Design decisions#

Why Whoobe is built the way it is. Each page states the decision, the reasons, what it costs you, and when it would be revisited. They are short on purpose; the long design history stays in docs/plans, and the pages say which plan to read for it.

Decision	In one line
One workspace per organisation	The tenant and the security boundary is a whole installation, not a row in a shared one
A simple stack	Node, PostgreSQL and S3: no queue, no cache service, no search engine to operate
Credentials stay on the server	Editors, browsers, sites and apps use integrations without ever holding the third party's secret
Client access without user accounts	Applications use Whoobe with a server key or your identity provider's token; Whoobe creates no end users
The data model is data	Content types are defined at runtime and stored as data, Pimcore-style, without a deployment
Draft is never public	One shared rule decides what the public surface may return
One renderer, many outputs	The editor preview, the published site and the app come from the same content tree and renderer
No automation engine of our own	Forms and the gateway plug into Activepieces, n8n or any webhook
Self-hosted and MIT	You run it, you own the data, no usage-based fee

How to read them#

Each page has the same parts: Decision, Why, What it costs, When to revisit, Where it shows in the product. If you are evaluating Whoobe, the What it costs sections are the ones to read: they are the trade-offs you accept.