Blockchain Innovators - Conor Svensson and Ben Edgington

Hi it's Conor Svensson here, founder and CEO of Web3 Labs. This is a conversation I had with Ben  Edgington ETH2 researcher and product owner for Teku at ConsenSys. Ben comes from a very technical  background having started his career working on supercomputers at Hitachi where he worked for two  decades before he got sucked into the blockchain vortex and joined ConsenSys in 2017. In our  conversation we cover a lot of ground on Ethereum and the transition off proof of work to proof of  stake which is something he's very involved in and has seen firsthand how it's evolved during the  past three years. Ben's a very well-known figure in the Eth2 community and if you have an interest  in keeping up with this fast moving space, I highly recommend you check out his newsletter  'What's New in Eth2' which is available at eth2.news. Thanks and I hope you enjoy the  conversation as much as I did. Hey Ben, it's great to have you here. Conor, it's wonderful  to appear on your podcast. I've enjoyed listening to a few episodes and yeah I'm amazed to be here.  Well, thank you. You've started to do a fair bit of podcasting yourself as well in the last year I  guess? Yeah we had this thing going with CoinDesk which was nice. We had about a nine month run but  then my co-host moved on from CoinDesk and we didn't pick it up again. So if there's a podcast  slot open, I would gladly fill it! So it was a good experience, I enjoyed doing that. I'll tell  you what though Conor, it's exhausting having to have an opinion on every topic! I'm usually kind  of like 'yeah whatever' for most things but if you've got to appear and talk about something, you  need an opinion. So I found that prepping for the podcast week by week and trying to work out what I  really think about this was quite an interesting process. Yeah, absolutely, and one thing though,  no doubt you do have a very significant opinion on or at least awareness of is the Eth2 ecosystem but  rewinding before that, you were at Hitachi for just shy of 20 years until roughly four, well just  four years ago. What was it that made you jump from that world into blockchain? Because you've  gone really, really deep in terms of what you've been doing and we'll of course get on to that  in a bit more detail shortly but you worked in this very large corporate for so many years  there must have been something pretty exciting that really sucked you in? Yeah, right down the  rabbit hole now! So I started my time at Hitachi doing very technical things. I spent about eight  years working on super computers doing very low level optimizations as well as parallelization of  code and lots of kind of computer sciencey type of things. I'd come out of academia to do that.  Hitachi pulled out of the supercomputer business in Europe and I kind of fell into a generic  techy sort of role doing all sorts of things like data center infrastructure  and cooling. I worked on IPV6 routers for a while, biometrics stuff, a bunch of fintech security,  just loads of random things and you know, as is the way, sort of drifted into management and then  ended up sort of moderately senior. So my job title was Head of Engineering for one division  of Hitachi Europe and we were mostly doing fintech stuff, so mostly working with banks in the City of  London and across Europe, biometrics and other security software. Yeah, it was a kind of quiet  life really, could quite happily have driven my desk for another 10 years and retired, you know  retired well. Around the beginning of 2016, all the sort of sales people were coming back  and reporting 'yeah, all our clients want to talk about is something called blockchain,  can you find out what it is?' They were talking to the banks and it was hot, hot, hot at that time.  So I did a bit of digging and came across Bitcoin, I thought interesting, but not that interesting,  read up on Ethereum and was just blown away. This captured me, it you know entranced me and I just  started spending evenings, weekends, every spare hour just reading everything on Ethereum,  learning everything about Ethereum, hacking around with smart contracts and just generally  learning, reading the yellow paper and learning from the bottom up.  Eventually, I got the opportunity to make that a full-time job. I realized I was bored  of management, it wasn't for me. What excited me was doing the hands-on techy stuff and the day job  just didn't give me that that opportunity anymore. Yeah, so four years ago, as you say, I joined  ConsenSys and no regrets. It's been an incredible journey. And so much just happened in that time!  Going back to Ethereum specifically though, what would you say was the thing that pulled  you in? Was it the technology or the community or a combination of both because they were the  first second generation blockchain. But I think from my own personal perspective, I've always saw  parallels with the Ethereum community is the Linux community without the malevolent dictator,  but I'd love to hear your thoughts there. Yeah, community is super important and I made my first  sort of tentative steps, a couple of PRs to the Solidity compiler and they were so nice to me  and so welcoming that I immediately felt part of that world and that was great. Technology  was fascinating, what I liked about it was it was all about computing in a  resource-constrained environment, which is what I've done on supercomputers. You don't think of  supercomputers as resource-constrained but it is if you can speed up your code by 10% you can  save a couple of million dollars of hardware and so there's a lot of pressure to make them  perform really fast. So all the same things about memory constraints and micro optimizations were  happening on the EVM but you what really captured me was the drama! I was getting interested in this  through the DAO hack thing, I mean the DAO itself was fascinating and then the hack and  the adventure that that was and the drama around it just kind of hooked me completely  and and dragged me in. So it's a combination of all of that and it is a bit like living in a soap  opera sometimes but yeah it keeps me engaged, it's great! Yeah, absolutely. I remember going along to  an Ethereum meetup back in Sydney, I think it was a few weeks literally before like the DAO  hack happened, and the person there was talking about how this incredible thing's been written,  it's so few lines of code and it's basically like an automated investment vehicle and  we know what happened then! Yeah, I never invested in the DAO basically because I didn't really get  it. I just like to understand what I'm getting involved with. I don't like to ape into things,  to my detriment it turns out, but yeah I dodged a bullet on that one. But, yeah it was fascinating.  Yeah, yeah and so certainly with the NFT market now as well, that definitely doesn't make too much  sense there either in certain respects but there's some people who really do get it out there. Yeah, I mean it's fascinating how it's all evolved over the past four years. I reckon  that when I first got interested in Ethereum, you could pretty much keep up with everything that was  happening by spending a couple of hours a day on Reddit and then you were across almost everything  that was going on. Nowadays, it's impossible - completely impossible - to keep up, there's DeFi,  there's NFTs, there's a whole host of stuff at the protocol level which I try, that's my niche to  try and document some of that and never mind all the other chains, the roll-ups, the side chains  and everything happening there and then you've got the regulatory stuff and you know it just goes on  and on! So one of the things I've had to do over the past few years rather rapidly is kind of close  down my areas of interest. I used to try and know everything and now I know a lot about very little!  But that very little is very significant given everything that's happening with Ethereum and  it's not really correct to call it the transition to Eth2 so to speak, I guess  it's the whole merge as it's being referred to. But, when you started at ConsenSys you were  within the protocols division and then, I think it was roughly a year after you started there,  you began putting out 'What's New in Eth2' which was like a newsletter and as you said,  there was so much information and it was so hard to keep up with it all even back then  you just wanted to start sharing some of what what was happening in the wider community. So  when did you start working on you know, move across to Eth2 bits? It was obviously  before you started writing the newsletter. Yeah, it's an interesting sort of history.  So I landed in PegaSys which is a protocol engineering group of ConsenSys, it was quite  small at the time. I was about number 12 or 13 through the door, we're now 70 plus people. I  first started working on enterprise Ethereum stuff because you know, with the corporate background  and the fintech clients that kind of made sense but it just felt too much like doing what I was  doing before and it wasn't the passion that had driven me in it which was much more about the  open permissionless mainnet world. So with the permission of PegaSys, which was very generous,  I moved over to really to kind of pioneer PegaSys' mainnet work there. At the time ConsenSys had  no relationship with the Ethereum Foundation for historical reasons and somehow I found  myself on a call with Joe Lubin and Vitalik, I think it might have been just the three of us  somewhere in January '17 or December '16 - December '17 or January '18 somewhere around  there, and Vitalik was saying 'okay we need some help, we need some protocol research'  and it just felt really right, really opportune. So from then on I devoted myself to sort of  building PegaSys R+D, built it up from a couple of people up to 20 plus now and a lot of that  work was around Ethereum protocol research, mainnet stuff as well as some enterprise stuff.  So a year into that I did various workshops. There was a workshop in Taipei on sharding,  which is great, and then we sort of scoped out what Ethereum 2 would look like in July in Berlin  of 2018 and later that year, around September/ October time, I kicked off the project which  became known as Teku which is our Ethereum 2 client. It was a research project at that time,  we we didn't have a goal to make it a mainnet production client. It was really part of our  being involved in Ethereum protocol research and having a seat at the table. We're writing  the code so I managed to get a team together. Joe was a great help in moving people around and that  worked out really well and we built this thing. After about a year, so we're up to late 2019,  we sort of looked at what was there and thought 'actually this is okay, this has got promise, it's  more than a prototype or a research project, this has got potential to become a mainnet client'.  So we have a decision-making process in PegaSys called SPADE, I can't remember what it stands for,  but it involves a lot of consultation with just about everybody. So we did an exercise,  Rob Dawson who's now CTO in ConsenSys led that and we gave it the green light. We gave Teku,  it was called Artemis at the time - we had to change that for the usual tedious reasons but  it's now Teku - and we transitioned that to product development. So we're now a product  team and I'm product manager for Teku which is a mainnet Eth2 client or Beacon train client  or ConsenSys client, whatever terminology you prefer to use. In terms of like how the space  has evolved as well, certainly originally with this whole migration with Ethereum  going from proof of work to proof of stake it was viewed as being a number of  phases that need to happen with phase zero being the Beacon chain launch which happened  last December and then this notion of sharding and then the actual application level, the cut over  so to speak, but that's that's all got sort of ripped up in the last 12 months and I presumably there's a lot of specifics into why that happened that you could tell the listeners about?  Why there was such a sudden change there and why we're in the trajectory we are now.  One of the things I enjoy doing is stepping back a bit and sort of comparing the world I'm in now  with the world that I was in when I was working for a massive Japanese multinational corporate  and the different ways in which things get done and you know it actually accomplished a lot,  no criticism of the way they do things. But, how things are done in Eth2 world or Ethereum world  is very different, very bottom-up, it is a lot of cat herding. We have the Ethereum cat herders and  we're reliant a lot on people coming up with new ideas, coming into the ecosystem bringing ideas,  taking leadership without necessarily having leadership conferred on them  and there isn't a sort of hierarchical top-down structure. Vitalik will come up with some ideas,  he'll throw them out there and then it's up to people to pick them up or not pick them up  and and things like that. So it's the cathedral and the bizarre model,  which I've written about a few times, this is Eric Raymond's kind of classic essay  on open source development and we're fully in the bazaar and it's noisy and it's chaotic  but yet somehow, really incredible work emerges from it as it becomes ordered and  order emerges from chaos. So yeah, it's fair to say that the next generation of Ethereum,  the future of Ethereum roadmap has had a few pivots over the last couple of years.  Partly because new technologies and new things come to light.  So two things which kind of deflected us from the older road map, so we had a  phase zero, phase one, phase two road map where we would deliver the Beacon chain, which we did,  and then we would deliver data sharding and then we would deliver executable shards,  execution environments and after that we would turn off proof of work and migrate Ethereum 1  over to Ethereum 2, a sort of multi-year project. This got truncated so we delivered phase zero  and now we're doing proof of stake straight away, this is the next thing we're doing and after that  we're doing data shards. Then we sort of basically have delivered the roadmap. And a couple of things  changed that - one was the emergence of rollups as a technology, layer 2 technology on Ethereum.  So they give us a lot of what executable shards would have given us. They take out of the protocol  layer the management estate and the management of execution. They just rely on the protocol for  data. So the on chain you have the data, off chain you have the execution and the state management  and that solves a lot of problems for Ethereum. We can store data on chain okay, especially  when we've got lots of shards. It was the state management and the execution that were expensive.  So this roll-up ecosystem that's emerging makes a lot of important parts somebody else's problem,  which is nice from a protocol developer's point of view. We can simplify the base protocol and then  it facilitates an ecosystem where people can run many, many experiments and have  you can have many different types of roll-ups and some will succeed, some will fail  and you can try and iterate many different solutions. Just stuff we cannot do at  the protocol, there we have to be really, really conservative at the protocol level,  it's impossible to break anything - we mustn't do it. So that's been a big breakthrough. The other  one was my colleague in ConsenSys R+D, Mikhail Kalinin, just envisioned what we're calling  'The Merge' now, which is we take the Ethereum proof of work engine so the Eth1 clients that  everyone's familiar with like Geth or Besu, and we basically just turn off proof of work in them  and plug in proof of stake. We already have proof of stake running in the form of the Beacon chain  and Mikhail envisaged a very straightforward way to do this. So this has become the plan  and the pressures were on, right? I mean the narrative around proof of work is getting  increasingly sort of toxic. This thing about NFTs are destroying the planet and burning down the  rainforests and things. Yeah, it's about time we deliver proof of stake, it's been too long.  Absolutely, and just to go into some more details as well about those two specific areas,  like the roll-ups versus the shards for instance, you have a number of different networks that have  been emerging, what's more broadly being described as these layer two networks for Ethereum,  Polygon being the most popular one but there being a number of others. In your minds and certainly  Vitalik mentioned this too I think, zero knowledge roll-ups are the preferred technology for scaling  but is it fair to say they're not quite at a point yet where they're readily  available and so hence a lot of the layer two technologies that exist right now aren't quite  fulfilling that more utopian vision of where it's going to get to but we're not too far off?  Yeah, I have a kind of layperson's knowledge of this stuff because, as I mentioned earlier, I've  had to focus and roll ups are some of the things which is sort of adjacent to what I focus on but  not core. Though I think I have some intuition for how it goes and there are basically two paradigms.  One of which is called optimistic roll-ups. They rely on something called fraud proofs to remain secure so anyone can challenge the execution on the chain if the roll-up operator  does something which is incorrect then you can challenge that with a fraud proof and there's  a sort of crypto economic mechanism to reward correct fraud proofs and punish bad ones and  so on and punish bad operators. Zk rollups have a sort of validity proof thing going on whereby  they post proofs to the chain that they acted correctly and the proofs are very easy to verify.  This is a much smoother, a much more effective way to operate. With the optimistic roll-ups there are  things like you can have a week-long withdrawal delay, if you want to get your funds out,  your NFTs out to give people the opportunity to generate the fraud proofs and make sure they can  get on chain without being censored whereas the zk roll ups don't have this issue. As you alluded to,  this zero knowledge cryptography and it's not the zk part that's actually used,  it's the succinct part of it but that's getting a bit technical. This cryptography is very new.  It's really been built in the last couple of years and is developing you know,  just super fast. There's this cambrian explosion of different techniques to apply it. It's far from  being mature yet. The EVM, which is Ethereum's Virtual Machine, the engine that runs stuff  is not designed to play nicely with zero knowledge proofs and the SNARKs and the STARKs  which we have. So, if you want to compile your Solidity program and run it on a zk roll up,  this has a lot of complexity and overhead behind the scenes and people are working on that.  There are domain specific languages like Cairo and Zinc, which allow you to write directly in kind of  zero knowledge specific languages. But this is all so new Conor and is really rapidly evolving. It's  hard to know quite where we are or where we're going to be in a year's time but from a technical  point of view, the zk stuff has the most promise, it's the most scalable and it's the most efficient  of the technologies out there as far as I can see. Then when we talk about the convergence with  the merge with the Eth1 and Eth2 clients, how will that kind of pan out longer term because nearer  term you have your various different clients to Teku and Besu say in ConsenSys. Is the case for  the proof of stake and proof of work networks? (I'm trying to avoid using the Eth1  and Eth2!) You don't have to worry about that with me Conor, I publish What's new in Eth2,  you know I own the website eth2.news, I mean you know, I'm fighting to preserve this technology!  It's a terminology. Do you think that it will make sense longer term the clients to merge  into one because coming at it from a very simplistic and probably naive perspective,  ultimately it's like the consensus mechanism that's being replaced with the merge,  is that too simplistic of a view of this or do you think that's feasible longer term?  Yeah, it's definitely feasible. It's something we're thinking about in PegaSys. So the  initial design is you can choose any of the available Eth1 clients as in execution clients  so that might be Geth or Besu or Nethermind or Erigon and you can pair it with any of the  available Eth2 clients or ConsenSys clients which is Teku, Prysm or Lighthouse or Nimbus  currently, Lodestar is doing good and they can combined act as the joint Ethereum 2 client.  So they're well encapsulated, we've defined an API by which they speak to each other  but of course, you're getting two components from different sources, they've got different stacks,  they've got different documentation, different metrics, different logging, everything. So  we're in the happy position in ConsenSys of having both an Eth1 and Eth2 client. We have Besu and  Teku, they're both Java, they're both Apache 2.0 and so we're exploring what it might look like to  not exactly unify, but perhaps bring them closer together so that if you want to run an Ethereum  2 client or a future Ethereum client post merge, you can just deploy this one thing and it's all  in one package with consistent documentation and consistent metrics and it's all there. With Teku, our target users are primarily institutional, the professionals,  the power users who are hosting thousands of stakes. That's our sort of user persona as we  plan our product strategy and so thinking through from their point of view, they can  dual source their stuff, they can take Geth and Prysm and run them together but if it goes wrong,  then who do they go to? Who they shout at? Who do they point the finger at?  This is always a classic issue in purchasing, you know having accountability for the thing. So we're  thinking we might not get much technical advantage from combining the two clients but there might be  a nice point where you've just got a single source and so if it goes wrong,  you can just come to PegaSys, we'll fix it. You don't have to worry about which bit's gone wrong  and stuff like that. So I think there's not so much technical advantage but I think  there's some nice usability and marketing advantages to sort of combining the two. So  we're definitely looking at it but it's not going to be for a while. Yeah, absolutely. So  going a bit further into the whole topic of the merge, you've just got back from Greece recently?  Oh yeah, you can tell by the tan! We spent 16 hours a day in a windowless basement  which was a shame (!) but yeah, we got basically dev teams from most of the clients,  Eth1 and Eth2 side and from the Ethereum Foundation and a couple of ConsenSys R+D teams.  We all got together in an undisclosed location in Greece and spent the week basically  building a merged testnet. So what this means is getting the Eth1 clients  represented there and Eth2 clients in all the combinations I previously mentioned,  working on a single testnet and running Ethereum on proof of stake or simulated proof of stake.  It's the first time we'd done anything quite at that scale. We'd had a sort of hackathon earlier  in the year where we'd done an early version of this but this was a huge leap forward and we  reckon we saved about three months of remote work by getting together in one place and you know,  after two years of not leaving my house it was kind of nice to get together with people again!  Yeah, I hadn't flown anywhere for two years, which is unheard of in the last like 25 years, so  that was kind of weird. Yeah, yeah. So it was a great success. We set some milestones,  you know, fairly easy ones. First milestone - pass all the spec tests. Second milestone - run your  client with one other client, so if you're running an Eth2 client run it with one of the Eth1 clients  and vice versa. Then milestone three is kind of run combinations, so you've got multiple  clients on the same network. Then I can't remember four, but five was run a test node with basically  all possible combinations and keep it running for several days. So we did that and yeah,  it was great, a really good experience. Previously of course, there's been times where client teams  got together but I think presumably it's the scale of it this time around was different  because you had both Eth1 and Eth2 as well as some researchers around it, correct me if i'm wrong  but previously it was more you'd have Eth2 focus teams working together? Yeah, yeah these things  have been developed pretty much in parallel. So the Beacon chain and these sort of Eth2 plans  have, until quite recently, been done quite distant from the historic Ethereum network stuff  so until a couple of months ago, Ethereum 2 or Ethereum on proof of stake just hadn't even been  discussed at the regular all core devs call and we started this Beacon chain almost a year ago,  now running proof of stake and so on. But, it is in parallel to the main Ethereum network and the  two sides - the traditional Ethereum 1 developers and the Eth 2 people - really didn't have much  point point of contact. That's all changed and this workshop we did was the sort of tangible, the  practical evidence of the change but we're also getting the proof of stake, the merge, discussed  on recent all core devs calls and now everyone has agreed from all the client teams, Eth1,  Eth2, this is the next thing we're doing. We're not going to do any other upgrades to Ethereum  before we deliver this merge, this proof-of-stake Ethereum. And in terms, of  course, the question that everyone's asking is - when that's going to happen? Previously,  I think six months ago or so, the speculation was towards year end this year maybe Q1 next year,  is there a feeling from the community on that at the moment? Within this year was always over  optimistic. The timing is very slippery in this space Conor and I'm very hesitant to say anything  definite just because nobody is in control of it, right? If everything goes perfectly smoothly and  there are no hitches and all the governance is happy and there's nobody in the community you  know, waving a red flag then maybe towards the end of Q1, beginning of Q2 but realistically,  it never goes as smoothly as you hope and we are a community of optimists, we wouldn't be doing this,  we wouldn't be aiming to change the world if we weren't optimists.It sometimes kind of feeds over  into our timing estimates, so I'm hesitant to put a definite date on it but I would be extremely  disappointed if we didn't deliver in 2022 and I'd be a bit disappointed if we didn't deliver  in that first half of '22. So that's pretty much as specific as I'm prepared to get. Yeah, yeah,  I know it's certainly exciting times, especially seeing the two communities really working together  and like you say, there's not going to be any other big changes until this happens. That's going  to incentivize a lot of people to make it happen in as timely and but safe manner as possible.  So say if we catch up again a year from now and the merge just happened,  what's next? What's going to be next for you? Are you going to be deep looking at sharding or  are you going to be starting to switch your focus somewhere else because you've been so  focused on this for so long now. Oh personally? I mean, yeah the Eth2 effort rolls on. So there are  plans to do a post-merge cleanup which will enable withdrawals. So currently you put down the stake,  you don't know when you're going to get it back, nobody can actually take their  earnings out of the Beacon chain yet so we need to enable that, that'll come a few months  after the merge. Then sharding is the next big thing which gives massive scalability and then  there's a whole bunch of other exciting stuff so you know there's a whole career to be had  being involved in that. We'll see how it goes. I've got no particular expectation.  There's a certain amount of internal encouragement within ConsenSys to sort of pick up again the  management/ corporate side of things. You know, I've got 20 years of it under my belt  so getting involved in the more sort of corporate level stuff especially as ConsenSys is  looking more like a traditional software company, it is attracting some serious investment  and is becoming a big player. Honestly speaking between you and me, I hope  nobody's listening (!), I'm not overly excited by that. You know, lesson learned, the techy stuff  is where my heart is. So I'll probably find something like that to to fill my time  but it's also like a, I think, a classic problem in terms of as career technologists, so to speak,  with regards to being able to remain hands-on but also grow within organizations. I think that you  almost see this classic thing where if someone starts off as a developer, they become a mid  level developer, senior developer then maybe a lead developer and then it's management and  management and management and management and if you're very lucky, you end up in a place where  you have that and I remember working for an investment bank a number of years ago,  we were in a fortunate position for a time where we had managing directors who were still coding  but that was really an exception to the norm and as you say though, there's so many people,  like yourself, like people who work in my company as well, who really want to stay very hands-on but  you do feel a lot of the time that organizations as they grow they often have that classic problem  where the opportunity to continue to be hands-on and this isn't of course reflecting on ConsenSys,  but it's just more of an observation about the technology industry more generally.  So I think it's really important that we can find ways to champion people to have careers like  you're doing now, where you've come in and you've got all of this significant experience but also  very significant breadth of technological experience and you want to see people,  I think it's important for the technology community to have more people to look up to  like that. Yeah, it's a classic dilemma isn't it? You know, I don't have any great answers! I'm  very thankful for ConsenSys facilitating me over the last few years. Largely, I've  carved out my own path and have been allowed to do so which has been terrific. I think  in more traditional organizations you don't necessarily get the opportunity to do that.  There's been a huge amount of support and I think other people are finding that as well. We have a  lot a high degree of autonomy within the company and yeah, I've learned to really value that.  Yeah, I don't have answers to the questions of what do you do with elderly technologists! What I am convinced is that people are most productive when they are excited about what  they work on and I tried to get excited about management stuff, making power points and things,  but ultimately it didn't get me out of bed in the morning whereas the things that I do now motivate  me enormously. Yeah and you can completely see that when it's moving at such a rate as you've  mentioned earlier on. The whole ecosystem has grown so much in the last while. On reflection  as well, do you think that what's happening right now in terms of blockchain, if we're going back  probably the last five years of activity and we know that the next five to ten it's going to be  equally as transformative, people certainly have parallels with the growth of blockchain adoption,  it's growing at double the rate that the internet grew at in terms of user adoption. Do you feel  that there were other moments in the last say 20/ 30 years where you think there's been other  pivotal technologies beyond the internet and crypto and blockchain or do you think,  because I often hear people talk about how this is like a once in a generation thing that's happening  right now? I'd love to hear what you think of that statement. Yeah, it does feel like it doesn't it?  I mean, I sat on the sidelines during the dot-com boom. I was working at Hitachi at the time and  my colleagues were all off going joining crazy startups and having you know,  the time of their life and I was very risk averse and watched it all all happen from the sidelines.  Part of my thinking was, when this blockchain thing started to look like a paradigm shift,  I don't want to miss out again, this is my last chance, I'm not going to sit and watch it all  happen. Nobody died during the dot-com boom, everyone survived they might have had a wild  ride but everyone was fine, so risk is low - just dive in. I don't know about other technologies AI,  machine learning, things like this are quietly changing the world. The mobile devices, the whole  thing was inconceivable when I started my career, that I could wander around with the world in my  hands, access to the world's information, do video calls on my phone, you know just  astonishing and these kind of revolutions are happening all the time around us. But  I don't know what it is about blockchain that feels like a paradigm shift. This sort of web  3 narrative is really interesting. The internet of value rather than the internet of information,  lots of memes around that. Yeah, it's going to be really interesting to see how it shakes out and  in five years ten years time what impact will we have actually made on the world. There's a lot  of noise and hype at the moment but seeing what actually lands is going to be really interesting.  Yeah, absolutely. And just one other question on that, if you weren't working in protocols,  which area do you think would interest you most or captivate you most within blockchain stuff? So,  I initially interviewed for ConsenSys Diligence to be a smart contract auditor. PegaSys grabbed me,  which was fine, and I don't know whether I would have been a good auditor or not but  I do love that level of detail. So the way I like to understand things and how I kind of  ended up doing what I'm doing now is, I understand stuff from the bottom up.  I need to know about the bits and the bytes about the machine code and the executions and  the packets on the wire and build up from there. I'm not a person who can look at an app and then  dissect it and go down the stack and that's why I have trouble with kind of modern computer science  where they graduate and they don't even know what hexadecimal is and I'm like, really? So that lends  itself to the protocol work that I'm doing now and also to the kind of smart contract auditing  side because it's very, very detailed. You really have to know what's going on under the hood. You  can't make any assumptions. So I don't think I would be a DeFi Degen or anNFT artist or any of  that, but that would have been a decent path as well! You'd be one of the key people making sure  it all works and it's secure so it's certainly a very, very important place in those worlds.  So if people want to connect with you and keep up with what you're doing, of course, you've got  the Eth2 news but what's the best way for people to reach out or find you?  Yeah, eth2.news, so I write every two weeks on What's New in Eth2, that's the title. So it's just  a roundup of what's going on, fairly technical but occasionally I let an opinion hang loose in there,  and I've been doing that for three years or so. You can find me on Twitter, I am @benjaminion_xyz,  don't ask!, and yeah Twitter is a decent place to contact me if anyone wants to get in touch.  Wonderful! Well Ben, it's been an absolute pleasure to have you here today. I look forward  to catching up again soon. Yeah, great fun talking Conor. I mean there's so much that we could cover  but appreciate it, thank you for considering me an innovator, I'm proud to join the ranks of the  others that you've spoken to and humbled as well. Yeah, good conversation, thank you. Thanks Ben.