Last week, I attended the @ethglobal hackathon in New York and failed miserably. My idea was to build an app that allowed customers to order food at restaurants by collecting @lensprotocol publications. This would obviously require a mobile app, and that's when I realized how hard it is to build anything mobile, especially when you don't have a lot of time and just want to try things out and build a quick prototype. I got stuck trying to decide which technology stack to use: React Native, Flutter, or a PWA / responsive app? Tried a few things, but once I started to actually code something, I ran out of time and couldn't submit anything decent

But I came out with some lessons learned:

-- Desktop apps have been good enough for purely financial applications, but if crypto is transitioning into the consumer app space, it needs to go mobile. Also, mobile apps provide access to QR scanning, geolocation, access to contacts, camera and mic, motion sensor, push notifications, NFC / bluetooh, and more, opening the door to a whole new set of experiences

-- Stepn, @friendtech, and POAP / NFC apps and the new developments with AA and services like @privy_io have shown the potential of mobile apps, but there's just not enough people building mobile-first dApps. Why?

-- Building mobile dApps is hard. Building for mobile is generally harder than building for desktop, but when it comes to dApps, it's even worse. Debugging, wallet connection, building a decent authentication flow, and a usable UI are some of the challenges. Even more if you have to go through the process of publishing to an app store

-- Mobile is not ideal for hackathons. Desktop apps are faster to build and are easier to integrate with many protocols, allowing to apply to more bounties. Also, they say it’s difficult to find good UI designers in web3 hackathons, but it's even harder to find mobile app devs. Even with decent SDKs or RN libraries available, most web3 devs simply don't care / don't want to learn mobile dev. In fact, in the last ETHGlobal hackathon, only 10% of the 300 submitted projects had a mobile component, and very few had a decent UX