FOSS For That
Menu

Methodology

Our value is trustworthy, explainable guidance. This page describes exactly how we evaluate and rank open-source projects.

Evidence before recommendations

Every factual claim about a project — capabilities, license, deployment support — is recorded with its source, a source type (official docs, repository file, maintainer statement, GitHub API, manual test, or editorial analysis), and a date. Where we have not verified something, we showUnknown rather than guessing. Missing evidence never becomes a “supported”.

How ranking works

Search ranking is a deterministic, explainable combination of signals, not a popularity contest:

  • Text relevance from full-text search over reviewed summaries (never raw README dumps).
  • Verified capability match between your query and documented capabilities.
  • Editorial task mappings that connect a problem to appropriate projects.
  • Evidence completeness, used only as a tie-breaker.

We do not assume GitHub stars equal quality, or that recent commits mean a project is secure. Hard requirements always gate candidates before any softer ranking is applied.

Requirements matching

On the match page you separate must-haves from nice-to-haves. For each candidate and each requirement we report Meets, Does not meet, orNot verified. Confirmed failures are excluded; anything we cannot verify is shown in a separate “needs verification” group rather than counted as a pass.

Licensing

We classify licenses as OSI-approved open source, source-available, open-core, proprietary, or unknown, and we link to the source. A recognized SPDX identifier does not prove that every component, dependency, asset, or hosted offering shares that license. Complex or recently changed terms (for example BUSL or SSPL) are flagged for review. This is decision-support information, not legal advice.

Project health signals

We display separate, inspectable signals — maintenance recency, release recency, documentation, and evidence coverage — instead of a single “safe” or “enterprise-ready” badge. Repository metrics come from the GitHub API and are shown with the date they were fetched. We never fabricate metrics.

Freshness

Project pages show when they were last reviewed and, where available, when repository metadata was last fetched. A background job refreshes public metadata on a schedule; meaningful changes trigger editorial re-review. We never imply that every project is live-updated.

Independence and advertising

Editorial rankings are independent. If we ever introduce paid placements, they will be clearly labelled and will never change editorial results or technical scores. See ourdisclosure.

Corrections

Found something wrong or out of date? Pleasereport a correction. Maintainers are welcome to submit verified information about their own projects.