# PHP Alive and Kicking Episode 34 — Derick Rethans > [!info] Ported from an automated deep-read of a locally recovered transcript (no platform captions existed). > Sourced 2026-08-16. Only text inside `> [!quote]` blocks is verbatim; names and technical terms may carry recognition errors from automatic transcription. Feature status, vote counts, and release plans reflect 2026-08-04 and need current primary-source verification. > [!question] Lingering Questions > - Which PHP 8.6 features discussed here (duration API, partial function application, TLS session resumption) were actually accepted and shipped after this recording? > - What minimum evidence should a deprecation RFC require about transitive (not just direct) ecosystem use? > - What bus-factor target is realistic for a critical PHP service currently held by one or two maintainers? > [!tip] Article Ideas > - Compare PHP's and WordPress's approaches to triage, deprecation-evidence requirements, and contributor succession — both are mature, compatibility-committed ecosystems facing the same bus-factor problem. > [!link] Cross Reference > - [[3. Resources/Permanent Notes/Maintainer Attention Is the Scarce Resource in Open Source]] — Rethans's triage-queue account is a direct instance of this claim > - [[3. Resources/Permanent Notes/A Published Deprecation Schedule Makes a Fast-Moving Dependency Safe to Adopt]] — the deprecation-evidence argument here is the producer-side mirror of that consumer-side claim > - [[3. Resources/MOCs/Open Source Governance MOC]] > [!quote] Vocabulary > - *duration API* — a PHP 8.6 date/time addition distinguishing calendar operations like "add a month" from "go to next month," which aren't identical > - *path/branch coverage* — code-coverage granularity finer than line coverage; can reveal gaps line coverage misses, at real cost (a proposed Xdebug implementation reports ~10% extra overhead on an already expensive operation) ## Bib - **Source**: [PHP Alive And Kicking, Episode 34](https://www.youtube.com/watch?v=JXW9Z8uKvQM) — PHP Architect YouTube channel, guest Derick Rethans - **Reader doc**: [[3. Resources/Readwise/Articles/PHP Alive and Kicking Episode 34 — Derick Rethans]] ## Author's thesis Mature open-source language stewardship is mostly maintenance work — triage, compatibility analysis, infrastructure renewal, documentation ownership, review, and trust-building — not new syntax. Rethans traces this across PHP 8.6 feature work, Xdebug direction, php.net/documentation modernization, deprecation governance, infrastructure migration, and contributor succession, arguing each area survives on the same discipline: evidence before removal, accessible entry points before deep trust, and staged access rather than default openness for sensitive systems. ## Key arguments (≤3) 1. **Triage is foundational, unglamorous work** — over 900 open issues at one point; Rethans entered PHP through years of bug reproduction, and the episode treats this as central maintenance, not a side task junior contributors get stuck with. 2. **Deprecations need a positive case, not just staleness** — "you need to have a very good reason for removing something"; ecosystem scans must account for transitive dependencies, not just direct use in top repositories. 3. **Contributor trust grows in stages** — triage and report validation are accessible entry points that don't require internals depth; security and infrastructure access grow through review because they can't be open by default. ## Claim atoms ### Open-source health depends on sustained triage, not only feature development > [!quote] ID:rw_01kz7vdprw21x94xkxx5hh4yx9 > there was over 900 issues, Where/why: [09:37](https://www.youtube.com/watch?v=JXW9Z8uKvQM&t=577) — a growing, uncategorized queue is presented as evidence of insufficient triage capacity, not merely a backlog to clear. ### Backward-compatibility decisions need ecosystem evidence and a stated benefit > [!quote] ID:rw_01kz7vdnpgc3xcgkvkfcd7571t > I think you need to have a very good reason for removing something. Where/why: [46:12](https://www.youtube.com/watch?v=JXW9Z8uKvQM&t=2772) — deprecation governance argument; removal cost gets transferred to users unless the benefit is concrete. ### Correct diagnostic output can justify measured performance cost > [!quote] ID:rw_01kz7vdwysgqh085yq8fb7bd19 > I prefer correct output over quick output. Where/why: [30:10](https://www.youtube.com/watch?v=JXW9Z8uKvQM&t=1810) — Xdebug coverage-accuracy tradeoff; a ~10% overhead is acceptable when it closes a correctness gap line coverage misses. ### Contributor growth starts with accessible work and expands trust through review > [!quote] ID:rw_01kz7vdtsxbt2f9fr150jcsrm7 > you don't need to know the internals so deep for that Where/why: [61:43](https://www.youtube.com/watch?v=JXW9Z8uKvQM&t=3703) — succession argument; triage as a low-barrier on-ramp that builds toward deeper trust. ## Permanent candidates - **"A growing, categorized issue queue is evidence of insufficient triage capacity, not just backlog"** — extends the vault's existing maintainer-attention claim with a specific measurement angle. Promoted below. - **"Deprecation needs adoption evidence and a stated benefit before it transfers migration cost to users"** — connects to and extends existing vault deprecation-governance claims. Promoted below. ## Processing checklist - [x] Connect into the relevant MOC — [[3. Resources/MOCs/Open Source Governance MOC]] - [x] Create two Permanent notes (triage-as-evidence, deprecation-evidence) - [ ] Verify PHP 8.6 feature claims against final RFC and release records before citing elsewhere ## Links - [[3. Resources/MOCs/Open Source Governance MOC]] — maintenance, deprecation governance, and contributor succession as governance infrastructure - [[3. Resources/Permanent Notes/A Growing Categorized Issue Queue Measures Triage Capacity Not Just Backlog]] — new Permanent note this source generated - [[3. Resources/Permanent Notes/A Deprecation Needs Adoption Evidence Before It Transfers Migration Cost]] — new Permanent note this source generated