From web2 to web3 pitfalls and learnings | Nuno Loureiro, Ethereum Foundation | ETHTaipei 2023

uh welcome so we're going to be talking a little bit of web 2 and web3 from design perspective so yes we are going to talk a little bit about design it's not a topic on this conference that's why I have to adjust a little bit my my slides but I hope it's going to be a fun ride for you guys now um so every talk about design needs to have an apple reference this is mine we all know this has a classic and we all love this hello from Apple so hello my name is Nunu I come from Lisbon Portugal all the way to talk a little bit about Design This Is My City Sunny Lisbon by the ocean on the other side of of the globe English is not my first language so I might be completely nervous about speaking in English but bear with me a little bit I might do some typos and some stuff so bear with me sorry about that quick intro so I'm only designer at ethereum.org for lithium Foundation this is mostly where I work on every day is an open source Community Driven platform that aims to learn and onboard everyone to ethereum this is what we have been doing for the for the for a while now I joined a team one year one year in something and um we do have a lot of pages and we are Community Driven mostly on codes on content on design and one other thing that's really important is about translation so I would take this opportunity to highlight that some of the pages need translation and we want help from everyone if you want to join the community if you want to help a little bit just go on our Discord go on team.org there's a bunch of pages to learn how to help this is a good example that the learn about ethereum is not even translated in your own language here so be sure that you could you can help a little bit you can on board and be able to to help others within the ecosystem um again you have pages that is not translated again to help people to jump into the our Discord and help a little bit topic um we all have been seeing these three axes divided so we believe that we are transitioning from web to web through at some point where are we now and how can we learn from the web 2 to web free and mostly how not to do the same mistakes again I put up a bunch of a table a bunch of words to try to describe the evolution of the web from my perspective and I I did not wanted to to do a full statement of what is web3 comparing to the other ones but I think it's an evolution and you can see that it's once you read once you read and write and then you read write own that's mostly the most important one being in a foundation when you build a house imagine like a house a foundation it's where you build where everything started on web 2 we get walls and on web3 you could have window roof where everything starts to be cozy so you can have non-webbed one it was mostly on desktop web tube was mostly on mobile and I think web3 will mostly on Keys like in the broader sense that it's something that we believe that's on our side on the second one Ingenuity it's more like everyone everything was built out of naive knowledge and was Russell around web 2 was a lot on Venture so Venture Capital Drive web through to where we believe it is and I think web3 is based on community that's the main Drive of it so it's reserved selfish rewarding to inform to publish to sign to connect to login or to have a contract so those terms try to make a stand on how those those three phases of the web as we know it's are divided to I visually I kind of did a small example of how how did they go like it's one to one one too many or many too many I think this diagram tries to explain on a visual uh metaphor that how can we perceive what is web3 and how can we take the most advantage of what is web3 um for all of us and will help us build a better um well in adoption I this is like a rough estimate doesn't mean anything but the adoption line we are still on the beginning so we still need a lot of work to do a lot of building to do a lot of technical hurdles we'll have to overcome uh so might at some points uh everything that we are building today might be different in one year there's a lot of things new protocols new things coming up but we need to to have a sense that what we are doing today can have an impact later on and some decisions and focus on design decisions you'll see that they will have some impact and they will endure a lot longer and one example is this one do you remember what this is did you all use this probably some of you are not old enough to use this I did use this and we used to save before hard drive things on on this small device and now as you see we all use the same icon when you want to save something we use the same icon even knowing that's the same artifact is no longer in use and it's a rare thing it's a vintage thing at this point and there's a lot of folks that do argue that we should use different icons but at this point it became the pattern became a user's pattern and this was a design decision somewhere in the 80s someone say okay I use this icon to represent I'm saving something and even today 20 years later we still use that so imagine that could happen and the most thing to wallets we now perceive a wallet being an icon of physical wallets and we use that in our own interfaces right now probably in 10 years from now the word wallet might not even exist but we might use the icon to represent what is a wallet and might be completely different what what the wallet looks like today so this is a good example of how we use design decisions on today and that might endure a little bit later so you have to be careful about that this is mostly be be aware that we have to be questions about some decisions so if I had an alternative um title to to my talk would be how did we mess up and I'd say we because I'm old enough to be on web 2 and I do remember building the whole thing how we can how we messed up on web 2 and how not to mess up for web3 and that learnings be important so I will focus mostly on accounts while it's insecurity for the small part of the talk uh I think it was the right audiences the right targeting to to focus on that so I changed my slides a little bit and we always take these tests to begin with we all agree that we need to improve our experience on web3 so there's a common ground for everything and we has designers needs to step up to agree on something and to step up a little bit our game to create better experiences we all endure roughly things and I will give you an example of one thing that we that we are doing that is a really mess up do you know what this is we all use these kind of screens when you want to log in in web 2 screens and this is really bad this is what we call back in the days a Nash Car interface because we do use all the different logos to sign in this is this is bad this is this isn't fatigue for people this is silo dependence this is centralizing the logging The Experience like if I'm using one service I'll be depending on that service Twitter with the new Twitter from Elon Musk they killed the sign in with Twitter and it was a mess for the ones that had uh account login for third parties using those those kind of things so this is a bad example how we messed up in web too and what are we doing exactly the same thing we are doing the same mistakes again because we are doing silos again we are using branding we are using I'll use this or use that to be on board and we need to be careful about this uh probably will be different again in later on probably we'll have some sort of sign in with ethereum might be unified that's worked all together on this but this is a good example of how can we not do the same mistakes that we did in the past and because it's a ux pattern like okay let's put a bunch of buttons okay you want to log in which wallet do you want and it's it's bad this is decision fatigue when try to be your first time using some sort of service and encounter this kind of thing what what would be your decision like you lost half percent of your users just based on this because it was it will be there first experience on web3 so main takeaways from this part is like we need to work together and agree on standards agree on how to use and open the process like focus on the journey is not because others are doing like we did in the past because it's established uh patterns with focus on this on on on how we can do better and how can we agree on doing better not because I'm doing and I'm a big Corp on on web3 or the other are different projects smaller projects you need to to use the same patterns I think we can improve that experience a lot so let me just take a speed up this was just a small pause like we all know this this is my daily lives we all know some of those things can be tough so so the second part I think focus a little bit on security I think um we on web 2 we're always targeting how easy to use other services how can we do a one-click kind of onboarding like one click and you're in like so easy and we count the number of plugins or the number of users we've converted I think we need to rethink our metrics for web3 we need to rethink how to use this and one example that I will focus now is like when you use the username password um no way of uh logging to to a service and some of the wallets right now are falling back to username password because it's a known pattern it is not necessarily a good one so this is just a quill a quick um way of showing that we force users to use complex passwords to log in when we just could just learn how to use a seed phrase a seed phrase on instead of a complex password it's a lot safer it's a lot better and for us as humans a lot easier to memorize than the random thing with characters like give me 16 characters but just to four six eight phrases words in one seed phrase so why didn't you we help users to begin with why did we make it so complex for folks to have a password like it can be more secure now you have to remain things using password managers second author second second uh author Factor authentication so we have to use a lot of different things when you could make it a lot easier and this is a design problem this is not an engineer problem we as designers as ux designers did this wrongly to begin with we didn't understand that we could make easier interfaces for folks if they had access to the knowledge of okay if I memorize five words and put it all together it's a lot easier to onboard to um to memorize the password and to be able to to log into a service so why don't you use this so this is a bad ux and some of the wallets right now are working on defaulting from seed phrases on and and try to put on top of its smart contracts that are blinded by um by username and password and I'm arguing that not a good thing so after all seed phrases can be good and I might say that even private private Keys public keys are a good thing they are secure they are a good way why don't we educate what I don't guide users to understand what those are in a sense that the cute learn from it improve the experience and be more aware of the risks of using those so we need to to guide users on how to use it best practice probably not use the jargon and not use complex words that most of the the normal users don't know and need to to Google something to understand what they are thinking about so I will give you a small example of a metaphor that might be a little bit too much but I will try to to simplify as much as we can but so we'll learn learn how to how to drive we need to have a license how to drive right we need to go have driving schools you have to learn a common ground not to not to like put others in Risk not with myself in Risk so we need to learn signs from streets they are Universal right everyone needs to understand this to be able to drive so we are guiding towards and and a technology that has its risks that is dangerous for people to you to work on so you need to create a momentum of how to teach them to use this kind of tools to use defy to use dabs to use wallets to use all that thing they need to understand the risks and probably it's a communication problem it's a design problem above all um and he's learning this is an example to extreme but I'm not saying that we need to take a license to use web3 of course but we need to be sure that folks knows the risks if uh I do something wrong I might lose my funds I might lose my nft I might lose whatever I have value in there the missed Transaction what happens if it runs like this goes through what can I retry it can I not retry it so it's always try an error and I think you can simplify a lot to that throughout the process and above all that's a design problem um so my main takeaways in here is that the learning curve can be the likely experience in the sense that we can design a Learning Journey for the users they don't need to be on board in one click they're not not they do not need to use your dab just by okay converts happening I need to transfer funds from one one chain to another chain or from one dollar to another wallet this is like let's do it let's take it easy take one step back understand what's the struggle from the user and make the interfaces work better for the users and not better for the business itself and that's the difference between web 2 and web3 and about responsible like each user needs to be responsible for for what they use like we all have used keys for our house as we don't give out the keys from our house to a stranger like you don't give up a lot of these things we need to be be aware that there's risks using this technology and once you're aware of that I think we all can work two towards a better future um uh they don't need second they don't need to understand all the details like you can abstract some of the chain complexity but they do need to understand uh some parts of it again for the car metaphor you don't need to work to know how an engine works in detail but you need to work to know how a break Works how steering wheel works how do you put air on a tire those things all of us in society needs to understand those in order to to go about and to drive or to be able to to understand how things work for web 3 I think we need to start doing the same you need to understand a little bit more of what's involved what's the risks what's in it you know it cannot be just simple social media post something and like something there's a lot more value involved a lot more risks involved that we all need to be aware of that that's mostly my message for for you guys today um we do have new technology coming out we do have a new blank covers coming in which it's a kind of abstraction this is like a rough notepad that I took from a cousin brass abstraction on security and onboarding mostly we can redefine the user Journey using account obstruction when all the fixtures will be available and all the all of us on dabs all of us on wallets can start using them so this is again re rethinking the process using new technology that will come out really really soon and we all might enjoy better experiences this is one example of how can we improve the journey and just to finish my talk yesterday vitalik was conclusions was this slide and just to be sure that I was targeting the right thing this was what I am trying to pitch more it's the last one we need to adapt the better ux for our product effects so let's learn from the past improve the the future thank you so much for listening to me uh I hope you enjoyed it's a little bit uh different topic and thank you so much [Applause] regarding seed phrases have you seen a sea phrase system that works for an international audience that mainly uses a non-latin alphabet I did not and that that's the real problem to onboarding to be more accessible for sure but suit phrases can be random random phrases you know that you just need to memorize something they don't need to mean anything if you're aware of flooding characters at least a little bit you can memorize that even knowing that they don't have a meaning uh it's not the same thing I know it's not the same thing but you're true should be more accessible for sure hi um I'm still like highly new in the industry and I kind of feel the most difficult thing for me to understand was the wallet and I realized that now like I get to know a little bit more like wallet the term wallet is actually for me it's very um limited understanding towards what a wallet really is so I think for me it's more like an avatar like what do you think Wally really is or like how can we redesign wallet for um a newbie to understand faster while it's really involved to be accounts above all and the counter abstraction will be that so uh wallets the term wallets was not a good term to begin with that's the beginning of the talk when I'm talking about the discount thing it's just because of that we use wallets because we use tokens are coins and coins are inside wallets but we've never hold coins in our wallets like we do the money itself it's always on chain so the the wallet by itself it's not a good metaphor to begin with so it was bad to begin with a lot the same way it constrains your perception of what is a wallet it still Mis uh uses a lot of our perception because we are used to our wallet holds money in it and it's on my pocket it's not true a wallet is not that but we account obstruction that will evolve to be a an account you will have some sort of accounts either you could call it someone will argue that will be a smart account smart account because you can do a lot of different things but the the evolution towards sign in with ethereum might be one way to solve this problem because if you start to sign in with ethereum or sign in and with your wallet and it will start to be an account mostly and not a wallet that's what I think will happen in the future and might help you might help other users as well an account it is then okay thank you any other questions you've mentioned that there's like a need for a better standardization amongst like lists of crazy numbers of wallets and all those kinds of things how do we get there because it seems like the path that we've like their calls to actions from people like you but how do we go from this world now where it just seems like the the default is n plus one standards as opposed to just having like everything unify like do you see a better path than just hoping for it or calling for it yeah there's no there's no right answer to that there's not one way but I think we are at the point to start dividing on that and start to do workshops one thing that I was uh talking with someone yesterday that probably start we need to start have designer hackathons to solve those problems you know just technical hackathons we can solve a lot of problems and new ideas for new business on hackathons like the one that happened before these and all the academics that you've been into it but there's not that many designers on hackathons and probably you don't solve these kind of things and we need to agree on standards we all base the way was built upon standards we all agree that HTML is one standards like everything is based on standards and we do use those tenants to evolve the roadmap eips are standards like we do a spec they evolve we need to do the same for the wallets or whatever that is and we will agree on one way of doing things and we all should use that one thing that I think that might be the way with sign in with ethereum but I'm not sure hey uh thank you for the sharing so I was wondering you as a designer what is your view about the ability for users to make mistake in onboarding right like because in blockchain there is this concept about reversibility mutability you make a mistake it's it's gone for right but in web 2 sometimes you can ask for a refund things can be reverted and feel a certain sense of security but in web 3 when a new user approaches it always feels it's a atomic mistake right they're really going to lose everything and that thing really scares a lot of people so just wanted to hear your thoughts about that topic yeah that's that's a big evolution with the kind of abstractions I said when you get smart wallet smart contract wallets they will be able to have different mechanisms to ensure that kind of trust right now imagine that we are in a in a phase like the beginning of stuff it's still really raw like if I'm mistake giving you some funds the funds will be locked some someplace else and I'll be doing a lot of these like I think technology will help but my main pitch here is that technology by itself it's not enough we still need to use really good user research design research whatever you want to call it to to provide that trust within the interface itself so because when you open a wallet like imagine metamask and you press sends and you sign something like everyone gets butterflies on on the Starbucks right it's like everyone needs like okay this is going through like me go to readers can just confirm it uh we all do that and that's an oral full uh experience so the experience by itself is not a technology prop it's evolved with technology is solving the tech Parts but we need to solve that on the wax and design part as well and all the dabs all the ideas that we are all working on D5 mostly needs to work on that as well you need to be trustful the interface needs to be need to have some some interface by itself like if it is something full of emojis and unicorns jumping and all that stuff does that gives you trust to use that services like imagine someone like that or it's uh used to use traditional finance and it goes to a really boring website to do a bank transfer then imagine that on on web3 it's just unicorns and emojis are over the place is that a good thing like we need to ask ourselves these questions and be critical about it thank you any other questions okay once again thanks Nino for giving us such an inspiring talk foreign