Product Decisions This Supports
- Build vs. buy on the rendering layer. Any roadmap item requiring server-side HTML generation — transactional email, PDF/report output, static marketing pages, admin dashboards — forces a templating decision. This package is a candidate artifact in that evaluation, not a recommendation.
- Legacy containment vs. migration. If it already appears in a dependency tree, the real decision is whether to freeze it, fork it internally, or schedule a migration to Twig/Blade. The archived status and 2019 last release make this a decision with an obvious default.
- Dependency-policy calibration. 32 stars, single-maintainer signals, and a health score of 0.14 give leadership a concrete example of the "long tail" dependency class — useful when setting an approval threshold for OSS adoption.
- Licensing posture. MPL-2.0 is file-level copyleft: modifications to its files must be published under MPL, but linking into proprietary application code is generally permissible. That is a legal-review input, not a blocker.
- Reference implementation for an internal spike. If engineering wants to build a minimal template compiler, reading a small working one is cheaper than designing from a blank page.
- It does not support decisions about differentiated product capability. Templating is commodity infrastructure; nothing here is a moat.
When to Consider This Package
Consider it only when all of the following hold:
- Scope is an internal tool, throwaway script, or one-off static page generator where the blast radius of a defect is near zero.
- The runtime is a legacy PHP version already pinned in the environment; no PHP 8.x support is required.
- You intend to vendor or fork it and accept maintenance ownership yourself.
- Legal has signed off on MPL-2.0 obligations for any distributed artifact.
- Someone on the team has read the compiler and confirmed output-escaping behavior.
Look elsewhere — Twig, Blade, Latte, Plates, or Mustache — when any of these are true:
- Any new production surface. Archived since 2019 means no security response. Template engines are a textbook server-side template injection surface; an unpatched engine with undocumented sandboxing is unacceptable for anything touching user input.
- You need a maintainer to call, a CVE feed, or a support path. There is no one to call.
- PHP 8.x+ compatibility is required. Deprecation drift alone will cost more than the engine saves.
- You need docs, IDE support, framework integration, or hiring-pool familiarity. Twig and Blade are known quantities to candidates; this is not.
- Escaping and auto-sanitization are undocumented or opt-out. Verify before, not after.
- You need template inheritance, slots/components, or i18n. Mature engines have solved these; a 2019 micro-compiler likely has not.
The honest read: health score 0.14, 32 stars, and archived status together mean this is a reference artifact, not an adoption candidate. A proposal to add it to a new service is a signal the templating decision hasn't been made deliberately.
How to Pitch It (Stakeholders)
To executives: We evaluated a small open-source PHP templating library as a possible dependency for HTML rendering. It is competently scoped but effectively abandoned — last release 2019, archived, ~32 stars, one maintainer. The license (MPL-2.0) is workable. Recommendation: do not adopt for new work; if it exists anywhere in our stack, contain it and plan an exit. Doing nothing risks slow accumulation of unpatched rendering code in a security-sensitive layer; migration cost is small and bounded because mature alternatives cover our feature set. No roadmap change required.
To engineering: It compiles templates to HTML and is readable enough to borrow ideas from. Don't add it to a new service. If it's already in a repo, vendor it, pin it, and file a migration ticket against Twig/Blade in the next tech-debt cycle. Before any continued use, confirm output-escaping semantics — there is no maintained security posture behind it.
To legal: MPL-2.0, file-level copyleft. Modifications to its files must be released under MPL; linking into proprietary code is generally permissible. Relevant only if we fork or distribute.
Decision-log headline: Commodity capability, abandoned upstream, no differentiated value — decline to adopt; migrate off if present.