David Dal Busco - After All, Why Build on Web3?

I'm uh David. Um yeah, I'm a look at me. I'm a developer and um I'm not really like a crypton native like I'm really like I like the tech. I like what's the idea and I use the crypto but it's not my background. And also since it's it's clu like it's a wonderful conference but I have to say I'm not that familiar with it at home.

So yeah, I come from uh the internet computer from another perspective, but still I wanted to share what I think about what I'm still building and yes. All right. So when we had contact about me doing a talk here, I was a bit thinking about what I can speak about, you know. And this year when I open Twitter, it's all about AI, you know. It's like just like 90% of my tweets are about AI agent and so on.

And even here at the conference, you hear a lot about MCP server and so on. And you know at at the end of the day I asked myself okay it's AI everywhere and is it still relevant to code in 2010 like building maybe but coding is it something I should still do and honestly there are data everywhere everywhere but who really cares about the data in the end of the day it's like I don't know like I always like to think that I'm not sure my mom really care where are things stored So yeah, does it make sense that I continue building on web 3? And when we go to the beginning for me, as I told you, I'm not like a cryptonative, but what web three promised to me was the ownership. That was really something that interested me, like being in control of my project, being in control of my data, of everything. like I'm I'm I'm I'm the one that rule my own little kingdom.

And also I really like that it's an open infrastructure that you can just query and interact with everything. Everything is public. No closed door, no private things. and also like it the fact that it's not a lock thing while I'm not really sure that it's the case on web three like it's yeah we probably derived a bit and also of course I don't really have to explain why I think it's really important to have no gatekeepers and I don't want anyone to disallow me to do anything so all right No. Yes.

So when I look at the status that's actually the web I got and I mean those product are great. There is no question asked. I mean beside that I really don't like the UIX of uh AWS like I mean they do a great job and so on but it's not really what I was promised and it's still like I don't know how many% of the web is based on this infrastructure even on the blockchain we know so yeah I mean that's what we got and So what am I compromising? I mean like I I I I I talk about ownership but at the end of the day I build on this cloud. I mean it's not my cloud or I talk about permanence but I mean yes I can trust but to some extension the service can disappear you know like you never know what happen.

is like not super sure and I say trustless but I have no control over that infra or I have a control to some extent only and I talk about decentralization but yeah I mean yeah it's not really that decentralized or not at all so yeah it's like a bit uh I should take a course to learn how to to make that clicker work oh yeah So in addition to that I also spoke about AI and I mean yeah it's nice I use copy lot it me I can use v 0ero to spin an application like in five minutes I can let cursor v my my app we just spoke about it and uh yeah it's really great but what who owns the app that that is spin at the end who who runs what I've built where does it cleave? And um so should my my question like you know when I think about this thing is like should I just give up maybe it's fine like that and let's continue you know or uh maybe it's too complex to do differently and I have all those those question honestly and um but still I mean of course I I hope Lucy not give up and I still I still want that that that fact. I really like that concept of ownership on on on on my project on my data and I still want unstoppable apps. And of course, I also want a DX that doesn't suck because it's not because it's web three that I should have like a clerstone DX. And on the other side, I don't want like I'm I'm really super lame at DevOps.

So, I don't want to host my own server because I don't know how to do that. and I don't want security all or a fragile tool chain right so that's why I try to kind of resolve that on my small level and I built something called Juno and uh it's a platform and a set of tools that helps developer build entirely on the blockchain on the internet computer right and when I mean entirely it's back end and front end so for example here you the the UI of the console that developer use to to configure their project that URL itself comes from the chain. So yeah, so it works. It runs runs regularly on the IC and you have various services like a little let's say key store database, not really a database, more a key store. You have a file storage which can go up to 500 giga.

So I mean it's like you can put 500 giga in a smart contract. So it's like kind of nice. There is an authentication. You can write some serverless function if you want to I don't know like do some crosschain feature and this kind of things. And yeah that's that's that's that's all my point.

It's like every developer that use the tools is the sole controller of the things. So me David that provides the platform cannot do anything. So yeah as I said it's like Juno is the platform I build but the superpowers come from the internet computer like I mean I don't even wanted the chain you know so basically I provide smart contracts so that developers can spin their own smart contracts and all the smart contracts can interact together basically and yeah it works like that. So a developer going to jumps in Juno and when they create like let's say an account it actually create what I call the mission control which is basically a smart contract wallet and from there the developer can spin their project and what is kind of really important like I said is this part here has no control at all on the things. So only the developer or the developer mission control can stop, start, delete, upgrade, deploy and so on.

Me as a provider I cannot do that. And um the project let's see we have all used the ETH it is cluge app right this last day. So let's say I want to host that app not on the I don't know where it's hosted but I want to host that on on the blockchain. So what I would do is I would spin a satellite and in that satellite I would just deploy the app ETH cluge and then everybody at the conference would access that app here because it's available on the web and every everything comes from that satellites which is actually like a it's a wasam container and basically you put your app within that was container that run indefinitely on the on the chain. So I going to show you really quick how is it like when you spin uh something on Juno but at midday like just after this talk I going to give a one hour workshop because I'm way more uh it's I rather like to code than to speak and uh I will show you the steps like a bit all the features but just to give you a sneak peek.

So basically to install Juno it's like any front end framework you like either you have an app and you just init Juno in your app or you use a template you say blah blah blah I want next blah blah blah I want serverless function go and it's done right because it's all based on JavaScript and then once you have the your environment locally you can run an emulator because on the internet computer there is no test net but you and basically run a let's go a test net on your machine. So you kind of run all that local environment with everything and even the the before we saw the console you know even the console of Juno you can run it locally. So you really have like an environment locally that mimic production. So you can do everything just before going to mainet which is kind of nice. And then as I told you there are various features like a notification kind of small database file storage analytics and so on.

And in your front end app I give you a JavaScript library and you basically just use function like for example this uh set document is like you save a payload that contains hello and hello from Juno on the chain in a what I call a collection messages. And the thing is like it's mutable. So you can update you can really interact like you would interact with any um NoSQL database basically because it's the way I programmed it right and um on the back end side so I give you basically you don't start with a blank smart contract. You smart you start with a smart contract that has already features. That way you don't have to implement the wheel every time.

But if you want to go a step further, then you can basically supercharge the features and implement your own stuff. So if for example we take again the example of the ETH clues app, let's say we want the ETH clues apps to allow people to send notification between them, right? So what we can do is the the internet computer has a features that allows to send message from the blockchain to the web two world. So basically you can extract that and write a function that say okay when someone saves something in that collection that in send a notification to web two and then the web two going to ping the the phone of the other person right this kind of thing it's a bit abstract that's why there is a workshop and um for the deployment I don't have web two infrastructure so there is two ways to deploy on chain so either you are super strict about what you do and you really want full control and you just work with a CLI on your machine where you can deploy or the new things which I personally like to do is I like to set up GitHub action so that GitHub builds and deploy but you can decide if you trust GitHub or not or if you trust any runner. So you can basically there is three level of reload permission and you can go and say okay I trust GitHub I'm agree that GitHub can deploy my front end and upgrade my back end or maybe you can say I don't trust GitHub and I just want GitHub to build and propose me an upgrade or propose me to deploy and then through the CL or the console I can go and approve those chain and you those it's like it's like having a mini governance to deploy things which is kind of nice because then again you don't compromise on who control things or at least you can choose and uh yeah like I said it's you can host your front end in a smart contract and that's where we kind of compromise a bit uh with all the premise of not using web two is you still need like a URL you know like you need a domain name like juno build or whatever And it's it's where we still have to configure a DNS but maybe in the future you never know and there are also little other features like for example I don't use any analytics like plausible and so on don't speak about Google analytics there is like a smart contract on Juno that takes care of doing analytics so basically if you go to juno build then it sends the events to a smart contract and then those data aggregated from there.

I don't I don't know if it's useful, but it's kind of nice to have to be fully on web three because because why not? I mean, and uh yeah, there's still a bit of touch of crypto even though I'm not one native. It's like as I told you, there is wallet. It's also interesting to know that everything is powered by cycles on the internet computer. It's a reverse gas fee.

It means that there is no chance that you get like a $10,000 bill at the end of the month. It's basically you pay it in advance and then it's consumed. So if you pay I don't know like $10 when it's consuming then it's finished, right? So then you can monitor and so on but there is no surprise which is really nice. And also obviously like there is no way comparison with gas fee and so on.

It's way way way cheaper. I mean it's really totally different. And of course you got access key and you got cost chain feature like I'd say like for example you can wrap and wrap it on on the IC and so on and uh yeah what is this for? So I think that's like the whole talk it's based my point of view but I think it's good for dev that I read of backend glue because I don't like I say I don't like too much deop so I make it light. I think it kind of not bad for india because maybe I'm one and uh I think it's really useful for people that are a keen of certain level of security and like me that who wants to to control their um their project.

So why is it interesting for you? Maybe because at the end we are disclos why why why would you try that stuff you know so like I said maybe you have a project and you are running some whatever tools and you still have a front end that is static style render and use that on I don't know verel why not move it to juno I mean you can just move your front end if you want maybe you want to try to build a full stack app on chain because it's kind of cool in the workshop I going to showcase some some some example of what you can build. Uh those famous crunch chain feature and maybe the talk was not too bad and it kind of sounds a bit cool hopefully. And uh yeah, so to come back to my opening is like I feel like at least on Twitter web three and blockchain isn't that cool and hype anymore and but it doesn't mean that it's not worth it or it at least to me it means something and I think it makes sense to continue to build it because if we don't do it then I'm not sure who will do it. So yeah, thanks for your time.

[Applause] Thank you so much es especially reminding us what we actually building towards right we're building every time we had this discussion just before we started. It was so fun. We talked about how AI is coming in but we still need the human element to make sure everything rolling that I saw you just brought in very fast. So the way you build your your product is it onchain with do you still use the gas fees and everything when you do the transactions?

No that's the thing on the IC there is no gas fee. So, so it's like kind of every operation has a definite price and uh like even like if you do a like an ICP transaction I don't know if it's like I don't know what's the price exactly of the course but it's fixed and it's like I don't know 0.00 whatever cent

so it's pennies on the dollar basically.

Yeah. Yeah. That's good because many times when you have something built with blockchain that goes in, you have so much gas fees going through all.

No, not at all.

So that's why I want to make sure because that's the love about it. We love web three, right? We want to build on it.

So I can give you an example. So I I used to have my own website uh hosted on Firebase and I never moved it to Juno because my idea was always like my personal goal was to the day Juno is decentralized because it's not yet decentralized I will move my own website there but this year Firebase changes price change their price and I started to have like a bill monthly where I had to pay like 20 or $30 for a website you know I was like why should I do that and then at some point I was just like it and I moved my uh my website to uh to uh to Juno or to the internet computer and I think I don't know like I will maybe pay $10 or $20 for a year but maybe maybe less you know so I think it's really like even like to host a website is really uh at at least at the moment really interesting. Yeah, because all web two website providers are figuring out we are gonna die soon but we need the money before we go.

Yeah. Yeah. Probably. Yeah. Yeah.

Do we have any questions from the audience coming in today?

I think they had a too much party yesterday so I don't think they have much to say.

No worries. But yeah, thanks for joining uh even if 11 a.m. seems to be early inclusion detox right now.

Yeah, exactly.

But it was really love having them and thank you again David. A big hand everyone for David and thank