 Welcome to today's webinar, Unlocking the Potential of Decentralized Storage. Today's workshop will break into a conversation with two leaders in the space of decentralized storage. We'll talk about the benefits of decentralized storage before diving into Filecoin, the world's largest decentralized storage network. We'll discuss real-world tools and practical applications, explore and implement decentralized storage solutions, the event supported by an award from the Filecoin Foundation for the Decentralized Web. My name is Billy Bicket and I'll be your event host. Why we're here, who we are, we're a technology capacity building organization called TechSoup based in San Francisco. We've been doing this work since 1987 and essentially we're a global network that bridges technology solutions and services for public benefit. I want to just give you a couple of breaths on the goal of this learning series for those of you who are new. This is an ongoing learning series about the decentralized web and the potential it has for civil society. Our goal of this series is to create the conditions that enable the builders and makers of public good technology and those civil society leaders out there to harness emerging tools for the public good to do it early, effectively and responsibly. And so we host these webinars pretty much on a weekly basis. We have office hours. We have a number of different activities to support anyone interested in digging into the web potential. In terms of our community, just a couple of notes on how we think about community for those of you who are new here, those of you who've returned. You've heard us say this before but we feel it's pretty important to just make it clear that we're a super inclusive community. We welcome everyone. We're here to support each other. We're really focused on building more agile and resilient mission-driven nonprofits. And technology is just one of our tools. You're all invited to participate in chat. At the end of the session you'll have an opportunity to ask questions of the panelists and we encourage you to ask as many questions as you like. For those of you who want to go deeper and do a deep dive, there'll be an opportunity to do that at future sessions as well. And then finally, we treat each other with kindness and respect. So please consider that guideline as a guiding principle. Without further ado, let me just introduce our expert guests here and then I'm going to pass the mic to Hunter who's going to kick us off. But first let me introduce Ian. Ian Davis is a social impact engineer who works at Filecoin Foundation and Filecoin for the Decentralized Web. He's a platform engineer with over a decade of experience building and managing distributed infrastructure. He's currently on staff at the Filecoin Foundation where he leverages his expertise in decentralized identity and data, storage solutions to advise and assist social impact projects in meeting their missions. Hunter Trusseter is the head of global social impact programs at the Filecoin Foundation for the Decentralized Web where he leads a team focused on building a more equitable internet. He came to Filecoin Foundation in July of 21. After a 15-year career as a diplomat in the State Department's Foreign Service, Hunter's final assignment in the State Department was as the acting director for Senate Affairs, serving as the senior most career official leading the department's engagement with the U.S. Senate. He served in Asia, Europe and the Middle East with a particular focus on building global programs addressing online harms and human rights issues. With that I want to pass the mic to Hunter. Thank you Hunter for joining us and thank you for your ongoing support. Thanks Billy, I appreciate that. First let me break out of my script here and just flag for folks because I just put these on. If you see my hand crest above the camera and you're wondering what are those things on his hands, my wife is from Wisconsin but I'm from California so we keep the house at her temperature and my fragile hands can't handle the cold and I was actually lost feeling in my hands as I was trying to get ready for this so I had these little special mittens that I put on so that's what those are just in case you're wondering. But again thanks Billy and good morning everyone. So I've been looking at the chats and just thrilled to see such a diverse group of community leaders here with us today. Now before I dive into the substance I'd like to lay out the structure of this call so that you all know what you've gotten yourselves into this morning. First as Billy mentioned you've got two speakers here today myself and then my colleague Ian Davis. I'll be covering things at a higher level while Ian will drill down into specifics including providing a demo of how Filecoin distributed storage works. For my higher level portion of the call I'll introduce myself and then I'll give you an overview of the two decentralized technology foundations that Ian and I represent as well as a brief description of the work that our team conducts. From there we'll dive into the substance. I'll talk a bit about what is this thing that we call decentralized technologies then I'll go into more specific details about the benefits of decentralized storage specifically. After that I'll turn it over to Ian for his remarks and the demo. Without further ado my name is Hunter Chessiter. As Billy mentioned I'm the head of Global Social Impact Programs at Filecoin Foundation and Filecoin Foundation for the Decentralized Web. Because that's a bit of a mouthful we'll be referring to Filecoin Foundation as FF and Filecoin Foundation for the Decentralized Web as FFTW. We often combine the two and just refer to them jointly as FF slash FFTW. FF focuses specifically on the storage protocol Filecoin including the governance of that ecosystem. FFTW on the other hand seeks to support and empower all open decentralized technologies. Think of FFTW as trying to be the tide that lifts all D-Web ships all of my colleagues work for both organizations. Speaking of colleagues and work let me rewind slightly to finish my personal intro. I oversee the social impact programs at FF and FFTW. I've been with FF slash FFTW for a little over two years. As Billy mentioned prior to this role I was a career diplomat in the U.S. Foreign Service for 15 years. During that time I focused on online hards and human rents which is what led me to this particular foundation. I live in Reno Nevada with my wife who I mentioned is from Wisconsin and is much more resilient when it comes to the cold than I am because I'm a native California. We also live with our two Alaskan Malamute dogs and a large cat that has been very unhappy on a diet for the last couple of years. So it's now the medium-sized cat. In my non-workout hours I'm a volunteer first responder for my camp. On the social impact team we fill a somewhat unique niche in our ecosystem. While the majority of our colleagues work most closely with other D-Web and Web3 professionals, social impact focuses on users who are not naturally or organically drawn into this corner of the technology space. In effect our role is to expand the tent to include new organizations and users. This is part of the reason why I'm so excited to meet with you all today. I love talking about the benefits of distributed technologies and how they can provide alternatives to big technology companies while ideally upending some of the flawed systems of the current Web. Now with those wavy words alternatives to big tech and upending flawed systems let's dive into the meat of it. To begin with I think it's important for me to define what I mean by decentralized technologies. Heavy emphasis here on the eye part of that sentence since different people have different and entirely valid definitions for these terms. For me decentralized technologies meet three characteristics. A they are permissionless. This means that there is no central authority that decides who gets to access the technology and its services. If you want to use the technology you can without having to seek permission from a usually faceless entity. If you'd like to stop using the technology you also can without having that same faceless entity chasing you to keep its ad dollars rolling in. B they are trustless. While a term of art in the tech space I personally think the term trustless can be confusing. For those unfamiliar it refers to a system in which there is no need to rely on a third party for trust. This means that interactions are handled directly between two or more parties rather than through an intermediary which is responsible for verifying the data. This is important because in traditional technology the intermediaries are all too often companies that seek to control and monetize the interactions and see they are open. This means that users can peer under the hood to see how the technology works while also being able to control how they interact with the technology. Would you like to build an extension that improves how you engage with the technology? Have at it. Would you like to explore the code to better understand how the project works? Here's our repo on GitHub. Those are my cornerstones for what defines decentralized technology. They are permissionless, trustless, and open. So then why is decentralized technology something that you should be interested in? There are many benefits to decentralized technologies. I'll share a few of them along with examples of FF and FFTW projects that are reaping these particular benefits. Now bear with me, please, as I focus on my foundation's projects. It's less about being self-centered and more about not wanting to talk about other people's work for them. So the first benefit, you control your data. A single company is not unilaterally responsible for your data. A lack of competition and control over a massively popular platform does not empower an individual tech company to determine which messages receive which audiences. This centralized control is a real threat to organization's data. For example, not long ago, a Rohingya group had been using YouTube as its video archiving solution. YouTube received a takedown request and subsequently deleted the Rohingya group's account. The group irretrievably lost years worth of video and records because it had not realized the need to archive its data redundantly. One of FFTW's goals with projects focusing on archiving communities' data is to ensure that they are never vulnerable to this type of data loss. We see Filecoin as an ideal way to ensure that you maintain control over your vital archives and do not risk losing everything through a single deep platform. A second benefit is traceability and immutability. For those unfamiliar, many decentralized technologies are built around blockchain protocols. These blockchains operate as distributed ledgers where records of transactions are stored across a network of computer nodes hosted throughout the community. Attempts to change a record can be easily spotted and rejected. Additionally, these ledgers are publicly available, meaning that trust in their veracity is extremely high. How could you use this traceability and immutability to further your work if you're in the business of, say, recording and exposing atrocities committed by governments or security services? It can be an ideal solution. In one instance, our project partner Starlin Labs worked with Hollis Systems and the Atlantic Council to submit evidence of Russian war crimes in Ukraine to the International Criminal Court. The ICC's bar for attacking evidence is understandably quite high. With digital evidence, there is always concern that the data may have been altered. By leveraging the traceability and immutability of blockchain technology, Starlin Labs was able to submit a novel cryptographic dossier of evidence to the ICC. And we've seen similar successes with projects focused on police violence in Latin America, potential war crimes in Syria, and a government overreach right here in North America. A third benefit of decentralized technology is their redundancy and resiliency. Archiving your data on distributed storage ensures that there is no single point of failure. Take, for example, our collaboration with Internet Archive on its Democracies Library initiative. Democracies Library is a project to archive open data so that it is freely and easily accessible by everyone. Sorry, I should have said archive open government data. FF slash FFTW is working with IA to ensure that these open government data sets are stored redundantly and resiliently on both IA servers as well as distributed storage providers. To ensure this ensures that backup archives of the Democracies Library data are protected on servers in multiple locations around the globe. Building on the theme of making government data openly accessible, here's a fun fact that I learned from one of our project partners, University of Maryland. Most geospatial data is open data. This is because most geospatial data is collected using taxpayer-funded government resources such as satellites. If most of this data is open, why is most of it also so difficult to access? The answer is that governments may label the data as open, but these same governments do not often have the expertise or resources to build easy to use repositories to access this data. And this is where private or profit companies come in. They slap a watermark in the geospatial data and then host it behind a paywall. Now, while legal, this is certainly against the spirit of open public data, but these companies are simply following the dictates of the centralized internet incentive structure. Decentralized technologies have the ability to upend these incentive structures. For example, smart contracts can be used to automate payments and revenue sharing to an artist. Now, regardless of the form, these incentive structures can be designed to scale without the centralized control that we so often see from traditional tech companies. In the case of our collaboration with University of Maryland, we're able to build an easy to navigate platform that hosts the type of geospatial data that was all too often previously hidden behind paywalls. We're able to populate the platform with new data sets through an active collaboration with the geospatial data community. Speaking of the power of leveraging community data, another of our collaborations is with the Human Rights Data Analysis Group, known as HRDAG. HRDAG works with community groups, truth commissions, and others to bring data about human rights abuses into the light. HRDAG obtains this data from a variety of sources, including community repositories. HRDAG then pushes for the data sets being made publicly available to ensure that the truth is spread as widely and as openly as possible. This effort benefits from the resiliency of distributed storage technologies. By archiving its data with distributed storage providers, HRDAG ensures that its data is redundantly protected from loss. I'll now turn it over to my colleague Ian Davis. Ian is going to demonstrate how you might replicate HRDAG's use of distributed storage to make your data openly and resiliently available to the public. Thank you. Over to you, Ian. Great. Before I get into exactly how HRDAG uploaded their data, I'm going to give a little bit of a background on how decentralized storage works and how it's different than traditional storage. This should hopefully be accessible to non-technical folks, but if there's anyone technical in the audience who has deeper questions, feel free to keep them in mind, ask them in the chat. I'll be happy to answer them once we're done. So how is the decentralized web or how is the decentralized web different than the centralized web? It's an alternative network architecture for the web that relies on peer-to-peer connections between different computers rather than all consumers of the data connecting to one centralized location that manages a bunch of computers behind the scenes, like Google does. Instead of connecting to a single server identified by a URL, clients join the network and request content using what's known as a content identifier or CID. And the content identifier is really what makes decentralized storage networks unique. So how content identifier works is if I have this picture of a cat, we take at a code level, we take the picture of a cat, we look at the data, and we create a unique identifier that's unique to that data. So if anything changes, let's say someone goes in with Photoshop and is, you know what? I want this cat to have red devil eyes instead. That's going to change the content identifier, and it's going to reveal to the network that this is actually a different image, and you'll still be able to fetch the original image of the cat using the original content identifier. And this is what Hunter was referring to earlier as trustless. You don't have to trust that when you request cat.jpeg from the network, you're getting this exact picture of a cat because instead of requesting something by name, you're requesting it by this identifier that you know is guaranteed to only change if the file itself has changed. And when you request this identifier, the file might actually be stored on a variety of different nodes in the network. So there's something called the gossip protocol. Engineers love giving real-world names to technical concepts. So all of these computers are gossiping with each other like a knitting circle, and they're all asking each other, do you know where I can find this cat picture that's identified by this content identifier that I've just been given? And one node might say, I don't have it, but I know this node over here has it, and then you connect directly with the node that has the file, and you can download it onto your own node. Now, this is how the system works behind the scenes. Most people, when they're interacting with the decentralized web, probably aren't running their own nodes. They're interacting with a provider like Web 3 Storage, which we'll go into more detail later, who interacts with the network on their behalf, uploads data for them, and then they can access it through something called an IPFS gateway, which we'll go into more detail about during the demo. So you request this file, you connect to a node that has it, you get delivered the content back. Where does Filecoin come in? IPFS is just a decentralized storage network. It doesn't, by default, people are just storing files on IPFS nodes out of the goodness of their heart or through some sort of paid service to host these files. What Filecoin does is it adds an on-chain storage, basically a blockchain, to track storage contracts for long-term archiving of these files. So if you want to make sure that a node keeps a copy of a file for an extended period of time, you can use Filecoin. And what the sort of blockchain network does is it has a proof system that every 24 hours it verifies that your file is still provably being stored by one of the nodes. And this means that indefinitely you will always, you can ensure that the content is being archived by this network. Anytime somebody asks for a CID, anytime someone asks for that cap picture, they're going to be able to get it back. And the cap picture example is obviously a fun example to use in explaining this, but as we go into the demo, HRDAG has really used this to archive some of the most important human rights data that they have. If we're looking here and I've got a couple of windows here because as we go into the demo, I'll need to be putting some data into my window here. But if we look here at, this is a blog post by HRDAG that's talking about some of the data sets that they've collected. So here we have both DUF and centralized servers. So they're hosting it on their servers the same way they do with the rest of their data, but as part of a partnership with us, they also created this IPFS decentralized web version. And so if you look at, this is actually a different website called an IPFS gateway that's just giving insight into the network. And later on, we'll go into more detail, but there are multiple IPFS gateways. You can choose which gateway provider you want to use. All of them will give you access to the same content because the content is identified by this identifier here known as the CID. So to start off, we're going to take this PDF of the results of the Colombian Truth Commission as part of this dataset. And we're going to go over to the Web3 storage service. This is at Web3 storage or Web3.storage is the URL. I've already created an account. You can actually see I already uploaded one file earlier. The file I uploaded earlier is actually just a small picture that I use to identify myself on online forums. One of the important things to notice here is this storage providers line though. If you look at, this is actually redundantly stored with eight different providers right now. As I upload a file, it takes a little bit of time for Web3 storage to make all of those storage contracts to ensure that the file gets persisted long-term. So we won't see these storage providers listed yet on the data I'm going to be uploading today, but I wanted to make sure that I showed what it would look like in a few weeks once all of those storage contracts are made and the file is persistently replicated. And so now if I go to upload files, it's a simple drag and drop interface. And so over here in my UI or in my file browser, I've already downloaded a version of the Colombian Truth Commission data. And I've got this results paper here. So I'm just going to drag it over here. As you can see, it's now calculating the content identifier and uploading the data. It's all complete now. So if I go back here, I'll see in this files list, okay, great. I've got it. As I mentioned earlier, the storage providers are still, they're still making some of these contracts. It looks like the Filecoin deals will be active actually within 48 hours. So the timeframe that I mentioned earlier is actually much faster than I thought it was going to be. And if I click here, it'll take me to, like I said, a version of the IPFS gateway that's hosting this file. And if I click on it, it's very easy to see the file again. You can use these IPFS gateway links on your website just like HRDAG did. Nobody has to download any IPFS software. They can just interact with these gateways the same way they normally would and trust because the content identifier is the same as the one that you identify, that you uploaded, that the content they're requesting from these third-party gateways is actually the same content as what you uploaded. But sometimes you want to upload more than just a single file. So now I'm going to go to my terminal and I'm going to use something called, something that the WebFeed storage folks make, which is a UI. And if I look at this HRDAG folder here, you can see I have actually all of the data that was available in that HRDAG blog post in one folder. And I'm going to do just W3 up and then the folder name. So it's a very easy command. It's now taking all 43 of those files and it's going to put them into one content-adjustable archives, put them onto IPFS, and then eventually back them up on Filecoin. We're going to have to wait a little bit of time for it to do it. Oh, it was actually a little faster than I thought it was going to be. And now here's one of the gateway links. How these links work is each gateway provider changes here. So sometimes you have W3S.link, other providers are dweb.link or ipfs.io. And then you have slash IPFS. And then you have what I referred to earlier is the content identifier. And so if we click on this link, we can now see all of the data that was just on my laptop is now published on the decentralized Web on IPFS. Over time, Web3 Storage will, as I said, create those Filecoin deals to make sure that this is persistently stored. One of the things that's great about this is HRDAG uploaded this data and I took the exact same folder as them, generated a content archive. It ends up and I'm just re-uploading the same data. So I'm actually not uploading a different copy of the data. I'm uploading it. It generates the same CID and I'm just personally choosing to host a redundant copy of this data now because I think that it's very important data and I want to make sure that this data stays accessible. So in addition to all of the providers that HRDAG has worked with to make sure their data is replicated, I as an individual person without changing the CID can choose to replicate the same data as well to make sure that it stays available for long term. And if we click into one of these folders, we can see, looks like it's having trouble loading. There we go. We can see that I've also got the same PDF of the Columbia Truth Commission that I had before. And sometimes the first couple of times you load content, it'll take a little bit longer using the system because it has to do that gossip protocol. It has to go ask all of the other nodes where it can find these and download it before it can give these to you. But once you go through that initial request, it's going to be very fast because this gateway is going to keep a local cached copy of the data. And so that's the crux of the demo, but I want to show you two more small features or tools that the WebView Storage team makes that you might want to use to take this even further. Since I know there are some makers in the audience, I wanted to mention NFT.storage. NFT.storage is the same underlying technology. They don't actually make NFTs, but what they do is they provide you persistent storage for whatever data might go into your NFT. So if I wanted to make an NFT of that green photo of myself that I shared earlier at the beginning, I could very easily use this service to upload some data, get one of those persistent IPFS URLs, and then put that into the NFT. One of the things with NFTs is you want to make sure that the NFT represents something immutably. And so IPFS, because of the immutability of content identifiers, has emerged as the best way to store data long-term for NFTs. And then if you're very advanced, the WebView Storage team actually releases, as Hunter mentioned, it's an open ecosystem. So they release everything that they use to build applications on top of WebView Storage as a set of component libraries. So if you wanted to build the ability to upload data to the decentralized Web directly into your website, or you wanted to build a custom tool that was provided a unique interface for your workflow. For example, we've been talking to Starling Labs about their workflow managing evidence and managing web captures. And so if they wanted to build a tool that was uniquely suited for storing web captures, they could take some of these UI components from WebView Storage. They could build their own custom application hosted at their own website and domain that still was able to use all of the same easy uploading and easy storage technologies that are at the core of WebView Storage, IPFS, and FilePoint. And so with that, I'll stop sharing and open it up to questions. Thanks, Ian. Real quick before we dive into questions, one thing that I want to touch on, it's not really a question, but I think it's an important detail to note about how our Foundation approaches this work is that as excellent as Ian is at presenting these details, there's absolutely no expectation that coming out of a demo like that that anyone on this call would know how to just dive in and use the technology. This was a demonstration on the efficacy of it rather than the practical use of it. And so we want you to know that literally our jobs is that we are here to help. If you are interested in this, if you have any questions, please don't be shy about reaching out. We would love to help walk you through how you might utilize these technologies. With that, Billy, I'll turn it over to you. I think that we've got some questions. Great. John asks, what assures IPFS are good actors hosting your data that that might be personally identifiable information? I'm sorry, Billy, would you mind reading that again? My audio glitched for a second. Yeah, no problems. John asks, what assures IPFS are good actors hosting your data that might be personally identifiable information? I think the question is there, how do we ensure that we're protecting PII? Billy, is that your read of it as well? Yes. Yeah. And also, how do we ensure that the data that's being stored is being stored with good actors? One of the wonderful things about decentralized technologies is that they give you choice. You are not really bound into a system where an easy example is Twitter versus Macedon. If you are sorry, X versus Macedon. So if you're just using, if you're using X, you really have to only, you really only have to follow the rules that are at this point a little bit fickle that are being defined by that one technology companies. Whereas if you're on Macedon, which is an open decentralized alternative to X slash Twitter, then you have the ability to decide which particular space within Macedon you would like to use as your social network. And so you have a number of different alternatives that define their approach to social networking in different ways. In the same way that if you're using distributed storage, you get to define your requirements for the storage providers and you get to go out there and ensure that the storage providers that you're working with meet your particular requirements. Now, as far as it applies to PII, so you might say I want storage providers that are going to be respectful about my data in X way or storing it only in these geographic areas, which is something that we see that the geographic locations is something that we see with people who are storing government data. Sometimes they don't want that government data to be stored. Even though it's open government data, they sometimes don't want it to be stored within certain jurisdictions that they might consider to be problematic. Now, when it comes to PI, the default for FilePoint and for many types of open distributed storage is for open data. They're in the name, right? Open distributed data. PI can be a particularly difficult thing to store in that environment. But for people that are looking to protect things like PI, the default is encryption. So what you do is you would encrypt your data before it goes into the storage platform. And there are a number of different ways to go about that. I'll turn it to Ian to talk about some of the different ways that we've seen people approach the encryption of data, but that really is the solution when you're working with things like PII that you certainly don't want to make openly publicly available. Yeah, like Hunter said, this network is really designed for persistent, long-term storage of open data sets. HRDAG actually puts a lot of work into making sure that they remove any personally identified information from their data sets before they publish them on the decentralized web, because they want to make sure that the public results of their data are persistent for the long term, but that individual information about people that can be used to dox them or retaliate against them isn't made public as part of that sharing. There are, like Hunter said, active research groups that are looking into encryption. One of the services, there's a group called fission.codes that actually runs a Google Drive-like service that makes heavy use of encryption and actually has a way of ensuring that even they, as your data storage provider, cannot see your data. In fact, nobody can. They create a set of keys for your account and only you have access to the data, you can potentially share it with other people, but they and nobody else on the network who doesn't have access to your keys can decrypt and read any of your personal information. So that's one solution. We've also seen some people starting to experiment with what are called private IPFS networks. So by default, the network I just demonstrated is a global public network. So anything that is posted onto it is freely accessible to be downloaded by anyone. Encryption means that if somebody tries to download my data, they won't be able to read it, but they'll still be able to download it from the public network. So people who don't think that is enough have started doing private IPFS networks. So the data is only shared on a network that's accessible to, for example, me, my coworkers and my friends. And additionally, on top of that, it's encrypted. One of my friends might be able to download an encrypted blob of my data, but they're my friends. So I'm going to trust that they're not going to try to break that encryption. And yeah, so that's really where those are some of the solutions people use for storing private information. But again, like Hunter said at the beginning, we really recommend people mostly focus on the use cases for public data, public and open data, because that's really what the technology is best, most used for and most useful for. Great. We've got another. Yes. Billy, if you don't mind if I could prioritize a question here, because I think that it segues so naturally into the question you'd posted, which is what sector of civil society that the types of orgs have the most to gain and the least to use and using decentralized storage based on your experience. If you don't mind, if I could jump to that one, please. I think it's just such a natural segue. So really for me, the heart of that question is what types of organizations will benefit the most from distributed storage, like FilePoint and building off of what Ian and I have just said. For us, oftentimes, it really comes down to organizations which are in the business of making large data sets openly and publicly available. I'll reference an organization that we've been working with that we didn't mention during our presentations, which is an organization that works with community groups, largely minority and frankly, oftentimes disempowered or disenfranchised community groups throughout the country that have been struggling to make their community archives resilient. And these community archives are oftentimes posted in churches because these not only but oftentimes posted in churches because these churches are the nerve center, the heart and soul of these different communities. And so what has happened is that when these archives are stored in a single physical location, those archives are very vulnerable. And unfortunately, in some cases, these buildings have suffered damage. It could be flooding damage. It could be fire damage. And these groups which rely on their cultural heritage have lost access to those archives. And so what we're doing is working with this group to, for them to then make digital copies of as much of the data as they can for these community groups and then upload that into FilePoint storage. Now, this is not intended to replace the local community repositories. Those are vitally important to these communities. What it does is it ensures that they are resiliently protected and they are very biased towards making this information openly accessible, not only to their communities, but to anyone who is curious about learning about their community. So it really comes back to those groups which are in the business are focused on making information openly and publicly available. And that's of course, along with the other examples that we've used, like the Internet Archives of the World, which all they're trying to do with the DeRocchi's Library Initiative is take these government data sets and make them as easily and publicly available as possible. The UMDs of the world, which are trying to make it easier for people to access these large geospatial data sets. Thank you, Hunter. Actually, can I ask a follow up question to that? And then I'll ask, I'll get back to the chat questions. My follow up question is for community archivists working in local communities around the world that have an interest in the resilient nature of the system. Is there a resource that we can point them to help them get started with moving towards decentralization outside of all the other resources that you make available and transparent on your GitHub and your many various program web pages? Actually, I'm going to turn it here because he's so talented. He is aware of multiple hats. It's like the burden of success. And so he's been supporting one of the other teams in the foundation called the UX team, which has been working hard to make these things more kind of user friendly and easily accessible. So Ian, I think that you probably have some research that you can point people towards. Yeah, I don't actually have any amphithep in my head, but there are there are plenty of resources currently linked on our website, which I shared earlier, ffdweb.org. There's also if you the file plan docs can be a little can be a little technical if you want to get into the technical details of how some of the stuff works. But you can also if you're interested in doing community archives, this is my personal passion. I was doing community archive work for a long time before I got involved with the foundation and actually got involved with foundation because I felt like decentralized storage was one of the problems I would often see with community archives is that small community archives just really didn't have the funding or the technical capability to manage data hosting infrastructure themselves or manage custom applications. Many of them don't have engineers on staff and even one engineer like me trying to volunteer with large a large number of community archives can get spread then really quickly. So I've always been excited about the power of the decentralized web. Just take some of the operational burdens off of community archives. So if you're building a community archive and you want to potentially leverage FFS or Filecoin feel free to send me an email at Ian at Phil.org. I'd be happy to find some resources or talk you through how to build a community archive using this technology. And actually if I can add to that take that maybe one step further. We're not we're certainly not here today to pitch our foundation but we are here to help you. So let me pitch our foundation a little bit. We're not just in the business of providing technical support. We're also in the business of providing resource support really affecting me in funding. Know that if you are keen to use these technologies reach out and we'll work with you both in drawing on Ian and our other partner engineer to provide that technical support but we'll also explore the potential for us to provide some financial support as well. We are a foundation. This is what we do. So if I didn't come here today to pitch it just know that if you're interested in this please reach out. We're always happy to hear from you and we'd love to talk about a potential collaboration. Thank you Ian for being so generous and open with our community and thank you Hunter for that comment. I think that that might wake some folks up. Okay let's see other questions here. If you have another question at the top of your mind please drop it in the chat. I see that Heather had a question a little earlier about checking for redundant redundant data across other CIDs. Does the program check for redundant data across other CIDs specifically was the question? I can take this one. It depends on what you mean by redundant data across other CIDs. One of the things that's great about this is that redundant copies of the same data are all identified by the same CID. There by design the idea is that multiple nodes might each be hosting separate copies of the same data identified by the same CID. There are obviously each node hosts multiple CIDs many different copies of different files. There's been some work sometimes if you think about something like Google Docs. Something that we might consider as humans to be the same document actually changes over time and some systems built on top of IPFS will store the full history of different CIDs that represent the document at different points in time. So as someone's editing a document the content is going to change it's going to get a new content identifier and people are going to potentially store redundant copies of each point in history. But one of the benefits of the system is the redundant copies of the same piece of data aren't actually identified by different CIDs at all. Great. Thank you for that. Heather if that was helpful and clear give us a thumbs up. If you have a follow-up question please drop it into the Q&A icon. Billy if I could you mentioned Heather and Heather had a great question earlier that we talked about a little bit in the chat but I'd love to talk about here just in case not everyone's been keeping an eye on the chatter has moved on and Heather asked about the environmental impact of this type of technology and I think this is a hugely important conversation. It is no secret that certain types of blockchain technologies and Bitcoin I think is the one most often referenced in this case require a significant energy load in order to function and that's because they are proof of work and basically they are increasingly complex mathematical problems that are had to be crunched by seriously energy intensive computers and just to keep it simple. And we are keen to a provide a better alternative to that in proof of space time rather than proof of work which means that you don't have to just sit there and just consistently and constantly crunch mathematical problems but we're also keen to be very open about what the energy impact is of FilePoint and so we've got this website that we dropped the link in the chat earlier but I'm going to drop it again just to resurface this and this is FilePoint Green this is the project within the overall FilePoint ecosystem that seeks to make basically show our homework. We want to be completely transparent with everybody about the energy load of FilePoint but the thing that I'd like to say is that proof of space time is literally orders of magnitude less energy intensive than something like proof of work. Now I don't have the numbers right at the top of my kind of the top of my brain but because it's been a little while since I've died into this and FilePoint Green is a different team but just know that we are investing in this constantly and the goal here is to get to a place where we're not just energy or kind of carbon neutral but actually carbon negative now again not my specialty so I don't know exactly how someone gets to be carbon negative but I know that this is something that they're working on. And just to reiterate Hunter's point something like Bitcoin their proof of work system is doing useless work in order to generate new blocks they're basically computing these large cryptographic numbers that are only being used to compute new blocks. In the case of FilePoint new blocks are only created when somebody's provably storing new data and so all of the incentives and the sort of proof systems of FilePoints are directly tied to the useful work of storing data and making additional capacity available to store data. And the FilePoint proof system that checks every 24 hours to make sure that a particular file is stored actually doesn't even check the full file. They use a special cryptographic process that only checks small bits of the files and are able to prove that the entire file is still there by randomly choosing small bits of the files to check and that saves a lot of energy because it means that you don't have to download a multi terabyte or multi petabyte file just to check that it's still there. Great thank you Ian. Dorit has a question. Their question is they have 50 years of hard copy data they'd like to store it. They're thinking is this a case where all of the documents would need to be scanned and then saved into the system is that correct? You usually yes if it's hard copy we're going to have to figure out a way to digitize it. Depending on where you're located one of the things we could do is we could work to connect you with one of our system project partners and that archive is a great example of this to help you with the digitization of that but yes if you do want to get it into FilePoint or really any sort of digital storage we're going to have to figure out a way to make that leap from analog to digital. And I know the Internet Archive in particular loves working with regional archives and regional libraries. There's a lot of regional libraries that have microfilm or microfiche or old archival methods and boxes and boxes of this data they no longer personally have the capacity to store and so the Internet Archive loves receiving if you chat with them they'll take boxes of this stuff they'll handle the digitization process and then through our partnership with them that data will eventually end up on IPFM. So if it's not something that you personally have the capacity to do reaching out to the Internet Archive to see if they might have the capacity to take some of your hard data might be a way for you to get that on to FilePoint with minimal effort on your part. Great and we literally talk with people at IA weekly sometimes daily use us as your resource for that if you don't know anyone at IA just reach out to us and we'll make that connection and we'll facilitate that process for you and we might just help provide a little bit of a funding bump up to IA to cover it just let us know what you need. Amazing that is very helpful any other questions we're going to do a last call for questions it's never it's never bad if a meeting ends a few minutes early we've got we're moving to next meetings the program series was designed for you to jump in to these sessions and then jump back into your next meeting last call for any other questions from the audience. Okay oh John with a just before the buzzer here has Filecoin ever participated with proof of humanity projects sorry about that John I saw that earlier and I missed it so yes once again Filecoin team the question is has Filecoin ever participated with proof of humanity projects. So it's actually not a term that I've interacted with in the past proof of humanity I'm if I had to guess I would say it might have to do with digital identity but John if you're able to in the chat just expand a little bit on that term I don't know Ian, Caitlin if either of you have interacted with projects that specifically use that term. Yeah I can talk more about it we so proof of humanity systems are trying to solve what's commonly called in web three that the civil resistance problem or proving that multiple people interacting with a network aren't just spot accounts or clones of various people that's not really a problem that our network faces we're not trying to build right now a personal identity system so if somebody creates multiple wallets to interact with our system sometimes a given storage provider might be owned by one person but might actually have multiple wallets to represent different computers or nodes on the network sometimes when uploading data to the system we actually encourage people to create separate wallets to uniquely identify different data sets rather than different individuals involved in the process as far as I know we haven't participated with a lot of proof of humanity projects just because that's not a core concern for our network we're not trying to deduplicate people and in the periods where we do need to attach a real world physical person we just go through a traditional sort of KYC know your customer know your client company process to verify who the person is on the other end great there's one other question that emerged here from John I think this is a follow-up what do you mean by human rights data yeah I saw that thanks it's always trying to define a term with any group is always from her difficult process and it's easy to get bogged down in that and so we don't want to tackle that because part of it is that we as a foundation don't believe that it's our job to tell anybody how they define their data both not just something in terms of whether or not it's human rights versus not human rights data but also the value of that if an organization or community or group feels that their data is valuable then to us it's valuable and so we don't seek to apply a lot of labels like that I think in something in particular like human rights I actually am not sure how many people are on the call now but let's say it's 30 of us I think if you were to ask all 30 of us you'd get probably 30 nuances of how we define human rights data so we don't seek to put those labels on we rely on the groups that we partner with if they feel that it's something valuable or if they feel that it applies to human rights then to us it's valuable and it applies to human rights great one one other question from Dorit once the data is saved and stored to locate it is this a simple search or data stored on IPFS I'm assuming yeah you want to take that yeah so I actually just posted an answer in the chat as well but right now there's no centralized search service the way there would be with something like Google but as long as you know the content identifier or that CID you can go to any IPFS gateway I mentioned IPFS.io was one of them dweb.link is another and w3s.link is a third one but there are many if you look in the IPFS documentation that you can use and all of them if you give them the CID it will behind the scenes ask the network for it and fetch that content for you it's very easy to retrieve them once you know the CID and we are actively working with some of the core developers of this to create better metadata standards so that we know what's in each one of these CIDs and we can provide a richer search experience like what you would expect from something like Google. And that's specific to searching for something but in the case of something like democracy's library it is hosted there so if you wanted to access it you just click the link just like a normal link it just takes you to the distributed storage version of it rather than the decentralized storage version of it a lot of the repositories that we're building there's no expectation that you're actively searching for it you just go to that repository where it lists storage data. Great we have one last question from Heather and then we're going to wrap up maybe a silly question but is this like DOI address for libraries? Is that a comparable reference? I'm not actually sure got it yeah it sounds like a quick search of what DOI identifiers are it does sound very similar so it sounds like DOI identifiers are libraries assigned permanent identifiers to a piece of content or an article in order to identify them in library systems so yeah I would say that a CID is similar to a DOI identifier but instead of a human choosing a unique identifier for a particular piece of content the cryptographic system automatically generates this identifier which means that two people independently even if they've never talked to each other for the same piece of content will generate the same identifier I assume the DOI system because there's someone involved who's assigning the DOI number to a particular piece of data means that there has to be some coordination among librarians to make sure that everybody agrees on what DOI identifier references what piece of content thank you I want to thank Heather and Lynn here for raising this question because you've just given us now a new wonderful analog that we can reference when we're talking to people I certainly did not know about the DOI address system I'm going to reference it now yeah thank you Heather Lynn John everyone with the questions everyone who showed up for turn visitors double thanks and big thanks Hunter and Ian and Caitlin for joining us and supporting this program your generosity with your time and resources and your willingness to engage with our community is really remarkable and we're really grateful once again big thanks to Hunter and Ian for joining us being so generous with their time