The story of a Dapp developer | Fabian Vogelsteller (August 2024)

I've been around since uh 2015 in ethereum um how many people have been around since more than 3 years in the ethereum space all right okay more than five okay good good then you might have heard about the M browser actually that was a the concept video that uh Alex fand the ux design of ethereum did and he created this video in November 2014 and he basically came up with this idea of you know let's create a web 3 browser you know where we can visit Dent applications um and uh you know and he literally came up with lots of things I mean you already see like a tradable company transparent democracy pretty much Dows right um marketplaces obviously and exchanges and crowdfunding and escrow and Family Trust and pretty much not even everything we have built yet in a video in 2014 you can find on YouTube um which is um it was quite a foresight I would say that Alex had here um and this was before the network was started so that was really just inspired by the ideas that were around you know uh through the white paper so I joined then and 2015 in January because uh Alex and uh some people in the them they were looking actually for someone to help with building this idea and that's what we did we built the my browser which was also the etherum wallet and then people called it the misted wallet and it got all very confusing uh because the idea was that the eum wallet is One D app in the M browser and then we also released the version that was Standalone and then people were confused anyway it was probably a very good-looking wallet at the time and I just found a video that had literally uh lots of those cool screenshots that I haven't seen in over 10 years myself um so but you had like these uh accounts and you could give them little names and there were also smart contract accounts I mean believe it or not but the etherum wallet had smart accounts um because you couldn't see any transactions thanks to I guess Gavin's Choice uh probably um you can I mean and also saving data on on on your note right um You can't see past transactions so we had to actually deploy a smart contractor to make that happen and Gavin actually wrote that smart contract he wrote the first multis which also became then the official multis of eth for very long time and that's what we used in the Mis browser so you literally got a multi smartcon account out of the box you just needed to deploy it that means you needed to get ether which made it all very bit more complicated also the M browser and the eum wallet ran a full note that was pretty fine for the first few months syn sying was kind of done in a few minutes then as the months passed that got a little slower um yeah here we see some nice transaction screens um you know I made it here with like a progress spot as was going through but I mean we got outco competed by probably the more ugly version of the wallets of today that have tremendously changed in UI since 2017 as you can see um but the idea was really like that uh you know the the big idea of the metamask was to not run a full no and just use an RPC endpoint and then you basically get your instant login um and making it a browser extension which was pretty useful too we kind of had to make up all the way of how you talk to an app in a browser we had to make this up for the M browser and that's what also the metamask used and that was in you know 20 15 and also in the same year um I took a proposal from vitalic created a proper specification and called it a the request for comment to get some comments in that turned out to be pretty quickly the idea of the Sentry no the the specification of ethereum for tokens um and pretty much you know um blew everything out of the water um when it comes to ethereum so esc20 then it kicked off this whole wave of standardization and it kicked off the whole wave of tokens and the rest is history right defi icos pretty much all of this came out of this um and a lot of D apps um because of a single standard so this was kind of like my work at ethereum working you know on the M browser also web 3js which I didn't add slide four um proposing this that was really no work I mean that we just took like a proposal changed it up you know and wrote a spec that was one hour of work then we discussed a little bit and out came the the whole mess that the the is today um but there is actually a standard that I then proposed like two years later we're going here like forward in time and that was a bit more by coincidence in a way so I I had a workshop um you know with like a Consulting workshop and U the idea was to think about a smartphone account um and it came up with an idea and after I got home I thought okay I should make a proposal and I created ERC 725 um for the ones who know their standards history that's in 2017 there was the exact proposal after 7 to1 which is now the nft standard so I came up with this AUM identity standard I call it actually a proxy account right there I called it a proxy account um later this was translated probably to account obstruction I called it the proxy account um and the idea was basically use a Smart contract instead of as a as an account instead of like just an EA um but that was also a bit more than the idea of of what we had in the M browser which was a mulick um so I proposed that standard and they kind of like created this whole little battle between uport which came from consensus because they wanted to make the theum standard and then they were all got pissed that this didn't suddenly got attention and I had actually no I interest to get any attention on this I just thought I put it out there and then somebody does something with it so you like kind of tried to explain why this is not a good idea blah blah blah um on the end uh didn't happen okay there's a bit of a double double slide here um so what happened is it was kind of lingering for a little bit um but I also realized you know then in 20178 like we have this big problem that kind of we have all of this Cool Tech and obviously icos were a thing you know and Def was a thing but I mean we didn't really have any users and funny enough we are still today at the same position right we barely have any actual smart contract interactions except for defy trading you know and maybe lending something and swapping something but where are the Dows the actual Dows not like those 10 billion Dows you know that hold a protocol token um but where are the people right and the problem is really it's really hard to use and if you make a bunch of calculations you can probably come up with this assumption so um that nobody's using blockchain um I mean even in in in crypto conference it's hard to find people to actually do something with a D app then in 2018 I decided to leave the etherum foundation in August um with this nice little emoji and walking out of the room here um and that's because I started my own project called luxa and uh the idea was here really to change this whole thing up you know how can we make this useful and yeah years 75 account what I showed there was quite an important thing um long story short 23 we started the network and it's actually exactly like ethereum so it's actually a clone of ethereum it's not an L2 sorry it's an L1 but it's the exact same thing like a theum and run with 20,000 validators and why did I do that and the answer is standard proposals that's actually the the the the the reason why really luxu exists in the first place and probably as You Realize by now there's thousands of standard proposals and there's a whole like discussion battle and everything and probably a lot of people in ethereum kind of like don't like me so much that I went somewhere else started a whole new chain and on top of this make all standards completely somewhere else than where everybody is debating the answer is it's really hard to get adoption around a whole set of new standards that fundamentally work different and change the foundations if we do that on all outdated standards that would just wait up on made up on a whim and then somehow everything H happened from that and we have a network effector on something we literally can't change so you probably know how long sometimes one single standard debate you know takes I think uh 4337 you know it's a bunch of Standards combined into one because it was too complicated to get agreement around even one of those right just make a monster standard and say this is it um so what happens if if you have um and that basically this whole process took 5 years um but what happens if you have a lot of new standards ideas and you want to like fix all the stuff like the tokens and the nfts and you know allow the there you know any kind of metadata that you can attach and be more flexible and inform each other and all this kind of stuff so it gets more complicated when you want to build foundational building blocks that are not just all kind of like a version of esc20 with some one extra thing um and even more complicated is and we have that discussion right now in ethereum if you come up with a whole smart contct account so when I started this whole thing in 2017 it was just one basic idea uh turned out that that is a very good fundament for actually creating a real smart conon account um and uh that's what Universal profiles is I mean not to go too much into technical detail but it's actually funny because I just wrote yesterday Jeffrey by accident uh or by by coincidence actually Jeffrey wilker he's the founder of go ethereum um he's the one who hired me building the M browser and yesterday I just saw his message popping up in Discord because he was writing in his in his Discord which I'm also in about his game GD online the game he's building and uh he's the one who literally in 2015 in March told us that we should not expose a public key in the M browser to users or to D apps they should only see a smart Conta account and they should never see the private key or the public key of the private key because that's stuff that's not really you know persistent and ideally they can change it and it should be a fe a FAL and not like something you really use so 9 years later like for real 9 years later there's a smart account uh and as has been lots of uh you know obviously directions I mean Kos is you know build a multi sick you know and and a modular account system um but there is this benefit of being like in this space where you don't have so much noise and you have to debate every single change with lots of people and like it's a very slow progression specifically when you don't want to do something more complex where you have multiple contracts you have to interact with each other and permission systems and key managers and and all of this stuff um yeah it's sometimes better to just step out you know go to a different kitchen cook there make the meal and then come back and say here look try that you know and I mean these smartcon accounts work equally on ethereum and um we re build the browser extension that we will open up to eum as well and on the end the beautiful thing is if you do it the right way it looks under the hood or looks for the DF developer just the same way it looks for for an EA right it's the same exact thing so it can this is the beautiful thing about smart contct solidity and and all of this stuff you can make things look extremely simple um but you can also make it then look nice and it then actually becomes a real account um and that's kind of like the story you know how I came from like accidentally proposing something and then really going down the rabbit hole into standards proposals and uh created a whole set of them now um that can also be very useful and will hopefully be very useful to ethereum and any of the other evm chains because it is a smart conon account that is a wallet at the same time it's your account it can be extended and can have information can have everything that you need in one uh account one smart contract you know as the core how I call it the root identity of yourself yeah any questions okay question running hi just simple one you said that it's long to go through the standard creation what's here the process how you shorten the time from initial to sort implementation I mean I I skipped over that slide it was from 2008 18 to pretty much 23 so it's not short it's like six years of like it's quite a while I mean it took a minute to build this um how to shorten this I mean I guess if this would have been in the ERC space I would be at like 10% of that complexity by now and still I wouldn't have an agreement on nothing so yeah I can you go into detail why you think you had this like screenshot with the message we said it was useful to expose the public key can you go into detail why it's useful No it should not be use you should he should not expose the public key so the idea is like rather than giving a d app a public key which the user can lose right where you really don't have any ability to attach information or do anything kind of complex give them a smart contract that hopefully can have multiple public keys or multiple private Keys controlling it and so on so on and and uh and he had this idea in in March 2015 literally I mean when we were in Amsterdam that's what he said and Gavin had the exact same ideas you know and um they already kind of could saw saw see it from the beginning that that's the way you should should how should you know a d app should interact with people not through a single key but through a more complex system but I also knew at the time that it's just we have no standards and we would like not launch the Mis browser in any due time if you will come up now inventing that thing so it had to be just the key initially and the multis yeah you were talking about like how it took very long to propose standards isn't that a tradeoff of decentralization or why you think it needs to be like I guess it could be expected it takes longer now than it used to take so what would it still be your main criticism is it main mainly criticism about not being able to agree or it to Long can you maybe so the thing is I mean consensus in agreement is important but if you really want to create a foundation that's well thought out that is not just like you know something simple like a token but something like an account right an account should be manageable over time ideally it should be not changing that much because otherwise DFS have a hard time dealing with it um it should be you know reactable should react to whatever something it should hold information so smart coners can read information on it so it needs to have all of it needs to be extensible you have all of these properties and if you just really talk I mean we have some ideas you mod account system and ethereum and so on but it's a very early in in in its design and really not very that much uh standardized and that already three years or four years since they're thinking about it right so sometimes it's important to just start from scratch think it through yourself come come up with an idea then test it out implement it what took so long is really that all the different pieces needed to be standardized we didn't want to make one custom account and say this is it you know we wanted to make a system that is completely flexible Works anywhere for any purpose and and is is yeah it's just unbiased and that takes time and then you have to build a lot of things on top of this to make it kind of visible I mean Kos safe knows that you know you need to build a Kos safe you need to build an app and you know an extension and all of this stuff to really make it easy usable right we had to build the same stuff and transaction relay service that you know you don't even pay gas and la la la la so you build all of this stuff on top of this interfaces that it looks nice and only then people can really see oh okay that makes sense you know oh that's how they account can look like that's how can work it takes time to and imagine now you have a bunch of debates along the way that would be I mean we had a bunch of debates internally but at least there's one little dictator who could say this is how we do it so sometimes you need one cou and not like 10 you know and thank you very much for the presentation as well I guess my question kind of spems off the the past as well um not to make this entire evening a retrospective um but I'm curious what what is your takeway what's like the if you have one more slide in the presentation what does it say um in the context of you know I took what a lot of what you said to be about technical de and decisions that were made in the past and you know things that impact that down the line but also just mentioned you know maybe it's better to have one shaft a bunch of two shafts in the kitchen um where does that fall in the line of you know eum today get a lot of critiques about decentralization and the way eips are developed written and implemented um where is this tradeoffs where is the Spectrum how do we actually come together and think I mean I guess the The Meta point is that we have technical debt in all of our democracies in world in the EU you have massive Tech Deb to even establish an organization I don't even know how Department of centralization exists in Germany as an organization ocracy um so what's the the name is a so what's the trade-off between going outside as you did versus working inside and how does everyone participate in that system I mean so sometimes it's necessary to step outside and to do it different you know and to make it I mean that is the only way sometimes things can move forward it will otherwise take another x 10 years and uh I mean there a lot of money and a lot of people and a lot of teams in the eum ecosystem and we count ourselves totally into that as well but if you always try to really just like build up on the first idea which is in this case usc2 which everything else is literally like somehow bound to then you are restricted and restrained in what you can do sometimes it's important to step out do do it different or do it right rethink the foundation and then come back in and then ethereum and other chains can learn from for example Luxo where you see okay it works there oh wow this is like can do this and this and this and the tokens and all of these new features that are possible now maybe we should also start using this account system or should use uh these new tokens for new projects and that's how you can then inform the old but transforming the old I mean I personally don't believe that's possible in that sense I mean not in any kind of reasonable time frame so it's sometimes extremely important to just step out and Luxo is is literally ethereum I mean the chain is is ethereum it has a different set of validators um started purely as a POS and there has no proof of work phas but apart from that it runs every theme client runs Luxo as well um which I'm very thankful for by the way not going to lie but um because that just gives takes away the effort of making the actually chain um but um it's we can then innovate on what what I know you know which is the standards and the and the front ends and being thead DF developer I mean okay I was the only one I was the DF developer you know um that's what I knew how to build interfaces and how to build uh developer tools and that's literally what you know it's combined in these Universal profiles um okay one last question here and then hi um I didn't quite understand Luxo very well like you said you're hoping ethereum could benefit from what you did with loo but it's not an L2 it's l L1 like how exactly could ethereum you know to make use of lxo what's your you know what's your plan for the future yeah maybe uh I mean this is whole belief that l2s are somehow closer to ethereum than any other evm chain I mean not going to lie but that's complete I mean an L2 is nothing else than a separate Network separate user separate Network it just happened to write its state into ethereum but if you build a bridge that bridge is probably not very secure either right I mean we have a benefit here um that there is I mean gnosis has the same benefit is it's the same consensus algorithm like a theorem so you can even make a bridge that can two-way verify Zer knowledge proofs so it's even a better Bridge than any L2 Bridge can be um how it benefits through the standards you know I mean eventually I mean these standards run on any evm chain right this is a complete unbi totally open for everybody to use standard our brow extension will open up to ethereum uh already kind of open up to base uh will open up to other chains um you don't get all the belts and whistles of having all the other cool standards that all the projects will will use so you miss a little bit out on some cool automations that you would get but you will have a proper account that you can manage with permissions and recovery and all this kind of stuff as well so it's more about you know over time the thing can grow and because all of this is extensible people can add now standards on top of a New Foundation like they're adding on top of the foundation called esc20 right that's literally what everybody's adding on in some way um now we can add on a different Foundation but you know doing this on another chain you can learn first you can see watch does it make sense does it work you know what can do people build what can be done and then they realize oh oh I couldn't build that here maybe I do want to use that so it's Open Standards on the end for everyone awesome thank you so much a big round of applause for you more questions