Epicenter - Learn about Crypto, Blockchain, Ethereum, Bitcoin and Distributed Technologies - Trent McConaghy: Ascribe – The Internet of Ownership

Episode Date: May 11, 2015

The idea of using the Bitcoin blockchain to represent and track the ownership of off-chain assets is almost as old as Bitcoin itself. One of the startups that is pursuing this vision is Berlin-based Ascribe, which allows creators to register (‘ascribe’), transfer and archive digital art. Ascribe CTO joined us for a fascinating discussion of their ambitious goals to enable a so-far elusive secondary market for digital art, help creators monetize their work and ultimately retrofit an entire ownership layer to the internet. Topics covered in this episode: Why ownership in the internet is broken How Ascribe wants to realize the vision of Project Xanadu of an internet with bi-directional links and baked-in copyrights How the SPOOL (Secure Public Online Ownership Ledger) protocol works The components of the Ascribe tech stack How Ascribe will crawl and similarity-match the entire internet to retrofit an internet of ownership Why Ascribe focuses first on the digital art market Episode links: Ascribe SPOOL Protocol Cointemporary Project Xanadu Wikipedia Rewiring the Internet for Ownership with Big Data and Blockchains Slides This episode is hosted by Brian Fabian Crain and Sébastien Couture. Show notes and listening options: epicenter.tv/078

This episode of Epicenter Bitcoin is brought to you by Shapeshift.io. With no account or son it's required, it's the easiest way to buy and sell like coin, counterparty, dogecoin, dash, and other leading cryptocurrencies. Go to ShapeShift.io to instantly convert your old coins and to discover the future of cryptocurrency exchanges. Hello, welcome to Epcenter Bitcoin, the show which talks about the technologies, projects, and startups driving decentralization and the global cryptocurrency revolution. My name is Sebastian Kutjou.

And my name is Brian Kavanaughan. We're here today with Trent McConaughey. He's the CTO of Ascribe, and I've known Trent for quite a while. He is also in Berlin, and he works in a really interesting startup that uses the blockchain to sort of solve the problem of ownership in the digital space.

So Trent, thanks so much for joining us today. Hi, it's really a pleasure. Thanks for having me on. Yeah, I mean, Trent gave a talk recently at a Bitcoin meetup, and it was, I mean, we talked about a scribe and what they wanted to do a few times before, but the talk I thought was really good and sort of really, you know, both forth this vision that you guys have. So I thought it would be great to have you on because it's also, I think it's a really good use case of people doing something with the blockchain,

that goes beyond the usual money transaction context. So, well, Trent, to get started, can you tell us a little bit about ascribe and what the problem is you guys are trying to solve? Sure. So in general, it's a problem that exists with digital property and it gets even worse when it's on the internet. And the problem is that it's really difficult

for creators of digital content. to get compensated fairly for their work. And it's also difficult for consumers to consume work, works sometimes and all that. But it really comes down to the creators. How do you get paid when you create stuff? And Ascribe is really trying to address that in order to help the creators get compensated

in order to help audiences get the content they want, in order to help connectors, marketplaces and so on, have authentic digital content. So that's kind of the problem that we're attacking. and it really is an internet scale problem. And so we've actually, some of the technologies we're deploying our internet scale. There's really kind of two major pieces to the approach.

One of them is easy, secure legales, and the other thing is visibility into usage of the content. So the legal's, the easy part of the legals is basically formalizing the existing copyright rights that creators have, and the secure legales, this is basically recording transactions in the blockchain, so time stamping, so leveraging the time stamping aspect of the blockchain to record these transactions that are already legally binding from the contracts.

So maybe before we dive into the specifics of a scribe, I'd like to spend a little bit more time on this problem. I mean, I think to some extent we're all a little bit aware of it. you know, I mean, everybody will have heard or will be aware of file sharing, right? There's a bit torrent and then, you know, we all know the problem of piracy. And then we know what the record labels, for example, do to try to prevent that, which then again can be a problem for consumers. But can we spend a little bit more time on fleshing out exactly what the reason is?

is and why this is not working at the moment? Sure. Definitely. So kind of throw a rock and you'll find a problem when it comes to ownership of digital property in the web. You talked about piracy and all of that. A lot of the time people don't want to be pirates. They're happy to pay fair price for work. But it just isn't convenient.

So for example, when iTunes came out, actually, as the years have worn on with iTunes and Spotify and the like, Piracy has gone down of music simply because the experience was good enough that people were willing to pay a fair price for that. In another example, I live in Berlin, but my German is really pretty terrible. So I want to see new release movies in English online, just as soon as they come out on DVD. But I can't legally here in Berlin. I can go to iTunes, Germany, but it will only give me the German version. I can try to

sign into iTunes in my Canadian account, but it notices my ISP and doesn't let me. I can try to use Netflix, but it doesn't have the new movies. And there's really, you know, Amazon also, it has just the German versions. So the only way that I can actually get

a movie that's English, that's new release, is if I VPN or torrent it. And I don't want to be a pirate, so that's really a problem. There really should be a convenient way for consumers who want to view content, who want to pay a fair price, to use it.

So, you know, there's many, many examples. That's a couple. And, you know, you talked about DRM. In general, DRM is sort of this idea that goes back decades, this idea of trying to take a lock and put it around a digital file. And then this, you know, magical, mystical hope

that if you put a lock on that file, maybe people will never, ever crack it. But, you know, that's kind of fighting the physics of bits as soon as one person has cracked the key. Then it's gone for good. You know, with DVDs, Sony and a bunch of other companies had spent millions to create a cryptography, DRM for DVDs.

Within a couple days of DVDs being released, it was cracked. And at first, Sony and the other guys, they thought, oh, yeah, yeah, must be some big, well-funded corporation. No, it was just a guy. His name was John. He cracked it. He's just a kid, right? DVD John.

So the point, and what the point is, Once that was cracked, it was out there for good. So in general, you know, DRM has been an idea for a long time, but it's really against the physics of bits, right? Copy, copy, copy, copy, as Amanda Palmer puts it. It really shouldn't be trying to lock down the content itself. It should be about trying to protect the ownership of the title to that content. So it's the difference between access and title.

And that's really kind of, you know, everyone has been getting that confused. And I describe what we're trying to do is really help people to nail down the title itself to transfer ownership of the title, etc., etc. Yeah, you mentioned something that hits really close to home for me, man. It's like trying to watch American or North American content in English legally. It's just impossible to do it in Europe without going through pirated content. So there's one quote that always sticks to my mind, and I can't remember who said it exactly, but it was some podcast somewhere.

I heard someone say, like, this is so absurd. like, I want to watch this, but there's some idiot with a law degree that is preventing me from doing so because I don't understand why I'm pressing myself. So, but yeah, it is a huge problem. How does, what has changed now and how does blockchain technology potentially solve that where we didn't have solutions before? Right. I mean, so the ascribe solution leverages blockchain as part of the overall technology. and legal framework to solve this. It isn't just blockchain.

It's a combination of things. So the blockchain, it's really sort of three things. There's the legal side, and then there's the visibility side. And together they constitute the value. So really, to be able to establish ownership, I mean, as a creator, as soon as you create a file, a piece of art, whatever, you have the copyright rights. but it's really hard for people to transfer that

and if someone else claims those copyrights, what recourse do you have to say that you had it first? And sometimes artists, they will actually take a USB file and mail, a USB stick, put their file on that and mail it to themselves back and forth and back in the mail just to sort of have that proof. What the blockchain does is it's sort of like taking and putting a file into a USB

and mailing it to yourself, except it's much, much more convenient. So you get that sort of legally binding time stamp that you had access to that file at that point in time. And then the second part is, you know, when you're transferring ownership, what that comes down to is bequeathing a portion of your copyright rights, the beneficial use rights, to the purchaser. And then they can transfer it and so on and so forth. Traditionally, this has only been the realm of the big corporations, you know, the guys who can hire teams of lawyers to keep away us Canadians from watching, you know, the latest new release. But what a scribe is doing, thanks to partly blockchain, thanks to the right

legals and other technology, is rather than just making this the realm of the enterprise, it can be the realm of the individual. So go ahead. You mentioned transferring ownership. And we were talking about this before the show. And something that I've been thinking about since doing this research is in a digital realm, is ownership really desirable? I guess is one part of the question or discussion we can have.

And the other part is, do we have this notion of scarcity in the digital realm that we have with physical art? Or should there be this notion of scarcity? And by that, I mean, so you have a piece of art. Okay, so I've got this mask on my wall here that somebody made. Okay, well, I can touch it. I can feel it. It's nice. And I know that there was an artist that worked on this.

And that gives it more value to me than if it was some, like, knockoff from made in China. With digital art, you don't really have that because it's just bits, right? And I would go to as far as saying that if I look at the Mona Lisa for the first time, if I look at it as just a JPEG, I'll still be moved by it, I'll still be inspired by what the artist was trying to put out. So it seems like as soon as you get into the digital space, there are a whole bunch of things that don't exist in the in the physical realm. Yeah, so to answer your question,

there was sort of two questions there. Can you have scarcity in digital? And what about the idea of property and digital at all? And this idea of property and digital at all, it actually goes back to the very idea of intellectual property. If you think about the word intellectual property, intellectual means ideas and property means, well, to own something, right? and it goes back hundreds of years.

In fact, one of the very first copyright cases was in the 1600s where people were fighting over a book and 3,000 people died in a battle over a book actually in Ireland, of all places. And it was about the copyright rights. And so the question is, is it desirable for someone to have to be able to own an idea? Well, if you're the author of that book

and you just spent the last five years of your life writing it, how do you feel if everyone has, just as much right of you as you in order to make money from it. And that's not very fair. So the idea of owning your ideas that you put hard work into as a creator, that's the core idea of intellectual property. Copyright is about works like books and music and so on. There's inventions, which is what patent law covers. And then there's trademarks, which is logos and these sorts of things. And each of them is sort of a piece of overall intellectual property, the idea of owning ideas. So now the question is, can you translate these existing intellectual property rights

into bits? And the answer is actually yes, simply because they're already ephemeral. They're already, you know, not tangible. It's an idea. You can't really put your hands around it and grasp it. So it's, you can use bits to record that you have title. but it's always been something that is it's not tangible. So I think that's how it relates. And then just sort of your final question sort of had been related to the idea of what about multiple editions? You know, what about having five or ten editions?

You can do that with, you know, books and otherwise, but does it make sense for digital? And, well, if you look at it, they're actually, the idea of additions goes back hundreds of years and there's always been a debate, but at the end, editions have always emerged. you go back to the time of Rodin, there's actually 20 copies of the thinker out there in the world. And there was one cast for all those 20 bronze sculptures of the thinker.

Photography, there was debates for actually decades on the idea that you could have multiple prints. But that's been resolved. Etchings, all of these things, there's actually been limited editions. And part of the reason actually is going back is simply to help compensate the creator. So it's basically the creator is bestowing the copyright rights in order to get comments. If you could have a million copies or a million editions, then it wouldn't be as valuable. It might not be possible for the creator to get his copyrights. Right.

So in terms of additions, I guess it's with additions that I have the most problem with, I don't see the value in having additions in the digital realm because it's just bits and you can make as many as you want without any work. you know well I would suggest like that let's come back to this sort of general question but I think let's talk first about the sort of the example of the digital art market because I think

there it's an interesting example and I think it kind of makes a lot of sense there and it becomes much more complicated to wrap your head around this when you talk about a song for example like do you have an interest

owning a song if all you want to do is listen to it, right? But let's talk about the digital art market, because I think that's a great example, and that's also what you guys are focusing on at the moment. Yeah, sure. So digital art is our initial market. We're also spending time with photography and 3D, but focusing on digital art.

So in the world of digital art, which is basically pieces of art that are created digitally and transmitted digitally, so like an image or a movie, these sorts of things that when you would transfer ownership it's actually ideally transferring it as a file as opposed to

putting it onto say a VCR and gluing the VCR lid shut which is actually what people have done in the past so in the world of art the most important thing for value of a piece is provenance the ownership history

and so let's say you have something that it looks 500 years old but you don't know who the owner was for 50 years, the value is going to be a lot lower. And if you have it and you don't know the provenance at all, then the value is essentially zero. Sometimes you can even recover the providence. You know, there's this example of the Faberje egg where someone bought it in a flea market for $5,000 or so, and they tried to sell it for scrap metal, but no one would buy it.

So they ended up Googling around and discovering that it was a Faberje egg dating back to the time of Peter the Great. And then it got sold in auction for tens of millions of dollars. So in that example, the provenance was recovered, and thanks to a sufficient documentation on the internet in the archives of the Armitage, etc., the providence is recovered, and the value shut up to tens of millions. In the world of digital, and with files, there hasn't traditionally been providence. And because of that, there hasn't been a way to really lock down the value of these

digital files that gets transmitted. And because of that, the digital art market hasn't really taken off. So the overall size of the art market itself is, depending here you ask, 50 billion, 60 billion, 70 billion, somewhere in there. So it's bigger or par with movies, with internet advertising, with video games. It's that sort of size of market. But digital art is, depending what numbers you look at, it's somewhere between 100 million and a few hundred million.

So it's really a fraction, which is kind of amazing, given that it's 2015, you know, this age of social networks and digital everything, except for the art market. And really, people in the art world have called it an elephant in the room problem that, you know, it's 2015, digital everywhere, yet we don't have digital art in a broad sense. And it's because this provenance problem has not been properly solved until now, of course. So basically, by using the combination of transfer of copyright rights embedded in a contract and recorded on the blockchain, that, enables digital provenance in a perfect kind of fashion. As long as the blockchain is around or mirrors and copies are made, then that providence is around. And that's basically provenance that did not exist before.

So it solves this very, very root-core problem of digital art. And hopefully, you know, it will make a pretty big impact on the art industry. So if I can be faced that, right? So the problem before us, let's say you created this digital work of art and then I'm the artist and I want to sell that. well with art the concept of scarcity is kind of very central right if you're an art collector you don't want to buy this thing if a million other people can have it as well or if you're not you know if it's not rare in some in some way and of course with digital files how could it be I mean it's I mean maybe the artist can promise you he only gives it to you and a few other

people, but then you kind of have to trust that, you know, trust is word. And then you don't know what these other people are going to do, maybe because once they have to file, of course, they could give it to other people. And if there's no record of who owns this, then like, why would anybody pay for this? It just, it makes sense that this doesn't work, right? And then, but of course, if you bring the blockchain in, you sort of registered a piece of art in the blockchain and say, oh, there's only five of those. And, you know, now you sell it to me and I can check in the blockchain that I own one of them and they're only five and I can transfer it on and that person then has it and it's clear I don't have it anymore so that kind of yeah

sort of solves that problem and of course it separates and I think that's also very interesting when we talk about the distinction between the physical and the digital right in the in the physical one you have sort of the file or the the content the object and the ownership of that this is the same thing right like either you have to pay and you also own the painting. But then digitally this could be different, right? So I could own this work of art and someone, and I have it, the file, the physical, digital file, and someone else also has the file, but doesn't own the art, right?

So you're sort of separating those two, which is very interesting. And then I think that's also where it gets confusing, because in some instances, you're like, why would I pay more for owning it? for owning it when I can just have it. But of course, I think in the digital art market, it's totally obvious because if you're a collector, well, then you care about ownership, right? That's very central.

Yeah, that's exactly it. You hit the nail in the head. It's really about the difference between access and ownership. And interestingly, you know, if you have a piece of digital art, you actually want people to share it, make a million copies, because it's actually a proxy for value. If you can see that there's a million shares of your work, it probably means that there's someone out there who's going to want to pay money to actually have the title to that or one of the titles to it, right?

So it's kind of cool that this distinction is not only useful to really establish ownership, but also as a proxy for value. Doesn't that sort of break the model where by having ownership you're in a position of power because you essentially control the distribution of that piece of art being able to display it in a gallery, for example, or sell it to the highest bidder, whereas if you can just have copies online, whether or not you own them or not, people are still going to be able to look at them and to experience the art, right? Yeah, so in general, like, I mean,

you can't fight the physics of bits, right? I mean, you can try to take a file and lock it down or water market or something, but, you know, ultimately, all of that's going to just, you know, break. And we've seen, you know, that happened with the attempts of DRM over the last decades and it hasn't worked.

To say to own something, it's a funny concept a bit. It's a bit like the word consciousness. When you use the word consciousness when trying to describe the brain, there's a lot of different interpretations of what it means, and it ends in a lot of, and sometimes arguments, etc., and strong

philosophies, etc. So ownership of physical is pretty straightforward. You're holding it, etc., it's yours. In digital, it's sort of as Browner's pointing out, it's a little bit different. But the cool thing is, when it comes to media, whether it be digital art or 3D designs or music, all this sort of stuff, you can actually unpack the word ownership into something that actually has a legal meaning, and that's copyright law, right? So copyright law is designed exactly to protect the creators and anyone who gets the copyright rights. So to say that you, you know, if you create something, you can say that you own it.

And that translates exactly to saying you have the copyright rights to transfer. ownership, that means as a creator, you're bestowing a portion of the copyright rights to whoever buys that piece. And then, you know, as person A sells to person B, person B sells to person C, that's actually a transfer of the copyright rights. So the title, and to your point of sort of distribution, you can't, you know, stop the flow of bits in a sense. If someone has a copy and they want to, you know, spread it throughout the world, you know, you can't sort of plug the internet that way, right? You can try, and, you know, big corporations in the past have tried, but it doesn't work very well, right? But if you are a copyright owner and someone out there has taken your work and is trying to sell it on a website,

you actually have legal recourse to approach the person on that website and tell them to take it down. And if they don't, you can have legal proceedings against them. A wiser strategy would simply try to monetize it to say, hey, you know, you've been using this. How about you give me 50 euros or 500 euros for your efforts? and there actually are companies out there who do this for certain times of media. It's just sort of a business-to-business sort of thing right now, though. We see that these sorts of things, if you have the right technologies like blockchain and really easy to use legals, then you can bring it to the individual.

So in a sense, just like PayPal brought transfers of money between individuals and then Bitcoin too now, think of us as a PayPal for only. ownership, a PayPal for digital IP, where you can transfer digital IP from person to person without needing an army of lawyers behind you. Our show today is brought to by our friends at Shafshift. Shafshift is the fast and easy way to trade altcoins. They now support over 25 of the most popular alf coins, including Dogecoin, counterparty, Dash, Monaro, and so many others.

When you want to trade all coins, forget about using an exchange. That's so 2013, man. Just go to Shafshift.io and get it done in less than one minute with no account or son-up required. Their currency conversion tool makes trading all coins as easy and convenient as using Google Translate. Here's how this works. You head over to ShapeShift.io, you tell them which currency you want to sell, and give them the wallet address for the currency you want to buy. Let's say you want to sell some Dogecoin, they get some dash. You then simply send the Doche coins to the address they provide you, ShapeShift converts it for you, and puts a dash directly

into your wallet. It's that smooth. Long gone are the days when you had to give an exchange your passport, telling you shoe size, wait six confirmations, just to buy some popcorn. With ShapeShift conversions are instantaneous, and your privacy remains protected. So head over to Shapeshift.com and start trading some altcoins today. We would like to thank Shapeshift for their support of WebSenter Bitcoin. So we've talked a little bit about DRM,

because it's kind of obvious that if you own the title, of some, let's say, music, then locking down access and charging for access to listen to that music, is a great way to monetize, and that's why people want to do this DRM thing, even though maybe it's hard to do it,

and there's a part where information wants to be free and that sort of stuff. But it seems like it would be a great fit or a sort of a natural fit if somebody developed a DRM system on top of a scribe, right, where perhaps the software would change, ownership on the on the blockchain and you know for example am i allowed to watch this movie

and you know if so i can see it if not i can't and that perhaps it also i mean it's a strange idea to think about but because then you could have maybe a secondary market let's say i buy some hollywood movie i watch it and then i can transfer the ownership to someone else so they can watch it, I could maybe sell it on. Have you thought about that? And what's your view of the idea of sort of putting a DRM system on top of a scribe? Right. So when I think of what DRM, I sort of think about, like, taking a file and putting a lock

around it or maybe putting on some, like, hidden watermark or something. But, and I think that's, you know, there's other ways, though. What you're talking about is it was a bit of that, but the other half was sort of a way to monetize and sort of a rental framework, right? so people can rent stuff from you. And so this is something that we actually have lead users for, and we're building as we speak, to support the rental. And rentals and loans.

So, for example, I'm an artist. I want a piece of work to a museum, and I want that loan to actually be recorded on the blockchain. So this functionality, you know, this is very important because, you know, if you loan to MoMA or Louver, then you're going to, the value of your work will go up. And then a rental is just simply a loan with dollars attached, right? And you can think of it as also related to ownership or more generally copyright rights.

So doing a loan means you're bestowing a certain amount of copyright rights for private or public viewing for a limited period of time in a limited context. And so that is exactly, you know, part of what a scribe is helping to support. If people want to build systems where you're, you know, you're like a video rental store or some sort of, of loaning, renting marketplace on top of a scribe, great, right? A scribe is simply a service. Our main way of people interacting with us is via the API in that sense. We have a web app, we have an API.

But for this sort of thing, it would be our API, and we're really just a service. So, yes, definitely, we've thought about it a lot. Well, let's talk about this idea of the Internet of Ownership. And in the talk you gave at the meetup in Berlin, you talked about this really interesting a project called XanaDu that kind of had this vision a long, long time ago. Yeah, I'd love to. And if you don't mind, I'd like to sort of backtrack a couple, like take a few minutes just to elaborate because it's pretty cool.

So if you sort of think about why ownership of digital property, especially on the internet, is a mess, you kind of, first of all, have to go to more near-term internet history. and that's the World Wide Web. And in 1989, Tim Berners-Lee drafted the concept of the World Wide Web in March of 89. By Christmas of 1990, he'd actually implemented the whole first cut, the first web server, the first web page, HTML, all these sorts of components that were needed to build the World Wide Web. And then in the 25 years since that happened, the World Wide Web has exploded.

That's where we have Amazon and Facebook and all of these other billions, literally. of websites now. But at the core of the World Wide Web, it has a challenge. If someone can take my file and use it and I don't hear about it, they can even link to me and I don't even hear about it. And this is because the World Wide Web

only has unidirectional links. So it only points one way. I can link to your web page, but you don't know about that I'm linking to your web page unless you happen to be on my web page. And so that's right. kind of the core challenge of the world web. It's these sort of very mindless links and it has no sort of rights management built in. Now going to the Xanadu as you'd mentioned, you know, we'd been

digging more into this and whatnot and we found that there was a vision going back to 1960, a guy by the name of Ted Nelson, and he had this project called Zanadu. And in fact, Ted Nelson coined the term hypertext. And Zanadu was the original hypertext project. And the vision was basically a hypertext place, just like World Wide Web, where you have documents linking to documents, but having rights management and attribution built in. And his design was simply bidirectional links rather than unidirectional links. And that way, so if you link to my web page, I know about it directly because I have linked back to you.

It's just built into the protocol. And the second thing is copyright was built in, baked in. So if I'm established as the copyright holder, if you link to me, then and you start making money, I'm going to have a trickle of dollars coming down to me as well. So that was kind of an amazing vision, actually. And this goes back to the 60s. And, you know, Ted Nelson, you know, after he had that vision, he actually spent decades working on it, refining it and so on. Started a company in the 80s to roll this out. But it was really hard to build.

It was actually pretty complicated. And so when the world web came out in 1990, it kind of just took off so much more quickly because it was so simple and kind of left Zanadu in the dust. But the vision of Zanadu, this idea of proper attribution and rights management built in, that's still an extremely worthy vision. And we really love that vision. You know, that's basically where, you know, you look at all these different problems on the internet, and it comes down to the fact that, you know, there isn't built-in attribution.

You know, throw a rock at a vertical and you get a problem, whether it's music, etc. And Zanadu really was trying to address that, right? There's this great quote from Ted Nelson. HTML is precisely what we were trying to prevent ever-breaking links, no rights management. So anyway, that's the idea of Zanadu. We think it's pretty cool. And others have tried to solve this since, like, you know, for example, like WordPress has pinned back. So there's been other technologies developed proprietary or open source that enable this sort of like,

ability to see who's linking back to you. But yeah, having that embedded into the core technologies, I guess would have been pretty useful and valuable. Yeah, yeah. No, I agree for sure. So how does that tie into Eurovision, this idea of bidirectional links? Right. So directly, actually. So you can kind of summarize

what Zanadu is trying to do and what we're trying to do with we say, where's my stuff? And you can break that down to the where's and my stuff. And the wares is really, Zanadu solved it with bi-directional links that are sort of like created automatically

by the protocol on the fly when you create a new page. And then the my stuff that Zanadu had was baked in copyright. So just sort of everything built into the system. And there was these ideas of a bunch of things like transclusions and metastructures and so on. That was the Xanity design.

With Ascribe, we're also boiling down the problem we're trying to solve as, where is my stuff, but we're solving it slightly differently. So the where's part, rather than baking bidirectional links into a protocol, we're crawling the web. So we're swallowing the web and then automatically computing those bi-directional links for media. And so, I mean, several people have done this over the years, several groups. You know, in 1998, Larry Page was in the shower and he asked,

what if I could swallow the internet? And it was that question that led to him and Larry, sorry, him and Sergey, developing the backdrop algorithm which evolved into Google. And, you know, Google search itself is about, you know, understanding the links back and forth, you know, the page rank, all that. They were doing it for discovering content. We're doing it for attribution. And it really ties to ownership much more directly.

And it's not about a web page per se. It's actually about the media itself. So that's the sort of wares and the where's my stuff. The other half, the My Stuff part, is two parts. And I hinted at this in the beginning. There is the Easy Legals and the Secure Legal. So Easy Legals is a terms of service that has the copyright stuff built in. So when you register a piece of work, you are claiming that you have the copyright rights.

when you transfer ownership, you are bestowing a subset of those copyright rights to the next buyer, such that that buyer can resell, they can use it for public display, et cetera, et cetera. So to summarize, you know, how we relate to Zanadu is, Zanadu had this amazing vision to keep rights management to solve the where's my stuff question. We have the same aim, but with a scribe, we're trying to do it in a much more sort of pragmatic way. not trying to force new protocols right now. We're working with the existing protocols of the internet and then having this sort of overlay on top, sort of rewiring at the sort of the very top level to make it really pragmatic. And of course, you know, it's still hard, right? I mean,

to swallow the internet and then to relate that to content, that's actually an internet scale machine learning problem. But that's exactly what we're good at. You know, in a scribe, there's a dozen of us, you know, eight engineering and of the engineers, five of us have been doing big data and machine learning for a long time. So, I mean, this also seems sort of a crazy, ambitious project, right? And I think when you talked about that, I was like, is this really possible? And so with Google, like, one of the things people often talk about is that if somebody tried to build a better search engine than Google now, it would be extremely difficult because

of just that enormous server infrastructure and expenses that they've had. So you could like new startup costs would be huge. And you know, it's like of course still very risky. Are you able to build something better? So can you guys really do that? And how can you guys call the entire web and then find those links and attributions? And I presume also it means you would have to, you know, account for different file types, right? or like, let's say somebody has the same work image,

but in a different, you know, compressed versus not compressed. Like, I mean, you would have to recognize all those things. So how is that really possible? Yeah, so first of all, I totally acknowledge that this is crazy ambitious. I agree. And I also acknowledge that it's really hard. And we see, though, that solving ownership on the Internet, it's not an easy problem, but it really is worth solving, right?

It matters to the future of the internet. It matters for compensating creators as well as for the consumers, etc. So, yes, it's ambitious, yes, it's hard, but we're going for it. And sort of as a team and a scribe, we have the background to actually go for it. You know, the problems that I was solving in my previous work, we were actually looking at where we had 100 billion or even a trillion data points that we had to analyze in the span of 10 minutes or an hour on a laptop. off and we were doing it.

And it's not just myself in the team. There's other people on the team who have been doing really hard problems in semiconductors and in other big data realms. So if you have the experience and the audacity, then you really can go for it. And we kind of have both. So to your question of different file formats, et cetera, this isn't about just searching for the hash of the file because someone can change a single bit and then the hash is different. So you can't go based on file names or hashes.

you actually have to do a content-based search, right? And then basically the content-based search, you have to do this for different media types. You have to do it for images, for movies, for 3-D designs, and so on. Right now, we're doing it for two types of media. We're doing it for images, and we're doing it for 3-D files. Actually, I think we might be the very first company in the world

that's doing a 3-D-file similarity search. And we'll be doing for more as we go along. And so we're doing this, and really, you know, we have a given image we're comparing against when someone registers an image or they want to track what's going on. We actually compare that against the reference database of images that we've crawled

and depending how you count it, depending sort of how deep in the web you go, it could be 10 billion images, it could be 500 billion. And this is possible thanks to cloud infrastructure. So we're using Amazon very heavily. Actually, it's kind of funny, Amazon sort of tracked us down in Berlin

and they're like, who are you guys? Why are you using such a crazy amount of compute power, you know, like this tiny little startup in Berlin that's using more compute power than like super established startups? Like what's going on, right? But, you know, this is possible now if you have, you know, you leverage technology in the right way. You have the right team to really pull it off. That's what we're doing. And it really, you know, it's not just, you know, sort of image search or media search. It ties to ownership. And this is really the key. There's sort of this multiplicative value of when you actually combine those two things. because then you can start to ask questions like, okay, this is the file that I own. When has it been used in the last month by others? How many people have used it? Who's using it? What value are people getting?

And those are really useful. And then you can make a call yourself on what do you want to do about it, right? Do you want to just let go because virality is a proxy for value? Or maybe you want to charge license fees? Or maybe you want to do a take-down request. But it's up to you. And we're giving that power to the users, the individuals. And we think that's really powerful.

So does that mean if I'm an artist and now I'm going to register my digital work of art or something on ascribe? At some point in the future, you will provide like analytics services where you can see, yeah, what's happening with that? Where does that go? Yeah, that's correct. And it's in the very near future. So we're actually already doing this for beta customers, businesses, where they are using our service to basically compare their content against what's out there on the internet. And we will be rolling this into the web app for individuals in the next two to six weeks.

Actually, maybe sooner. So. Yeah, so a few months ago or a year ago or something, I was approached by a very large publisher who were interested in DRM and publishing and whether the blockchain could be some help there. And the interesting thing is that they weren't necessarily so much again. against the idea of people sharing the content and being passed on because, you know, they have, for example, for the newspapers, they know that an average three people read them, right? So they get passed on and family members or they're in a restaurant or something.

And then that's really valuable information for them to sell their advertising. And they were like, oh, if this could happen in the digital realm, that would be okay too, right? So if people pass it on, but we need to know. We need to have the analytics there and understand, like, where does it end up? what do people do with it in order to sell ads effectively. So I guess this is sort of, it ties in a bit there too, that you might be able to provide some of that intelligence. And then that's certainly really fascinating.

Definitely, yeah. You know, what we're doing is it's really sort of pro-creator across the board. So creators that are individuals who want to monetize individually, but also, you know, larger corporations that are proxies for creators to help that way, you know, journalists working for newspapers, etc. And so we do work with, you know, larger businesses as well to help them to monitor the Internet for their content or other content. You know, the ascribe solution scales to that, right? So, you know, we're working with companies that have tens of millions of users.

I'm curious, so you mentioned a while ago about the data analysis that you were doing. What, are you using Erlang for that? I do love Erlang, but we are using Python. Okay. Let's maybe get into some of the technical components of a scribe, and maybe you can describe a spool. Sure. So spool is the protocol, the blockchain overlay, for the transactions related to ownership. So when we developed spool, we actually looked around at all the different other sort of protocols

that had been emerging that stood on top of the blockchain, things like open assets and counterparty and all these. And, you know, those are pretty cool. We like that. We talked to some of those guys, but they didn't have quite the feature sets that we needed, and they didn't sort of address the law in the way that we needed to address the law here.

So in the feature sets, you know, we wanted things like unique additions. We want, you know, if you register where you have 10 unique additions, you need to be able to distinguish edition 3 from edition 4. Whereas these other protocols, they would say, okay, I've got 10 e-gold, right? and you transfer three e-gold at a time, but it doesn't matter which three in a sense. And then we also wanted things like loans, consignment,

you know, having someone else sell on your behalf, unconsignment, rentals, these sorts of things. And so these different features, the other protocols simply just didn't support those. And when we looked sort of, if you merged our protocol with another one, we looked at that as well, there was very little overlap,

really just sort of only the transfer part. So we decided that we would just have our own protocol called Spoole, Secure Public Online Ownership Ledger as a blockchain overlay, and the specification is actually quite simple. We've actually put it online onto GitHub. It's github.com ascribe slash spool slash spool. We'll put it in the link in the show notes as well.

Okay, cool, yeah. And so that's the general idea of sort of the features, and the main features are registering and transferring ownership. The other aspect, though, that sort of sets it apart from other protocols is that it's really not about transferring tokens that are sort of like a coin that's colored or something. It's really much more about time stamping ownership transactions. So it's much more of saying, I declare that I have this ownership now. This is not law. Law is law. The code is simply code, but it's code that could be used as evidence in a court of law.

So that's really how we're leveraging the blockchain. And then when you transfer ownership, it goes from, you know, one address that owns it. So there's sort of an address that owns a given addition, say addition number of two or three. So when you transfer, there's a transaction where it's going from the existing input to a new input to a new sort of address. And that's the transaction for transferring ownership. And sort of as this goes along, as you transfer from owner A to B to C, you have this sort of providence that emerges naturally. And it's really easy to look on a blockchain explorer and stuff too. So you can go to blockchain. Info or whatever

and see this chain of transactions going from person A to person B to person C. But all of this going back, it's not about some sort of token that gets passed along. It's much more about recording actions that can be used as evidence in a court of law. And actually is a really cool aside. The blockchain has already been used as evidence in a court of law. Maybe you guys heard, Andrea Santinopoulos was telling me it was used in the Silk Road trials, actually. Yeah. So we're like, thank you, Silk Road.

But it shouldn't really be surprising, right? Like hard drives have been used as evidence in courtrooms for, you know, decades. So something like the blockchain, which is, you know, even more, even less hard to tamper with, it should be obvious. So we're happy about that. Go ahead. So just to sum up how this works, a scribe is sort of a certificate authority then in that case. there's a hash that's embedded in the blockchain for the proof of existence of this file, for instance.

It could be like a picture or a 3D file. And then for every addition, you would have a transaction that is issued to the addition holder. And then as they transfer that, they're transferring ownership essentially. Yeah, exactly. So as long as you have the private key, you're the owner of this file. Exactly, exactly. So maybe I'll just describe, to be very precise, in that registration transaction, the input is an ascribe address. There are several outputs. Let's say you have three editions. Then the first output would be a hash of the file plus the title of the artist, plus the title of the work, the name of the artist, the title of the work, the year. And then you would have three outputs for each, one output for each edition. One output for each edition. One output. for change and one output for an operan, which is like a verb saying register or transfer, whatever. So that's what the transaction looks like. And yes, exactly like he said, Sebastian,

if you have the private key to say the address of edition two of three, then by definition, according to the terms of service, you are the owner of that piece of work. And yeah. So how does the loaning an addition or piece of artwork? Yeah, well, so I mean, when you're transferring ownership or loaning, these are all about bestowing or transferring a portion of copyright rights from person A to person B. So a loan looks a lot like a transfer of ownership. It's just more copyright rights that are more limited in scope. On the blockchain, the transaction is shaped very similarly. So you have the existing owner of the rights and then to a loan.

You basically have Satoshi's flowing from the existing owning address to a new address where they're getting the loan. And there is also change and there's also an operan that has the verb of loan. So but does that mean, I mean somehow it has to go back to the rights owner, right, if you loan it? And I guess, I mean, ideally, you would not have to rely on the person who is loaning the art to send it back to you for a transaction. Ideally, you could say, like, I loan it to you for three months and then at that stage, it automatically gets sent back to me. Does that work like that? Yeah, you're exactly right. And it's precisely because we're not relying on.

It's not, you know, coins that have to flow back and forth. This is all timestamps of actions, right? So inside the operater, it says loan, but it also says the start date and the end date of the loan, right? Because it's really, once again, documenting a contract that has already been signed that they've already agreed to. So it's not about if someone steals the private key and runs with it, you know, you're kind of messed up. It's not that at all. It's simply used as a way to document these different transactions related to ownership. So does that mean, for example, you could also restrict and what the person,

does that loans the object, for example, could you say you're allowed to loan it on or not? Is there some granularity you have there? So right now, we have sort of good defaults. So typically it's going from an existing rights holder, someone who has an addition or the artist, to a single entity. And they don't have the right to reloan just how our sort of default contract is. However, we've already began doing custom contracts for people that like to use us. One of our customers, for example, they're using it. I won't get any detail, but.

And in general, why not have custom contracts for the businesses or the individuals that want it, right? And as time goes on, too, we'll start with good defaults, but we'll also make it easy to sort of switch the contracts. Just like with Creative Commons, you can go to Creative Commons and choose one of a few contracts in the sort of easy to use fashion. We kind of see that with us too. Today's magic word is ownership, O-W-N-E-R-S-H-I-P. Go to Let's TalkBitcoin.com to sign in, enter the magic word, and claim you're part of the listener reward. So Spool is the open-source protocol that makes all this happen. A scribe is the for-profit company that is creating this protocol.

and maintaining it, right? Yep. Can you talk about the broader, like, full tech stack of what a scribe is going to be offering? Sure. And so just as an addendum to the spool stuff, in the next month, we will be open sourcing the reference implementation of spool in Python as well.

And we've actually been having chats with other people in the Bitcoin community about using the spool protocol themselves for ownership. I mean, it's sort of a complementary protocol to the other protocols out there, but it stands on its own and it's useful. So, okay, so I'm going to talk about the tech stack here.

I'll start at the top. There's two ways you can interact with Ascribe. There's with the web app directly. So if you go to ascribe.io, click sign up. Then you can enter your email and a password, and then you're in, and then you can register work, transfer ownership, all of that, it's really designed to be simple.

Incidentally, by the way, too, when you're using it, you don't have to know about Bitcoin at all. We provide some details down at the bottom, but there's Bitcoin being transacted. You don't have to worry about it. Then the other way that people can interact with the scribe is via our API. So this is typically for marketplaces or galleries with higher throughput, these sorts of places.

And so they just use our API. So both the marketplaces and our own web app consume this rest API, it's a standard restful API, we'll be making that public shortly too. And that is being served up by our ascribe servers. And then the servers themselves, they kind of have three pillars that they're kind of doing under the hood. And so pillar one is, and it all relates to answering this question, where's my stuff? You know, the where's part and the my stuff part.

So on the wearer's part, in order to, we do the automatic discovery of bi-directional links, and that comes down to the web crawl and then the machine learning similarity search. And, of course, it's accessing this database, this global database or file system, whatever you want to call it, called the Internet. So that's one part of the stack. The other two parts that the ascribe ownership servers are talking about. One of them is the terms of service, which we actually take very, very seriously. we've gone way over way to do a good job on this, working with lawyers on multiple continents.

And in fact, even though we have a small team, we already have a full-time lawyer who has lived and breathed copyright law the last 10 years. And he really, really cares about this stuff, about copyright, about privacy, all of these things. Actually, his last gig was doing a class action lawsuit on behalf of Canadians to Facebook for mucking around with Canadians' privacy.

So, you know, he's really fantastic. We're very happy about that. The third thing that the servers are sort of serving up besides the web call and the terms of service is they're wrapping the blockchain stuff. So there's sort of three pieces of that stack. There's the spool, which is the Bitcoin overlay. And then that is, you can think of it like a dialect of the Bitcoin protocol itself, which of course is the API for the blockchain itself. So that's sort of the overall tech stack. As far as users are concerned, there's really two ways they can interact.

There's the web app and the API. The web app is really for individual creators, artists, graphic designers, photographers, writers, musicians. And in that, they can register their work, can sign it, and archive. You know, we'll store those files on the cloud for free. They don't have to worry about that. And individual galleries can use this too for consignment and collectors as well. So you can think of it as your sort of digital wallet. of your digital art or your 3D files, whatever.

The rest API is really targeted towards marketplaces. And there's actually sort of two benefits for them. There's answering the where's my stuff question for themselves. And as well for their users, they can pass that information to the users to say, hey, this is how much your stuff has been shared. So yeah, that kind of, and then the third sort of, you can call it an interface, if you will, is the spool protocol itself, right?

Which like I've mentioned, yeah, we've opened source to we've gone underway to make it simple. And it's really about the transactions, recording the transactions, such that if there's ever a dispute of ownership, you've got extra evidence for a court of law. It's a really intricate system. I mean, you seem to have quite a few pieces there. I guess that's the value that you're providing is having all those pieces together. I suppose anybody can use this protocol to build.

similar business models. Yeah, and we encourage people to do so, actually. You know, I think it would be difficult to build a company where you're just hashing a file to the blockchain, right? You know, high school kids can do that. Our initial prototype was, you know, sort of that tiny, and it was, you know, straightforward to do. It was sort of a weekend project, right?

So, but it's really all this extra value that we provide, the web crawl stuff, giving that visibility. and reconciling this idea of hatching to the file with actual copyright law, with the realities of the world, and not just a copyright license, but being there, coming to bat with the legals to back it up, with the resources to back it up and so on. So it really is about an overall solution for ownership.

Another way to think of what is you can think of PayPal or Stripe as payment processing. Think of us as ownership processing. So let's talk a little bit about some of the concrete projects you've been up to. I mean, there's a few times you mentioned to me the website, Contemporary, which I think is really cool. Yeah, sure. So Cointemporary is a website and an art gallery that every 10 days, they sell a new piece of work, a piece of artwork. And it's run by a couple artists in Vienna, Valentine Rory and Andy Boot. and the only way you can buy it is with Bitcoin.

So here's an example where you can go online and you can spend Bitcoin and buy art. Now, up until recently, they had been selling physical art, but we've partnered with them, and they are now selling digital art as well. So the first round of digital art, they started about two weeks ago, selling 50 editions of some work from Valentine.

And this past week, they've actually been selling work from a quite well-known artist named Harm van Dendorpel. And it's actually a screensaver, but that's digital art. And so that's been pretty cool. We've been pretty excited about that. The guys have been fantastic to work with. And actually another sort of really cool aside related is these additions of Harms work

that are being sold right now, Mac Vienna, which is a 150-year-old art institution for including contemporary art, they actually bought copies of additions of harm's work. So we actually have seen instances of blockchain ascribed work being bought by these major art institutions already. So it's happening, right? It's pretty exciting. People have been dreaming and talking about, you know,

blockchain and art for a while. You know, we've been talking about it almost two years, actually. And it's happening now, right? With successes like Mac and with Cointemporary and so on, we're just really happy to be a part of this and to help bring proper digital ownership to the art world. So where do you see this going in the next 12 months? Or let's say one to two years?

Yeah, so right now we're spending a lot of energy in serving the art world. So we're working with a lot of digital artists and galleries and other art. larger art institutions, museums, galleries, gallery management software, these sorts of companies for digital art. And actually, interestingly, some have approached us and we're helping them on physical art as well. It turns out that having really, you know, better provenance for physical art can count for a lot as well. And on the side, by the way, we do certificate of authenticity.

That's cryptographic, so a cryptographic COA. And collectors really like that. So that's what we've been doing. But also we've been working with marketplaces and individuals. for photography and for 3D design, where they also have these challenges. And quite often they started with one challenge, like, oh, I don't have visibility into what's going on. But once they start working with us, they realize, oh, wow, this ownership thing is really awesome. Give it all to me, right? I want it all.

So we're working with sort of these three initial verticals and really proving it out with revenue, actually, which is kind of exciting for us as a company. And we're going to be really, you know, driving that. But as we go forward, we're going to be expanding to other verticals as well. But really always sort of art as our core as a sort of heart and soul, which really fits for Berlin too. Berlin is a company that is art plus tech. So solving a problem that is at the intersection of art plus tech is really exciting to us. And, you know, a scribe is kind of like that.

Now going beyond, though, we see that, you know, as we go address vertical to vertical and really build line share there and usage, then, you know, as you cover, as you cover the you start to address all the different media types on the web, all the different pain points in the web. So fast forward, you know, one year, five years, ten years, we see that, you know, we're already internet scale, but we see that we become sort of this horizontal layer for the internet, really an ownership layer for the internet. Just like Facebook started out as a site for Harvard students to like, you know, connect to each other and poke each other, but it grew to basically being this very horizontal social layer for the internet. We kind of see ourselves similar.

We're starting out solving some very specific pain points in certain verticals, but growing, serving needs of users, of creators to ultimately become this very horizontal ownership layer. And we think that's really awesome because the internet needs it. It's a hard problem. The internet needs it. And we're working with people who have been architects of the internet to actually deploy this. We'll be announcing some of those names in the next few weeks. I can't tell you now,

but we'll say soon. So that's kind of the vision. It's, maybe you can clear this up for me. It's unclear to me how, if a scribe grows, right, and you become sort of an authority for copyright and ownership rights online, how do we avoid getting into another situation where we have these large authorities that essentially have the monopoly on the market much like certificate authorities have now? Yeah, so you talk about it being horizontal, how is it horizontal and how does it, how does it protect the interests of artists? Right, for sure.

I guess those are two questions. What about the concern about centralization of an authority and then protecting the interests of creators? So right now in the registered transaction, we are sort of acting as an authority, but anyone else can also run the spool protocol and they can be the issuing address. And we're actually working with marketplaces

where they want to also be an issuing address themselves. And so we're evolving a protocol to support that. But it's always useful to know that there's a trusted point of origin. So we're not trying to lock that down ourselves per se, because that kind of goes against the vision of, you know, decentralization and supporting creators and all of that. You know, we can provide a lot of value that there's no need for us to lock that down, right?

And then the question of like, you know, serving creators. So in ascribe, we talk a lot about our values as a company, you know, we're wildly ambitious, but we also have to be cognizant. our number one value in a scribe is creators first. So we're probably going to run up against people who don't like that. You know, collectors saying, hey, I really want to serve my needs. But you know what? They've had a pretty good run with piracy.

And we want to see, you know, we want to reduce the number of stories we hear about poor starving artists who are creating great work, who people are stealing their stuff. And, you know, we just want to see them get compensated. Whether it's via additions, whether it's via tipping, whether it's via streaming. you know ascribe service will support all of those it's really about getting the creators properly compensated for what they do cool well I think this is a great great note to end on and it's certainly

an awesome and extremely audacious vision you guys are pursuing but I mean I think it's very exciting and I think we'll be following closely where is elites and what it can do to the internet and just sort of our experience with digital content into future. So, yeah, super excited about what you guys are doing,

and thanks so much for coming on today to discuss it with us. Thank you very much. It's really my pleasure. Thanks for the great questions. Okay, well, we'll have links for all the interesting readings from Spoo Protocol to ascribe, website, contemporary, etc. in the show notes. And for listeners, thanks so much for listening. You can follow us on Twitter with Epicent of BTC, and we'll be back next week. Thanks and I'll see you then.