Text of peps.python.org › pep-0598 as Styx crawled it on Nov 19, 2026. The page may have changed since. Only its words are kept, without pictures or layout, and only the first part of a long page.

Open the live page

PEP 598 – Introducing incremental feature releases | peps.python.org

PEP 598 – Introducing incremental feature releases PEP 598 – Introducing incremental feature releases Author : Alyssa Coghlan <ncoghlan at gmail.com> Discussions-To : Discourse thread Status : Withdrawn Type : Informational Created : 15-Jun-2019 Python-Version : 3.9 Table of Contents Abstract PEP Withdrawal Summary Example Future Release Schedules Proposal Baseline feature releases and feature release series Incremental feature releases Maintaining interoperability across incremental feature releases Feature complete release and subsequent maintenance releases Proposed policy adjustment for provisional APIs Motivation Caveats and Limitations Design Discussion Why this proposal over simply doing more frequent baseline feature releases?

Implications for Python library development Implications for the proposed Scientific Python ecosystem support period Implications for simple deployment environments Implications for complex deployment environments Duration of the feature additions period Duration of the unreleased pre-alpha period Why not switch directly to full semantic versioning? References Copyright Abstract PEP 602 proposes reducing the feature delivery latency for the Python standard library and CPython reference interpreter by increasing the frequency of CPython feature releases from every 18-24 months to instead occur every 9-12 months.

This PEP proposes to instead reduce the frequency of new baseline feature releases (with the associated filesystem layout changes, bytecode format changes, and C ABI compatibility breaks) to occur only every other year (2020, 2022, 2024, etc), but to combine that change with a new policy and approach that allows the introduction of backwards compatible features in the initial set of point releases within a given release series.

PEP Withdrawal This PEP has been withdrawn in favour of the rolling beta release stream proposal in PEP 605 . However, the concerns raised in this PEP are likely to apply to any other “Long Term Support branch” proposals that allow feature backports to improve the developer experience of supporting such releases (such as the EL Python draft at [3] ), so the ideas presented here may provide useful design suggestions for such proposals.

Summary The proposal to keep the current CPython release compatibility management process, but go through it more often has significant practical downsides, as a CPython feature release carries certain expectations (most notably, a 5-year maintenance lifecycle, support for parallel installation with the previous feature release, and the possibility of breaking changes to the CPython-specific ABI, requiring recompilation of all extension modules) that mean faster feature releases in their current form have the potential to significantly increase the burden of maintaining 3rd party Python libraries and applications across all actively supported CPython releases.

It’s also arguable whether such an approach would noticeably reduce the typical feature delivery latency in practice for most end users, as the adoption cycle for new feature releases is typically measured in months or years, so more frequent releases may just lead to end users updating to every 3rd or 4th feature release, rather than every 2nd or 3rd feature release (as often happens today). This PEP presents a competing proposal to instead slow down the frequency of parallel installable feature releases that change the filesystem layout and CPython ABI to a consistent 24-month cycle, but to compensate for this by introducing the notion of build compatible incremental feature releases, and then deferring the full feature freeze of a given feature release series from the initial baseline X.Y.0 release to a subsequent X.Y.Z feature complete release that occurs ~12 months after the initial baseline feature release.

A new feature_complete attribute on the sys.version_info structure will provide a programmatic indicator as to whether or not a release series remains open to further incremental feature releases. Alternate implementations of Python would also be free to clear this flag to indicate that their support for their nominal Python version may still be a work in progress. For compatibility testing purposes, and to maintain pickle compatibility in mixed version environments, a new sys.feature_limit attribute (and corresponding CPython CLI parameter, --feature-limit X.Y.Z , and environment variable, PYTHONFEATURELIMIT ) will provide a way to limit the runtime availability of features added in incremental feature releases.

The existing cycle and the new cycle would be synchronised on their feature freeze release dates, so the full Python 3.9.x feature freeze would occur in October 2021, 24 months after the Python 3.8.0 feature release, with the initial Python 3.9.0 release taking place in October 2020. Example Future Release Schedules Under this proposal, Python 3.9.0a1 would be released in November 2019, shortly after the Python 3.8.0 feature complete release in October 2019.

The 3.9.0b1 release would then follow 6 months later in May 2020, with 3.9.0 itself being released in October 2020. Assuming micro releases of 3.9.x were to occur quarterly, then the overall release timeline would look like: 2019-11: 3.9.0a1 … additional alpha releases as determined by the release manager 2020-05: 3.9.0b1 … additional beta releases as determined by the release manager 2020-08: 3.9.0bX (final beta release that locks ABI compatibility) 2020-09: 3.9.0rc1 … additional release candidates as determined by the release manager 2020-10: 3.9.0 (BFR - baseline feature release) 2021-01: 3.9.1 (IFR - incremental feature release) 2021-04: 3.9.2 (IFR) 2021-07: 3.9.3 (IFR) 2021-10: 3.9.4 (feature complete release) 2022-01: 3.9.5 2022-04: 3.9.6 2022-07: 3.9.7 2022-10: 3.9.8 (final regular maintenance release) … additional security fix only releases as needed 2025-10: 3.9.x branch closed Feature complete release numbers would typically be written without any qualifier (as they are today), while the baseline and incremental feature releases would be expected to have a qualifier appended indicating that they aren’t a traditional CPython release ( 3.9.0 (BFR) , 3.9.1 (IFR) , etc).

The Python 3.10 release series would start being published the month after the first Python 3.9 feature complete release, in parallel with the final 12 months of routine Python 3.9 maintenance releases: 2021-11: 3.10.0a1 … additional alpha releases as determined by the release manager 2022-05: 3.10.0b1 … additional beta releases as determined by the release manager 2022-08: 3.10.0bX (final beta release that locks ABI compatibility) 2022-09: 3.10.0rc1 … additional release candidates as determined by the release manager 2022-10: 3.10.0 (BFR) 2023-01: 3.10.1 (IFR) 2023-04: 3.10.2 (IFR) 2023-07: 3.10.3 (IFR) 2023-10: 3.10.4 2024-01: 3.10.5 2024-04: 3.10.6 2024-07: 3.10.7 2024-10: 3.10.8 (final regular maintenance release) … additional security fix only releases as needed 2027-10: 3.10.x branch closed In this model, there are always two or three active branches: 2019-04 -> 2019-10: 3.9.0 pre-alpha, 3.8.0 pre-release, 3.7.x maintenance 2019-10 -> 2020-05: 3.9.0 pre-beta, 3.8.x maintenance 2020-05 -> 2020-10: 3.10.0 pre-alpha, 3.9.0 pre-release, 3.8.x maintenance 2020-10 -> 2021-10: 3.10.0 pre-alpha, 3.9.x feature releases, 3.8.x maintenance 2021-10 -> 2022-05: 3.10.0 pre-beta, 3.9.x maintenance 2022-05 -> 2022-10: 3.11.0 pre-alpha, 3.10.0 pre-release, 3.9.x maintenance 2022-10 -> 2023-10: 3.11.0 pre-alpha, 3.10.x feature releases, 3.9.x maintenance 2023-10 -> 2024-05: 3.11.0 pre-beta, 3.10.x maintenance 2024-05 -> 2024-10: 3.12.0 pre-alpha, 3.11.0 pre-release, 3.10.x maintenance … etc (Pre-alpha and pre-beta development occurs on the main git branch, all other development occurs on a release specific branch with changes typically backported from the main git branch) TODO: this really needs a diagram to help explain it, so I’ll add

The rest of this page is on the live site.