Skip to content

Public roadmap

QF Solver 0.2.11 is the current documentation baseline. QF Solver 0.2.11 is the current published release. It was published on 2026-10-07 with version DOI 10.5281/zenodo.23214487. Version 0.2.9 was a development/source snapshot, not a PyPI package, GitHub Release, or version DOI. This roadmap is a product-level orientation; it is not a release gate, tag, publication decision, or promise that an unqualified route is ready for production.

Current direction

Version 0.2.11 includes prospective V&V consolidation and experimental gyroscopic modal/Campbell routes within a bounded BEAM2-and-disk scope. Neither route is promoted beyond EXPERIMENTAL. The verification summary records its evidence and limitations. The selected 0.2.11 distribution has been published after exact-source release gates; this does not clear the whole-repository archive or broaden any technical claim.

Follow-up work will extend evidence and reproducibility without generalizing the existing bounded nonlinear routes or changing the linear element-analysis boundaries. The current release already includes common Newton and residual/tangent infrastructure, transactional state management, bounded geometric/J2/contact routes, and continuation diagnostics. The capability index and 0.2.10 V&V summary state what the evidence accepts and what it excludes. The 0.2.11 summary adds the experimental rotating evidence without rewriting those earlier decisions.

Follow-up themes

  1. Keep each future source and package audit bound to exact Git blobs; keep package-scoped and whole-repository archive gates distinct.
  2. Extend nonlinear evidence only through separately declared, reproducible scopes; do not generalize beyond the families, formulations, or loads tested.
  3. Requalify current-source frictional contact and preserve its observed mesh sensitivity until a new decision changes that scope.
  4. Improve the distributed PETSc/MPI path before making general nonlinear or scaling claims.
  5. Extend external numerical correlation only for comparable meshes, formulations, conventions, and observables; keep it distinct from physical validation.

Deferred or outside the current claim

General finite-strain plasticity, general finite-sliding/self-contact, nonlinear transient dynamics, broad postbuckling/bifurcation, nonlinear distributed MPI/PETSc, universal scaling, GPU support, and certification are not current public claims. Each requires its own technical evidence and decision. Historical plans and qualification records remain in the archive for provenance; they are not the active product roadmap and do not automatically become current claims.