Transcript of the GEO presentation by Arturas Vileita, The Graph’s Community Talk #17

Next up, I would like to invite Arturas to take the stage to tell us a little bit more about Geo. Hello, I'm Arturas, and I work as a product manager at Geo. We initially announced about Geo this year in June during the Graph Day, and we've worked quite a bit on the project since then. On October 25th, we announced fully publicly about this project, and in the upcoming few weeks, we will have the actual first version of it launched, so people will be able to play around with it and see what it's about.

As a product manager, I can talk more about the concept behind this project and the idea, not as much on the engineering side but in general. So, as you probably know, the term web3 has two meanings. It was first, in 1999, the semantic web was a big thing that people were talking about. In 2006, web 3 was first used to describe the upcoming curation of the semantic web, the transition from web 2 to web3 was thought to be a transition between just the members we understand it right now, where all of the data is just put in different websites with no constraints of exactly how the data is structured in place, to that data being structured in a way that it would be machine readable. Only later on, the web3 as a blockchain-based internet came about.

There are basically these two different web three terms floating around which mean different things. The interesting thing about Geo is that we are actually building this sort of Internet on the blockchain, so we are web3 in this modern terminology. We are web3, but at the same time, we are also making it in a machine-readable way, so we're building it a semantic web. So basically, we're building it web3 squared. That's a small joke about how we can look at it.

The thing is that all of the data that we will be placing on the decentralized knowledge graph of Geo is going to be with triplets, which is just a really simple way to add data. It would be very simplistic; we just basically have an entity, an attribute, and a value for every fact or claim that we want to make, and that is machine readable. We can have an unlimited number of claims about any fact or entity. To give you an example, whatever hobbies you personally have, it might be history, for example, or astronomy, or whatever, and let's say you are interested in history, particularly in World War One, or maybe specifically, you're interested in Ancient Rome or Greece, or whatever. The question is, if you have a hobby and if you want to have information about it, if you're looking for it, where exactly do you get this information from? So you read books about those topics, maybe find cool articles about them as well on the internet, maybe you can find some podcasts or YouTube videos about them, and you sort of grow your knowledge of that thing this way.

When it comes to Geo, you can take because it is a decentralized knowledge graph based on facts, and each fact is an entity. So let's say you get the book that you like, and in every book, there are obviously different components. It has chapters, some mentions, maybe examples of something, maybe it has a sources list, maybe have certain claims in each book. You can find all of these things. Adding all of the claims and all of the facts mentioned in any book or an article into Geo's system means every single fact is now an entity onto itself. Now you can add an unlimited number of attributes and values to those.

For example, if you find a book about Ancient Greece or Rome as a historical book and there's one specific fact, like for example, in Ancient Rome, there was a certain living standard there. You can tag that statement with many attributes such as Ancient Rome, living standards, historical facts. Now this fact is just a part of the general knowledge graph, and maybe you find a similar fact in some other article or some other book which says the exact same thing. Now you add that fact as well, assign the new entity ID as well, and you can add the same attributes. That is also a historical fact about Rome. So when you will be searching on Geo for historical facts about Rome, now you will get both of these facts even though they're from different sources. This way, basically, we can merge knowledge on any topic from all of the sources that are available in the world to us into one systemic structured place, and we can navigate them by using universal search terms.

If you're looking at information, for example, what is the timeline of Ancient Greece, or what is the best product to buy under these circumstances, or how should they vote on the upcoming bill or something like this, you can easily immediately get all of the arguments for and against these things from all of the sources neatly listed for you to read. This basically changes even how we experience the internet because right now, if you want to find some information, we have to basically open up 20 tabs, read through many different articles and websites. You don't know exactly if you find something that's not in correlation, some things that are mutually exclusive. There are different statements about the same thing. You don't know who exactly to trust and how to address those contradictory terms. On Geo, you don't have to do that because once you're searching for something, you just get immediately all of the statements from everywhere, but also, if there are contradictory statements there, you can set up a debate mode immediately, and you can assign that this entity has a relationship with another entity. So if both entities are facts which are saying different things about the same thing, you can assign the relationship as being a conflict, and now you can address it on Geo as well, in the same structured way.

We are trying to make it that everything would be an entity, and every entity could have an unlimited amount of relationships. Basically, we would be recreating a human brain into a computer but not a human brain of one person but instead like a human brain of the whole society into one place. We can use it to understand topics in a much more structured and complex way. The same as we can use it to resolve conflicts and misunderstandings that the society is facing, like different approaches to the same issues that we want to solve and so forth. We can use it to communicate in general and cooperate in solving these problems. One way that we're doing it is we're going to have spaces in Geo, and each space is just a community that is actually working on the same topic. So for example, you can have a space just for history buffs, and everyone can just add all of the data about history in one space. Maybe someone else has a health space where people who are into biohacking and want to optimize their health can add all of the data from all of the best sources they know about these topics. Each space can have its own democratic system, like how exactly do we vote on any fact, who decides which facts are going onto the knowledge graph, and how do we value these facts? Any space can decide on their own. So this is sort of a democratic experiment as well because each community can adapt the best form of governance which is best for that specific community of experts or enthusiasts.

That's the gist of what we're trying to build on a conceptual, general way. In the upcoming few weeks, we should be launching the beta version. It's going to still have only the readability function for now, so you will be able to search through the data on different topics. Later on, as we launch the first version, we will be adding these different features, which are the relationships between those facts and entities. You can find out more right now on Twitter for Geo and geobrowser.ao on our website. So that's a short introduction. If you have any questions, I'm happy to answer. Thank you so much, Arturas. Yes, I think people have a lot of questions, but you can ask them in the chat or reach out to Arturas on Twitter. Thank you so much.