Brief summary
Establishing who someone is and how much they can be trusted is hard — even more so on the Internet. Here, our podcast team explores some new concepts in digital trust, such as decentralized identity, privacy protection and identity ownership.
Podcast transcript
Mike Mason:
Hello and welcome to the Thoughtworks Technology Podcast. My name is Mike Mason. I am one of your regular hosts and I'm here today with Zhamak, who is also one of your regular co-hosts. Hello Zhamak.
Zhamak Dehghani:
Hi Mike. Great to have another episode together.
Mike Mason:
Yeah, it is great. So today, we are going to be talking about trust and the challenges of establishing trust between things and entities and people, especially on the internet. With the current design of the internet, we're going to be talking about various initiatives and efforts that are in place to progress how we deal with trust. Today, we've got two guests on the podcast. I'm going to ask them each to introduce themselves briefly here. So we've got Ken Adler. Hello Ken?
Ken Adler:
Hey there. How's it going? I'm Ken. I've been working on the OpenNet and then later the internet since the late '80s and I've been involved with identity-related technologies pretty much that whole time. I ran a publicly signed PKI in the early 2000's which got me all ready to read Satoshi’s white paper in 2009. And I have been a CISO in a previous life. I joined Thoughtworks about four years ago with the DPS group, where I was the security pillar owner of DPS. And over time, just because of how Thoughtworks and our clients and the needs, I've increasingly been focused on identity and the related authorization type of engagements. And I am an early member of the Trust over IP Foundation, where I serve both on the enterprise and the patient ID ecosystem task force.
Mike Mason:
Awesome. Thank you, Ken. And we will be getting to some of those topics in much more detail. We're also joined by Shangqi Liu. Shangqi, do you want to introduce yourself, as well?
Shangqi Liu:
Hi everyone. I'm Shangqi from China. Currently, I'm the head of Blockchain at Thoughtworks China. And also, I pair with Mike and Zhamak working in the Thoughtworks group creating those off the radar. I think there's been quite a big trend on the digital identity recently. So also, my work has a lot of engagement in into that. So I'm so glad to be here to share some thoughts and experience around this decentralized identity.
Mike Mason:
And although I kicked off the podcast today, I would say I'm the least knowledgeable person on the podcast. So I'll be asking, hopefully not too dumb, but entry level questions when everybody else gets a bit too technical here. So Zhamak, why don't I hand over to you to kind of introduce the topic and kick off.
Zhamak Dehghani:
Sure. We could start in so many different places. I think it's such a rich domain. But I wanted to kind of go a little bit further kind of backed away from technology, maybe start at a more philosophical place as, what is trust and how do we trust each other, at least in this scope of digital world, right? On internet. How as people or users of services with trust services, how the services trust us and reflect on that a little bit and how we've established it. Because, as Shangqi you've mentioned, we are at such a pivotal moment to redefine some of those fundamental assumptions that we have made about establishing trust.
Zhamak Dehghani:
And to be honest, I think the internet is a little bit broken when it comes to trust protocols and approaches. If you think about it, the way we establish trust is protecting a bunch of person we identify of attributes behind a password and trusting that those personally identifiable information that I share with service providers to prove who I am, I'm not sure I'm uniquely the only person I know, that information has been breached probably multiple times. And a lot of people know about my, I am assuming, about my social security number and so on. So, assuming identity based on a password protect, that set of attributes, that identity controlled by central or a federated set of services, I think it's rather flawed and should be challenged.
Zhamak Dehghani:
So maybe we start with a little bit of definition of kind of trust. What do we mean by, a service trusting a user, or a user trusting a service. What does that mean to each of us?
Ken Adler:
Trust is really, it's both an emotional and a logical act. And trust is the belief in the truth, the reliability and the ability of others. And belief, we get really philosophical about this. What is the definition of belief, and it can be confidence based on something, it can be faith, and it can be just hope. So again, it's really the belief in the truth and the reliability and the strength and the ability of others. And I think, in sort of this scope that we're talking about, it's helpful to think about trust outcomes. What is it that we're chasing here, and I think if you take it at the highest level, to be a bit mercantile with it, trusted eCommerce interactions. And underneath that, you sort of have trusted ecosystems. And then underneath that, each individual organization, within the ecosystem, you want to believe that there's a trusted governance structure, something that you can trust, you can look at, that one way or the other, you have defined this trust in this other organization, hopefully as a logical act. But certainly a lot of trusts, especially these days is more on an emotional level of brand trust, things like that.
Shangqi Liu:
Yeah, I remember, there's a book called the Liars and Outliers, it's quite famous in the information security space. And it's interesting, because it’s Chinese, translated into the title is, our trust. So basically, that's a book talk about trust. It discuss different angles around the trust, like how do you steal trust from the social perspective, from the moral perspective, the reputation perspective, and then the system and mechanism perspective. But all of this is not as simple. Yearly, technology is the way we are really trying to establish trust with certain set of technologies to do that. But the answer is not that simple.
Shangqi Liu:
Yearly we see trust is more than a simple mechanism. We need more collaboration. So we need to consider different parties, different roles in your ecosystem. And based on this assumption, to do our assess on the trust. That's my understanding on the trust.
Zhamak Dehghani:
Yeah. One of the, I guess, find definitions of trust that I found, it's very hard to actually define it, is from this author, Rachel Botsman that does a lot of work on trust, and who you can trust, using one of her books. And she says, "Trust quite simply is confident relationship to the unknown, establishing relationship transaction with uncertainty." And the way I think we have traditionally try to, I guess, get established that relationship with unknown is by turning that unknown to known by defining this concept of digital identity, right? Authentication is such a foundational pillar in any transaction we do, is that you say, you are who you say you are.
Zhamak Dehghani:
So if we pick then identity as a way of, kind of making the unknown a little bit more known, I like to kind of walk down, the historical landscape of how we have defined identity or digital identity traditionally, and perhaps what your thoughts are, where it's broken, and how we can address some of the gaps or the broken bits. And Ken you've been around this topic for a long time. And you can maybe bring some of that historical perspective for us.
Ken Adler:
So first identity, let's define identity, right? So identity in the biggest scope is really sort of the oneness of a person, right? It's everything that the person conceives and union belief is that there's sort of two aspects to your identity. There's a sort of the inwardly focusing identity anima, and then there is the outward focus aspects of identity persona. And as we know, what we're talking about here is really about the persona, the way that you project yourself in different environments and contexts.
Ken Adler:
And historically, the identity and the trust tying them together, historically if you go way back, it's really very social based. And the mechanisms that you use to establish both the identity and the trust, are physical things. Your face, the way your face looks, for your identity, and for the trust the expressions that you may put out. And then over time, guess both your identity and your trust has changed to be more sort of institutional based, where, certainly trust you trust in the church, and somebody else is a member of the church, and so trust is done that way, and then you can get into identity politics and all of that-
Zhamak Dehghani:
You really went far back, Ken. You went from the face to church. Come forward, come to the digital world.
Ken Adler:
Absolutely. The digital world is really where you're relying much more on knowledge, where you're dealing with both physics and math. And what we're talking about today is really about programmable trust and programmable identity. And one of the things that's interesting about this, is that it does allow you to become less centralized. But I still might, I would caveat that you would, there's still a heavy institutionalized aspect of the crust and the identity there. As far as actual technologies, techniques, obviously there's a variety of things, when you're talking about the identity of a person or identity of a service. So username passwords, services, their certificates, certificate authorities, and in modern times, in more modern times, there's a heavy reliance on cryptography, to identify individuals and services and such like that.
Mike Mason:
So Ken, just as the outside guy here asking you a random question, I have no idea how you're going to answer this one. But I remember back in the day, you could get Unix systems to talk to other Unix systems using basically completely unencrypted protocols, where if you were able to open a port on this particular socket from your machine to another machine, it would trust you, because it had been told to do that. So I could log in over there as me because some network service somewhere was doing that. To me today, that seems hopelessly naive and it obviously is because we're now talking about all the crypto and stuff that was layered on. I'm curious, was that ever thought of as not hopelessly naive? Or all the unencrypted stuff that existed on internet version one, was there a notion that this is not going to be good enough?
Ken Adler:
Well, you know what they say, this is why we can't have nice things, there's a lot of jerks out there. And certainly there's a saying about the internet was built without an identity layer. And because really, the technologies really weren't there at the early days. And thank God it wasn't built with an identity layer that we didn't get stuck into something that was not ready for today. But now, we've got to right that wrong and build an identity layer.
Zhamak Dehghani:
Shangqi on the identity story, I wonder what your thoughts are on the advancing from kind of centralized way of establishing this identity to federated an application of that and where you see that that's going?
Shangqi Liu:
For me I think we have been in the age of federated identity, because previously I ran into issues when I had to create another one email address, trying to have siloed user accounts and to remember different password. I still think it's very hard to remember different passwords. And now, like federated identity, I can use my Gmail or WeChat ID in China to log in different systems. I think it's much better than others. However, I still feel like even I can use one logging account but my information still stay siloed in different commercial internet companies. I cannot easily extract my information and to do some information share. I just feel it’s hard with the current setup.
Zhamak Dehghani:
Yeah, I think just so many things are gaps that exist within, even with a federated way of establishing identity as in relying one identity holder to give access to that identity on my behalf, to a variety of services. Of course, that's removed a lot of friction but still, there are a lot of challenges. And one of those challenges, as you mentioned, is sharing information across the boundary trust boundaries, right? How do you know that now, this information that you've shared across this trust boundary with another organization, how that's going to be used, and consent management, and that whole category of not having visibility to the information that you've shared and how it's being used, and being able to not only, I guess, give consent, but also reveal consent in some situations.
Zhamak Dehghani:
What are the, and I know we're kind of building up to this decentralized identity story. And so self sovereign identity. But I want to pause here for a minute, on some of the kind of challenges that exists with federated or centralized identity.
Ken Adler:
Yeah. I agree with you. There's a number of challenges that we've had throughout this sort of whole progression from centralized identity where identities in a single database in the monolith to federated where we're getting a little bit better, to hopefully, the future of decentralized. But main problems on the first two is, first is the most obvious thing that there's an intermediary involved, there's an institution involved, and that institution can change the rules of the game on. And it can take away your access, can change how it uses that information, you give it to them, how they share it, how they sell it, or whatever. So you've got that issue, you've got an issue that ultimately, it's a scaling problem.
Ken Adler:
And it's not only a scaling problem, because it's not going to be as big as Google is, it's not going to cover every place, everybody. But it's also the more you get stuck into one and you as an individual, there's a switching problem. So, the cost to switching and giving up one becomes larger. The other thing is typically using the technologies that we use for centralized and federated, which is username passwords as your credentials, they're not independently verifiable. They're not cryptographically verifiable. And as you mentioned, there's other things like to use federated identity, of your verifier of your website, and you want to start using it. You already are using Google and Facebook, but now you want to use Amazon's. Now you've got to integrate and do Federation, with Amazon and there's a barrier to allow more players in there. And of course, it allows for tracking so that central entity can tell where you've been and all the different places that are requesting authentication.
Zhamak Dehghani:
Yeah, I think the last one I've been particularly annoyed. Recently, I think I've clicked on maybe one or two ads recently working from home, buying things to make myself happy and now internet is unreadable for me. Every page is just littered with related ads that on a different platform maybe on Instagram, I clicked on one ad and now they're all popping up on, I don't know my YouTube and I have to change my browser, like move to brave browser because I just can't read the internet anymore without seeing stretchy pants all over it.
Mike Mason:
Here's a question. You mentioned usernames and passwords, right? Password seem like a bit of a rubbish idea given that it's 2020. But what do we replace them with? What is the post password will look like?
Ken Adler:
I was going to say the same thing, great segue. It's all the stuff that we're talking about in the rest. So certainly there's been a long history of security professionals, screaming, kill the password. And basically in one form or another, a lot of it comes down to this asymmetric cryptography where you have two sides, two related keys. And as long as you protect your private key or whoever is in control of that private key is the assumed identity in this particular case.
Ken Adler:
So I always think, when we talk about decentralized, I think one of the existing technologies that we're familiar that sort of give a hint towards decentralized identity, I think about SSH access. People are very familiar with SSH, right? SSH into a Unix server, and you've put your public key up there in your Unix server, but their private key normally, or in the canonical case, you're generating yourself. And you're holding on to that. So that sort of hints at where we're going with this centralized identity.
Zhamak Dehghani:
I heard this interesting, kind of metric on passwords, it takes from Christopher Allen, who's the co author of TLS SSL, and he's a big advocate since 2016 and maybe before on self sovereign identity. He mentioned that if you have an 18 character password, which is probably a good secure password right now, it takes 25 dollars on Amazon to break it. So it's just that level of protection is just the bar is coming lower and lower, with more compute power.
Mike Mason:
Is that true, really?
Zhamak Dehghani:
That's what I heard.
Mike Mason:
25 bucks for an 18 character password, wow that's incredible.
Zhamak Dehghani:
I have to double check, hopefully everything is coded before publishing this.
Ken Adler:
I think you can buy it if someone's gotten it already. If they fished it from you, you can buy it. I'm not quite sure of that study there.
Zhamak Dehghani:
Shangqi sorry, I interrupted you there. Take us there, we want to jump into decentralized identity and self sovereign identity. Tell us more.
Shangqi Liu:
Yeah. I think for most of us, the decentralized identity, we learn this could be adopted first from the crypto space and if you see, like Bitcoin, easier the recruiter of currencies people regard their simple wallet address or public key, in some format of public key as their identity. That's interesting because that's quite different from how people usually looks like identity. People think of identity as some profile in my Facebook with some information, but actually all you need is a unique ID that could identify yourself, plus a lot of other profile information like you can make this ID as your Bitcoin address and to transfer money in.
Shangqi Liu:
The key difference I would say, between the centralized identity and the decentralized identities is the wallet address is purely generated a bios. So like Tencent, you could generate your private key locally without connecting to the network, without connecting to any commercial internet company. You can even try to use your brain trying to figure out one, two, five, six lines of numbers to calculate your private key. That's quite a powerful thing. That means you don't need any sort of hideout circuitry to generate your identity. You only need some mechanism of mass to do such. I think that's quite a disruption to people's mind that you actually could own your own identity without third party.
Zhamak Dehghani:
Interesting. You used the language ‘own’ there. One of the things, I really like that now that you as an organization, or as an entity, as a person, you control the generation of that identity and you're not at the mercy of some other organization, whether tomorrow you will have that identity or not. But there is a warning around the language of ownership as identity becomes a property, and if you're a property, then you're subjected to kind of a state, federal, whatever property management. So there's a little bit of a warning around, kind of using the ownership. But I do completely agree that's one of the foundational principles of this essentialize identities that is under your control.
Mike Mason:
But then surely the trade off, like on blockchain, the point of it was to enable decentralized, not under government control, exchange of funds. And if all I'm doing is generating a key pair, and my public key becomes my wallet address, that's great if all I'm trying to do is store funds in there. But if we then say that's my identity, it doesn't mean anything, because I just made it up. And I could have 10 of them. So where does the identifying portion start to actually come in?
Zhamak Dehghani:
That's a really good point. I think that's a good segue to maybe talk about kind of this different layers of this protocol, that together all of those layers considered together, would give you that sense of identity or sense of verifiable attributes or characteristics that as a person or organization you have. Just on the definition, maybe just pause for a minute as the definition of, I've googled this before, so for disclaimer, the definition of a decentralized identity from Drummond Reed, who Ken Adler knows, the founder of Sovereign and big evangelists, I guess in this space, is lifetime, portable identity for any person, organization or thing that does not depend on any centralized authority, and can never be taken away. So I think that's just kind of sums up what decentralized identity is. And if the readers or listeners haven't read Christopher Allen's 10 principles of self-sovereign identity, you should go and read it. It's just beautifully put 10 principles to just define what this thing is, and why.
Zhamak Dehghani:
And then I think that now that the question is, how does this generated long number digit character thing would actually turn into representation of me as a person or organization? Ken, do you want to have a go at that?
Ken Adler:
So I think there's a couple things here. Maybe one way to think about it is let's just zoom out a little bit. So first thing is that, yes you're going to, let's start with connections. So first thing to understand is that in this scenario, when you're talking about self sovereign, decentralized identifiers, there's a lot of peer-to-peer stuff going on. So first of all, the DID itself is not a public key, per se. In fact, the DID standard doesn't even say how you generate that key. But the thing is that, it is globally unique, and that it's associated with a dead document, which contains the metadata. And they don't say where you got to put the document, there's a couple different ways you can do it. But generally, you don't try to put it on the blockchain or on the public ledger.
Ken Adler:
So the relationships between me and you, for example, or there's four of us on this call, so there's three other people, the idea that I use for my communication with you and Mike is going to be a different did than I use with my connection with Zhamak and with Shangqi as well. So as an individual, you don't have just one digital identifier. You have large numbers of them, many of them and they're pairwise and your keys and your public keys, all of that stuff is off chain. It's not on any kind of blockchain or public place or semi public place where people can see it. The only things that goes on the public chain are the identifiers of an issuer of a credential. With someone that is like the government that wants to issue your driver's license using this stuff, has something that other people have to verify that this person, that this organization is who they say they are.
Ken Adler:
So you zoom a little bit more in and you say, Well, what do I need with decentralized identity? I need protocols for the secure connection, how do I establish these secure connections between me and Mike? And then I want to exchange data over that connection. And so I need some protocols and some standards around sort of watermarking that data, so that data can also be verified cryptographically, that I was really the person that actually sent that, this data to you, and that it wasn't messed around in transit. So you need these protocols around the data itself. And then if you want it to be useful outside of just you two, and especially you're using public issuers, you want to have some kind of public key registry, or public registry, which we typically talk about as being either blockchain or some other distributed ledger technology, or it could be other things. It could be other peer-to-peer kind of networks.
Ken Adler:
But the trick is that, it shouldn't be a central identity, anybody who's commissioned on to that should be able to verify things independent of a centralized identity. So maybe that's a good first shot at it.
Zhamak Dehghani:
Yeah, and maybe to just put that in context, Shengsi, I know you guys are working on a digital wallet implementation. I know COVID was a big driver for establishing some of this infrastructure, do you want to share maybe some of the work that you're doing to put this kind of stack of technology that gives us this decentralized identifier and a set of attributes that can be verified about us into context?
Shangqi Liu:
Yeah, for sure. Actually, recently we are working on the digital wallet, which is one of the key features is decentralized identity. We are following up very famous initiative, which is the COVID-19 credentialing initiative in this space, trying to use decentralized, I would say now decentralized identity, but the verifiable credential is very important part of the decentralized identity. Using the verifiable credential to provide a format of health certificate for normal people or for healthy people, because in China, you go everywhere, you'll need to provide your certificate to prove that you are healthy, you're safe. Otherwise, you cannot go to a shopping mall, you cannot go to your office.
Shangqi Liu:
So currently, we're still using a very traditional technology trying to extract data from your service provider and to extract your traffic information to know you haven't been to the dangerous area or like the dangerous area that has a lot of COVID patients. So it's important to provide such kind of claim to prove that you are safe, you will not infected anyone else. However, we need a better way, which could build up a better privacy if you wanted to leak less personal information. I think verifiable credential is such a technique. So basically, it is established on decentralized identity. And like for individuals, you could have your own identity. But also you'll need the collaboration of some health authorization like CDC, which can give you the claim you'll need to request the third parties help to issue you a claim in a format of verifiable credential.
Shangqi Liu:
This verifiable credential is a digital format with certain fields and there are ways to do the cryptographic verification. Third party could verify that this credential is signed by the Health Care Center and this verifiable credential could be transferred And installed at your mobile wallets, basically, you can carry it everywhere. So from now we have these different solutions of digital certificates, but none of them can be verified cryptographically cause basically, they are just a digital paper, you still need to link back to your authorities to do the check. By the way, this verifiable credential actually, we can just use the, like the authorities public key to do the verification, to verify if this certificate is still valid.
Shangqi Liu:
And another amazing thing about this verifiable credential is that it allows you to leak the least information that you require. Like if I want to prove I'm healthy, all I need to prove is I'm healthy, I haven't been to the dangerous area, dangerous place. But I don't need to provide every place that I've been to, which is a I believe is a covenant status. So I think the verifiable credential could help to leak the least information that that we need.
Zhamak Dehghani:
Interesting. If I summarize, it looks like we're moving away from establishing trust by providing a whole lot of information overly unnecessarily maybe, and then protect those with password and really move back to the physical world, right? We have a wallet, I have a wallet, and I carry my driver's license, and maybe the certificate that I've done the tests. And I know my bank cards, there's a whole lot of certificates that are claimed about myself that I carry my wallet today. And we're talking about the super wallet now that carries that, and a whole other set of certificates that I have been given by issuers or institutes. But the fun part about is that, in different places, like I go to a mall, and they need to verify that I am healthy and I'm trustworthy and I don't have COVID, and I didn't go to dangerous places, I can only provide the bits of information that is relevant to that verification and not more.
Zhamak Dehghani:
So it's a privacy respecting way of kind of sharing information by cryptographically, verifying that the actual issuer that gave me, there is a trust with that issuer. That's kind of like a super wallet that I now carry all of that information in.
Shangqi Liu:
Yeah, exactly. I think when it comes to the trust, we do not need to exchange error data, we just need to exchange the information, we need to prove that we are trustable.
Zhamak Dehghani:
Which makes me wonder, so what is missing to have this really as a widespread implementation and get that widespread adoption. I love this kind of idea of narrow waist of internet or narrow waist of protocols, right? This hourglass kind of shape of technologies. We're in the middle of the hourglass and narrow waist, you have technologies that are enablers, right? IP was there and on top of it like TCP and an HTTP and all of these applications and maybe today's like the HTTP in that narrow waist on top of it, having that kind of standard, you can build so many tools and applications on top and so many technologies to support it.
Zhamak Dehghani:
I wonder what is in that narrow waist of new trust stack that would enable the wider adoption of verifiable credentials and digital wallets as a way of establishing trust?
Ken Adler:
Well, I think you can look at it, sort of in two stacks here. First of all, there is the technology stack, and that is emerging now. Man, it's sort of like the early days of the internet when, yes, there were a number of standards or standards were coming out. But there's a lot of proprietary stuff floating around stuff that hasn't become standard yet. So therefore, it's proprietary or non standard. So there is a lot of motion in the industry. However, having said that, if you wanted to get something and up and running and demonstrating, you could do it today with the standards and they semi standards and other hacks that are around to demonstrate the thing and that's going to continue to progress over time.
Ken Adler:
I think the interesting thing is that at the end of the day, it's not just about the machines. It's not just about cryptographic trust, but there's this human trust aspect of it. That Shangqi was talking about, also about, hey you trust the government, the health care providers stamp that, hey, you passed your test and stuff like that. So there's a whole bunch of governance that has to take place and has to be widespread and accepted in order for these different ecosystems to be created. And that's one of the exciting things around the trust over IP foundation. Is that the trust over IP Foundation, which is highly associated with a number of the other sort of main organizations, and we can talk about the, do a quick rundown of the different organizations involved in this industry.
Ken Adler:
But the trust over IP Foundation has a dual stack, they have a technical stack, and they have a governance stack. And the stacks are divided in four layers. And it's very nice, just like the IP stack is in four layers, it allows you to swap things in and out that each layer. But generally speaking at the first two layers of the stack, you're dealing with cryptographic trust, and to higher level, you're dealing with human trust. So one interesting analogy I heard was the analogy of roads, cars, passengers and cities, I thought that was a really neat analogy. So the roads is layer one, and there's regulations about how the roads are constructed, or how the railroads are constructed. And so that would be your public utilities, and all the regulation, not regulation, but if you wanted to belong in this ecosystem that you would follow, you would use an Indie blockchain, or you would use an Ethereum blockchain or whatever.
Ken Adler:
Then the next one up would be the cars. And as you can imagine, there's a whole bunch of regulations around car manufacturers, the safety, regulations, and in different places, those things are different. So, somehow you'll need governance, if you want to perform or want to interact within an ecosystem of people, you'll need to make sure that you're using the right cars. And then you go up higher, and you think about the passengers. Here's what the analogy is a little bit thin, but I guess, to me it would be like, okay, you don't drink on this subway. So there's regulations around how the passengers and then finally, at the highest level is your city's right? Is how, is your city regulated?
Ken Adler:
So this environment, you're dealing with your public utilities, sort of your server to server, your DID. The third level would be around your VCs and how you encode your data, what kind of schema you use within your VCs, your verifiable credentials. And the highest level is sort of the ecosystem governance of, if you want to belong to the local ecosystem put together by the local Chamber of Commerce, that you agree to behave in certain ways, all up and down the stack. So maybe that's helpful. So the short answer is the governance. Integrating both the human trust and the machine trust to me, in a unified governance stack is pretty interesting.
Zhamak Dehghani:
You're probably one of the very unique individuals that would consider governance, interesting and exciting, but I can't imagine how necessary it is in this space.
Ken Adler:
With emphasis on necessity. As opposed to necessarily interesting.
Mike Mason:
Okay, well I think we are running towards the end of time on the podcast. What I will ask, first of all, where can people go to find out more about this stuff, if they're interested?
Ken Adler:
Well, the way I look at the organizations, there's a whole bunch of different organizations involved, so let me try it this way. So if you're for just community around identity and where ideas sort of bubble up and come, there's a great thing called the internet identity workshop IIW, it's been going on for 15 years. That's where OpenID connect came from, and a lot of this stuff came from their self sovereign. So IIW, there is around technical specifications, there is the decentralized identity foundation DIF, and so those are working groups coming up with specifications. And then generally speaking, those specifications, then to become standards, they push it through the, they go to the W3C. So there's standards in W3C, there's the verifiable credential standard already out, and the DID, soon to be out through W3C.
Ken Adler:
If you're interested in actual software, open source software, and implementations around here, a really good place to go hang out is hyper ledger, which is with the Linux Foundation, and they have a variety of software there. And if you're looking for a public utility, that's out there that you can connect into a network that you can connect in and start using today. There's the Sovereign Foundation and the Sovereign Network. And the last place are reserved for really being the first place, the trust over IP Foundation. I found that a very useful place to get a hold on it, because again, on the technical side, the trust over IP foundation doesn't try to create new tech, it really is an accumulator, and a curator of technology coming from all of these other entities and others, including the IETF and so on.
Ken Adler:
Plus, that's where all, even though there's governance activities happening in different places, like in sovereign and DIF, but really where the winds in the sail around establishing governance frameworks and templates for them, that people can go off and create their own governance frameworks around, that's trust over IP. So those are sort of the main places that are, there's a lot of information around this topic.
Zhamak Dehghani:
Shangqi, if I understand correctly, I think your team, the blockchain group in China has put a fair bit of open source software out there as well. Is there anything that worth for folks to look at?
Shangqi Liu:
Yeah, recently we just released our mobile wallet, which is called the T Wallet, shortfall servers as well as trustworthy. And we do follow the W3C DID standard, as Ken mentioned. So basically, there's a bunch of things here. And another thing I think I need to point out is that, recently in our tech radar, we have blips that mention decentralized identity, trust over IP, and we gave out a reference. So for anyone that hasn’t subscribed to the Technology Radar, please go to see the radar.
Mike Mason:
Okay, awesome. Well, I'd like to thank very much, my co-host Zhamak for having all the actual information here because I just asked some questions. And to Ken and to Shangqi for being on the podcast today. Thank you.
Zhamak Dehghani:
Thanks everyone, it was wonderful.
Ken Adler:
Thank you.
Rebecca Parsons:
And on the next episode of the Thoughtworks Technology Podcast, Neal Ford and I will be joined by Stuart Holloway, who is very active in the Clojure ecosystem, is a member of CogniTech, which was recently acquired by NewBank and we'll be talking with a story about the state of the Clojure ecosystem, how the languages evolved and some interesting developments in that ecosystem. So please join Neal and I next time on the Thoughtworks Technology Podcast.