Their Apps are Dead, Long Live Your Apps

If you'd told me even a decade ago that by 2025 you could ask anything with an internet connection to make you software in scarcely more exhaustive detail than some people dictate their elaborate coffee orders, I would — of course — have believed you. I wouldn't have thought it would then actually work, though…

Maybe you're one of the precious visionaries who saw this coming clearly years ahead of the rest of us (in which case I congratulate you on the increased value of all your Anguilla domains), but by and large I think that the last few years have been a wake up call for how quickly this technology is changing the world.

We've had our wake up call now, though, so the question isn't what's happening, it's what comes next? In a world where apps can be generated on demand, what changes?

---

The Backwards World of Apps

By and large, we still think about apps as the top-level entity that we work in. We take notes in one application, we do data analysis in another. We have video calls about a project in one application, and then we do the project management in another. Everyone knows this sucks, so we spend huge amounts of effort building bridges between them with importing and exporting, or integrations piping data from one to another, and the sea of "do everything in our all-in-one environment!" offerings is deep.

This is doing everything backwards. Our data doesn't serve the apps, the apps serve our data.

Moreover, it's not just any work, it's your work. You should own it, not the apps. Why does it live inside them?

In the past, making the apps was the hard part, so we ate that pain — but that's just not true any more. Today, if you want a tool to do most things, you can have it made inside an afternoon, in just the way that works for you. Yes, there is still a way to go in between today and fully replacing enterprise-grade software made for thousands of collaborators in highly restrictive environments, but the writing is on the wall.

---

So What's The Holdup?

Odds are if you're reading this near the time of writing, I'm preaching to the choir. You're probably already making stuff with Claude Code, or Lovable, or maybe you're even turning your nose up at those tools because you've built your own multi-threaded agent orchestrator. That's fine. This next part is still relevant to you, too.

So why isn't everyone already using a suite of their own custom-made tools and toys? We think it's because while making a vibe-coded tool work is easy, connecting the whole toolbox is hard.

How many projects have you abandoned? Every maker likely has a graveyard of old tools and toys that were custom built to solve something that week — often in a way that we're quite proud of. If these projects all lived above, and worked on, the same shared foundation then there would never be any lost work. Lost apps, sure, but that really doesn't matter any more.

---

A New Foundation

We see a future where you keep all your work in Spaces that can go wherever you go. Use your data in apps as you see fit, whenever and however you like. That's why we've built what we've built.

Rool is a new way to flexibly store and use data, designed specifically for the world in which apps come and go as you need them. The information isn't stored in a classical database, but as objects in a Space curated by a specialised AI agent, that can flexibly permute things so that it works for whatever task you need. There is no more "incompatible" data. If you want something to work, it will do.

All you need to do is ask whatever agent you're vibing with to integrate the Rool SDK and it'll work straight out of the box. Everything you create will work together as a single connected toolbox.

---

Built for Peace of Mind

We've (of course) made it rock solid when it comes to undo-ability, so you can't break or lose anything that can't be undone. It's Google login for ease of use, and while we're in the early stages of beta testing there's a lot of free tokens you can burn on whatever you fancy creating.