GRTiQ Podcast: 16 Chris Wessels

The following podcast is for informational purposes only. The contents of this podcast do not constitute tax, legal, or investment advice. Take responsibility for your own decisions, consult with the proper professionals, and do your own research.

Chris Wessels (00:00:22):

I think that as blockchain begins to eat more and more of both our fundamental institutions, but also the kind of products and services that we just use every day, they increase the total addressable market of The Graph.

Nick (00:00:44):

Welcome to the GRTiQ Podcast. Today I’m speaking with Chris Wessels, an Indexer at The Graph. In addition to his Indexer operation GraphOps, Chris is a well-respected voice within the Indexer community. He is also a Wave 1 grant recipient from The Graph Foundation for his work on the Indexer agent. Our conversation is incredibly insightful. We discussed Chris’s departure from FinTech, his thoughtful description of what DeFi is and how it might impact the world. We also have a fascinating conversation about Ethereum, Layer 2 blockchains and gas fees. We concluded by discussing Chris’s long-term vision for The Graph and where we currently are in its evolution. I began the conversation by asking Chris about his background.

Chris Wessels (00:02:03):

I’m a technologist. I’d say within the realm of technologists I’m definitely more on the generalist side of things, have always just been fascinated by interesting problems and trying to understand and reason about them. I was born in South Africa actually, and very early on I was absolutely taken by computers and the vast world that existed within them. And so from quite a young age, I started playing with the computer and that led me into programming quite young. First starting with silly little scripts back in the days of Windows BAT files and really just things to tickle my imagination and understand this new world better. That led me into games, and at first it was kind of playing games with friends, but very quickly I think I became more fascinated by how these games work than playing the games themselves. So, I started writing games at quite a young age and then from games I went into websites and then from websites into networked applications and all fueled by this desire to understand and just curiosity.

(00:03:22):

Yeah, so this takes us to very early high school and at this point I’d started a business and we ran essentially a web agency in South Africa and across high school completed about 20 projects of varying complexity, e-commerce websites, brochure sites, things like that. And was I suppose from a young age I’ve always been interested in technology and computers, but I’ve also always had a bit of an entrepreneurial spirit. And so after finishing high school, I went immediately into the software industry and professional programming at first in Cape Town, but it wasn’t long before I felt like I had exhausted the opportunity space in Cape Town. And so at 21 I moved to London to seek out bigger opportunities and more interesting problems to work on. I’ve been in London now for about seven years and in that time I’ve mostly worked on startups and within the startup ecosystem spanning quite a wide range of industries, both as a founder, as an employee and as a consultant.

(00:04:38):

I went through Entrepreneur First in London and since then founded a few businesses along the way, raised lots of venture funding, hired lots of people and built out teams. Failed a lot of times. And it was really in the depths of building out a business within the regulated financial services sector that I really kind of encountered DeFi for the first time seriously and was really engaged with it with a depth of knowledge of how the traditional banking and payment system works. And it just completely blew me away and I really knew that I needed to focus on it.

Nick (00:05:22):

I want to come back to DeFi in a minute, but can you tell us when you first got involved or interested in Crypto?

Chris Wessels (00:05:29):

So, I first encountered Crypto painfully early actually, so it must have been maybe 2013 or 2012 when I first started mining Bitcoin on my laptop CPU back in the day when that was totally a viable way of mining. And then thought, this is awesome, and proceeded to forget about it for the next couple of years and format my hard drive and lose all of my coins. I also had the opportunity to invest in the Ethereum ICO, so I was around then and engaged enough to see that happen. I bought a pair of speakers instead. So, painful decision in hindsight. So, I suppose all through that time Crypto and really what was Bitcoin back then was definitely at the periphery for me and something that I was aware of and tracking from a distance, but not something that I’d really engaged with.

(00:06:32):

Fast-forward many years and I found myself in London as a CTO of a banking as a service company and had really spent the time engaging with the traditional payment system and working with the card networks and the banks and even the European Central Banking level. We got regulated, we did all the right things, at least in hindsight to deliver what feels like quite a simple product. And it was at that time that I seriously engaged with DeFi and back then it was really the very early days of MakerDAO and kind of Uniswap B1 coming shortly thereafter. But it just really hit me how through a number of factors, partially just that the technology was fundamentally superior for building and representing these kinds of systems systems. But I think also the lack of regulation, which in my experience within FinTech really stifled innovation and served a strong function of keeping the existing group of entrenched businesses in a strong position of power and making it really difficult for innovative competitors to come in.

(00:07:56):

It just struck me that DeFi was just this incredible change over that and in a matter of weeks of deeply engaging with the technology, I went from the opinion of what we’re doing here is working on the forefront of finance and FinTech is an amazing force to change the landscape to actually realizing, well, we’re working on the past here, quite literally what we’re building will be leapfrogged by a new class of protocol built on blockchains like Ethereum. And I think subsequently I’ve come to believe that it really will be Ethereum at the center of a lot of this. And so yeah, that was the point at which I decided I need to take a step back. I need to focus on this because now the Pandora’s box is open, I can’t close it.

Nick (00:08:53):

I want to ask you more about DeFi and it’s a little bit of a jargon or buzzword in the crypto space. How do you think about what DeFi is?

Chris Wessels (00:09:01):

I think it always depends on how much time there is to have the discussion and how genuinely interested the other person is. For me, I like to think of the implications of blockchain almost from first principles. And in that sense, I think understanding first principles allows one to understand the impact of blockchain on finance, and that frames what DeFi is, I would say.

Nick (00:09:29):

Well, I love that answer. So, let’s do it. Take us through what DeFi is starting from first principles.

Chris Wessels (00:09:36):

I think the key realization for me was that, I think when most people think about blockchain, they think about it as a digital technology and it is a digital technology, but it’s also so much more than that. I think blockchain is fundamentally a complete revolution in social technology. And this speaks to what I was saying earlier about the breadth of definition of what technology is. I think so many people think technology equals digital technology, but if we kind of wind back the clock on humanity, something like discovering and being able to manipulate fire was an incredibly transformative technology for humankind and really shifted our trajectory in a way that was completely unimaginable. And similarly, if you’re talking about small groups of humans and tribes and coordinating socially to serve some broader purpose, you run into this idea of Dunbar’s number, which, if anybody doesn’t know that, is a number that was proposed by I think a British anthropologist.

(00:11:00):

And it is theorized as the maximum number of people that a single individual can have a meaningful relationship with. And it’s theorized to be a cognitive limitation. And Dunbar’s number is 150 or that’s commonly used as Dunbar’s number. So, as you have these groups of humans and they’re attempting to coordinate and function together as a small society, we invented another piece of technology, a social technology, which is the institution. And the institution is this really useful piece of abstraction. It allows us to dissociate the role of something from a group of human beings or human beings that we connect with on a personal basis. And we formed this relationship with this abstract thing that is the institution and that technology is really what allowed human beings to scale societies beyond Dunbar’s number.

(00:12:10):

So, institutions are an incredible technology in the grand arc of humanity, but unfortunately, and maybe this is somewhat informed by my experiences and perspectives growing up in South Africa, but I believe that unfortunately given enough human beings and given enough time, institutions naturally evolve to become misaligned with the mandate that gave them their stature and credibility within society as a whole and ultimately transition into extractive layers that no longer really serve the public good or even private interest that they were architected for. And so that is for me where blockchain comes in and I see blockchain as a fundamental revolution in social technology.

(00:13:18):

And yes, that’s enabled through this digital technology, but the implications extend so far beyond that, because we can move the operating frameworks and rules by which these institutions on mass, we’re talking about the most substantial abstractions in the world that we live in, we can transition those rules and operating principles from the meet space, which is subject to the corruptibility of human beings into smart contracts and into essentially code where that code runs at a global scale and is not corruptible and can be designed to serve its participants in a way that cannot become extractive or is far more anti-fragile in that sense. And I think that was one of the major revelations I think that caused crypto to become a one-way street for me is I realized that this was an opportunity to work on something that had enormous implications in the context of humanities development and arc as a species.

(00:14:40):

And so I think that is the frame that I like to think about when I’m trying to assess how blockchain can impact any single industry. And so DeFi is obviously the impact of blockchain and the … I think DeFi is what happens when blockchain begins to eat finance as it exists in the world today where it largely exists as a group of interlinked institutions run by human beings and unfortunately in many layers incredibly corrupt and extractive and not serving the public good mandate that they were originally created for.

(00:15:29):

And practically DeFi is about permissionless inclusion. Anybody can access these services from anywhere, nobody can prevent them from doing so. And that’s not just on the demand side of the market where you have customers that are looking for these services, but it’s equally on the supply side, which is really something that it was a stark realization for me that the existing financial incumbents essentially have regulation as a moat and a moat that on the surface seems like it’s there to protect consumers, but in practice does an incredibly good job of insulating the incumbents from competition. I find the implications of that on many of the most important industries and institutions to be incredibly exciting.

Nick (00:16:28):

Well, that might be one of my favorite answers to what DeFi is I’ve ever heard, but I want to ask a follow-up. So, a lot of the listeners to this podcast aren’t particularly working in Crypto, so they’re sort of outside the space. So, for the average everyday person, how will DeFi impact them?

Chris Wessels (00:16:48):

You can come at it from different perspectives. I think fundamentally I’ve heard the sentiment that Bitcoin is DeFi, and I’m not sure I’d agree with that entirely, but I do think that it hints at one of the most powerful components of what DeFi is, and that is access to monetary systems and services and assets that have hard properties. And a particular set of reading that really kind of informed my perspective on this note is Ray Dalio’s writings both in his books around Principles and how to Navigate Big Debt Cycles and then his more recent work on the Changing World Order. If you zoom out and look at the history of money and financial services as a set of institutions that are really there to serve the needs of human beings, money is really just an abstraction that allows us to coordinate more easily.

(00:17:57):

We have a really bad track record of building systems that are anti-fragile and that serve the interests of the everyday person in the long term. And particularly given our environment now, I think we are in an incredibly interesting time in terms of the international monetary order and the hegemony of the US dollar. And I think at the base layer DeFi and blockchains just as a way of having everyday people having permissionless access to an asset class that is not subject to the same manipulations as our existing systems is huge. And that may not be the case if we were at the beginning of a grand supercycle or some kind of reserve currency cycle, but particularly where we are now where the US dollar and the US is really being challenged in its role, I think this is an important thing for everyday people now.

(00:19:12):

And then on top of that, I think that people don’t just want idle assets. So, much of the value that exists within financial services isn’t just about representing and trading assets, but about the productive value and the real utility that can come from those assets. And that for me is what DeFi is. And I think that the fact that everyday people will be able to access real productive utility on their assets without anyone being able to prevent them from doing so or changing the terms under which they do that is incredibly exciting and is incredibly empowering for the individual in this evolving environment. But there’s so many layers to it.

Nick (00:20:08):

Well, I believe you, I know that’s true, and I think you did a really great job there. So, let’s turn our attention back to you a little bit. Why was it when you had, it seems like a lot of opportunities and you were on a career track, why did you decide to pursue a vocation in crypto?

Chris Wessels (00:20:26):

I think it just appealed deeply to me in terms of being an interesting and fascinating intersection of all these different disciplines and really an opportunity to understand more about the world in a way that kind of had practical benefit and real world implementation benefit. But other than the interesting challenge of the problems, something that really attracted me to the space was the culture of the community and the ecosystem. And I do think it’s important to say that crypto is an incredibly diverse bunch, and I certainly don’t agree with every subset’s perspectives and positioning, but something that I have always valued tremendously is how much the community values innovation and rigorous intellectual discourse. And I think something that really exemplifies this is how maybe 50% of the key contributors, at least in DeFi are anon or pseudo anon folks. We have no idea who they are, where they are.

(00:21:54):

And it always just struck me as an incredible thing when you can have an ecosystem of people that are so unconcerned with things like who you are or what you look like or where you’re from, and just so focused on progress and on innovation and on advancing the discussion. And I cannot think of any other industry that wouldn’t reject that with a degree of pretense. I can’t imagine finance or any number of other industries having a serious conversation or engagement with a bunch of anonymous people. And so for me, that really just exemplifies where the values within the community are. And I think that that environment is predictive of hugely impactful and innovative change at a global scale.

Nick (00:22:52):

Do you remember when you first became aware of The Graph?

Chris Wessels (00:22:54):

I think I first encountered The Graph somewhere around mid 2019, maybe early 2019. This was kind of after I had begun to investigate DeFi more in depth. And it struck me that there was this incredibly interesting activity burgeoning within DeFi. We were making the very real transition and progress from Ethereum mostly being the rails to move around ICO tokens that had basically no value to building real applications that created this productive utility for assets. But it was very difficult to, in concrete terms, reason about this activity and measure it and track it. And so, one of the projects that I had considered working on was essentially something to solve this problem, something to essentially stream changes as they come off of the blockchain and index data in a way that would make it accessible and easy to represent, even in the case of dashboards, to help educate and bring awareness to this activity that was burgeoning. And it was in that context that I first came across The Graph and then sometime later towards the end of 2019, I joined Consensus.

(00:24:28):

I’d made the decision that there was absolutely nothing else that I wanted to work on more than blockchain and the Ethereum ecosystem and Consensus seemed like a really fantastic place to be able to impact and have real input into the development and evolution of the ecosystem, but also retain quite a broad perspective of what was happening in the market. And I landed up working on a similar set of problems within Consensus around extracting data about the real activity and utility that was being created in these burgeoning DeFi platforms and delivering it to, in the case of Consensus, institutions that wanted API endpoints to fetch this data and stick it into their ML algorithms or whatever and crunch it. And so it was again at Consensus that we really looked carefully at The Graph, and I’m grateful we landed up going this way because it really helped me to be convinced at the value of The Graph.

(00:25:35):

But we landed up working on essentially an alternative to The Graph, a real time data pipeline that was capable of streaming data off of the blockchain as blocks were mined and transforming it and normalizing it into these higher order data schemas that could then easily be queried and leading the teams that built those solutions and really getting into the meat of the problems that needed to be solved to do that. Whether you’re interested as we were in quite a holistic sense, or whether you’re an individual project that’s looking to materialize information about your own system, it really gave me a firsthand experience that the challenges associated with doing this.

Nick (00:26:26):

I have a follow-up about that. So, what did those challenges or that experience teach you about The Graph?

Chris Wessels (00:26:33):

I think it made me appreciate the legitimacy and difficulty of the problem, and I think there were two or three realizations that I had about The Graph that were enough for me to become convinced that it was going to be the solution that won in this layer.

Nick (00:26:59):

So, then I guess you decide at some point you want to make the move and get involved in The Graph community. What was behind that move?

Chris Wessels (00:27:07):

So, I think first and foremost, I had firsthand experience of the problem that The Graph was trying to solve. And by that experience I knew that it was a real problem and that there was tremendous value in whichever solution ultimately won that race. So, there was that. I think the next thing was that around me, all of these incredible DeFi protocols were being built out and The Graph really had unparalleled traction in DeFi. It was like, I have this DeFi protocol, I want to materialize higher order information about the activity within it. What are my solutions? Well, it really came down to The Graph or roll your own. So, they had incredible traction and I think traction, there are a few better substitutes than traction. And then I started to do some thinking around the problem and had some realizations about The Graph’s approach that made me very much a believer.

(00:28:12):

And I think the first was realizing the incredible power of building a credibly neutral open standard for this problem or as a solution to this problem. And I, a few years earlier, had spent some time building out essentially a Kubernetes alternative. This was before Kubernetes had actually launched. If anyone doesn’t know, Kubernetes is a container orchestration platform and really revolutionized how we deploy software. And Kubernetes itself was built on top of Docker, which is this piece of software for making containers. And it really struck me that what The Graph was doing was very similar to what Docker had done, and Docker is really just a way of taking some components that you need, kind of like your runtime environment to run a given piece of software and package them up into this little blob of data that you could then ship anywhere and run on any server.

(00:29:24):

And it really totally changed the game for how software is deployed and orchestrated. What was curious about Docker is that the underlying primitives, the underlying bits of technology that made Docker possible had been around in the Linux kernel for a decade plus. And all it really took to create this spark was for Docker to create an open standard, truly open source software that was in the community for the community and eventually by the community too to solve that problem. And they had incredible traction and success with that. And I think that generally when you have credibly neutral open standards, if you can rally your community around that and you know can essentially rally the kind of collective brain power trying to solve that problem around an open standard, then you create this snowball and everybody wins. It doesn’t matter if you have competitors that are working on the standard.

(00:30:36):

I think it’s a matter of the whole becomes greater than the sum of its parts. And this was something that I recognized about what The Graph was doing specifically in terms of subgraphs. So, subgraphs were this open standard, this open packaging format essentially to bring together all the individual pieces of the technical problem. Things like your smart contract ABIs and your mappings and your schema, creating this open package to the put all these things together into one open standard that would allow anyone to develop subgraphs and solve this problem. So, that was the first thing. It’s very difficult for any centralized competitor to compete with a credibly neutral open standard. And then the second slightly more practical thing, which is really just an implication of the first is that this effectively executes an inversion of control.

(00:31:42):

And what I mean by that is before The Graph, there were solutions, companies that effectively offered this solution as a service where you could go to them and say, “This is my protocol, these are my contracts. I want essentially this schema, and I want you to materialize that schema and maintain it and make sure that I can query it and I’ll pay you for that.”

(00:32:12):

And that worked, right? And we still have companies that operate that way successfully today, but a key challenge there is that it puts the onus of developing the solutions for that individual customer or product on this centralized data company. So, you go to them, you say, I want this solution. They ask you some questions and they go away and they build it and then you get it and you can use it. And I think when I realized that subgraphs as a credibly neutral open standard were that, that’s awesome first and foremost, but it also then repositions that responsibility from developing essentially the components of the subgraph, all those things it packages up instead of requiring a centralized entity to do that work, you have an inversion of control and whoever is actually building the product, whether it’s a DeFi product or an NFT product or whatever, who naturally has the best understanding of their own domain, they understand their system the best, they probably understand their customers the best.

(00:33:32):

It allows them to develop all of the components required to deliver that data solution. And not only does that mean that the quality of that component and the likelihood that it meets the brief goes up rapidly, but it also removes the centralized bottleneck, right? If you have a thousand of these projects all coming to the same company at the same time, there’s going to be a huge bottleneck and the ability for that business to deliver all of those projects is going to be severely limited. But in this model with the subgraph inversion of control, individual developers, individual projects and companies are empowered to solve that problem themselves and not have to solve the vast technical problems at the periphery like database management and index optimization, and they don’t have to worry about any of that. They just build the few pieces that are critical to getting the right data off the chain and putting it in the right format. And that little package then can be shipped off to any Indexer who can serve those requests and those requirements.

Nick (00:34:48):

How would you explain your rationale for deciding to become an Indexer?

Chris Wessels (00:34:52):

For me it was a matter of leveraging existing skills, specifically around large scale systems and infrastructure and running and maintaining big systems that serve lots of requests. These are skills that I’d developed in my professional career and feel tremendously relevant in the context of what it means to be a good and effective Indexer. So, there was a really good natural fit with my skillset, but I think beyond that, being a core participant in the network has always felt like a tremendous opportunity to impact the evolution of what I see as one of the most important protocols in the web3 space and zooming out or thinking even more holistically if we think of things like The Graph as these new age institutions, a new way of collaborating and coordinating at a global scale around a common objective, we’re really talking about working within systems that are completely new and innovative in the context of our history as a species. So, the opportunity to be intimately engaged and involved in a protocol like that at this early stage is just absolutely tremendous and something that really attracted me to getting involved as well.

Nick (00:36:28):

In addition to being an Indexer, you also received a Wave 1 grant from The Graph Foundation. Can you tell us a little bit about that experience?

Chris Wessels (00:36:36):

Yeah, I was, and that whole process was really fantastic. Generally, I found that interacting with The Graph Foundation has been incredibly accessible and they’re generally a very friendly and welcoming group of people. I applied and went through a small series of interviews just so that they could better understand the scope of work that I was proposing, and generally they were very supportive and happy to offer additional support in any way that they could. I would highly encourage anybody who is interested in contributing to The Graph to have a look at the Foundation grants website and get a feel for the process involved. It really is quite light touch and focused on funding the right valuable project, so there really isn’t a lot of red tape or anything like that.

Nick (00:37:34):

So, tell us a little bit about the project you received the grant for

Chris Wessels (00:37:37):

My grant is focused on Indexer agent, which is a component of the software stack that indexes run. It is responsible for interacting with The Graph Protocol smart contracts on the Ethereum blockchain. And in short, the scope of the grant is to do some work which optimizes the gas efficiency of Indexer allocations, and that work has already resulted in about a 23% gas cost reduction for Indexer reallocations.

(00:39:37):

In preparation for this interview and trying to learn a little bit more about your background, I stumbled across this notion that so many of my prior guests, it was really a Ethereum that drew your attention to the crypto space. And so I guess I’d like to ask is that a fair characterization? Was Ethereum the hook that brought you into the crypto space?

Chris Wessels (00:39:57):

So, my entry, like I think many of us was Bitcoin, but the problem with Bitcoin is that all you can really do with the Bitcoin blockchain is move around coins from one wallet to the next, and people don’t just want to hold assets, they want to make real productive use of those assets. The problem with Bitcoin in that context is that to do anything of productive value with your Bitcoin, you need to move it off of the Bitcoin blockchain and in the case of the Bitcoin ecosystem, mostly into any number of fully centralized, fully custodial services to make productive use of your Bitcoin.

(00:40:57):

And that always just seemed totally antithetical to me because the properties provided by the blockchain are entirely useless in that context, and we effectively just recreate the same system where the assets that are being used productively are just held and controlled by a small number of institutions. What really blew my mind about Ethereum was this ability to construct more general purpose applications that were capable of creating utility and productivity for assets and even beyond assets, if you’re talking about things like nation state scale voting or any number of other coordination problems, you can create real value and you can encode the operating principles and frameworks of how that value functions and created and is sustained and have all of those rules be subject to the same scrutiny and immutability and security as just the asset itself.

Nick (00:42:15):

Okay, I want to come back to gas fees for a minute here. In the context of Ethereum and certainly within the context of this podcast, a lot of listeners have interest in the future of gas fees. How should we think about that?

Chris Wessels (00:42:28):

So, people obviously want ultra low costs for transacting, and that’s fair and it’s something that we should seek to optimize for in the long term. We need interacting with these decentralized applications and protocols to be as cheap as possible so that access is not restricted to those that can pay lots to interact. But transaction revenues are actually a really important component of the incentives that ensure security for our blockchains and the vast majority of that security today, and this is less the case for Ethereum, which is another great indicator of why it’s such a strong network, but generally the vast majority of these incentives are provided by block rewards. New currency issuance to pay for that security. And longer term we want transaction revenues to be high enough to really provide a strong and consistent set of revenues that ultimately result in incentives and security spend by the actors like miners or eventually proof of stake stakers that provide real security to the network.

Nick (00:43:43):

Well, I’ve never really thought of gas fees in this way, so I really appreciate that answer. So, then how do we resolve this conflict between the idea that gas fees help secure the Ethereum network and this constant push within crypto to somehow lower gas fees?

Chris Wessels (00:44:03):

So, I think the answer is Layer 2s, maybe not Layer 2s in isolation, things like Sidechains as well are aligned with this vision, but I really think Layer 2s are going to be the major catalyst for the reduction of gas costs for end users. What makes Layer Twos really interesting is that they move the computational aspect of transacting off of Ethereum Layer 1, and all of the data relating to the transactions that happen still gets settled onto Ethereum Layer 1. And this means that Layer 2 still inherit the incredibly strong security properties of Ethereum Layer 1, but because that computational aspect, which is really the majority of the gas cost that is incurred when transacting is moved off-chain, you’re able to pass that set of gas cost savings onto the user. So, as we see Layer 2s launch in the coming months, it’s going to be really exciting to see applications migrate into these new environments and to have users and capital follow.

(00:45:18):

And I think Graph protocol is just as many other apps a prime candidate to move certain aspects of the system’s operation into Layer 2. That will drop gas costs for all participants in including Indexers, Delegators, Curators, but without any reduction in the economic security that exists today on Ethereum Layer 1. Now, beyond that, I think there’s quite an interesting opportunity that emerges if you follow the train of thought, which is as applications and users migrate onto these different Layer 2 solutions, the data that sits underneath these systems, the data that is critical to understanding the activity within these products and protocols and users’ positions within them is then segmented or fragmented across many different execution environments.

(00:46:26):

And this further complicates the problem that The Graph is seeking to solve. If I’m a developer and I don’t have something like The Graph today, accessing the data or for my application on Ethereum Layer 1 and creating a data model to reason about it is difficult enough as it is, but as my applications and users migrate into many different locations, that data just becomes even more difficult to access, unify and reason about, and this really cements the value proposition of The Graph.

Nick (00:47:07):

Okay, we’re definitely coming back to that observation about The Graph and its value proposition, but before we do, I got to be honest, I’ve never really understood why gas fees exist, who receives them, why we pay them. Can you help me understand that?

Chris Wessels (00:47:22):

Gas fees exist, I think fundamentally to protect the network from spam. So, if transactions didn’t cost anything, then an attacker could feasibly overwhelm the network and bring it to its knees with zero cost. So, gas fees are actually an important part of preventing the network from being spam attacked in that nature. Where gas fees actually is going to change over time. So, today gas fees go to miners and they form part of the incentives that miners receive to spend money on equipment and energy and secure the network as a result. And this touches on what we were speaking about earlier around how transaction revenues are a really important part of security incentives and that cryptocurrency networks should really be striving to maximize those incentives in terms of transaction fees as opposed to block rewards which are effectively just subsidized with supply inflation.

Nick (00:48:37):

Great, that’s very helpful. And it brings up a second follow-up question, and I’ve never really understood the difference between Layer 1 and Layer 2 blockchain, but something you said there prompts this question, which is one way to think about it, that Layer 2 blockchains exist somehow to deal with gas fees or to speed up the transactions. How do you think about that?

Chris Wessels (00:49:00):

Yes, I think the definition of a Layer 2 is that it is a scaling solution that inherits the full degree of economic security from the Layer 1 that it kind of branches out of. And then subsequently batch settles updates into. I think Layer 2s really bring two transformational changes from a user experience perspective. The first is the one that we’ve spoken about, which is by virtue of moving the computational aspect out of Layer 1 and doing that in an environment that is a lot cheaper, that significant component of gas cost is removed. And so those cost benefits can be passed onto users of the Layer 2, and that just makes transacting a lot cheaper and significantly increases the economic scalability of the Layer 2. But the other very important benefit that Layer 2s bring is much faster state advancement. And what I mean by that is the rate at which state advances on Ethereum Layer 1 is the block time.

(00:50:27):

So, roughly every 13 seconds a new block is mined and all of the transactions in that block are committed into the state of the system. And so if I’m a user and I’m interacting with Ethereum Layer 1, I can submit a transaction and in the best case I’m going to get feedback about whether that succeeded or failed in 13 seconds. And as we try to build out web3 products and experiences that are not just on par but even exceed the experiences of web2, that 13 second feedback loop is just not acceptable. And so, one of the great things that Layer 2s do is allow what you could call block times to become very short. So, the feedback loop for indicating to a user whether the transaction that they’re attempting has succeeded or failed and then showing them the results of that can get cut down from 13 seconds all the way down to half a second or less. And from a user experience perspective, I think that that is as important and impactful as just reducing cost as much as possible.

Nick (00:51:45):

So, you’ve really got me thinking here about gas fees and L1 and L2 blockchains. I want to ask then, is your vision of the future that L2s will be more important? Are L2s the future of blockchain?

Chris Wessels (00:51:59):

Yes, I do think so. I think that the definition of Layer 2 is going to evolve slightly. I think fundamentally what makes Layer Twos just a superior scaling solution to anything else is that they inherit that very strong set of security properties from Ethereum Layer 1. But I think if you play time forward, the vast majority of the applications that users care about will be located on Layer 2s and you know, might even see a future where users never really interact with Ethereum Layer 1 directly, they just have no need to. That’s not where the applications that they care about are located. And there’s no real downside to only using Layer 2s, because they’re fast, they’re cheap, but they still come with those strong security properties.

Nick (00:52:58):

So, if L2s are the future of blockchain, how does The Graph fit into all of that?

Chris Wessels (00:53:04):

If L2s are the future, then the data problems that The Graph solves today only become worse for developers and users trying to reason about their activity on-chain and what will become on many chains. And so the more that Layer 2s and other solutions like Sidechains are the future, the more that the value of The Graph as a network is cemented as a way of having a unified model for accessing querying and reasoning about that data.

Nick (00:53:39):

All right, Chris, this is all very helpful and I feel like I’m thinking more clearly about gas fees Layer 1, Layer 2 blockchain, but I need to throw Ethereum 2 into the mix now. So, in the context of the discussion we’re having, how should I think about the difference or the impact of Ethereum 1 and the move to Ethereum 2?

Chris Wessels (00:54:00):

I think that the community as a result of the messaging or positioning that was put out at some point has got quite hung up on what feels like a very stark difference between Ethereum 1 and Ethereum 2. I would say that the reality is more that Ethereum follows a roadmap that allows it to evolve and be upgraded over time, and really it’s the same network. The most fundamental change in Ethereum 2 is really just that we change the consensus mechanism of the network and we plug out proof of work and we replace that with proof of stake. And that comes with an enormous set of advantages in terms of hugely reducing the environmental and operational costs of maintaining that security. So, quite literally the cost to provide a given level of security to the network is going to drop drastically once proof of stake is launched.

(00:55:05):

So, that’s really fantastic for the entire network and it also brings a bunch of other really great properties like faster finalization of transactions and things like that. I think if you look beyond proof of stake and you try to bring together the idea of Layer 1 and Layer 2 and Ethereum 2, essentially what roll-ups do is use Ethereum 1 as it exists today as a data availability layer. And what I mean by that is the transactions that is practically a bunch of code that runs and figures out a result in answer are no longer actually executed on Layer 1 unless there’s some kind of dispute or something of that nature.

(00:56:00):

And all of that computation is moved off of Ethereum Layer 1 and it happens off-chain in the Layer 2 and therefore doesn’t incur all of the gas costs associated with that computation, but to ensure that security is maintained even though those transactions aren’t executed on Layer 1, the transactions themselves, i.e. something like Bob sent 10 ETH to Alice, those are recorded on Ethereum Layer 1 even if they’re not executed there, because that ensures that the state of the Layer 2 can always be reconstructed and validated and all of the economic security that goes into Ethereum itself also secures that data.

(00:56:50):

So, you can kind of think of Layer 2s as using Ethereum Layer 1 as a place to save and access data about what happened on the Layer 2 in the past. What Ethereum 2 brings post proof of stake and as we introduce data shards is many different data shard chains that each has a data capacity of its own. So, you can imagine if the amount of data in terms of bytes or megabytes or whatever per second of data that the chain can actually come to consensus on and secure, if that is X and Ethereum one is capable of X and that kind of naturally results in scale constraints for Layer 2, because Layer 2 is really just saving data onto Layer 1, but Layer 1 can only handle X amount of data per unit of time.

(00:57:54):

Then that creates natural scale limits on how much can happen in Layer 2, because all of that needs to be saved back to Layer 1. And as Ethereum 2 introduces data shards, each shard has a capacity of X. So, if there are 64 shards, then Ethereum as a Layer 1 network can support 64 times more data capacity in terms of megabytes per second of data. And that extends into what roll-ups are capable of because suddenly the amount of intermediary state, all of this detail about what has happened inside the Layer 2 that can effectively 64x as well, because we now have so much more space on the Layer 1 to save that data.

Nick (00:59:21):

I want to move to a part in the podcast where I ask guests to define important concepts or ideas. So, let’s start with The Graph itself. How would you describe or define what The Graph is for somebody who’s non-technical?

Chris Wessels (00:59:34):

So, I would describe The Graph as a fair marketplace that brings together actors on both the supply side, people who index data and can answer questions about that data and people on the demand side, people who are seeking those answers and the rules of this marketplace, which is also a protocol are enforced by the blockchain, and that means that all of its participants can be sure that the protocol will not become extractive and is a true public good and its objective and its governing rules and its operating principles are geared towards serving its participants as well as possible.

(01:00:21):

Beyond that, I think The Graph is a revolution for data markets waiting to happen. It’s really powerful when you can create open networks that also have strong incentives, because incentives drive so much behavior and I think the incentives inherent to The Graph protocol are going to drive a huge amount of innovation and unification in the availability and cohesion of data in terms of global data markets. And I think these kinds of pressures, competitive pressures and incentives are going to build better global data businesses and I think that that is really exciting when compared to our environment today where data is incredibly siloed and these systems that allow it to be productively used to generate insights are equally siloed. And I think The Graph is a real force to change that.

Nick (01:01:28):

So, if we were to compliment that definition or that description, how would you answer the question of what makes The Graph special or unique?

Chris Wessels (01:01:35):

What’s special about it is because it is one of these neo institutions like protocols as institutions, it can coordinate this marketplace in a way that optimizes for the outcomes that are important to its participants. It can be a public good in that sense. It can facilitate this commerce of people selling answers to questions and people buying those answers in a way that is not extractive at all and promotes rapid innovation and development and competition, which really leads to the best outcomes for that network’s participants and customers.

Nick (01:02:28):

Another concept I always ask guests to define or describe as that of a subgraph. So, let’s do that. How would you describe or define what a subgraph is?

Chris Wessels (01:02:38):

Subgraphs are an open standard that allow web3 developers to solve their own data needs. They do this by writing only the bits of code and technology that are essential to their specific data problem and then they package all of that up into the subgraph. That subgraph can then be loaded into the Indexer software to serve an open API that anyone can use to get the data they need. So, in a way, I think you can think of a subgraph like a international shipping container. What’s inside the container, the cargo and how that cargo is chosen to be packed and padded and made safe for transit can vary significantly. And this is also true of a subgraph. The contents of the subgraph will really be specific to the data challenges associated with that problem.

(01:03:44):

But like a shipping container, once that subgraph is packaged up, it can be loaded onto any truck or ship and treated like any other shipping container. So, this standardization and abstraction brings huge benefits because it effectively delegates each part of the problem to the party that is best suited to solve it. So, the specifics of the data problem are solved by the subgraph developer who knows the most about their problem and the specifics relating to indexing all of that data and serving requests for it at scale are solved by the Indexer, but each set of problems is isolated from the other.

Nick (01:04:32):

It’s crazy to think about and you just described it there, but there’s conflicts with the data models and the blockchain and how to query this data. So, all the way down to the smart contract, how do you think about that?

Chris Wessels (01:04:44):

When you are writing smart contracts, which is really what these web3 applications are, the optimizations that you are making are largely about cost reduction because blockchains, as we know are expensive to transact on and they’re not really the place to store vast amounts of data. So, when you are structuring the storage of that data on the blockchain, you’re making decisions that optimize the structure of that data to be as low cost as possible and the set of trade-offs that you would make if you wanted that data to be really easy to reason about and ask questions is quite different. But the implications of that set of optimizations if you were doing it directly on the blockchain are that it would be insanely expensive to transact. So, it’s like if you can recognize that the data model that smart contract developers are optimizing for is very different to the data model that you would optimize for if you wanted to very easily reason about and query the data are very different data models. And so the subgraph is effectively the glue between those two data models.

Nick (01:06:13):

So, how does The Graph Network fit into all of this?

Chris Wessels (01:06:16):

So, I think of The Graph Network really just as a coordination layer for all of the participants in the system. Practically the subgraph describes how to fetch the data of interest from the chain, transform it into the structure or format that best solves the problem that the subgraph developer was trying to solve. And so the Indexer takes this subgraph and runs it and as it runs, it effectively pulls that data off of the blockchain and transforms it and then it actually gets saved into a database that is operated and controlled and managed by the Indexer.

(01:07:08):

So, you could almost say that all of the data in The Graph Network is essentially held by all of the indexes that exist within it. And then you have mechanisms like the proof of indexing mechanism that allows the network to validate and ensure that if you have a large set of Indexers and they are saying, well, I’m indexing this subgraph and I therefore I am storing the data for this subgraph and allowing people to query it. The proof of indexing mechanism allows the protocol to ensure that all of the Indexers supplying that data have acted in good faith and truly have all of the data that they claim they have and that the integrity of that data is 100% and therefore when queries are distributed to those Indexers, the protocol is ensuring that they are serving the correct information.

Nick (01:08:11):

I’m going to also ask you about the role of Delegators. In your opinion, how important are Delegators to The Graph ecosystem?

Chris Wessels (01:08:20):

Really important. I think there’s naturally a separation between those who have capital to deploy into ecosystems like The Graph and those who have the technical expertise thus required to operate a competitive Indexer. And without that separation in the network model, I think that would really drive centralization and generally undesirable characteristics. So, I think separating people who have access to capital and are willing to deploy it into an ecosystem and are willing to make well-informed choices and those who actually operate the indexes just makes a lot of sense to me as a design decision. I think Delegators also surface a really interesting intelligence in the market. They signal to the market what sentiment or confidence is for a given set of indexes, and I think that’s also an important data point to consider in the broader perspective of who is doing well and who isn’t and how do we drive the network towards participants that are maximally additive and effective in their role.

Nick (01:09:39):

What advice do you have for Delegators when it comes to selecting an Indexer to stake their GRT with?

Chris Wessels (01:09:44):

I think there’s going to be a lot of tooling that comes out soon to make this kind of decision easier. I’m naturally somebody that’s drawn to the data and want to take a data-driven approach to it. And I think right now there’s a fairly limited set of reliable indicators around who is running a good operation and who isn’t. I would certainly say don’t bet solely on APRs or APYs, especially if they are subsidized by negative effective cuts from the Indexer. Those are never sustainable, you would question any business that operates at a loss. I think beyond taking a more holistic view than just APRs, I think at this stage community engagement is one of the best ways to assess the competency of an Indexer. So, I definitely look at community engagement and one of the best ways to assess that is to pop into Indexer office hours, which happens weekly on a Tuesday.

(01:10:52):

This is a great way to be part of the conversation or even just to listen in on what Indexers are doing, what are the active concerns within the network at any given point in time, what challenges are being faced and what solutions are being proposed? And I think you’ll find that Indexers that are engaged and active in that context really understand the challenges of operating an Indexer within The Graph protocol. And I’d say that that is a great signal to understand who is worth delegating to.

(01:11:27):

Outside of index or office hours, this can be challenging because the data isn’t always easy to get at, but as the network becomes more dynamic and network conditions shift more rapidly, things like new subgraphs or version upgrades being posted or things like that, I think who is first to move on those opportunities and who responds quickly I think is another great in indication of engagement and how competent an Indexer is. And yeah, then as we approach the post migration phases where we have real query demand coming through the network, I think the ratio of query fees relative to self stake will be a great indicator of how well an Indexer is leveraging their stake in the network to meet demand levels and get paid for queries.

Nick (01:12:27):

On this topic of query fees, I want to ask a follow-up question and it’s just something I’ve been thinking about, but consumers of The Graph’s services have been using the hosted service for free and now as subgraphs move to the main net, they’ll have to pay query fees. So, the question is, do you think there’s any reason to be concerned that maybe people won’t be willing to pay query fees?

Chris Wessels (01:12:51):

I don’t think so. Not in a way that represents any kind of existential threat. I think that there will be some adjustment period. People like getting things free, but I also would not underestimate the tremendous network effects that have been formed around the subgraph format. I don’t see many teams as an alternative undertaking huge engineering cost to attempt to solve this problem in some other way. I think they’re far more likely to accept that there is a real cost basis to operating these services and once the hosted service disappears, it’s not like there’s any other venue where you can really get these answers free. So, I would predict that the vast majority of applications will adopt the decentralized network. I think another thing to realize is that if you play through your head what a developer might go through, they’ve had this service free for some time and it’s met their needs incredibly well, they have sunk very significant engineering hours and funds into developing subgraphs and ensuring that they meet their needs and developing that competency and skillset within their team.

(01:14:19):

Then free queries begin to go away. And the choice that they’re presented with is, well, this is unpalatable, I want free queries, and I don’t think that they’re going to get free queries from anywhere else in the market. So, then the next step becomes, well, I could basically attempt to completely redevelop this solution, develop something that isn’t The Graph from scratch, that just makes no sense. I think the next step is what gets you to the point of believing in the decentralized network, which is, well, okay, I have my subgraphs, they’re in credibly neutral open standard, and The Graph node itself is completely open source. So, I can essentially set up a mini indexing operation, like minus all of the decentralized network components, but I can launch my own service somewhere in the cloud that indexes my own subgraph and serves queries for that subgraph.

(01:15:27):

And then I point my application at it and I don’t pay query fees, but obviously I have to maintain that service, I have to pay for the infrastructural costs and I need to develop competency and expertise in scaling and optimizing that service as demand. So, two increases. And I think you only get so far down that path before you realize that it is near enough impossible for an independent to compete with the decentralized network Indexers that essentially sync all of their time, resources, and in some case teams of people into running and optimizing these services and operating them in a reliable and cost-efficient manner.

(01:16:19):

And I think at that point, as a developer, you just realize that the net cost of deploying my subgraph to the decentralized network is actually going to be substantially lower than trying to cut out the query theme market and just run the raw infrastructure myself. And the decentralized network comes with a tremendous amount of benefits in terms of resiliency and redundancy, but also it’s highly geo distributed. So, latency to the end user for queries is going to be a lot better than anything that they could run. And just generally, it’s really hard to compete with a network that has collectively got billions of dollars at stake to do one thing really, really well.

Nick (01:17:09):

So, the fact that we’re talking about query fees and migration of subgraphs to the main net all speaks to the fact that we’re still in an evolving environment and Delegators constantly have this question or this concern of where are we in the evolution of The Graph. So, what would be your advice to Delegators who are curious about that, where we’re at in the evolution of The Graph?

Chris Wessels (01:17:33):

I think we’re incredibly early. I think The Graph is in an unusual position of having bootstrapped an incredibly capable product in a centralized context. The hosted service is the data backbone of web3, and that’s definitely an unusual position for a protocol to be in. But when we think about The Graph in the decentralized network context, we are incredibly early. I think the migration itself has been incredibly positive in highlighting some of the deficit that exists in terms of serving that data market in a decentralized way at scale. I think my advice to delegate is would be if you’re not in this for the long haul, then get out, because we are right at the beginning. The network is incredibly young and we are still at the phase of bootstrapping the network and validating that it is sufficiently capable and sufficiently tested to serve the very significant levels of demand that exist already in the market.

(01:19:10):

And systems like these need to be rolled out in a progressive and careful manner because we need to maintain the credibility of the network. And doing things too quickly would risk damaging that. I also, if you’re zooming out sufficiently far enough, I think that generally the conception of where we are is somewhat constrained to the single Ethereum blockchain context. And actually, if you zoom out and recognize that we are going to see the migration of applications and users and capital into these alternate execution environments like Sidechains, but more significantly I think Layer 2s, then you really begin to see that in the grand arc of The Graph and it becoming this unified aggregated data layer. We’re really just at the beginning.

Nick (01:20:20):

I love that answer and I love the way you articulated that. So, I guess the best follow up to a question about where we are in the evolution of The Graph is what’s your long-term vision for The Graph?

Chris Wessels (01:20:33):

I think that as blockchain begins to eat more and more of both our fundamental institutions, but also the kind of products and services that we just use every day, they increase the total addressable market of The Graph. And our lives already revolve around digital services. And there’s certainly no doubt in my mind that the vast majority of these services will in time transition to be blockchain based protocols and networks because that style of coordination just is hugely superior for maximizing the outcomes for its participants and users. So, as more and more and more of these digital services we depend on become blockchainified, they essentially fall into that total addressable market for The Graph. And I think we’re really just in the beginning of what that looks like. And today that’s largely, these applications exist on things like Ethereum and not much else, but in time I think we are going to see a diversification of where these applications run and where they store their data as well.

(01:22:06):

And what’s really interesting about The Graph is that in that context, it functions almost as this layer of connective tissue that is able to aggregate data and create a real time free market for that data in one place, even though it will be hugely disparate in where it is located in terms of the blockchains or execution environments like Layer 2s where these applications actually run. But then also, if you look at the evolution of the data requirements that any given application has, I think we’re definitely going to see an evolution of data storage beyond storing data on the blockchain. And that data, depending on the importance and trade-offs associated with storing that data, that data will move into other places like Filecoin or Arweave or other storage systems that are decentralized and have strong properties over our existing systems, but still occupy quite a wide spectrum of trade-offs that applications will make depending on the importance or value of that data.

(01:23:30):

And that in the same way that we talk about the fragmentation of capital and applications as they move into things like Layer 2s, I think the same is true for data. And as blockchain subsumes more and more of the services that we depend on every day, where data is located is going to become really disparate and The Graph is incredibly well positioned to just glue everything together and offer an aggregated and unified interface for querying that data in a meaningful way in one place without needing to deal with any of the complexities of where that data is located.

Nick (01:24:14):

Chris, thank you so much for your time. I really enjoyed this interview. For those listeners that want to stay in touch with you or learn more about what you’re up to, what’s the best way to stay in touch?

Chris Wessels (01:24:24):

Yeah, thank you so much. It’s been so fantastic to be on the show. I’ve really enjoyed chatting with you. I am available on Discord in The Graph protocol discord under Chris.eth. You’ll find me in the Indexer channel and a few others. And I’m also available on Twitter at undefinedza. You can find that in the show notes. Please do reach out. Happy to speak to anyone.