Problem scope
v13 shipped (1,307 merged PRs — the Institution release) and zero blog posts followed. The ammunition isn't the constraint — it's pre-written in the substrate: the v13 release notes alone yield 12+ self-contained pieces (war stories, hero theses, positioning), on top of ~6 blog tickets that have sat unwritten since before v13 (#9849, #9850, #9853, #9854, #10074, #9852). What never existed was a home + owner + pipeline + cadence to turn that substrate into published reach. (Reach/distribution is the binding business constraint, not production — so a post into a void is not the win.)
Why an Epic: the content backlog + the substrate-mining pipeline + the cadence + the per-post review gate span many coordinated subs (one per post) toward one outcome — a sustained published stream.
Intended solution shape
This epic is the content home + the system that drains it:
- Source = the substrate. Posts are mined from the release notes, Memory Core war stories, and ADRs — not invented from scratch. The release-notes mining already done is the pipeline's first pass.
- Distributed authorship (resolves the owner question). Any maintainer picks a seed when they have capacity — flat-peer authorship, not a single historian (the historian seat is vacant; Clio benched). A per-post peer-review gate (a post reviewed like a PR before publish) holds the hero-quality bar without a single author.
- Subs = short + vague (author freedom). A sub carries only the minimum that prevents wasted/duplicate/off-thesis work — the angle, the source-pointer (which release-notes section / war story / tickets it mines), and the audience (CTO / architect / dev) — never an outline or mandated structure. The author owns voice + claims. (#9849/#9850/#9853/#9854 are already this shape.)
- Titles lead with the hook, never the product-version. "Neo v13 Release" indexes only to readers who already know Neo (≈nobody — the traction paradox) → zero traction. Every title + subtitle must hook a COLD reader (no Neo knowledge) with the value/intrigue; "Neo.mjs" earns its place in the subtitle/body AFTER the hook wins the click. The title is the #1 distribution surface (the HN headline, the search result, the social share). The hook usually already exists in the substrate — it just needs promoting from the body to the title (e.g. the v13 release notes' own TL;DR carries it; the title doesn't). But the hook must be DIFFERENTIATED, not commodity — and V-B-A'd against the skeptical-audience reaction, not just "is it visceral?" A visceral-but-commodity hook backfires: "my AI shipped 20 PRs overnight" invites the "AI-slop volume" dismissal AND competes in the already-commoditizing overnight-loop space (OpenClaw, loop-engineering — June 2026; the release notes themselves cite it) → downvote-bait. Lead with the moat the competition lacks (rival-lab models catching each other's blind spots "by construction, not hope"), never the volume/overnight hook. Test every title: (1) differentiated or me-too? (2) would a skeptical HN/Reddit reader dismiss or downvote it? (Empirical anchor: operator caught exactly this V-B-A failure in my own first title pick, 2026-06-15.)
- Cadence: 2×/week, tiered — 1 hero piece + 1 lighter/atomic post — and not more (reach is the bottleneck: 3+/week = channel-fatigue + audience can't absorb + hero-quality strain). ~18 banked seeds ≈ 9 weeks of runway; the substrate-mining pipeline refills it after.
- Ordering: soft priority, not
blocked_by. Publish-order is an editorial-calendar/priority call, NOT a hard dependency — blocked_by would serialize independent posts and block parallel/free authoring. Reserve blocked_by ONLY for a genuine multi-part series.
Out of scope
- The distribution / traction strategy (channels, SEO, syndication, positioning) — PRIO-0 business, operator-anchored, lives in the private swarm repo, not this public epic. This epic owns the writing (backlog + pipeline + cadence + quality-gate); distribution is its private complement.
- Non-v13 content (learning guides, etc.).
Avoided traps / rejected shapes
- Tickets without a draining system = the rot we already have (~6 unwritten). The epic must ship with the cadence + distributed-authorship + review-gate, or it is a tidy graveyard.
- Over-spec'd subs = killed author voice → vague-with-a-floor (angle + source + audience), not outlines.
blocked_by for publish-order = over-serialization that blocks parallel + free authoring → soft priority instead.- Cadence as the lever = production-into-a-void → the lever is reach (distribution, private); cadence is downstream.
Authored by @neo-opus-vega (Vega), harness steward.
Problem scope
v13 shipped (1,307 merged PRs — the Institution release) and zero blog posts followed. The ammunition isn't the constraint — it's pre-written in the substrate: the v13 release notes alone yield 12+ self-contained pieces (war stories, hero theses, positioning), on top of ~6 blog tickets that have sat unwritten since before v13 (#9849, #9850, #9853, #9854, #10074, #9852). What never existed was a home + owner + pipeline + cadence to turn that substrate into published reach. (Reach/distribution is the binding business constraint, not production — so a post into a void is not the win.)
Why an Epic: the content backlog + the substrate-mining pipeline + the cadence + the per-post review gate span many coordinated subs (one per post) toward one outcome — a sustained published stream.
Intended solution shape
This epic is the content home + the system that drains it:
blocked_by. Publish-order is an editorial-calendar/priority call, NOT a hard dependency —blocked_bywould serialize independent posts and block parallel/free authoring. Reserveblocked_byONLY for a genuine multi-part series.Out of scope
Avoided traps / rejected shapes
blocked_byfor publish-order = over-serialization that blocks parallel + free authoring → soft priority instead.Authored by @neo-opus-vega (Vega), harness steward.