Small is a product decision
Fabric’s small apps began with a simple observation: a useful tool does not have to become a platform.
Lynx can care about links. Squidly can care about a compact public link page. Piper can care about creative management. Snipe can care about embedding. Their smallness is not an early-stage embarrassment waiting to be replaced by a giant suite. It is part of the design.
The infrastructure can still be serious
Narrow software still needs boring things done well: accounts, organizations, access control, billing, limits, responsive interfaces and clear upgrade paths. Sharing that infrastructure across apps makes it possible to keep the visible product small without making the underlying product careless.
That distinction matters. “Tiny app” should describe the surface area of the problem, not the quality of the plumbing.
Where the platform stops
The interesting question is where shared infrastructure becomes unwanted gravity. If every small tool has to inherit every concept from a central platform, the suite stops being small in practice even if each icon still looks cute.
The working rule is therefore restraint: share identity, organizations, entitlements and the pieces that genuinely reduce friction; keep the interaction model specific to the job.
Graduation without erasure
The lab adds another piece to this model. When a tiny experiment becomes a focused product, it can graduate to its own domain without erasing the path that produced it. fbc.ad keeps the project record. The product site gets to stay clean.
That makes “small” easier to preserve. A product does not need to contain its own museum.