Design Principles¶

Seven principles guide every TAPPaaS decision — which software gets curated in, how modules must behave, what ships enabled and what stays off. It is why the platform is called TAPPaaS — pronounced tapas, like the Spanish dish: a curated collection of delicious servings that together make a comprehensive meal. The principles are ordered the way our users evaluate platforms: risk first, then cost, then capability — and they serve as the evaluation filter for new modules.

1. Digital sovereignty — you stay in control¶

Your data, your infrastructure, your rules. TAPPaaS runs fully local, across a cluster, or hybrid alongside cloud services. No dependency on external vendors: you decide where your data lives and who can access it. Directly relevant for government, healthcare, and SMBs subject to GDPR / NIS2 — see Digital Sovereignty.

2. Security by design — zero trust¶

Security is not a configuration step; it is the foundation. Every component follows zero-trust principles: access is denied by default, everything must be explicitly granted. Designed to align with NIS2 and modern cybersecurity standards — without requiring a security specialist to operate.

3. Locked down and safe by default¶

Exposing a service to the network requires an active, deliberate decision. The safe choice is always the default — protecting admins and users from accidental misconfiguration, one of the leading causes of security incidents in self-hosted environments. Most platforms expose everything by default; TAPPaaS does the opposite.

4. Open source — no vendor lock-in¶

Fully open source (MPL 2.0). No subscriptions, no per-seat pricing, no egress fees, no cloud-provider dependency. You can inspect, audit, fork, or migrate at any time — and it meets open-standard requirements for public-sector procurement.

The same rule governs every component we curate into the platform. It must:

• be open source;
• show a track record of being secure;
• be established and sufficiently feature-complete;
• use open data formats — so you can migrate away from any component, or from TAPPaaS itself. No lock-in means not even ours.

5. Predictable cost — hardware versatility¶

Runs on commodity x86 hardware: new, old, or mixed. Existing machines can be repurposed instead of replaced, extending asset lifecycles and reducing capital expenditure. Combined with zero subscription cost, this points to a significantly lower total cost of ownership than cloud or traditional enterprise platforms. Honest nuance: older hardware draws more power — the net benefit depends on hardware age and workload.

6. One platform — lower operational overhead¶

One platform covers deployment, updates, identity and access, monitoring, backup and lifecycle — no stitching together ten separate tools. Configuration is declarative JSON, consistent across every module.

7. Enterprise-grade stability, without enterprise complexity¶

Built on mature, actively maintained open-source components. Works offline. Accessible from mobile. Scales from a single node to a multi-node HA cluster — enterprise reliability without an enterprise budget or a platform-engineering team.

What TAPPaaS is not¶

TAPPaaS is not a general-purpose PaaS for large engineering teams. It is purpose-built for prosumers, SOHO, SMBs, and local or regional government IT — organisations that need modern cloud capabilities without cloud dependency or cloud complexity. If you run a hundred-developer product organisation, you want a different tool.

These principles are the yardstick for module decisions and roadmap trade-offs. Where the story came from: Why TAPPaaS.