 When we want to build a new research organization, we have to ask ourselves the question what makes up a good research environment, what makes up a good research organization. And this is not a question that we've been asking as the first people. There are actually a lot of people that have been asking these questions for, and some of them aren't even in the room with us today, about what have been in history, research organizations have been very productive, research environments that have been extremely productive, and what were the key hallmarks that made them effective. And these are just some snippets. This is a slide actually from the last one in the conference, a slide for Juan. Then Ben Reiner has been talking about private ARPAs. This is a book I can very much recommend by Michael Nielsen, and then Adam Marble-Stewins and Sam Rodriguez work on focused research organizations. And when I think about what all of these works have in common, there are three elements that for me at least keep recurring, which is scientists have access to tools. They need to be able to basically test the boundaries of theory and tinker there and bump things against each other at the sort of cutting edge of technology. Then they need to be able to form dynamic teams, teams that consist of scientists from different disciplines, maybe even not only scientists but also practitioners that have very valuable micro-expertise. And then they need to be able to have some unconditional funding to get started and then ideally a smooth distribution of funding access as these projects mature. And I think one understudied example of where there was a productive research environment where the Cold Spring Harbor laboratory courses after World War II. So Cold Spring Harbor is actually not that far from here. It's like a 50-minute commuter rail train ride from New York. And after World War II, you had chemists, biologists and also physicists come together at Cold Spring Harbor and study the interaction between bacteria and phages. And that was at a time where the structure of DNA was not yet discovered or just about to be discovered. And by coming together only over summer with a diverse group of teams that had very specific micro-expertise, so for example Max Dellebrook was a trained nuclear physicist, didn't know a lot about phages and bacteria, but had a very quantitative way of looking at the world. These groups of scientists were able to ask some really interesting questions. Had the lab, you can see that in the background, to start some very easy first experiments and then would go back to their research institute, some of them in Europe, some of them in the States, and were able to write the grants that they needed in order to actually do the meaningful work that then led to a lot of breakthroughs. So rubbing it up, I think what scientists need to be in an effective research environment, they need access to scientific infrastructure to allow for this target tinkering, this access to like a new method that allows you to discover something about the world in a new way. They need to be able to form dynamic teams that might also just be very project-focused. So for example at DARPA, teams don't come together because there is a professor that has tenure and that's their lab, they come together because they're a particular project that needs to be solved or needs to be addressed. And then they need to be able to raise funding to develop initial infrastructure and then also accelerate the inventions basically with what Ben Reinhardt calls a smooth project size distribution. So you need to be able to grow a project in terms of the funding that it receives from, it is just an idea to this is a research project to this is already outside of the scope of a normal funding organization but maybe still not interesting enough for a venture capitalist towards this is a serious technology company. So then the next question is if we feel like we have answered some of these questions, what makes up an interesting and productive research environment, why do we want to do this online? And this is a bit of a personal story. So this is a graph that shows the movement patterns of inventors and on the y-axis you can see the net migration patterns where if the bar is below the zero line that means this country is churning inventors, inventors leave this country and an inventor is a person in this study that has published a patent and then on the x-axis you have different countries. And I was born on the left-hand side of this graph and then grew up on the right-hand side and then moved back to the left-hand side, got back to the right-hand side and today I'm back on the left-hand side and at some point you start to ask yourself well how many times do I need to move between these two different places and isn't there a lot of talent, a lot of potential inventors who do not have access to the means of invention and cannot be inventors because for them it's so far out of reach the next laboratory is so far out of reach the access to dynamic teams is so far out of reach the access to funding is so far out of reach. And we've lapped out we want to build a home for inventors online so that the location where you are becomes less important with regards to your ability to actually innovate. Which leads us actually to this overall goal that we have in the decentralized science movement the idea that within D-Sci we give everybody the opportunity to raise funds irrespective of where they're located and irrespective of whether it's a basic science funding project or if it has more of an entrepreneurial nature. We want to give everybody the ability to access laboratory services to run experiments no matter where they're based and we want to also build the infrastructure for scientists to share the data that they've collected in ways that reward both the inventor for the invention that they've made but also maintains accessibility for the commons. And when we look at the D-Sci stack today and I tend to think about decentralized science or science in general as this three-step process where you have funding agencies you have execution agencies places where actually the science gets done and then you have distribution mechanisms over the last couple of months really we have seen an explosion in projects like Science Fund, Gitcoin, Vitadao as agencies that can help you raise capital for a scientific idea that you have and then you have a lot of different projects that are targeting the distribution of scientific insight Research Hub, Adams, D-Sci Labs Molecule on the more translational side and Opsi for scientific data and what's really I think missing in the middle is a place where teams of scientists and tools of scientists interact where the funding has been collected and now you want to build your lab you want to build your web lab and you want to have access to particular services to generate new data and then you can distribute that new insight that you've gathered and that's something we're really passionate about at Laptown so the way that we want to provide that insight is by basically billing two, three products that are focused on the team side and the tool side and then an initial funding spark that I won't focus on too much today So how do we bring tools into an online room? How do we bring something like a cell culture bench into an online chat room? That's a really important question when we think about decentralized science and the answer that we have at Laptown is that we create a marketplace protocol where laboratories that are either for-profit or non-profit can share their services and offer their services in a very structured format and then we have team formation processes that are currently bundling under the idea of lab teams where we give these scientists a room to come together and also become discoverable by people that have micro-expertise So let's talk about the lab exchange first The lab exchange is a Dow government peer-to-peer exchange for data generating laboratory services So for example, if I have a small molecule that I've designed computationally the next question really after going through all the in-silico tools that are at my disposal within this group of scientists that I have found within Laptown is to then actually synthesize this small molecule and see what are the properties and test the binding behaviour to a particular target if I'm interested in that and for that I need to have laboratories that actually offer the service of synthesizing a small molecule and testing its behaviour and they need to be available on some kind of exchange So right now, the way that this exchange exists is a protocol is in a very early stage and the way that you interact with it the client is also very much focused on the first user group which are computation biologists that use the client to run in-silico discovery processes and then go through the very first steps of taking something that is only for example, a structural formula or a sequence of amino acids into the physical space so the way that you currently interact with the exchange is through a client that is still pre-release and there are some key functions that you can see here so there are basically two topics right now you can list all the available apps so for example, the Hello World example for the lab exchange is a reverse complement function you give a DNA sequence in you get a DNA sequence out but obviously also more complex stuff where there could be a laboratory on the other end on the exchange for example, folds proteins computationally and then what's important for us is that the data that scientists generate is also shared and at least visible to everybody so it's deeply integrated with IPFS and estuary in this case where from your command line you can upload a research file via estuary and expose it to IPFS and then basically you reference that when you submit a request for an exchange so how do you ensure quality in an exchange where people trade laboratory services there are really three mechanisms that we can make use of the first one is if you have a client so the scientist that's requesting a laboratory service from a provider say a laboratory the flow of capital from the scientist to the laboratory is not direct but the funds are held in escrow and there's an arbitration service which is not happy with the result with the quality in which the laboratory services was completed can basically appeal for an arbitration and the second mechanism is staking and slashing so the lab DAO plans to have a governance token lab that will be used by community members to basically collateralize the transactions that are taking place so for example if there's a transaction that's worth $10,000 such as synthesizing a protein that's doing to work to put something at stake that's proportional to the value of the transaction so if the lab is not doing the work in good faith then the ownership of the lab the membership of the lab within the DAO can be revoked and then as a third billing block not everybody can directly list on the exchange we're working through ways in which the DAO itself can onboard laboratories on the exchange in something that's taken to a token curated registry to ensure quality standards and also just overall compliance with regulators so when I open a request on the lab exchange the way I do this is I change the state of the smart contract and I basically provide some billing blocks that will later go into something that is called the lab NFT where the metadata is relatively simple I say I want to for example here generate a reverse complement and this is a standardized request where the whole DAO is maintaining a repository of different laboratory services that are offered and I say okay this is the user that's my wallet, there are some parameters in this case there are not a lot of parameters and this is my input file which is the FASTA file and then this is posted and once the laboratory claims that transaction says I'm on it and does the job and the additional data fields takes that metadata and mints it into an NFT and sends it to my wallet where the three additional fields are the identity of the provider any type of execution environmental information for example if it was a computational job and what computational environment was it done and then the output itself which in this case would be a FASTA file so just a genetic sequencing file and our hope is that through this open source community we could only maintain the development of new tools but we can also drive the emergence of standardizations around laboratory services overall because otherwise the conflict resolution cost between scientists is going to be so high so the hope is that over time scientists will converge on certain standards for how laboratory services are done such that we eventually end up in a state where something that is one off where we have a one-to-one relationship there is liquidity around the laboratory services and you have many to many relationships and eventually all the past transactions can be created if there were a knowledge graph which there are and you can develop more complex multi-step transactions where you actually plan complete experimental projects and can find someone to run them for you so this is how we plan to provide basically physical laboratory services to online research communities and then on the second side it's not only the tools that matter it's also the teams that matter to actually use them so we need to build tools that enable teams to come together online and form these dynamic teams so the question is how can we facilitate dynamic team formation and what I think is really important is that we have subgroups with subject matter stewards where there is a scientist or an initial project that is leading a group we need to have a forum for conversations where scientists can meet and that guide expert attention and then we need tools in which even an outsider that might just be browsing through the community see something that's interesting and can contribute their micro expertise and there's for everybody that wants to think a bit more about how do we design communities such that expertise is concentrated in a collective way and we have tools for collective intelligence Michael Nielsen has one chapter where he compares Kasparov versus the world and Karpov versus the world both of them were chess games and in both cases the chess master was playing against an online community but the tools that these online communities were using to make decisions around the chess moves are very different and in both cases the chess masters won was the way that the mechanism was designed had subgroups subject matter stewards and ways for discovery of micro expertise another question that we have to ask is how do we actually get scientists to contribute to a doubt this is still something that's very out there if you're an academic scientist you probably haven't heard a lot about what the type of work is that we're doing so we invest a lot of time in building onboarding flows that explain all of the centralized signs of our community and then something that we want is that laptop is the dow of laboratories where each lab starts out as a community of people that focus on a particular problem, focus on a particular project idea or develop and maintain a particular sort of repository but then over time develop their own culture start annotating and keeping track of their contributions among each other and this is just a screenshot of the coordinate graph which is a peer-to-peer system which we track contributions within our labs and then eventually even spin out as an independent sub-dow where this could really be the point where for example future research organizations meet or future biotech companies meet and then eventually branch off and focus on themselves and the project that they want to develop so how do I think we can further develop the d-size stack on this site I would like everybody to just explore new experiments in which we can fund signs I think reputation systems might be really interesting but also dangerous so that we don't repeat mistakes that we have with the current ecosystem where reputation is extremely important and the funding is very centralized and then on the execution side I think we need to just onboard more laboratory capacity and that's something that I'm really excited about and then on the distribution side not only think about the sharing of final research products such as a publication or an IPNFT but also think about how do you share hypotheses so for example an entry in a knowledge graph that ends with a question mark that is a shelling point for scientists to come together and share their opinions and with that I would like to thank you for your attention and look forward to the conversations afterwards