We all know what they say: when all you have is a hammer, everything starts looking like a nail. Well, this month, I finally took the plunge and tried out Pocketbase for the first time, something I’ve been wanting to do for a while.
And let me tell you, I’m starting to see a lot of nails that look just perfect for my Pocketbase-shaped hammer.
A Backend in a Box
Pocketbase brands itself as an “open source backend in one file”—it’s basically a single executable compiled from code written in Go which will set up a full backend when run. This backend includes various auth methods complete with email templates for validation, password resets and more, a SQLite database with support for migrations and binary file uploads, static file hosting, and a full management dashboard for schema creation, logging, rate limiting, CRON-like jobs…
This setup makes it stupidly simple to self-host. Drop the binary on a server, run it, and you have a backend which you can call via HTTP or the official SDK. It even sends out real-time events should you need them. That was always something that spoke to me strongly since I first learned about the project years ago—but it gets better still!
A Framework (still in a Box)
My go-to framework to build server-side apps and APIs has been FeathersJS for a long time, but recently it seems to be having a bit of an identity crisis. I must admit that I wasn’t impressed by the latest release and the next version seems a bit too much like an over-correction. So in a way, I definitely was on the lookout for a new default.
Pocketbase can fill this niche, because on top of everything it delivers out of the box, it’s also extensible both via Go and via JavaScript—which is a blessing for me, since I don’t have the time to learn Go at the moment.
When using Pocketbase as a framework, there’s a full set of hooks to manipulate pretty much anything that happens during an API call as well as from the management dashboard. There’s also a router to register custom routes, ready to handle a lot of what I can think to throw at it. Bindings for sending emails, using the official logger, manipulating the database…
Powerful stuff.
A Hammer with Something to Prove
Unsurprisingly, I am now building a first project with Pocketbase. I’ve been enjoying working with it so far, and I’ll hopefully be able to share more about the project soon.
There are some things I still have to figure out, like how I could best share a Pocketbase project in such a way that others could easily self-host it, but I’m confident there are ways to do just that.
And besides, as enamoured as I currently am with it, I still believe in using the right tool for a given job and like everything, Pocketbase is not a sliver bullet. It hasn’t reached version 1.0 yet and isn’t considered ready for production by the project’s author.
Nonetheless, I love what this project represents: that not everything has to be a product meant to scale to an infinite amount of users. That even individual people can build great applications which are useful for a lot of people, while at the same time being very easy to self-host and decentralise.
Your Data Belongs to You
I like to think of a world where every user can host their data on a small server set up by someone they know personally and trust, or by themselves. It’s not actually a novel idea, I just feel that it’s kind of been overshadowed by the current hype-wave. I remember multiple start-ups pursuing this idea just a couple of years ago, but they’ve seemingly all shifted focus or gone quiet.
Projects like Pocketbase make me want to push further into this decentralised world. Just give someone an executable and a database schema, make it trivial for them to set up such a backend and give their friends and relatives access to it.
I want to create great apps, not manage my user’s data. It belongs to them, anyway. But privacy-respecting doesn’t have to mean inconvenient.
Have you built and launched something with Pocketbase before? Do you have any insights and experiences to share? Feel free to let me know on Mastodon.