The Colanode Way
How Colanode is built and why it is important.
Colanode is a set of collaboration tools under one workspace. You can chat with others, write notes, build rich text documents, create custom databases, store files and more. It is open-source and you can self-host it.
Why
In the early days, work lived on machines we owned. Documents sat on our disks. If something broke, we fixed it or restored from a backup. Ownership was simple: your computer, your files, your rules.
Then the cloud made everything easier - and more distant. Software moved into browsers, updates came automatically, teams could collaborate from anywhere. But the price of convenience was control: data left our devices, lived on someone else’s servers, and became subject to someone else’s terms.
Over time, teams adopted a tool for every job - chat here, docs there, tasks somewhere else, files in yet another place. Information scattered across accounts and vendors. Search became shallow or siloed. Context got lost between apps, and the same work had to be recreated or glued together with brittle integrations.
When the network drops, work often stops. Many modern tools assume a perfect connection; a loading spinner shouldn’t be a workflow. Offline is treated as an edge case, not a first-class need.
Security and governance get harder too. Data lives with multiple vendors, each with its own policies, outages, and API changes. You inherit their risks and roadmaps. Long-term continuity - five or ten years out - becomes a hope, not a guarantee.
This is the pressure we’re pushing against: lock-in, weak data custody, offline fragility, SaaS sprawl, rising and unpredictable costs, and integrations that break when you need them most. We need collaboration that feels as smooth as the cloud, but keeps the ownership, speed, and reliability of local software. That’s the gap Colanode is built to close.
The Colanode Way
Here are some principles that guide how we build Colanode.
Open-source
Everything we're building is open-source and you can use it for free. It comes with one of the most permissive licenses, so you can modify it, extend it and build on top of it. There is no enterprise edition or features that are not available in the open-source version. Even the Colanode Cloud hosted offering is the same codebase as the open-source version. We want everyone to use Colanode and we want to make it as easy as possible to do so.
User Experience
We're trying to build a minimalistic and clean user experience that feels easy and intuitive for new users, yet powerful and flexible for advanced users. It should be effortless for users to find the information they need, or perform a task. Whether you are trying to use only one feature, or all of them, the experience should be seamless.
Developer Experience
We want to make it as easy as possible to self-host and operate Colanode in any environment. This may not sound obvious, but it is actually very important. If you've ever tried to self-host a tool, you know how complex it can get. Many dependencies, complex documentation and frequently vendor lock-in.
The Colanode way is to keep dependencies to a minimum and use established and open-source components. Right now, all you need to host Colanode is a database (Postgres), a fast key-value and message queue (Redis or Valkey) and an S3-compatible object storage (this won't be a requirement in the future, you can use plain file storage or other storage providers). You can run these components on your own servers, or any existing cloud service (most of them provide them as managed services).
We try to keep the same stack for all new features that we add. We take this philosophy so seriously that sometimes we might build something that already exists, just to not add a new dependency or vendor lock-in.
Local-first
In Colanode, all the data you have access to is stored locally on your device. When you make any changes, they are applied locally first and then synced to the server in the background. This means you can work offline, when you have no internet connection or the server is down.
Another benefit of local-first is that it makes the app work really fast. In typical apps, when you open a chat, hundreds of messages need to be fetched by the server, including reactions, users and other metadata. When you open a large document, the app needs to download all of it. This is the reason most apps feel sluggish. In Colanode, all interactions are local and don't require any network request. Everything loads instantly, you rarely see a loading spinner.
There is another benefit of local-first that is not often talked about. Since everything works locally, the load in your infrastructure is very low. While other apps need to optimize the infrastructure for all different pages and handle hundreds of requests for user interactions, in Colanode we optimize only for the sync engine and the rest is done locally. This significantly reduces the types of requests we need to optimize for and makes the traffic predictable. The benefit of this is that we can offer very attractive pricing for our managed Colanode Cloud offering.
Craftsmanship
The only way to keep high standards for a long time is to be obsessed over the little details. We're obsessed over making self-hosting as easy as possible and not adding too many dependencies. We're obsessed over the way we name the files, variables, functions, components, database tables, types, etc. We're obsessed over the way we build the UI and the UX. We're obsessed over the performance and the security and try to gain maximum performance over every line of code. We know that sometimes some details might not be the most important, but over time they add up and make a difference.
Colanode is not perfect and never will be. We're not trying to be the best at everything, but we're trying to be the best at the things that matter. We're not trying to be the most feature-rich, but we're trying to be the most feature-rich that is easy to self-host and operate.
Long-term thinking
We're here for the (really) long-term. We're putting a lot of effort into making an architecture and system that will last for years. Thinking from first-principles, each design decision is made a a building block for what's to come.
We would rather build something that will bring us more users and value next year, than something that would bring us more users in the next week or month, but is not compatible with our way of building and operating. We don't like hacks and we would like to avoid them as much as we can.
Let's build together
Join us in rethinking collaboration from first principles. Follow the work in the open at github.com/colanode/colanode, keep up with updates on the blog at colanode.com/blog, and learn how everything fits together in the docs at colanode.com/docs. You can try Colanode for free today at app.colanode.com.
Contributions - code, docs, feedback, testing - are welcome. Open a GitHub issue to report a bug or propose an idea, start a Discussion if you want to explore a direction, or send a pull request when you’re ready. If you’re self-hosting, tell us what worked and what didn’t; your real-world setups shape our roadmap.
Written by Hakan Shehu - September 2025