Making Of fylepad

A road from 0 to 100+ stars.

On September 13, 2024, I randomly decided to start working on a project called Journal. The idea? A web version of Ozan. Pretty simple, right? Well, there was just one tiny problem — I had no clue how to build a desktop app using web technologies like Vue and Tailwind CSS. I mean, sure, I'd heard of Electron, but I wasn't sure if it was the right way to go.

Anyway, I started experimenting. At first, I thought, Okay, let’s just make another Markdown-based note-taking app like Ozan. But then I stumbled across Tiptap, a rich text editor similar to Notion, and thought, Oh, this is nice. I gave it a try, and it worked so well. Then I started wondering — why are all those apps using Electron when it's so bulky? I kept working on my little browser-based app until I came across an article about Tauri — an Electron killer? that's based on Rust. It generates ridiculously small binaries, which was exactly what I needed to make my app feel different.

I gave Tauri a shot. And failed. Then I tried again. And failed again. At that point, I was ready to rage-quit and just assume I needed to learn Rust. But for some reason, I tried one last time — and BOOM. It worked. I had a full-fledged desktop app without any unnecessary bloat. And the best part? It was only 6MB. Compare that to an Electron-based app, and it's like comparing a sports car to a minivan filled with bricks.

Now, if I wanted to build a true Ozan clone for the web, I'd need a database. But since Tauri doesn't make embedding SQLite all that easy (my skill issues), I took a different route. Instead of going full database mode, I turned the app into a notepad alternative — just create a file or open existing one, edit, and save. It started as just a Markdown viewer and editor, but then I thought, What if the app remembered everything even after closing? So, I added state persistence using Tauri's native APIs. Now, you could write notes, close the app, and open it later — everything was still there.

By Day 2, things were getting exciting. I added tabs, so you could work on multiple files at once, and the app would save all your tabs automatically. Not bad for just two days of work, huh?

On Day 3, I focused on smaller details — support for different fonts, tables, LaTeX, and a PDF export feature. The PDF export is honestly hilarious because I was too lazy to implement it properly. So, instead of writing a real export function, I just made the app open the print dialog, letting users save the page as a PDF manually. Work smarter, not harder, right? I also added focus mode, dark theme, and other minor tweaks.

At this point, I was pretty hyped to show it off. So, I launched it on Product Hunt anddd wellll got 7 upvotes haha. Even Ozan got 13, so I was expecting more, but hey — baby steps. The response wasn't mind-blowing at first, but slowly, the project started gaining traction. And today? It has 102 stars on GitHub. Not too shabby!

I also have to give a huge shoutout to Gregory Benner — this guy took the project to the next level. He made it truly cross-platform, wrote GitHub workflow for multi platform distribution, and even tested it on his Mac. After his contribution, the project exploded from 40 to 100 stars in just a month. That's when I decided, Alright, let's make this look legit. So, I completely redesigned the app, heavily inspired by Peerlist.

And now? fylepad is popping up on different sites as a productivity tool. Whenever I post about it on social media, people seem to like it. I guess I should keep working on it.