. And these could be represented by things like soulbound tokens, for example. So instead of just having academic affiliations who want to have things like industry affiliations, guild affiliations, DAO affiliations, what are the communities that you're associated with, what are the values of those communities, and how do we measure those? I think we also want to look at for people who are involved actively in funding, how are they making an impact, what DAOs and organizations are they a part of, what areas have they done research in, what areas have they maybe impacted. And this is also a hard thing to measure, but I think using impact certificates and actually Po-ops in some cases, you can begin to use some of those technologies to measure the impact that somebody's having. We could also look at things like GITs and pull requests in the context of data becoming increasingly, let's say if you're doing in silico work in the context of biology, you can begin to look more in real time at how somebody is contributing data to a certain research area and begin to measure an impact that way. And there's also things like measuring DAO reputation. So if I'm a contributor in a DAO, am I somebody that people are delegating boats to? Am I somebody who's being rewarded via things like coordinate? Have I been elected a steward? And then also looking at other factors like value captured, value generated? Am I deploying funding? Have I received funding from certain people? We could also look at token holdings. We could look at how active I am as a governance contributor in a DAO. But in the end, there's all of these, I would say new interaction patterns that are happening via the internet that are actually producing a lot of really, really rich data that is quite different from how science works today. And what we want to try to figure out and where we haven't gotten to actually in the context of this short session is really thinking about how those things are all weighted, how they're measured. And I think the most interesting question here, if we think about creating like a holistic reputation system that would be akin to something like impact factor, it's really challenging to do because I think the needs of scientific communities are very different. So for example, I might be somebody that from a reputation perspective is a really good developer or is maybe just really good at data collation and data analysis. How is my reputation weighed against somebody who is maybe also publishing and also funding research? And so thinking about dynamically how these systems interact, I think will be a really interesting technical challenge. One of the things that we're thinking about was like if a DAO has a certain set of needs, so for example, feed a DAO might have a need for longevity researchers that have worked in X or Y fields and have, for example, experience with certain assays and certain data techniques, that is very different from another organization that is maybe just an investment focused DAO and wants to know if somebody has been particularly good at deploying funding or making good bets. And so I'll cap it here. So like the thing that we're thinking about is like in the context of how can we create dynamic reputation systems where the organizations themselves are specifying what they're looking for, whether it be a series of soulbound tokens and DAO affiliations or specific credentials, and how can we have a system that basically adapts based on the skill sets that individuals have. So sort of like everything we've talked about taking a dynamic approach to this reputation system that is not one-sided, meaning that it's not just the impact factor of researcher, but the needs of an organization juxtaposed against the specific skill set that someone has. And I think this could be really interesting because going back to an earlier example that you made, Paul, if there's like a 16-year-old somewhere in the world that ends up being an excellent scientific reviewer because they self-taught themselves, I don't know, like biochemistry and you know, on the internet no one knows your age, but they're making comments on papers or whatever that are really valued by the community. We want a way to measure that that is like independent of academic affiliation, for example. Or what if it was even just an AI bot? Or a dog. On the internet no one knows your dog. And in a blockchain network no one knows your fridge. On that note, thank you so much for sharing that, Tyler. Could your group also please upload your dog into the section? And then maybe I'll just go... Sorry, it's a bit buggy here. Cool. We'll go into tableware traffic public goods next just because that's up right now. From the Google Electric, isn't it? Okay, awesome. And we have about maybe we have about three minutes at the moment for each topic. I just want to make sure that everyone gets a chance to speak before they kick us out of here. Oh, Salva? Yeah, do you want to come up front? Cool. Yeah, awesome. Okay, so yeah, we're talking about impact certificates and hyper certs and retract your public goods funding. So I don't know if you guys can bring up there. We've got like a diagram. It looks like this. I think it's just a bit down further down. There we go. Yep. Okay. So the whole idea behind impact markets is to look at funding public goods and science as a public good. So the most important thing is to look at like, how do we value impact? That's the trickiest part. And then once we've got that, essentially the market, we need to find retroactive funding. So in this particular case, we're talking about, we've got this thing called the longevity prize, the Vincent sorting, we read it through get coin and sort of raise, say 250 grand. And so we've got like a source of retroactive funding available. The next question is kind of what do we want to, what sort of impacts is going to be something that can scale. So if the prize ends up being $10 million or $100 million or whatever, then we've still got impact being generated. Once you've got that, essentially you can do this thing called, you can mint an IPNFT or a hyper cert, they're kind of similar concepts. But what those do is they let you as somebody that can potentially deliver some impact, say you're a scientist and you've got a really great idea and you want to get funded and it's a public good, you can mint this IPNFT as hyper cert and then raise money basically from investors where before like, because it's a public good, essentially there's no way like under the traditional IP system that you can enforce a monopoly price. So yeah, I keep talking about generic drugs and things like that. So that's something I'm interested in. But which is basically if you can get an off patent drug and repurpose it to treat a new disease like so ketamine to treat depression, let's say you could actually create billions of dollars in social value with quite a little bit of investment there at Retractive Funding. So yeah, so basically once you've delivered the impact, whatever that might be, so for instance, improved clinical outcomes versus usual care, then you will get the outcome payment under this law from the Retractive Funding and the market basically figures things out. So if it's not enough funding, like if 250K is not enough, you can basically get philanthropy to put in the rest. So it's a little bit like get coin where like you sort of have little donations creating signals or little investments creating signals for a philanthropic funder to come in and fund the rest. You may rename like two or three Lego blocks that you see that are really important to make this. Yeah, I think what one could add is basically that every scientist could basically mint this impact certificate even if they have like other funding and then could sell it like one of our ideas was in a more like crowdfunding sense that you don't have to buy it for say 100,000 but you can put in 100 bucks or 1000 bucks and get like almost like a small share in this impact certificate which could be just like an IP NFT which could then be basically put into like a million pieces for example and then it's almost like also like a pop and kind of like collectible because you show that you funded research so it gives you not only potentially like the price but also almost like some status as being a science funder you're doing like impact so it's kind of like also a bit of philanthropy and it's almost like collectible so you can like collect all the little research you funded but it's also like a prediction market because you can potentially sell it so like as it could hit the price it could also grow in value so it's almost like you're signaling what you think is the most likely to hit the outcome. So the Lego blocks are like IP NFT's, ELC20's and kind of like deploying them just like on Uniswap and then people can create liquidity or short them. Just one more thing I like so for retroactive funders that's super important so maybe you could have an open like impact impact NFT scientific impact NFT to encourage people to become retroactive funders. Okay we have data and process up next. So I think some of the things we talked about was like so the first thing we went into was like how can we collect data for scientific purposes and then one of the things we cover one of the problems that we need to solve is like how do we collect how do you get gather people not just for collecting surveys not just for conducting surveys but also like how do you gather people to participate in clinical trials and all that falls under like a we need sort of like a database of sorts which can store metadata and allow people to like ask questions to this entire population and see narrow them down to specific segments and so that could be like a distributed database for example maybe and then you have to apply like operations on that distributed database to find your target audience and so other things we talked about was like how do we in throughout the scientific process you also need to like I didn't before you even start research you need to identify what is a problem that actually needs to be solved and so in order to identify a real issue that needs to be solved you have to make sure that is the actual problem and and you need to not not you can't trust a majority of the population to know what is an actual problem because certain populations may have like uh like there may be like groups of people who are healthy and groups of people who aren't and and the healthy people may uh outvote the unhealthy people so then how do you address the problems with the unhealthy people like maybe uh yeah like like uh because we do actually need to solve the problem for unhealthy people in a way uh yeah so what is the right thing to fund is the is a problem you solve also and then uh yeah more things on uh yeah so can you scroll down a bit like scroll down okay oh oh my no so I think other things are like uh yeah like so we talked about cloud labs and how can like cloud labs and like how can we conduct how can we have like databases of people of labs or like facilities that can help conduct research and then uh also like even currently what uh and then also another thing is like throughout the scientific process you also need to make sure your experiments are conducted uh trust in a trusted manner so how can we have chains of trust that go that go from like from the the ground maybe like from the individual level all the way to back to the researcher and then beyond the research actually yeah and then one final thing we touched on is like how do you duplicate uh how do you duplicate experiments or studies conducted and that will usually involve like open source studies and being able to query these studies so we'll need solutions to like uh uh catalog and index uh studies already conducted in the past and yeah there's a lot of problems in I think in in summary there's a lot of problems throughout the scientific process that requires individual tools and race uh tool very uh specific open tools that are uh globally accessible and in order to be able to uh increase the efficiency of the way the scientific process is conducted yeah I think that's summary of what we would think thank you so much okay any questions from anyone also maybe on the previous topic I would love to give more time for questions and discussion actually but we're also quite low on time and when I want everyone to be able to present um okay then last but not least we have publishing what oh yeah yeah sorry sorry okay thank you um so we're the publishing team um I guess for us it's it's really a question of building connective tissue also before I start we're also the penguin team because we have a penguin on our little thing here um but but the real real question we're trying to solve is that of like connective tissue and really linking everything together and the big question at the top is how do we amplify collective thinking within the context of scientific discovery and how can we make that process easier um for people um and this involves you know at the start we're just asking lots of questions um you know who are we designing this for are we designing this to replace from for publishers or for researchers um and also is there an incentive for publishers to participate so what we kind of realized during our discussion is that um basically what we're trying to do is replace publishers as middlemen and come up with a system that is more decentralized and equitable for the researchers that are involved um and another big problem that we talked about was actually um crypto history and and I think Brewster Kale had a brilliant talk on this yesterday or the day before from the internet archive um how you can just pull publications off of the record and you do what you really want to avoid that and you want things to be accountable 50 hundred years down the road so our our solution is really just to use the existing technologies that already are around like our weave and content addressing to build connective tissue um for linking between papers and publications um and the big sort of money or like the Lego here would be for these side would be some sort of publishing standard with the associated metadata that will link these papers together whether that is like referencing um whether that is like referring to people who built the system built or wrote the paper and I think that's where it kind of plugs into the reputation system that the other team presents earlier and also we discussed like how can we connect these papers to the ip that resulted from them and we were talking about how we could plug that in as well so you could imagine visiting like a wikipedia site that pulls from an our weave back end and has like really good user interface that could click and see what ip's are related to this how the reputation of the scientists who wrote this is and like also look at the references by following the our weave links and because it's our weave it's permanent um you don't have to like worry about the data disappearing any point in time um the we also looked at some other systems here like wikipedia is probably a great example for a linked um content system that has worked really well for for data um also stuff like dsi labs and um bio rxiv um we're doing research in adjacent spaces um but but yeah that's about it thank you thank you so much uh I really like how you can essentially already started creating connections between the different um yeah the different topics um team in the back benji that sounds like a bad excuse can you still talk about it during an upgrade and we we have about six minutes six minutes left okay Mike okay I feel powerful now um so uh we talked about uh decentralized clinical trials um firstly why would you decentralize them how would web3 work I think we came to a broad consensus that generally that generally speaking the things you'd initially want to look at would be clinical trials with things that weren't particularly dangerous and you can get over the counter um and it's okay to ask the general public to do um we talked about data issues in that um whether data should be stolen people's phone encrypts and put on the blockchain um and the issues that can happen when you've locked data up on people's devices and they get sick or ill or unconscious um and generally decide that in a clinical trial they should probably always consent and those exceptions don't matter too much um we also talked about how you would decide which trials to run and why um I think broadly we agreed there should be some vague oversight from professional people in that so you don't do something too dangerous um and lastly we talked about when down a big rabbit hole on incentives um the potential of uh micropayments and the pitfalls of people just taking the step counter um and attaching it to a piece of string and putting it in front of a fan apparently that works quite well um yeah thank you so much thank you okay so on the legal side we worked on a practical case um and I presented a problem to Jesse so I work for a foresight institute we are a sort of perfect problem child for this side we are a non-profit organization that is based in california that has existed for 40 years doing science really at the edge trying to accelerate science that's way too ambitious or too niche for legacy institution generally to to feel comfortable to fund so you we work on neurotech biotech uh trying to push uh longevity age reversal brain computer interfaces building dyes dyes and spheres because we also work on space um ai crypto um etc um and so we have this organization it's a non-profit and we want to doubt if I eat um and so there were two main categories of problems that I presented to Jesse one of them was uh how do we uh not jeopardize the existing organization and the second one was uh more focused on governance and how to create a data structure that is really aligned with the ethics of our organization and jesse was very helpful thanks lou so we really have two two sets of solutions to these two questions one is how do you make a dow from your non-profit that doesn't jeopardize the dow and its existence and the answer is what we did to vita now a fair launch with with no presale and that may rub some people the wrong way who want to raise money privately from investors before launching the dow but it's really the key to prevent an existential risk to your organization is to do a fair launch and what that means in practice is um having a team that's willing to bootstrap and and launch this organization and then get retroactively voted for their token based compensation after the launch of the dow number two is that there should be no direct claim on the treasury so you can't have a malloc dow style rage quit function that allows for the token holders to claim treasury assets um there has to be some kind of break in that chain in order to prevent the token from having a secured interest in the treasury of the dow in other words the token being a security and in that same vein no distribution of profits to token holders and so that leads to the question of like well well if you can't distribute profits to token holders token holders have no direct claim on the treasury then then what's the token for right how do you reward people for participating in this dow at all and and the answer is in part what we do a vita dow and the best example is to look at a university endowment a university endowment makes investments right it makes money it it makes money for the university and then that money is distributed to university administrators to students through scholarships to making buildings to make the university a better place no professor or student has a direct claim on profits that are made from the investments in the university endowment but the university grows and becomes a better place and becomes a higher salaries professors get paid more scholarships are handed out to larger degrees the more money that the endowment makes however again no professor has any claim on profits that are made from the university endowment and so i encourage you to think of your biotech dow and the money that it's making and the treasury that it has like a university endowment and the structure of the dow as a non-profit structured almost like a university and then later with our ip nft 2.0 protocol and the fractionalization that comes with that into friends and fam tokens which we can talk about at dinner tonight you what you do is you have a really interesting system whereby the token holders of your dow don't just have the ability to vote in snapshot on governance proposals which is like you know step one of dow participation and that's about as far as many dows get unless they have this monologue style raised grid function what you have with ip nfts and their fractions is the ability to drive value to token holders and allow them to accumulate portions of the ip and portions of the treasury assets direct exposure to interesting projects that they want to co-govern and you know if they kyc in a credit that they want to co-own yeah okay thank you just making a final announcement uh wait don't disconnect i put on top of the document uh so tonight in my in partnership with molecule foresight is foresight is hosting a decide dinner at teatro ripio uh and so we invite you all to join and continue this discussion i put the perfect thanks um otherwise the link is at the top of the document i put it just under the agenda in the google doc you can directly click there and register and join you join us that's it paul thank you so much yeah a massive thank you for everyone um that participated here i think we had some really really cool exciting insights come out of this uh you all have the shared doc what we'll try to do maybe from outside is like maybe correlate this into like a blog post of sorts or just a maybe a twitter thread um to share it with the community if you want to be tagged in it please feel free to put your twitter handle into the doc um yeah and uh look forward to maybe seeing some of you at the dinner tonight and remember build uh build lego blocks build railroads build roads that other people can use and that can cross integrate uh into the community because we still have such a big opportunity here to set those railways for everyone to um yeah to interoperate which is what open science and decides should all be about thank you so much