Knowledge Repos, Small Talk, and DAOs by Andy Tudhope
ETHBerlin·Wed, Oct 9, 2019, 12:00 AM
Speaker
https://ethberlin.com https://decentralala.com/ https://goerli.net
Transcript
that sticks all right so my name is Andy I work at status and this talk is about making sure that we can get aetherium everyone and everywhere and the question is how do we actually do that so it's status we view ourselves as the mobile window onto aetherium this means that's when you're sending messages through status you're talking directly with your friends they're going directly over whisper which is a peer-to-peer enter and encrypted dark messaging protocol that's part of the ethereum stack so whenever you're sending messages that's going directly to other people without any service in between earth and thus erratically different architectures any other messenger that currently exists at the moment but this is just the bare tip of the iceberg in terms of what we actually do because we also have a native wallets it means that you can transact seamlessly with anybody on ethereum and it's ready the combination of private peer-to-peer messaging with a public global immutable ledger that allows us to think about and speak about conversational commerce on steroids because it's not only messaging in our wallets but also the ability to interact with any DF that you that exists on ethereum at the moments and this is ready when I want to focus on in this talk because it's really about these conversational and social spaces when we build this new decentralized architecture that we're all going to be using that are most important to understand because they're going to be the most impactful in terms of building community and making this stuff sustainable for a long time in order to understand exactly why this is important we need to go over some history so back in the early sort of 50s and 60s before the computer had settled into kind of cultural significance as the solipsistic machine that you interact with on your own and you just sort of hack away in your own there were radically different ideas of what actually constituted a computer at all right so it was certainly there was this notion that a computer was more of a network method of advancing collective knowledge and enabling collaboration on scales never before seen and perhaps even as a means of enabling creativity rather than competition and collaboration rather than mass surveillance right so these ideas come from a very very long time ago if you go back to 1737 there was a guy called Andrew Michael Ramsey who was a freemason and he was the first guy to begin speaking about something called the univer dictionary and in Ramsey's imagination the universal dictionary was this book that would contain pointers to all of the different knowledge in the world and if you could do that and if you could make it complete you could basically generate a world brain right so that's these ideas get picked up by a guy called Charles cutter who is a library scientist and he's the inventor of something called the universal Decimal Classification system he was a contemporary of Melville Dewey and I don't know if there's any library scientists in the audience but the Dewey Decimal System is what's used in libraries across the world and these guys were thinking about how do we organize information a long time before like we had computers or the digital world anything that we know today and modernity and at the same time as these guys were working on actual library classification systems there are a whole bunch of science fiction writers perhaps most famously a guy called HG Wells who is also on the same trip of this hyper-connected super organism that was capable of intelligence and computation on scales that we had never before seen right and he wrote a short story called the world brain in which he really explored that and his ideas are then picked up by another guy called Taylor at the shardene who was a Jesuit priest who is actually accused of heresy by the Catholic Church because he sort of said that if we could implement these new ways of collecting and sharing information we could actually generates a sort of divine consciousness with unfettered access to information so this is what all these guys were actually thinking about in trying to build a long time before the computer got going in the 50s and 60s really so all of this led to a little-known but incredibly important figure in computer science as a guy called Paul Utley he was a Belgian information scientist and he he started in the late 19th century by building something called the universal bibliographic repository right and basically what he did is he hired a whole bunch of people to read as many books as they can and then write down on key cards a summary of that book and then the keywords and all the other different pieces of knowledge that it connected to and then hadn't had them stored in this massive bibliographical repository and writing and like he was the first guy to really say you know this universal book formed of all the books would be kind of kind of annex of the brain itself right a substrate for memory ubiquitous and eternal so it's kind of what Ramsey had been talking about except Paul Utley actually built us and was called the mundaneum and it existed in Belgium and was funded by the League of Nations right because they saw this advance of collective knowledge and the enablement of create that creative collaboration as a means of achieving and maintaining peace which is a really interesting thing particularly if you read the sovereign individual and think about the state of the world as it exists at the moment so the key idea in all of this history was these guys are talking about the fact that you can have this sort of top-down classification of knowledge but it's really really important to combine that with a bottom-up social space so while all these people were reading for Atlas mundaneum and sort of these experts labeling index cards and then storing them in the repository you also had this notion of social space so you could kind of track the different uses that any piece of content or any products or service had had and that would inform the context and the human meaning associated with that specific product service or content right so you you have these auxilary tables which existed below with like a plus or a slash or a colon and each thing means different things and this is the other key feature of this history which has kind of been left out of the web as we experience it today is that links themselves actually carry meaning right so it's not just something that takes you from one place to another in this hyper connected world of information it actually signals agreements or disagreement or its takes some kind of value as you move through the web of hyper-connected information which is really interesting that kind of brings us to the beginning of the modern history of the web which starts in the 20th century with a guy called vaniville she wrote an essay called as we may think and he proposed this machine called the mimics and you can see a sort of schematic diagram of it over there and basically the mimics was built on microphone because Bush had worked on microfilm in the war and you would store all the books that you could on this microfilm and then you kind of stand there with this really weird like device on your head and you would call two books up at the same time and then track your way through them and draw links and associations between the knowledge that you were working with and the idea is that you would have this completely bottom-up user-generated social space around information and that the trails that people had made as move through that information would be visible to everybody else right it's it's this idea of modern-day marginalia I'm a literary and linguistic students and we absolutely lose our minds for margin area because it makes encountering a book such a radically different experience right when you read a book that has margin ad in it you've got these like layers of knowledge that you can see as different minds have come to the book and left this like a chrétien of time in the knowledge that you're working with which is really really fascinating and it leads to a radically different encounter with that knowledge and it is kind of proof as Carl Sagan said that books and information are proof that humanity is capable of magic so that's what all these guys are doing and really the chief magician the guy that I most loved in this whole history was a guy called Doug Engelbart who worked at a place called Xerox PARC in the 60s and he was you know one of the most prolific inventors of all time he came up with something called nls which was the online system he was the author of this thing called augmenting human intelligence essay that you can get online and I highly recommend it and he was also the giver of something called the mother of all demos which was something that he did in 1968 where in one demo in the 60s he he showed off a graphic user interface a word processor a mouse and collaborative video calling all at once in 68 right it is unreal you can go and find this demo on YouTube I highly recommend it because it's just it's so far ahead of its time it's really quite crazy and in fact it's even sometimes in some respects ahead of what we have now because particularly that video called that he did with a collaborator in Palo Alto was fundamentally collaborative right they were sharing access to shared documents you could see their curses with their little initials below them and like co-editing it's in real-time through the video call right this wasn't like a shitty Google Hangouts where you have to do a screen share this was literally like that collaboration is built into the architecture of what Engelbart was working on it's really really important because Engelbart was absolutely obsessed with like groups of people and how do we like use these networked computer systems to enable people to collaborate and share information again on scales never before imagined and and and his sort of key question was how can we allow people to operate in domains of complex information structures and particularly his answer to this was potentially to develop what he called the knowledge repository of all knowledge repositories right and this is why I told you about Andy Andrew Ramsey and Paul at bay earlier because it's exactly the same concept just in a modern dialogue but if you have the knowledge repository which has links to all the different knowledge repositories in the world and allows you to navigate that it becomes in essence an annex of the brain itself right a universal book ubiquitous and eternal which is really interesting so the last piece of history that then gets picked up by this particular strain of thought was a guy called Ted Nelson who actually wasn't very technical himself but he was an incredible writer and he had a gift for coming up with the words so he was the guy who actually generated hypertext and he came up with transclusion and hyper books and a whole bunch of stuff about information spaces and structures and he he he proposed something called Xanadu which is being called the best vaporware never written which I feel like might appeal to people in this community right but you know even though he never implemented it it did allow for people like Andy fandom and various other Computer Sciences time to write what was called the hypertext editing system and that led directly to the work that certain berners-lee was doing on HTTP and the web as we know it today and I've left out a whole bunch of history and Mike particularly the military-industrial context that a lot of the stuff happened in but the question that I want to ask you based on all of this stuff is what does if varium look like as an information space and what can we build it to be so in order to answer that but just a quick summary of what the web that wasn't ready is you have this notion of two-way classification with bottom-up socials I mean top-down classification with bottom-up social space the notion of two-way linking in which links carry actual meaning and signal agreements of disagreements or stake value you've got the notion of visible paths through information and modern marginalia users must be equivalent to authors right to use these systems must be to author your way through them and finally and most importantly conversational and social spaces must be a part of the architecture itself right so the web that wasn't is really a creative collaborative conversation and I want to kind of suggest to you guys that Bitcoin in etherium is information spaces enable conversations about value that would previously not even imaginable and particularly aetherium with this generalizable machinery and language that allows for the trustus execution of arbitrary code on some shared thing is a really really interesting ontological framework in which to think about like what conversation actually means in the modern world particularly when it's so literally and closely tied to value itself you know so just to sort of emphasize that this trust this execution of arbitrary code on a global network capable of like deterministic communication that's agreed upon by a set of peers allows us to forge languages that help community practice and perhaps even further than that that require a certain kind of practice in order to be a part of the community at all right so the linguistic primitives that you use in order to enter these conversational and social spaces need to be of a certain form in order for you to be a part of the community or at all and like don't get confused by the technical language that I'm forced to use their like linguistic primitives are literally just the things that make up your language so like it's become socially acceptable when a friend asks me a question for me just to send a URL to like a link to somewhere that has the full and complete information and and that's now like an actual linguistic primitive that is part of our language right if you think about it for a little while what the URL allows us to do is a linguistic primitive is to share more information more accurately than one mind can hold in conversation in conversational memory at one time right and this applies to everything like memes on Twitter URLs and like if you extend that thinking into smart contracts and transactions the possibilities for the kinds of conversations that are intimately and uniquely tides of value become really really really interesting and we can start sort of thinking about bringing all of this work that Engelbart was doing with knowledge repositories and how you gather knowledge how you inform that with social and conversational spaces around these sort of decentralized autonomous organizations that are made up of these creative individuals and bring it all together into one thing so what does that actually look like well we've already got the top-down structure implied by the simple facts of blocks and then subtly indirectly by the consensus algorithms that we use in the hashes and the signatures in the various data structures that are suitable to our needs and then we've already got that bottom-up social space which is implied by the mere fact of transactions right so like the type kind and frequency of transactions that you get involved with are indicative of the kind of social spaces that we can begin to think about building with these linguistic primitives it gets even better though because if they're iam like Gavin would say the DEF CON one is a purely object orientated environments in which you know always which objects have sent you a message that allows us to do this notion of two-way linking fairly fairly easily and like we can actually stake value onto links so it's not just signaling agreements or disagreement but you can actually stake like a hundred S&T or a hundred ANC or you know two 'if onto links that you think strongly agree with your arguments and you want people to actually read that would be nice I write a lot of a lot of different articles and I know people never follow the links it would be nice to be able to signal like this is a really important link readers and then visible paths and modern marginalia are there because this is a public immutable ledger and you can see always what people are doing so a lot of this stuff is already in place users are already authors right when you use aetherium you author in new states always and that state is agreed upon by everybody around the world which is really fascinating I think that that's something that we can probably make more obvious to users through specific UX decisions and that's gonna come down to the designers in the room and finally this notion of conversational spaces right and they aim not a tracking or advertising but exploring and mapping creativity and enabling collaboration at scales never before possible what does that actually look like well there if you take a look at this is crypto kitties in in status and over here you've got a little chat icon at the bottom and what tapping on that chat icon does is it takes you to a whisper chat for crypto kitties this is a chat that actually exists on whisper and the topic is encrypted like and gossiped around so the topic is just crypto kitties and if you know anything about how whisper works it's gossiped around the whole network so you don't need just a back you don't need status to listen to the cryptic it is a unique chat on whisper it's just there as an inherent feature of the network anybody with any client that is listening to whisper and filtering for that specific topic will be able to find it and it means that crypto kiddies or aragon or get queen or whoever else has their own unique like social conversational space as an inherent part of the architecture itself right it's really really important that you realize that this is not just about status whisper is an inherent part of the etherium stack it goes together with the EVM and swarm and as soon as we enable this kind of stuff to happen all of a sudden you can have these like unique architectural conversational spaces happening behind whatever your dieppe is that's obviously not enough right we need to actually be able to have useful things happening in that chat we need to be able to prevent spam we need to have you know customer service methodologies we need to be able to share knowledge we need to be able to like interact in valuable ways with each other that extend beyond just helping community members and showing them what to do so if you look in the top right here this is sort of how the stack currently looks at the moment within status you have etherium at the bottom and then status is this wallet chat and browser and the browser allows you to in like access any d app that exists but like that the app is not native to status and if you like give it a try now you know statuses on main nets and you can take a look at it now you'll see that it often takes really quite a quite a long time to load an idea because it's just happening in a webview and so it's a really bad user experience and we can't control that so what we've done is we've built extensions which allow you to provide your own features directly within status right so we wanted people to build the apps that have a native feeling there should be no distinction between what is a status feature and what is an extension these things should be the co extensible right and and and extensions are so much more than just things that might help you in those chats which form a sort of architecture of the social space they're really all about kind of video streaming better decentralized exchanges custom asset screens and perhaps in fact like entire ephemeral DFT's that are called into being composed on your phone as you need them and then fade away as as you don't write and this is actually kind of what an extension looks like and it's really quite important because it's a radically new in different concepts so we've got a little gift here basically we're in our profile we're going we're gonna install an extension in status by putting in a URL at the moment we're just using a gist but this would definitely be hosted in ipfs or swarm we've successfully installed it if we now go to a chat you'll see that we if we type that command on there we have another command that has been brought in by the extension this is a collectible command and this extension allows us to send crypto kiddies in chat which is really quite remarkable I think and it's sort of indicative of just how powerful this stuff can really be right again it's not just about chat but this is the first place where we will see these kind of things having a huge effect because like what an extension is it's just leveraging multiple hooks that are defined by the host and hook into specific components of status it's composed very simply of views queries and events and all you need to do is specify what's called the ddn file which is like an extensible data notation so this is just like it's a universal data flow going between these queries events its queries events and views and it's hosted somewhere in a decentralized storage provider and you don't need to build it right you can just sort of specify it install it in status and off you go there's no bulb there's no need to ship it through the app store there's no need to ship it through the Play Store we really are escaping from a lot of these centralized solutions that force us to do things in specific ways and what's even better is that a statist user can loaded via the ENS username and they have tight and granular control over which extensions they use so that we can carry on ensuring security and privacy for our users so that's all that I have time for today thanks so much for listening and I just want to sort of encourage you guys to you know the the things that it is possible to conceive of with this technology and with these networks are nearly limitless right in the horizons of the possible stretched so much further than almost any of us can imagine so you know imagine and let's get to work [Applause]
Automatic transcript — names and jargon may be misspelled.