 y same powr structures except on the blockchain. Not any single one of you is going to be the revolution that you are all hailing to be either, but maybe if you work together you could. At the moment it's great because a lot of the ideas that we're having, are being swallowed up by big publishers. Like, you know, keep things private, because Elsevier, Wiley, Telefwilers, they have millions and millions of dollars of resources in Mamper that they can stick behind your ideas and implement them faster. Which is why maybe doing an approach like building it and then talking about it, like Artifacts did, is much smarter than talking about it, waiting for your movement to be co-opted and then wondering why that happened. But the rewards, so I'll try to get a positive note, but if you all work together and identify common standards and communities and processes that you can leverage to your strategic advantage and work in sort of like a step-wise way, then you might actually be able to combat basically the present system. To do this you actually need some sort of actual community governance structure and agreed upon by the people in this room. And ultimately, I guess what you want to do is create something so superior, like using web-nated technology, that we wonder why Elsevier, Wiley, Telefwilers, and journals and articles all existed in the first place, because we're still operating on like 17th century communication norms at the moment and failing to embrace the power of the web completely. People in this room are the ones who agree on how to change that, I feel. So, for me, the ultimate goal of blockchain for science and what you're all doing is to pull knowledge and resources to create a decentralized scholarly infrastructure. And by infrastructure, I mean building something that researchers won't know exists until it starts to break, until it doesn't work. So, for example, when a researcher joins a new institute, they're given a desk, a computer, an email address, a mobile phone, and they're told, like, this is where we publish, this is where your data goes, this is where your code goes, blah blah blah blah blah, and all of that can be backed up onto the blockchain, but they just don't realise it. And it's all taken away from publishers, and community, government, maybe at an institutional level or at a disciplinary level, whatever. But then you can actually adhere to these open science principles that we talked about before. You can have reclusivity, equality, accountability, freedom, fairness, all of these things that this wider open science movement is fighting for, rather than just recreating like a dozen different silos, all trying to compete with each other to be the new blockchain for science open scholarly publishing initiative. So, yeah, like, this is an idea I was going to flip by you. So, what, this is organised by blockchain for science. If you want to run an SDM association, they're basically like a pseudo governing body for the scholarly publishing industry. They dictate the future of the businesses, they have sponsored policy documents. There's a governance structure in place there. Why does blockchain for science become that? Yeah. You know, people have been saying that the open science movement lacks governance, and this is why a lot of the aspects of the wider open science movement are failing, and publishers still retain a lot of control. But why not actually create some sort of governance structure equivalent to what makes the scholarly publishing industry so successful from all of you? One of the people I didn't ask you to like give it like this. No, but I want to surprise you. Why not? Very good. Okay. You know, I think most of you recognize the logo on this, but you can do that. Address the real issue, which we're all trying to solve, and maybe haven't realised yet, which is control and governance of public research. Because like a lot of people have said, your question only looks brilliant by that. It's like, why are the contracts not available for publicly funded institutes with publishers? Because it's bullshit, right? Publishers currently maintain a tax on public education to make sure that their CEOs get new beach houses every year. And that is up to the detriment, to the control and freedom of scholarly research. And you can all help to fix that, I think. Thank you. Well, it was not straight. It was not straight. And I think we can go for beer now, yeah? Everything is there. No, no, very nice. Thank you very much, thank you guys. You didn't make a thesis like this. Yeah, yeah, yeah. Okay, I'll take some questions, but I have a better answer. I would like to, maybe before the questions, to say that the whole reason we're here today is not to have just a one-sided conference or on conference, but really to tackle the issue you just addressed. Is to go out of this conference together and join forces and try and really come up with good solutions and they will not only be technical, we will need to do a lot of marketing for our cause. And we need more transparency for what is actually going wrong because we know this within the scientific community. But a lot of people out there don't know the numbers of how much are we paying, our library paying, et cetera. I just want to add something at this point because it fits so well. So, Lamberd already said this and I discussed this course about these things yesterday. So, this is a slide that I got somewhere, from somewhere, okay? It is a picture that was taken of a slide in 2008 when there was the idea that there could be social networking for science, right? And when social network, there was a studio set in Germany and social networking was to show pictures, you being drunk, opening an internet, or your holiday picture, or you enter the group, the full focal fact didn't fit, sure. Or what's the matter, right? And it was insane to think of the use of social networking for science, right? And then there were people that came up with, well, let's build a social network for science. And people really got into it. And this was an analysis of the competition situation of Research Gate in the year of 2008. And we see, like, how many, we can count them together. It's like low resolution, but they are like 30, 30, even Springer Nature Networks was there. And they had like 10 times the amount of members at Research Gate at this time. You know, Mendeley is there, and what else is there? I like the academia and studio. You know, they're the ones, okay? And now the question is, like, these guys had to centralize and at some point cash out on the attention economy and build a silo in the container, right? And the question that we should ask us today is, is the blockchain economy world different and can be, like, built a system that the more people like working on the system, the more value a token or the system gets, without building a container and a silo on data. Like, the more people work with the serum, the more valuable the serum gets, the easier it gets, okay? If you can, like, have this blockchain economy idea, make working for science, then we have, like, really open science, okay? Okay. Yeah, is that okay? Okay. Now, now, now, John, John's, John's, I mean, both things are, like, open for discussion, yeah? And you, like, stay on the stage, okay? No, no, no, no, no. I mean, like, what do you say to this, like, notion or direction? Yeah. So, I'm really, really angry. I'm really, really angry, because I completely agree with what you said, and that makes me very frustrated. I like to disagree with people, particularly the conclusion, all right? The negativity and the critical aspects, okay, that's a kind of role. I think you are throwing a couple of things out with the buff, okay? So, one is a common misconception of blockchain. So, blockchain is just the tool that you can use. It's just, like, same database, so different, right? And when I use it, and we introduce DATs and the IDFS, and some of those double-buck, and all these other, sort of, protocols, that you can use, cryptography and the internet, if you like, to structure power relations between human beings. Yep. Now, when you structure power relations between human beings, you're doing a very different thing from just moving information around you, okay? That's why crypto-economics and everything comes into it, because you're basically immediately in the world of governance. So, using blockchain doesn't mean you've solved your problem. Oh, yeah. But it does mean that we can do very, very interesting things with precisely the governance structure that you set out in the role. You know, when you look at the open science community, it's like the open-source community, right? They hated blockchain when it turned out. I mean, I was trying to bring these guys together for 10 years, you know? It's like they really don't like each other, but if you bring them together in a convening like this, you can do amazing things when the values are up. So don't throw blockchain out as a really important governance tool. Mm-hmm. Okay. Yeah, it's like challenging us on this view. Yeah? It's just like, yeah, yeah, yeah. This was like... In other words, you can structure the structure that you put on the last slide, which we could have up again rather than this silly cartoon. Yeah, that one. And... This is just really... This one's right in the direct. ...in a sort of common governance structure to preserve the values of open science, okay? But it doesn't only have to be convened as a traditional legal structure like we might have done 10 years ago. Oh, I don't care about the legality. I just care about organisation. We should do it. Because like... Legality protects the soft culture. Look at open science, for example. People describe it as a movement, but like it's a movement without direction. And no one has... And no one has access to the likely governance. Yes. And because they haven't agreed on common standards, common goals, common pathways to get there, and you know, I know blockchain is still new, but these are the things you should be thinking about right from the beginning. That's what I'm talking about. Yes, which is all part of that. The governance dictates that it's... I'm not. I'm saying governance is critical. That's great. That's great. That's great. I think you're both completely agree. I think we're both... I don't see to be out. You agree. You both agree. I think we're both British. I'm just like... He's just took like a... I'm sorry, my point was that governance is critical. Yes. Absolutely. And I believe the lack of governance is why the open science movement, if you want to call it that, is largely failing. Yes, so let's use no and block chain for that. Absolutely. You've figured out right from the beginning. How that should look, which is what my little journal gave us. John, you gave a talk at blockchain for science for a living with talk that you talked about is the vocabulary, and you've taken open science so it's kind of strange for the blockchain to be stranger. I'm just wondering if you made progress yourself on simplifying the message in the vocabulary that you used as a comment. For open science or for blockchain? Over your community. Phew. Around open science. Well, we tried to fix it like four years ago by creating an open science ontology. That sort of worked. You know, it's not something which you just instantly solve overnight. There's only so much communication people can do. Yeah, progress has started to be made. Like I would say that the vast majority of researchers now broadly understand what things like open access are. Some of them might even get a bit of a string green and gold. But you know, I mean, I can even pitch myself in. Like there's still a general lack of training and education in these things, which is one of the reasons why I'm building the open science MOOC to make sure that this sort of education, people understand these critical concepts in research, are embedded right at the core of research and training at the beginning. A little bit of headway. Yeah, love it. Thank you for inspiring. I think one thing that I definitely take away from this is besides the fact that we agree on this ultimate goal of open science, I think that both of them are slightly mentioned. Yes, we should subscribe to this goal and adapt to this thing. I agree with that as well. Besides this, what I take away is that some kind of evangelism. This is the right word to pronounce it. So evangelism around a blockchain for science is maybe not... Yeah, how shall I put it? It's not somehow aligned to this ultimate goal, so to say, yes. But then again, I think that you also do not see it clearly at one or two slides how disruption on a technical level really happens, because when your argument is okay, you overestimate what you can expect from a researcher, then you seem to assume that to somehow subscribe to the concept of blockchain would be the necessary precondition to get something running. And this is not the case. I mean, look around. If you ask researchers, hey, are you an HTML enthusiast? You will hardly get people to say, oh yes, HTML, I'm completely in it. But still, all researchers today are very much users of HTML and producers, users and so on, as we all know, because it's boring. It's a kind of a technology that is built in a level that is hardly recognized anymore. And so saying this through maybe, maybe in one or two years or so, with some blockchain-based solutions. For instance, just imagine for a minute you could have something like research aid, but not as a web application, but as a protocol, yes. I mean, to have this user experience, but without having to rely on this single company and their contracts with Springer and so on. And this is very easy to grasp. If the user experience is otherwise the same as with research aid today, it's not hard to understand and to use it. And just imagine this. And we are, realistically speaking, just one or two years away of doing this. Yeah. I mean, again, I probably explain it better, but this is sort of what I meant by the scholarly infrastructure part. So the definition of infrastructure I use is Jeffrey Builders, where he says, you don't notice infrastructure until it stops working, like roads. And yes, so I fully agree with you that basically building it without them knowing is without researchers knowing. Because like then, you know, it's almost like, imagine like an analogy with the big deal, for example. People don't realize that they, the libraries pay millions and millions in subscription fees until for some reason they can't access the journals anymore. And then all of a sudden they're like, wait a minute, it's not working. Do you have the same thing? But you know, with the entire research process and just not with LCB. And Sid, do you want to write, we are maybe overselling it? Yes, it's true. I think it's totally fine, but like, you know, the risk is alienation. So when open access took off, you know, in Europe, five or six years ago, those who were open access evangelists got basically marginalized and threatened for it. Like Ross Mounds, for example, got blacklisted from the paleontology community for being too vocal in support of open access. It doesn't matter how right he was, he was just pissing off the wrong people. And I feel, if you go out waving your blockchain revolutionized science flags, people are just going to be put more off than they are attracted to it. No matter how well you describe how it's going to improve their lives and their research process. On that point. Can I go now? Thank you very much. And the research is also about research data. We have to switch from this big picture of you now to a very important challenge. Of course research is about data and James will talk a few minutes, right? Like how long, yeah, about like that, like how we share data, research data, maybe in a decentralized way and where we are, okay? And where we are heading to and what the challenges are at the moment. And then the, is it this one? And the, yeah. Okay, excellent. And then we'll have another coffee break. Okay, thank you very much. Thank you very much for that introduction. I'm going to be super quick. So I don't normally talk on this topic. There's two people called Dennis Burnoff and Owen O Carrigan, who we've done like a series of workshops of the last six months and they just can't be here today. I don't like good resilient peer-to-peer networks. Some peer stands up and fills the gap. So, the key thing of these workshops that were on conference last month in London was education, as John said. So if you didn't know I'd be a bass, that project, solid fare, we were providing a supportive infrastructure to get you going on these topics and then we're just at a stage where we're evolving some use cases, which we'll talk about. So, I wish I had your slide, John, because I was too lazy to go and map out our ecosystem, but this is the current one. And the thing about going to these events, I think myself and John has been there for three years now, is that it was at lunchtime with the Irish speaker today and last night we were currently at dinner. We've got all these tokens, we've got all these consensus mechanisms and all these new magic, pure view mechanisms. How are we going to get ourselves governed to do it? So our gambler for this Damhub project is let's do this for data and consensus around data structures and build up to these more exotic consensus mechanisms. So the first thought we should have is that anywhere projects have data, go and reach out to the other people who have got similar data and see whether you use the same standard, different standard, it's a different standard than how would you describe it. But the encouraging thing for me when I found out about the Damhub project was that Owen is a librarian at Cork University and he said, oh we've got these spare objects in open science and he's a big fan of semantic web, solid. And I said, yeah I've looked at all those but I don't think they're fit for the 21st century. He goes, well what else would you use? And I said, well we could use computational reference layers or computational scaffolding. He goes, I've never heard of those before. Yeah, you're right, I've just made them up in the last few years. So we're going to talk about the existing metadata world and also this new computational world that I foresee. So in London last month David talked about being open and transparent, I think six of you were at the event too. So you can, Owen goes into detail about what FAIR is and Matt from IPFS about these links being content addressable rather than URL addressable and all the videos are currently there. But our key focus right now is actually doing stuff. And so the first use case is can we take these existing metadata objects and have consensus over using them? So what is that consensus mechanism is still an open thing but I think it's a theme of these two conferences to build on that, to find them. And we are actually working every week to put forward ideas and so we'll share those. But ultimately thinking has to become deployed so we actually have a live test network at the moment. And so we're looking for other people to join that and bring their metadata or computational reference layers to the table. But I think you've heard this theme all day. It's that we've got David's minimum viable governance for a data standard. We've got Lambert talking about all these micro small data sets and all those need their own consensus mechanism. And we have to be better as John said. Not just being technically superior. It's actually more convenient in your day-to-day life as a researcher or in my case as a citizen scientist. So we've visited lots of data protocols sitting on this paradigm and it works for the old metadata ones and it also works for computational reference layers. So just to walk the talk, I have a citizen science heart rate project based on data coming from variables. So I've already set up our demo hub hub with our IPFS node behind it, it can be that. And every time we author our data to the network or scientific computations that are machine learning, you will see a list of all our work. So I encourage everyone to do that too. Right, the new stuff. So I'm not going to go through this today, I think we can talk about it tomorrow, but on this link here I've sort of thought, right, we're five years into the future, let's look back. And so that here is we have these science data chains and they allow us as a community to have consensus over the data portability and the description of the data. So regardless of our project, our peer review system, our token, we've all agreed that our data can be because data is the new order, right? So what's the difference between metadata and a computational reference layer? So our main use case for that is a thing called chemistry. So existing metadata standard for chemistry is driven by a non-profit called the Chemistry Institute of America, the CAS number. So we take caffeine. Caffeine can be marked up as a semantic language object, but in CAS's case it's just an abstract number that they've chosen for their database. And then you have to go and look up the subsequent science and experiments to get information about the structure of the molecule. So already in CAS we have things like an Inchi code and an Inchi code says, well, given the understanding of the laws of physics, we take the 3D structure of caffeine and we give out a cryptographic hash based on its 3D actual structure. And so that becomes the unique identity part. Well, the great thing about that, given there's one laws of physics, there is only one identity part for caffeine and so you simplify the metadata problem. So a chem chain in my project I'm a heart chain but that requires cells, cells requires DNA and eventually a cardiovascular chain. So all these systems will work together and tomorrow we can go through the demo. I'm not going to go through this in detail either but this is what our current thinking is for chem chains. We have all these users at the top, currently 200 million revenue, 130 million unique molecules in doing existence. And we're basically saying, let's make it a 20 million revenue market or a 2 million revenue market and we have this notion of a query exchanges and so we have Bitcoin exchanges but in the crypto storage mechanisms, you have things like Maze safe, foreign mega and you have IPFSs, space time concept and you also have all the debt project which is a pure reciprocal. So that idea here is that you can behind these storage circles say if you had unlimited capacity as a former on Maze safe, you would have the whole of the internet on your note, right? So they built into their mechanism a limit which may be 20 terabets and maybe I gave you whatever. So you can envisage that the query exchange, anyone is free to deliver the whole index of chemistry but you'll only get more in for say 5% of it or whatever the consensus mechanism is of the community. And John talked about blockchain but like John, it's a very broad route. This doesn't use blockchain at all in the sense of people work aluminium. Well it does in one sense is that I've proposed to these guardian networks, we use it as the last resort of imitability. But basically we have a model for saying that anyone is free to propose a new structure based on this inchy or inchy plus plus. And we've just said, when three other nodes in the network verify that you've made your molecule structure in accordance with this new way you format then you're up for grabs to be a candidate to the wider science community to provide evidence that this molecule actually exists. Of course that will come from all these cases because it's not a real chemical they will be using it. And so then we're going to harness all the data analytics to make a statement on whether the molecule actually really exists and what the community says about its properties. Yeah, so a very simple message. Community, let's start with micro little data projects and build up these more grand things. Start with the things we understand from the open science movement but I think there is potential on these computational ideas which we'll talk about later. But ultimately it's about governance and humans should be the centre of that governance then in our conclusion. And then we can put automation in. Here's what we've been doing as a community and you're all free to join us on a weekly basis. Thank you. Yes, this might be a stupid question but how do you hash any three dimensional continuous structure? You said it's basically a physical loss but it depends on the pressure and what kind of structure might decide the area. One scientist measures it that way and the other one measures it that way. You have measurement uncertainty. How is this going to be exactly the result? So this doesn't exist today. The closest we've come as a community and I've just joined the project is Inchi codes but you're right. For polymers and more complex molecules there's clashes and it fails. My back is saying you don't have the best cryptographers from the DAP project from IPFS made safe and I think it's possible now. And I'm just saying we might not be perfect but if you get closer to that it's far better than some abstract monopoly on index number that's just picked out like a core registration number. But the big idea behind these computational reference numbers today with the proof of work algorithm. The proof of work algorithm is securing the ledger. Inchi can imagine as we've built all our peer review and AI systems on top of these data protocols that's computational proof that these things are really being used in the world unless we have the compute power world going for a scientific purpose around the abstract notion of securing artificial scarce money. I realise this was a very specific lecture that's already taught me words in my head. If I'm a big chemical company I had the same sort of issues in medicine like AI systems for diagnosis. If I'm a big chemical company and I want to use some non-increaser for chemical entities the reason that I would choose what's the name of the big company the not-for-profit CAS system that has a large turnover is because I consume not because of the reliability in the data. So if it was reliability in the data we could probably use weekly data towards us in the ground. So how, I mean, that's the sort of legal advice you get if you're born in a large company. Choose this entity that will back up this statement because if it goes wrong and you kill someone you can see that. How does blockchain or any of this help back? So maybe we told the backstrap to the chemchains. A chemchains were a thought experiment so where did it come from? It came from the Berlin on conference that we ran in November was the last year. So Marcus Rusyn who works for Elsevue, right? I didn't know at the time he was talking about Inchi codes and I said to him, what are you describing here? Because I claim that all science should be in computational form. And so part of his day-to-day job at Elsevue is going around all this industry. And so they've come back and said if you can give us some legal framework that shows that the community will own this we would be more than happy for the community to own it. But what that legal entity is in structure we're continually working every week to find the solution that works for you. But it seems to be of academia, corporate land or the open science. The corporate seem to be the ones that want to liberate the world as much for economic reasons. So but it's early days but we are looking at finding a mechanism that if they did put the resources behind it it can be taken away if the project got taken over like we've seen with other startups. But just a comment on that thing the lawyers always say we want someone to sue that's what they always say. But the trouble is again it hardly ever happens if someone gets sued. But also that's what you have insurance for as well I mean. So it's just a bit of a straw man. I haven't received anything to sue right now. What's that? Recent idiot. No I would say it never happens I said it hardly ever happens but the amount of things that don't go forward because I don't know how much you know about going to court or anything but it's incredibly expensive you never get back what it's doing it's often used as a weapon than anything else but if you've got a good insurance company and it's insurance yeah I just you know there's a small risk there is a real risk so that's why you have insurance. And on that point one of the key applications of blockchain is to provide community insurance isn't that sort of useful? But that's why if it's properly decentralized then there is no one to sue but the trouble is a lot of people have a company or a structure there so yes you've got people to go after and that's one of the reasons why Bitcoin was successful because other people tried stuff before and some of them ended up in prison so there was no one to go after with it so that's the other way of really stepping back because yeah you might make a bit more money but you're the ones that are in the garden basically It's not an interesting question You don't have to take it on a Bitcoin just to have a different direction No I'm just saying to really predict what happens Yeah so going strictly legal open data for most insurance policy but the community government no one is making too much money yet that's the way it's going but real governance and taking all serious advice Perfect So we'll have a break now for sure I just want to ask you one question first an announcement so a quarter to seven we'll meet here and go together to the Luftburg for the dinner we have a reservation there it's at the Prater it's a different place from yesterday so a quarter to seven we'll go from here together or we'll meet at the Luftburg I put the link in the Twitter ground channel and on the web page and now I have a question we'll have a break now for sure and Alex and Shanguin Shanguin from Pluto and Alex from DEIP agreed to give their presentation tomorrow early on if everybody is fresh or we can have them tonight and we have like two other talks so the question is whether we have like two more talks or four more talks what do you think like who wants to have two more talks who wants to have four more talks okay you'll talk tomorrow when everybody is fresh okay great so we have like is this a Luftburg with the... it's 10 minutes it's 10 minutes walk from here it's not far okay now we have like a 15 minute break then we have two more talks and then we'll meet at the Luftburg where we're here okay okay excellent