GRTiQ Podcast: 156 Pablo Carranza Velez

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.

Pablo Carranza Velez (00:00:17):

As well and it makes the protocol more platform-like in the sense that more people can build on it. And I think that’s part of allowing it to flourish beyond the core dev teams.

Nick (00:00:57):

Welcome to the GRTiQ Podcast. Today I’m speaking with Pablo Carranza Velez, principal software engineer at Edge & Node, one of the core dev teams working on The Graph. Pablo’s been on the podcast before back in February 2023 when he joined for a panel discussion about The Graph’s move to Arbitrum during Episode 105.

(00:01:18):

I’ve invited Pablo back for a more in-depth personal interview. During our discussion, Pablo talks about his education and early interest in working in the space industry, learning how to program and becoming a developer, his interest in startups and then moving into web3 and joining the team at Edge & Node. We then spent a lot of time talking about two really important initiatives that Pablo is helping lead, the L2 migration, which is nearly done and by all accounts was a total success. And then Graph Horizon for which Pablo recently wrote a detailed forum post outlining the proposal and providing the community with a lot more context. As always, I started the discussion with Pablo by asking about his educational background.

Pablo Carranza Velez (00:02:02):

So I studied electronic engineering, which I guess in English it’s more usually said as electrical engineering. And it had a focus on signal processing. So I learned a lot about how computers work and how electrical circuits work. And then a few things about things like voice processing, audio processing, a little bit of image processing. So I had a bit of programming here and there, but it was a bit of a low-level hardware and a little bit of software kind of program.

Nick (00:02:35):

Why did you decide at that time in your life to pursue an education in electrical engineering? Did you have a vision for what you wanted to do in your career?

Pablo Carranza Velez (00:02:43):

I always wanted to do something related to space. So my initial thought was I would study mechanical engineering and try to get a job building spacecrafts somehow. And while I was doing the electric course for university, I saw the electronic sphere and they had these little robots and things there, and I just saw that and I thought, okay, this is very cool. I want to do that. I just went straight to the students’ office, I don’t know what it’s called, and asked to change courses to electrical. So it was a pretty impulsive decision, but I don’t regret it.

Nick (00:03:21):

If you had to explain signal processing to somebody that’s non-technical, and I think I’ve had a former guest on the podcast before that talked about this, I think it was Ani over at Semiotic who talked a little bit about his experience in working with signal processing, but how would you explain what that is?

Pablo Carranza Velez (00:03:37):

Good question. So there are a lot of things in the world that can be described as signals, and in general, I think it’s something that changes over time. So you can look at it and measure it, and generally we quantify it with something like an analog to digital converter or something like that. So you see this wave or a thing that is caught in a sensor and you can see it and you can do math on it. So signal processing is about understanding how to model those signals and how to filter them or how to do things to them so that you can either get valuable information or how you can modify them to get different effects or stuff like that. So for instance, a guitar pedal is something that does a form of signal processing where it takes in the signal from the guitar, it applies some filters, it applies some maybe a distortion or some non-linear stuff, and then it spits out a different kind of signal that is the processed version of that.

Nick (00:04:46):

And you mentioned that you had interest in space, maybe early on in your life you thought about working on spacecraft and things. Do you still have an interest in space?

Pablo Carranza Velez (00:04:53):

I do, yeah. It’s still an area that I love. I already worked on that for a bit, so I did fulfill that dream briefly, I guess. And it’s still an area that I’m quite passionate about. Maybe one day I’ll find a way to combine crypto and web3 with space. Ever since I started here I’ve been thinking about a space dial or something of the sort, but I still haven’t quite found how to mix the two things

Nick (00:05:20):

Well, people familiar with The Graph brand and the astronaut and the space know that you’ve probably found a home for yourself in web3. But I do want to ask this follow up question. And for listeners in the United States, they’ll be deeply familiar with this and maybe those that are outside of the US, but in the United States right now, there’s a lot of talk about UFOs, UAPs, aliens, spacecraft. As somebody who’s interested in space and the thought about mechanical electrical engineering type of topics, where do you come in on that debate? Do you think we’re getting visited by spacecraft from outside of earth or humanity?

Pablo Carranza Velez (00:05:56):

I haven’t seen any evidence that I would trust and say, okay, yes, for sure. I also think there are probably aliens somewhere in the universe, but I have no idea if they’re actually here or visiting in any way.

Nick (00:06:12):

I guess I should ask, are you aware of the fact that this is an ongoing topic in the US? Are you seeing this in the place where you live?

Pablo Carranza Velez (00:06:18):

I saw a few things on Twitter. I haven’t followed them too closely. It was quite interesting. If there’s anything real there, it would be pretty incredible, but I still haven’t seen anything that I found convincing.

Nick (00:06:33):

Well, it’s fun to see a public discourse about this, especially in the States. And for listeners that are curious, you can click around on Twitter and go look at some interesting things that are being discussed. Returning back to your story then, talk to us about what you did after university. How did you get started working professionally?

Pablo Carranza Velez (00:06:52):

When I finished… At my university, you had to do an internship to finish the degree. And that internship, I did it in a company that builds medical devices. So I did a little bit of EKG patient simulator thing and something that had to do with screens. I would go in car races, inside the cars that were used for races, which was quite fun. So that was, I think a first step towards the embedded systems side of electrical engineering and also a bit towards software, I guess, because a lot of the work I had to do there, I designed the small circuit for that with a team that was working on that. But a lot of the work I did there had to do with embedded systems programming. So that was my first experience. And then when I finished school, I have an uncle that had a company, he said he needed some software, a CRM kind of thing or data warehouse I guess.

(00:07:54):

And he said, “Is that something that you would be able to build?” And I thought about it for a few minutes and said, “Yeah, maybe I could.” And I really had no idea about any sort of higher level programming. I had studied C and stuff like that and assembly maybe, but so it was a bet that maybe I could. So I started researching. I did I think an online course on web programming, learned Ruby on Rails, which was I guess popular at the time. Not so much now I guess. And I just built it. I even contracted a few friends for the design and other parts of the software, and I think it worked out quite well. It was an internal piece of software for them so it didn’t become a product or anything like that, but it was an interesting first experience.

(00:08:45):

And then I started trying out this freelance software engineering kind of thing, and I contracted with some friends that already had a company and then we decided that we would join forces and start a separate company adjacent to theirs, I guess. So they had this company that was more focused on technology for advertising and a little bit of electronic arts and projection mapping and stuff like that. And we did this separate thing that was in the same office with them, but more focused on software projects and software and hardware sometimes. So at that time, I already started doing a few things that combined this web software development with a little bit of embedded, and I started getting into the internet of things kind of thing. I think it was before IoT as a term became popular, and it was a nascent field, but it was a fun thing to do, build machines, talking to servers and stuff like that.

Nick (00:09:47):

Let’s talk about that move you made then. So you studied engineering as you said, and then you make this jump. Maybe it was an organic move more than a jump, but you started learning programming and became, in essence, a developer. Was that a hard transition to make and did you find that educational background in engineering was well suited to make that shift?

Pablo Carranza Velez (00:10:06):

I think so. It was a natural transition for me. The things I learned in university had a lot of software in it and things like algorithms and data structures, so maybe not at the level that you would have if you’re specifically studying information systems or computer engineering. But I did have a good background I guess in those things. And starting with lower level software, lower level languages like C and assembly, it made sense to then progress from there to a higher level thing. It felt natural to me, and I think I always had, I guess a knack and a liking of mathematics. And so the software side of the things I studied in my degree always ended up being the things that I liked the most so it made sense for me to do that transition, I guess.

Nick (00:10:59):

A lot of listeners of this podcast are looking to maybe upskill themselves such that maybe they’re non-technical and they want to learn to code, and I’ve had guests, certainly members of the advocates program and participants in The Graph ecosystem who have done that. They’ve come in as non-technical and have learned to code. What’s your advice to listeners that want to do that, and as somebody who made that shift yourself or how would you tell them to approach that?

Pablo Carranza Velez (00:11:26):

I don’t think there is a single path that fits everyone. I guess it does take a bit of time. I think I was lucky, electronic engineering, it’s not software engineering, but it’s quite close so it wasn’t that big of a jump for me. I’ve seen other people that are more self-taught without having a degree and they’ve really made amazing stuff. So I think anyone that really wants to do it can do it. It does take a bit of time to learn the roles and get familiar with the programming language and understand the tools that you can use and how to use them and choosing the right tool for the job and stuff like that. But those are things that you can definitely build. Maybe it’s about starting with an easier programming language and something that has a quick applicability to whatever kind of job you want to do.

(00:12:17):

Or maybe it’s just about putting the time of practicing and doing programming problems and having your site projects and stuff like that. But I’ve seen so many successful cases and I’ve worked with so many amazing people that didn’t have more technical formal education. So I think it’s just maybe the main advice would be don’t give off.

Nick (00:12:42):

Good advice, and I think that’s come on in the past Episodes as well. I want to ask you this question about pursuing in your early career startups as I look over your background, you weren’t necessarily going to work for large companies, you were going into this startup venture world. What was the pull there? What was the motivation to pursue things in startup land?

Pablo Carranza Velez (00:13:06):

It was a mix. I think I always had more of an affinity to that. It might be a family thing. My dad was always building companies and doing that kind of thing. So I do think I have inherited that thing from him. Maybe his example showed me that you don’t really need to go for the giant company and build a career there. So it was always appealing to me the idea that you could be in a place where you can have maybe more impact, I guess. And so it was that. And then I think at university we had a few courses about startups or being an entrepreneur and in some of the projects like our thesis or final project at school, had to do with building an electronic product. So while we did it with a team, we thought about how we would build a business with that. And for a while we even thought about launching a startup from that. In the end, we didn’t.

(00:14:08):

But I think the combination of that family history and then learning those things at school and maybe something in my personality, I guess. I didn’t see myself working at one of the giants. It felt more natural for me to just either join a smaller team or just try to build something from the ground up.

Nick (00:14:29):

You’re clearly a smart person, thoughtful person. You’ve done some very technical type things so I presume you have a mind for analysis and contemplating deep questions. What have you learned about startups or entrepreneurship that you could share with listeners? Something about what makes a successful entrepreneur or what makes a startup succeed? Do you have any observations about those things?

Pablo Carranza Velez (00:14:51):

I wish I did. It’s not like I had a very long experience in successful startups of my own, I guess. The things that I tried, they worked for a while and then they didn’t. So maybe one of the main lessons is don’t run out of money. But there’s also, the other teams I’ve been part of, I saw different things that worked. And I guess what I’m seeing now is the main thing is trying to… It’s not going to be anything new, but the idea of trying to find product-market fit and really listening to what your users need and trying to understand what is the problem that you’re trying to solve. If I can look at my past experiences and see what things brought on success, they always had to do with solving things that people needed, I guess.

Nick (00:15:43):

Is there an idea from the past that you still are in love with, you wish you could go back and work on knowing things you know now or something that you still have conviction for?

Pablo Carranza Velez (00:15:53):

Well, there was this startup that it lasted maybe six months. I was trying to start it with a co-founder who was also, he was studying in the US at the time. And we didn’t have a lot of time and I was trying to fundraise and develop a product at the same time. In the end, I had to drop it. But it was the idea of doing nano satellites for radar imaging. Maybe the radar doesn’t quite fit in a nano satellite. They’re very small, I don’t know, 30 centimeters long, so maybe you need a bit of a bigger one. But I think after a while now there are a few startups that actually do that. So it’s probably too late. I think it’s called ISI or something like that. But I still think it was an interesting idea. It was one of those things where I thought I had an idea for something cool more than actually having something that really solved a problem for people.

(00:16:51):

So a solution looking for a problem, but it turns out there are actually a lot of applications for it. And I think it would be very interesting to try if I could go back in time and have maybe more funding and more time to develop that idea. And I would do things differently. I would try to build a bigger team from the get-go and not try to fundraise and develop at the same time. But I don’t know. I think it was an interesting one. I don’t know if it would’ve actually worked, but I do see some successful examples today that did that same idea.

Nick (00:17:23):

Remarkable. A little bit ahead of your time there with that particular idea. So if we’re mapping out your personal story here, and we’ve talked a little bit about your education and getting started professionally, when did crypto or blockchain or web3 enter your mind, and what were those first impressions of it all?

Pablo Carranza Velez (00:17:41):

So back in 2012, 2013, some people in my university were starting to focus on Bitcoin. That was the thing that was around at the time. And even some of the first non-financial applications of blockchain came out of people that I knew. So I always found that super interesting. And when I first heard of Bitcoin, it sounded like something that had a lot of potential. So I think at that time already, I knew that there was something interesting there and something that would have a lot of impact in society and in the world. But then I had my idea of focusing on other stuff. I had my startup at the time, and then I had these ideas about trying to build spacecraft eventually or do something related to space. So I didn’t really focus on that. And then I lost track of it for several years.

(00:18:35):

I knew that some other friends started joining crypto-related projects and then I started hearing about this thing called web3, and I wasn’t sure exactly what it meant. And then a few years ago, well, right before joining The Graph, I got together with an old friend for a beer and he told me about this interesting thing called The Graph that he was working on and what it was about. And it was quite interesting. So it got me. At the time I was also looking for a job change. I was doing fun stuff at Spire at the time, but I needed a change for various reasons. And that got me at least interested and starting to read about it in a bit more detail. And when I found out what it really meant, or at least what it means now for me, it was a little bit of a wow moment. Like, okay, you can really do amazing stuff with this thing called Ethereum and smart contracts and building new coordination mechanism sounds amazing. I want to do this now. And then I just learned smart contracts development and ended up here.

Nick (00:20:19):

I want to ask you a lot more about that move, and as listeners will soon find out, you’ve been very instrumental since making that move and have worked on some really important initiatives. But take us back. What was Spire and what did you do for Spire?

Pablo Carranza Velez (00:21:06):

Spire is a company that has one of the largest commercial constellations of satellites. So they build and launch nano satellites, and nano satellite means it’s something like a shoebox. They go from 10 by 10 by 10 centimeters to maybe 30 centimeters long or maybe a bit more. They’re called cube sets, I guess because they’re little cubes. Spire builds and operates a lot of these, like 100 or something. And they use these to provide data about things on earth with the idea of solving some challenges that people on earth have, things like shipping time estimation and weather prediction and stuff like that. And they also provide services to companies that want to maybe launch something on a satellite or do something with a satellite, but they don’t have the expertise to do the whole thing.

(00:22:04):

So Spire does this space services thing where people come with an idea or with a request for something they want to launch and then they build the whole mission and operate it then. So that’s what they do. It’s a very interesting company. They went public at the time that I was there. They did a… It’s not an IPO. It’s a company that buys a private company and takes it public. So they did that while I was there, and it was quite an interesting experience. It was the tail end of a startup becoming more of a corporation I guess. So it was quite a fun experience. And it was also, of course, my opportunity to actually launch stuff into space. I think only two or three months after joining, I was upgrading a satellite in space with code that I had written. So it was quite an experience.

Nick (00:22:55):

So going back to then this conversation you had with a friend, we’re talking probably 2021, 2022 time period here, and they’re describing to you web3, you’re becoming more familiar with what web3 is, but also specifically The Graph. And you mentioned that you had a wow moment. If you don’t mind, can you double click on that and just tell us what it was that sparked this wow or aha moment for you where you realized something new?

Pablo Carranza Velez (00:23:24):

I was familiar with Bitcoin and I understood the idea of the Bitcoin proof of work consensus. It’s like this, I don’t know, finding the magic number and everyone agreeing that that’s what it means to build a block on Bitcoin or whatever. I’m simplifying a lot, but so it sounded like an interesting primitive. And then I guess what I learned was how Ethereum works and the idea of using that to build this virtual machine that anyone can use and that anyone can verify and that the programs that we write to run on this computer can be seen by everyone, and we all agree on the state of this shared database. And for me that was like, wow, okay, I can see how this can be really useful. And then seeing The Graph as an example of how you can build such an impressive coordination mechanism from that was, okay, this really can be the future of human coordination.

(00:24:29):

I keep repeating the word coordination because for me, that’s the main thing that seems valuable here. That is usually we’re in this dichotomy between the individual and the state. So we’re either doing things for our self benefit or we’re forced to do things by the state because we, I guess agree that this is what the state needs to do and what powers it has. So here I see a primitive for how you can build stuff that is in the middle of the two things where we’re agreeing to shared rules to play this game and build new social constructs that are useful and that are more maybe opt in I guess, or that are more efficient and lead to better outcomes for people. It seems like such an amazing tool if we use it properly to build things that benefit people.

Nick (00:25:29):

Well, you said a lot of super interesting things there, Pablo, and I think the thing I want to unpack is most listeners are familiar with the status quo. Everyone’s got a computer, they’ve got an internet network. There’s a bunch of web2 companies that deliver services. They have a ton of market share. You could argue there’s monopolies almost in every direction. And then we’ve got this virtual machine in Ethereum where people can build and it’s permissionless and stuff. And I think my question is, and maybe going back in time to when you’re having these moments where you realize the value, how is the world better? Why is that better than what we’ve done now? It seems to be working well. You can argue people are getting the goods and services they need. Why do we need something to come along and disrupts if it’s not broken?

Pablo Carranza Velez (00:26:14):

Well, I don’t think things need to be super broken for there to be better alternatives. I don’t think the current scenario in terms of technology in the world is a bad one. I look at the history of humanity in the past few hundred years and I see something positive, a lot of better outcomes in terms of health and in general, things like poverty decreasing over centuries. Obviously there are a lot of things that are still wrong and people suffering in a lot of horrible ways, but I do think we have a history of things improving since the industrial revolution and technology. And in the past 100 years, it’s also been like that. And I think the internet maybe was built in a way that led to this more centralized state. And it’s not that it’s bad, but maybe it can be better.

(00:27:14):

There are things where if there’s a single company that controls the public square, then that company has a lot of power and it starts to resemble maybe the responsibilities or the powers that states might have. There’s always this conversation now about free speech, for instance, where usually in the US it’s the First Amendment. So you can have free speech and generally understood the state is not going to be the one that’s censoring you, but then what happens if it’s a private company that’s censoring you? Is it still free speech? Is it still important to protect speech in that case? And I think it is. So if companies are becoming more state-like, it makes sense that we need to be careful about how much power those companies can accumulate. And I think if we can build systems that are better, not because we’re forbidding companies from growing and creating value and doing useful things, but because we’re creating mechanisms that incentivize better outcomes where more people have power, where more participants can have a say on the rules of the game, then I think we should do that.

Nick (00:28:23):

Brilliant answer. So this is a longstanding question that I’ve asked guests over the last two and a half plus years that I’ve been doing the podcast. I’m going to ask it now. It seems like a great point in the interview here to ask it, which is, do you then see web3 and all the tech that’s growing out of this web3 blockchain movement, do you see it as just a evolutionary step as we try to refine and improve technology or do you see it more as a revolution, people standing up against systems, technology that failed us or something else? How do you contextualize that?

Pablo Carranza Velez (00:28:56):

I’m not sure if I have a single answer to that. I think it’s probably a little bit of both. There’s a bit of revolution and a bit of evolution. It makes sense that there’s the old internet protocols like TCP/IP that are more just purely technical and they involve some coordination, but it’s non-financial. And then the rise of the big web companies and then you see how blockchain comes in and Ethereum and smart contracts come in and they give you a way to combine the financial side with the technical and the coordination side. So I do think there’s an evolution there or a continuous history of things getting better. But I also understand that for a lot of people it’s more of a revolution in the sense that there are these things that are not working and we need to change them. So there’s also that other aspect of, okay, let’s fix things that are broken or that are not working. And I do think it has this more revolutionary vibe, I guess.

Nick (00:29:58):

So you have a meeting with a friend, you’re looking for work, you were at Spire and your eyes are open, your radar’s up to new opportunities. You hear about web3, you wrap your mind around Ethereum, smart contracts and a lot of things we were just talking about. In 2022 then you join Edge & Node, which for listeners that aren’t familiar, that’s one of the core dev teams working on The Graph. And in fact, what makes this core dev team unique from others is the founding team at The Graph after they launched the protocol spun into a core dev role, and that’s where Edge & Node was born. But now there is a whole bullpen of core devs and a lot of core contributors working throughout the ecosystem. Talk to us about going to work at Edge & Node and what your first position was there.

Pablo Carranza Velez (00:30:46):

I started as a member of the smart contracts team, and it was a small team, just three people when I joined. And we build the smart contracts that make the network work at the blockchain level. So the things that actually run on Ethereum are now Arbitrum One, and that means implementing the cryptographic mechanisms into code and then testing and applying them and monitoring them and making sure that things don’t break.

Nick (00:31:17):

Well. Pablo, you clicked on something there that I want to go back to and I think in some ways for some listeners, this’ll be an epiphany moment where they learn something new. But not only is web3 built on these smart contracts that we hear about happening on Ethereum and different things, but The Graph protocol itself is built on a foundation or a lattice, if you will, of smart contracts. Is that the right way to think about it?

Pablo Carranza Velez (00:31:42):

Yes, that’s right. The contracts are what defines the rules of the game in the protocol. If you want to participate in any way in The Graph, you need to follow the rules of those contracts. And maybe the main primitive there is staking. If you’re an Indexer, you want to provide a service, you stake some GRT as a warranty that you’re going to provide good data. And around that, there are a lot of other contracts that have to do with fee collection, dispute mechanisms, delegation and stuff like that. And all of those things are enforced in codes and written into the smart contracts. So when you want to interact with The Graph, you usually do it in one way or another through these smart contracts.

Nick (00:32:23):

Let’s talk about moving into web3 and going to work at Edge & Node. What was that transition like for you? Was it something new and exciting? Was it something overwhelming and you were thinking, oh boy, what have I done? Or a mix of all of it? What was that experience like for you?

Pablo Carranza Velez (00:32:37):

It was a bit of everything. It was super exciting. It’s nice. It’s a small team relatively, and I think that’s usually good, at least for me. Starting at a new job, you get to meet everyone and we have these weekly random meetings just for coffee with folks and getting to know the team, and I think it’s a brilliant team. So I really enjoyed the process of getting to know everyone. And at the same time, I had a lot to learn. I started off with one of our retreats actually, so I met a lot of the core devs in my first week, and there were already discussions about the move to L2 at the time. This was before deciding on Arbitrum as a scaling solution. So it was quite exciting to join there and have to learn all of these things on the go ’cause I didn’t really understand Layer 2s at the time, so I spent that week reading everything I could.

(00:33:40):

And then somehow I ended up doing a lot of the smart contracts work for that project, which was super fun. I always enjoyed learning on the job. So I think this was no exception, but there’s always a little bit of uncertainty when it’s like, okay, will I be able to learn this on time for all of the things that we need to do? So it was quite fun. I often complement with online courses when there are things that I don’t know so I ended up doing courses on game theory while I was starting on the job.

Nick (00:34:16):

It’s an amazing story and quite an entry into Edge & Node. So you join, you go straight to work basically on this L2 migration, and over time you write an important forum post that outlines the move to L2. And you actually joined me with some team members to talk about the move to L2. So long time listeners know there was a panel discussion last year where you and some of your colleagues came on and we talked about the move to L2. We’re further down the road now. L2 migration happened and it’s been a huge success and that’s part of the reason I wanted to talk to you, is how much of the network at this stage, at the time of this recording has moved to L2? Do you know?

Pablo Carranza Velez (00:34:56):

We’re almost there. So at this time, 95% of indexing rewards are being distributed on Arbitrum. And I think the last step we need is for subgraph developers, subgraph owners to finish transferring their subgraphs over to L2 ’cause that’s the final step. When most of the subgraphs that are actually being used are on L2, that’s when the council can finally do that last switch of 100% of rewards in L2, and then The Graph will pretty much run fully on Arbitrum at that point. It’s quite a challenge to have, I think, for a lot of participants for Indexers as well, to be able to maintain both protocols in parallel. So I’m looking forward to the time when that scaling effort is finally done and we’re fully on L2 and there’s no more mainnets to think about. But we still have those few steps left. What

Nick (00:35:51):

Surprised you most as this initiative got deployed? Did anything surprise or shock you? Anything like that?

Pablo Carranza Velez (00:35:58):

I think I wasn’t surprised. It made sense. It was quite amazing to see how the economic incentives actually work to drive outcomes as the stake. So when we launched the network initially on L2, it had no induction rewards and we thought, well, maybe a few people will be there to experiment and that was the purpose. And that’s actually what happened. There were a few early adopter Indexers that started staking a little bit there and answering queries, but as soon as induction rewards started going up, that’s when we saw stake going in and Delegators starting to transfer and everything. So it was quite interesting to see how as there was a rewards bump, then stake and delegation would start transferring and slow down as it approached the percentage that we expected.

(00:36:49):

So I think what surprised me was that it actually worked as we planned in terms of people actually adopting it, following those incentives. And I’m also not surprised that it’s taking a bit longer for subgraphs to be transferred. If you’re a Subgraph developer, I would encourage to transfer. I think there are benefits in being there, but it’s understandable that it takes a bit longer to go through the process and send your subgraph over.

Nick (00:37:16):

People in the community that aren’t as technical as you and additionally didn’t work firsthand on writing the forum post for L2 and doing all the L2 research and everything that you did for this campaign, they get confused maybe sometimes on what exactly the migration was. So sometimes on social media you might see a post of, we’re no longer on Ethereum or supporting Ethereum, we’re now at Arbitrum. And it’s like, well, wait a minute, what do you mean? What has moved? What didn’t move? In a really simple way for anybody that’s still a little bit confused on what actually moved and what’s this relationship between Ethereum or what’s going on here, how would you clarify that?

Pablo Carranza Velez (00:37:57):

That’s a great question. I think a big part of the confusion is in the difference between an indexed network and the protocol network. So in The Graph, you can write subgraphs that index a lot of chains. So not just mainnet or Arbitrum One, you can index mostly EVM chains, but there are a lot of blockchains that can be indexed by The Graph and therefore you can launch a subgraph that indexes any of those. So when we say we moved to Arbitrum, it has nothing to do with being able to index any of these chains. It’s simply that the protocol contracts where people are staking GRT or delegating or launching and curating to their subgraphs, they used to be on mainnets. They simply lived there. GRT is a token on the Ethereum network, and it still is, but the staking contract and all of these dispute mechanisms and so on, they were in Ethereum.

(00:38:53):

And it means if you want to do any of the actions that are part of participating in the protocol, you need to pay gas fees on Ethereum mainnets because you need to run these programs that costs money to run. And what we did is we launched a new set of contracts. There was a new staking contract and new dispute manager, new mechanisms for fee collection and so on, and it’s running on Arbitrum One, and there is a bridge between the two. And that means, so the GRT contract still exists in L1, but it can be bridged to L2, and then you can perform things like staking and the dispute resolution and the query fee collection and all of that by paying gas fees on Arbitrum, which is much, much cheaper. So that’s basically what we did.

Nick (00:39:36):

Well, you’re the new guy. You get to work on L2 migration, it’s a huge success. The network is moving. The rewards have made the move, so everything’s going well. And then you write this next forum post about Graph Horizon, another huge initiative that’s been getting a lot of discussion and attention within The Graph ecosystem. And in fact, longtime listeners of the podcast know the final Episode of 2023 was with Zac Burns, somebody who was on that team that seeded and developed this idea of Graph Horizon. But I want to go back to it a little bit with you. So let’s assume somebody hasn’t listened to that Zac Burns interview, hasn’t read your forum post or been aware of what Graph Horizon is. How would you describe The Graph Horizon proposal and what it is?

Pablo Carranza Velez (00:40:24):

So Horizon was initially Zac’s idea, actually. He’s the one that had that first realization. I think that is that staking is the core mechanism in the protocol and that it can be decoupled from everything else. So you can have this core thing that is just the staking part of things, and then you can build something that is more modular and more permissionless by using this core primitive. At least that’s how I see it now. I think since Zac’s original idea, we had a lot of conversations and we iterated a lot on the actual design. We keep iterating on it. Actually we keep finding things that we can improve in how that staking would work. But that is the core idea that right now the protocol is all mixed up in the sense that you have the staking and the rebates and the disputes and all of these things are working together. And they work quite well together because they’re all designed for the subgraph data service that we have now.

(00:41:27):

But there are a lot of ongoing initiatives about adding new data services, and that can already be done with the current protocol, but it’s a limited API in that you have to overload the meanings of some of the fields there and reuse some of the things that you might not want to reuse if you’re doing a complete different data service. So with Horizon, what we’re proposing at least is the staking part can be a separate thing where you can allocate to any data service, and that means you can put GRT as a warranty for any kind of service that involves data, and that works as the economic security for providing that service. And then the actual logic and the actual rules of the game for that particular service can live in a completely different contract managed by a completely different team that can specify their own rules. And that means there’s a lot more room for innovation where other teams can come in and do new things like other query languages or doing fun stuff with LLMs.

(00:42:34):

But they don’t necessarily need to follow the same rules of the game in terms of how much you need to stake, how the disputes work, how the verifiability works, how the query fee collection works. All of those things can be done at the data service level. Obviously part of our work in Horizon, we want to build the framework for how you would build those things if you’re doing your own data service. But I think that’s the main thing. And in doing so, we’re also taking a lot of the things that we have in the protocol today and decentralizing them further, I would say. Things like reducing their reliance on Oracles or stuff like that. So I think there are a lot of changes that are happening as part of this Horizon. So it’s an umbrella for a lot of things that… I think they go very well together and it makes sense for it to be a different initiative, but in practice there are a lot of improvements to the protocol that are bundled in this thing.

Nick (00:43:30):

Well, I want to encourage listeners that want to learn more about Horizon to visit the show notes. There’s a lot of things that you can read, and of course both Zac and I’m sure you Pablo as well, will strongly encourage community members to give feedback and to let their position and perspective on the Horizon proposal be known. I want to ask you this question, Pablo, about what you say to members of the community, members of the ecosystem that say, wow, there’s a lot going on here. We just moved to L2, Sunrise of Decentralized Data is in action. So all these chains, all these subgraphs are getting moved from the hosted service to the network and now there’s this proposal about Graph Horizon, which sounds like a V2 of The Graph and more movement. What do you say to people that are like, that’s too much, there’s too much movement going on? Assuming there is these people, but what would you say?

Pablo Carranza Velez (00:44:17):

That’s a good one. Yes, it is a lot and we’ve asked a lot from people in the move to L2. It took some effort, it took some time to actually go and move the stake for Indexers. It was quite a process for subgraph developers as well, for Delegators too. Everyone in the protocol, Curators as well, they had to do something to make this move. And it makes sense. It seems like a lot, but I think we really want The Graph to succeed and we want it to be the foundation of an open internet. And that means we need to keep moving, we need to keep improving, we need to keep building stuff. We can’t stay still because we’re building something that is meant to be revolutionary or an evolution of the internet, if we go back to your previous question. But we can’t do that if we just stay with the current version of the protocol.

(00:45:12):

There’s always things to improve, and I think if we see opportunities to improve, we need to take them so that if The Graph is going to fulfill on its promise of a decentralized universe of data or there’s different versions of the vision for The Graph, I think it’s important that we make it the best possible version of the protocol. Now, hopefully we can do that with the least amount of effort required from credible participants. So I think after the experience from L2, I would definitely take some lessons from there to try to make it easier for participants going forward.

Nick (00:45:47):

That’s a great point. I’m sure you and all the other core devs learned a ton about the protocol and ecosystem and these types of moves by virtue of the ones that have been recently completed and that will inform probably what Graph Horizon shapes up to be. What was your objective in writing that forum post? Were you trying to give an overlay or an outline of what Graph Horizon is? Were you trying to invite participation, feedback from the community? All the above? What was the objective with that forum post?

Pablo Carranza Velez (00:46:15):

I think what happened was Horizon had been discussed among core devs for quite a while now and we weren’t really giving the community a lot of visibility about what was happening behind the scenes. What were all the discussions we were having? How were we thinking about the future of the protocol? And there were some mentions of Horizon in public settings and it made sense to come on and say, okay, what do we mean with this thing that we call Horizon? And it seems like it’s going to be amazing, and it’s going to be a big change or whatever, but we needed to stay true to our values and then be transparent about what we’re trying to build. And obviously because The Graph is a decentralized protocol, we make decisions as a community, we’re not going to unilaterally do any of these changes. So it was about communicating the things that we’re thinking and trying to understand if people are on board with the ideas that we’re trying out and the things that we’re thinking.

(00:47:13):

And to me, it made sense to do a detailed forum post about it because I feel like writing it down… And it doesn’t have enough diagrams. That’s the one thing that I regret about that forum post and it’s a bit long, I know. It might take a while for people to go through the whole thing, but there’s a lot of content there of things that at least I think are important for the future of the protocol. So I guess there were many purposes. And for me, I think also internally in our team, it’s trying to understand if people are on board with a general idea, but also what they think about Brownfield versus Greenfield proposal and whether we want to go through the upgrades routes and just upgrade things in place or if we want to take the other approach like we did with L2 of deploying a new thing and then encouraging people to move over if they think the new protocol is better, I guess.

(00:48:11):

So that’s, I think, one of the main discussions. And then also, of course, whenever you’re changing something on a living protocol, there’s bound to be some discussion because any change will affect people, and we want to build things in a way that are net positive for everyone.

Nick (00:48:28):

Well, I think it’s become pretty clear for me, Pablo, in interviewing Zac and now interviewing you, that Graph Horizon is multifaceted. There’s a lot of things going on there and again, I want to direct people to the forum post to learn more. I think you did a great job going into the details. As with any community proposal, there’s a little bit of controversy or interest in a few of the topics. Of course, you mentioned Greenfield, Brownfield. People should learn what those are and express their opinions on that. There’s also these things like Delegators now being slashable in a Graph Horizon environment. And then this question about what happens to curation. I know the issue is far from settled, but how would you talk to listeners who are concerned about those topics or still trying to figure out exactly what’s going on there?

Pablo Carranza Velez (00:49:13):

These are tricky topics. It is hard to navigate making changes in a protocol that it’s already live and people have deposited their confidence in the protocol in a way. In the case of Delegators, they’ve delegated their GRT to Indexers and they’re expecting certain rules of the game. So we’re proposing a change here and it’s scary. Change is usually scary. In this case, there is this additional risk of your delegation cannot be slashed if the Indexer gets slashed. I think there are going to be some things that will be protective of Delegators in a way. So in particular, the slashing will first go to the Indexer stake. So if an Indexer provides bad data in some form, then they’re going to be slashed, but that slashing will only affect the Delegators if all of the Indexers take a slash first.

(00:50:07):

So if there’s a minimum stake of 100K, and even in the case, if we keep the 16 times delegation ratio we currently have, then that’s a lot of GRT that needs to be slashed before any tokens from the Delegators are touched. And the main thing that this protects is the economic security in a case of a very, very malicious Indexer that just sends a lot of bad queries, then you need to do a punishment that is proportional to that. And that means slashing a lot, but that can only happen if the Indexer is really, really bad. And the other thing that I would say is I think this will actually encourage decentralizing more of the delegation. If you put all of your GRT on a single Indexer, then well, maybe you are very, very sure that this is a trustworthy Indexer, but if you spread it out across 10, 15, 20, 100 Indexers that you think are going to be good, then if you make a mistake and one of them ends up being some sort of black hat entity that wants to do bad stuff and they get slashed and you do some of your delegation, then you’re still making rewards and fees from all of your other delegations.

(00:51:23):

And that should probably compensate at least. And if you look at other things like, I don’t know, liquid staking in Ethereum, like Lido and Rocket Pool, all of those are actually slashable. We don’t see a lot of slashing because maybe some validators in Ethereum gets slashed, but when that spreads out across a whole staking pool, the actual effect on a single person that stakes on each is still so small that their rewards end up winning. So I think that that’s on the Delegator’s side. I think the benefits from the protocol should outweigh the risks a lot. So for Delegators, part of the incentive for being a Delegator is helping The Graph grow. And if that delegation is securing the network a lot more than what it is today, I think it’s going to help The Graph grow a lot and that’s also beneficial for Delegators.

(00:52:16):

So that’s on the delegation side. And then the other one about curation and indexing rewards, I think that’s still an area of research. What has been shown by the work of the protocol economics team is that the current mechanism, it’s clearly working to a point in the sense that we have the Indexers that we have today and they’re providing great service and there’s good quality of service for the subgraphs that are incentivized by this indexing rewards mechanism, but it’s clear that it needs to change. It’s not the mechanism that we need for The Graph to reach its full potential because it’s driven by this token-based curation that isn’t quite effective at directing the issuance to the best subgraphs or the most useful ones. If you look at query fees versus curation for the subgraphs, there’s no good relationship between the two. And I think this is good evidence that it’s a mechanism that needs to evolve for us to achieve all of The Graph’s potential.

(00:53:22):

So I think part of the conclusion from the protocol team, I think is there’s no way to do something that is permissionless and civil resistant and that provides a meaningful reward if we don’t do something different. With the current tools that we’re using, we’re always going to reach a dead end trying to design something like that. So I think what happens with Horizon is we’re designing something that works without agents. You can get a subgraph index that you don’t need indexing rewards for that. We still need indexing rewards at least for now because it’s what’s sustaining the Indexers that we have and that’s important as well. We’re not going to just drop that one day to the other ’cause we don’t want to break things, of course. But there has to be an evolution, I think. As query fees and indexing fees take off and a lot more of the Indexer income comes from that, it will make sense to transition to something else.

(00:54:17):

And if we’re going to have GRT issuance, maybe find better ways to distribute that in ways that help the protocol grow in a more impactful way. So I think that’s the way, at least I’m thinking about this right now, and other people in the team are also thinking like this, but I don’t think anyone’s thinking that we’re just going to turn off indexing rewards from one day to the next, because that would be very bad. We’re breaking something that it is helping the network work at this point.

Nick (00:54:47):

How are you thinking about what comes next? So not putting a timeline necessarily on Graph Horizon, but as you think about it in your own mind, you authored the forum, clearly the community is engaged with it, people are posting and sharing their opinions. What happens next? How should we be thinking about that?

Pablo Carranza Velez (00:55:05):

That depends on the community’s response, on the council’s response. I want to talk with the council soon. I think that the engagement we’ve had with the forum post has generally been positive. There have been some concerns that were raised, and we’re talking about those in index or office hours and other community calls, and we’ll keep discussing in the forum as well. So that’s the process of reaching consensus as a community on what we want to do, but also, obviously that translates eventually into a council vote of some sort on whether we want to continue with the Brownfield approach or we want to explore the Greenfield approach, I guess, or if we want to try something else. But from the initial responses, I think mostly it seems that people are on board. I don’t know if the council will want to do a community vote for instance, that as I think was done once or twice in the past.

(00:55:59):

So that’s also a possibility. So I think there are many possibilities for how this could happen. In the meantime, our team is building and we’re trying to bring this set of ideas down into a more concrete form. What does it look like in terms of code? If we do the Brownfield approach, what are the actual changes that we need to do to the contracts? And how does that translate to things like function calls and arguments? And how do the rules of the game change more concretely? Or if we do the Greenfield approach, well, we actually need to implement that. So we’re doing that in parallel to the community discussion. And obviously if there’s feedback that leads us to change this, we can change the implementations as we go. But I think right now it’s about the community giving feedback and the council giving feedback. And from there, we’re going to be refining and iterating a bit probably.

(00:56:53):

And if it seems that there’s consensus on an approach, then we might end up doing some more formal graph improvement proposals. Because right now it’s a bit of an informal forum post just describing the general thing. But if we actually implemented the Brownfield approach, for instance, each of those changes might require a proposal that is a bit more formal and getting the actual approval for that. So I think over the next few weeks or maybe a couple months, we’re going to see more clarity depending on what the community wants, what path we’re going to take on the Greenfield approach, the Brownfield approach, and if there’s going to any changes to the actual things that are being proposed. But I would say keep an eye on the forums and obviously we’re going to keep communicating as we go. And if there’s discussions with the council and stuff like that, you’re definitely going to hear about it.

(00:57:47):

So the more participation we get from the community, the better. I really want to hear from people if they agree, if they disagree, if there are things that make sense or don’t make sense, and we can refine and maybe try to explain as well because there’s a lot of research behind this proposal. There’s about two years of the research and protocol team working to figure out what are the best mechanisms for The Graph to flourish. So part of it might involve explaining those things and showing a bit more of our work, which I think is important. And in other cases it might be, oh, there might be points that lead us to change some things or iterate on some of the proposals. So we’ll see what happens based on that.

Nick (00:58:34):

In your opinion, Pablo, can The Graph realize its full potential, can it reach that vision, that mission it has about indexing and querying for all of web3? Can it realize that mission without implementing something like Graph Horizon? Maybe it’s not exactly as it spelled out in the forum, it’ll evolve on community feedback, but something like this is necessary?

Pablo Carranza Velez (00:58:56):

That’s a good question. It’s hard to make predictions. I think the current protocol is a good one. It works. People can index subgraphs and get good quality of service and the mechanisms are sound in a lot of ways. So I don’t know, maybe it can. I think the reason why we’re proposing these changes is because we think, or at least I think it increases the likelihood of that a lot. We could take the current protocol and stretch it to the most that it can provide, and maybe that is sufficient for The Graph to be the foundation of the open internet. But I think the chances are higher if we build it on mechanisms that are more modular, simpler and easier for other people to build on as well. And it makes the protocol more platform-like in the sense that more people can build on it, and I think that’s part of allowing it to flourish beyond the core dev teams.

Nick (00:59:58):

In a conversation I had with someone the other day talking about this very thing, they were talking about the V2 of The Graph and was it necessary? And I was quick to remind them about all the version updates and upgrades we see with Ethereum, the most popular architecture, you could argue, throughout all of web3 that everyone seems to build on and they go through a series of roadmap updates. And when you frame it that way and you understand that The Graph protocol is just learning from V1 and continues to grow, it starts to make a lot more sense. And again, I want to encourage listeners to go read that forum post and share your opinion with the team at Edge & Node and the other core devs working on Graph Horizon.

(01:00:37):

Pablo, now we’ve reached a point in the podcast where I’m going to ask you the GRTiQ 10. I didn’t get the opportunity to ask you these questions because the last time we met we were in a panel and it’s impossible to do the GRTiQ 10 with panel discussions. So these are 10 questions I ask guests each week on the podcast, and I do it because it’s fun to get to know you on the personal side a little bit, but also it opens the door for listeners to learn something new, try something different, or achieve more in their own life. So Pablo, are you ready for the GRTiQ 10?

Pablo Carranza Velez (01:01:05):

Yes. Okay.

Nick (01:01:18):

What book or article has had the most impact on your life?

Pablo Carranza Velez (01:01:22):

I don’t know if impact on my life, but I really enjoy science fiction and I would say Hyperion and the whole Hyperion series has been really eye-opening in terms of visualizing potential futures for humanity.

Nick (01:01:37):

Is there a movie or a TV show that you would recommend everybody should watch?

Pablo Carranza Velez (01:01:42):

I don’t know if everybody, but if you like science fiction, then watch Battlestar Galactica if you haven’t.

Nick (01:01:48):

If you could only listen to one music album for the rest of your life, which one would you choose?

Pablo Carranza Velez (01:01:53):

I think that would be a pretty sad life, but I would probably choose something by Pink Floyd. I don’t know which one or something like that.

Nick (01:02:04):

What’s the best advice someone’s ever given to you?

Pablo Carranza Velez (01:02:07):

There was one particular piece of advice that was to say thank you more than sorry in a lot of cases. This was very concrete in work and business. There are a lot of places where maybe we default to saying sorry and actually maybe we mean thanks. Like sorry for delaying responding or, well, actually thanks for your patience, so you can phrase things in a more positive way. And I think this has been super useful because also internally, it helps to stand differently, I guess.

Nick (01:02:39):

Pablo, what’s one thing you’ve learned in your life that you don’t think most other people have learned or know quite yet?

Pablo Carranza Velez (01:02:46):

I don’t know if most people realize, but you can change your life if you are not happy. I think that’s something I learned over time. There were situations I was in that I didn’t really enjoy or I didn’t find fulfilling, and then I realized I could actually change it. It takes work. Sometimes it’s harder, sometimes it’s easier, but you can do it.

Nick (01:03:10):

What’s the best life hack you’ve discovered for yourself?

Pablo Carranza Velez (01:03:14):

Registering for races. I like running, but I realize that I’m very objective driven so if I don’t have a race to look forward to, I don’t train. So my life hack is I set myself ambitious goals and then I force myself to train for them.

Nick (01:03:32):

How about this one, based on your own life experience and observations, what’s the one habit or characteristic that you think best explains why people find success in life or how they find success in life?

Pablo Carranza Velez (01:03:44):

I’m actually not sure. I think there are probably different things that work for different people. Maybe there’s a thing about loving what you do, but I think there’s also a mix of practice and luck and the luck that comes from a lot of practice. But I don’t know. Probably perseverance. There’s probably many answers that could be valid. I don’t know if there’s one that fits all the success cases that I can see.

Nick (01:04:12):

And then Pablo, the final three questions are complete the sentence type questions. So the first one is complete the sentence, the thing that most excites me about web3 is…

Pablo Carranza Velez (01:04:21):

Building better coordination mechanisms for humanity.

Nick (01:04:25):

And this one, if you’re on X, formerly Twitter, and I guess now I should mention Farcaster, then you should be following…

Pablo Carranza Velez (01:04:33):

Vivid Void. Nothing to do with web3, but I think he’s very insightful.

Nick (01:04:38):

And then the final question, Pablo, complete the sentence, I’m happiest when…

Pablo Carranza Velez (01:04:42):

I’m happiest when I’m around people I love.

Nick (01:04:55):

Pablo, thank you so much for coming back on the GRTiQ Podcast and allowing me the opportunity to do an in-depth interview with you. I definitely appreciated that L2 panel, which seems like a lifetime ago, but it’s been incredibly fun to watch the success of that initiative. And now you’re working on Horizon and all the other things that you and the Edge & Node team are working on. So thank you so much for your time. If listeners want to stay in touch with you, keep tabs on the things you’re working on, what’s the best way for them to stay in touch?

Pablo Carranza Velez (01:05:23):

I think they can find me on Twitter. I’m not very active on social media. I’m on Twitter and Farcaster. I have the same handle on both of them. If it’s things about The Graph, then posting on Discord on the forums, you can usually find me there. But maybe over time I’ll become more active in social media or maybe not, but if you message me there, I’ll probably see it.