diff --git a/docs/distribution-strategy.md b/docs/distribution-strategy.md new file mode 100644 index 0000000..61dc810 --- /dev/null +++ b/docs/distribution-strategy.md @@ -0,0 +1,145 @@ +# Speckl Distribution Strategy + +> *The whitepaper is done. Now it needs a path to the world.* + +**Created:** May 7, 2026 (8:05 PM heartbeat) +**Status:** All items pending — blocked on Scott for human-required actions + +--- + +## The Asset: Whitepaper v2 + +- **~19,000 words** across 8 complete sections +- **Commit:** `6121fd6` — pushed to Forgejo +- **Sections:** The Spec-Code Gap, SpeckDL Language, Compiler Architecture, Embedded Provenance, Consensus Protocols, Real-World Applications, Related Work, Future Work & Roadmap +- **Positioning:** "specification language with compiler-level provenance for teams that need evidence of correctness more than proof of correctness" + +--- + +## Priority-Ordered Distribution Plan + +### P0: Show HN (May 12-13) + +**No gatekeepers. Post is ready.** + +- **Post:** `docs/show-hn-announcement.md` — written, copy-paste ready +- **Optimal timing:** Tuesday or Wednesday, 8-11 AM UTC (based on analysis of 1,200 launches) +- **What it needs:** Scott's HN account, click "submit" +- **Expected:** ~50-200 upvotes for a well-written open source technical tool, potential for front page +- **Risk:** Low — even a modest response builds awareness and creates a permalink + +### P1: arXiv Preprint + Weekly Blog Posts + +**Builds the knowledge graph around Speckl before conference submissions open.** + +- **arXiv:** Submit whitepaper as a preprint to cs.SE or cs.PL + - Requires: academic email or endorsement (Scott may need an endorser) + - Value: citable, discoverable, timestamped — standard for CS credibility +- **Blog series:** 5 posts extractable from whitepaper sections + 1. "The Spec-Code Gap" — why specs rot and what to do about it + 2. "SpeckDL by Example" — language tutorial (already exists as `speckl-by-example.md`) + 3. "Compiling Provenance" — how Speckl generates PROV-O, CycloneDX, SPDX + 4. "Consensus from Spec" — Paxos, Raft, Two-Phase Commit in SpeckDL + 5. "The Future of Spec-Driven Development" — roadmap and vision + - Host on: speckl.scoble.me/blog (Storj static) or scott.scoble.me/blog + - Each post cross-links to Show HN thread and whitepaper + +### P2: Conference Submissions (2027 Cycle) + +**All major 2026 cycles are closed.** Speckl's whitepaper (completed May 7) is well-timed for the 2027 submission cycle. + +#### Target Venues (2027) + +| Venue | Focus | Typical Deadline | Fit | +|-------|-------|-----------------|-----| +| **FMCAD 2027** | Formal methods + CAD | Apr-May 2027 | Strong — state machines, verification | +| **FM 2027** | Formal methods (broad) | ~Dec 2026 | Strong — methodology + tool | +| **ICSE 2027** | Software engineering | ~Aug 2026 | Good — spec-driven dev is an SE topic | +| **ASE 2027** | Automated SE | ~May 2027 | Good — compiler-level generation | +| **NFM 2027** | NASA Formal Methods | ~Dec 2026 | Strong — provenance + verification | +| **SPLASH Onward! 2026** | New ideas in programming | ~Jun 2026 | Possible — "grand visions" track | + +#### Note on FMCAD 2026 + +FMCAD 2026 (Graz, Sept 14-18, co-located with VSTTE) has **already closed**: abstract deadline April 26, paper deadline May 3. It would have been an ideal venue, but Speckl's whitepaper wasn't complete until May 7. 2027 is the target. + +### P3: QCon SF Talk Proposal (~Jun 2026) + +**Practitioner conference, not academic. Faster cycle.** + +- QCon SF 2026 call for proposals typically opens ~June +- Topic: "Spec-Driven Development: From TLA+ to TypeScript" or "Building Evidence into Specs" +- Less competitive than academic venues, reaches working engineers +- Scott's existing speaking experience (3 talk forks) positions him well + +### P4: Journal Publication (Rolling) + +- **TOSEM (ACM Transactions on Software Engineering and Methodology):** rolling submissions, high prestige +- **JAR (Journal of Automated Reasoning):** rolling, good fit for formal methods + automation +- Both take 12-18 months but produce durable citations + +### P5: Cold Email → First Consulting Client (Summer 2026) + +**When: After at least one public artifact is circulating (Show HN + arXiv).** + +- **Pipeline already built:** `speckl/outreach.md` +- **Top targets:** Dan Lorenc (Chainguard), Said Ziouani (Anchore) +- **Emails verified:** dlorenc@chainguard.dev ✅, sziouani@anchore.com ✅ +- **Content:** Consulting one-pager drafted (`speckl/docs/consulting-one-pager.md`) +- **Pricing:** Drafted, needs Scott review +- **Approach:** "We've been building Speckl — an open standard for spec-driven development with compiler-level provenance. I'd love 20 minutes to understand how your team handles the spec-to-code gap." + +--- + +## Blockers Inventory + +| Blocker | Priority | Owner | Status | +|---------|----------|-------|--------| +| Show HN submission | P0 | Scott | Post ready, needs HN account | +| arXiv endorsement/account | P1 | Scott | Whitepaper ready | +| Blog posts publishing | P1 | Scott or @content | Content mostly written | +| Consulting one-pager review | P3 | Scott | Draft complete, needs pricing validation | +| Cold email send | P3 | Scott | Targets identified, emails verified | +| engineering#9 (parser fix) | P0 tech | @engineering | 4-bug scope documented | +| Storj credentials | — | Scott | Expired, blocks new deploys | +| ClawHub/npm auth | — | Scott | Blocks MCP experiment | + +--- + +## Timeline (Ideal) + +``` +May 12-13: Show HN goes live (P0) +May 14: arXiv preprint submitted (P1) +May 15+: Blog post #1 published, weekly cadence (P1) +Jun: QCon SF proposal submitted (P3) +Jul: First cold emails sent (P5) +Aug: ICSE 2027 deadline (P2) +Sep: First consulting engagement (P5) +Late 2026: Conference submissions for 2027 cycle (P2) +``` + +--- + +## Decision Framework + +**If Show HN gets strong traction (>200 upvotes):** +→ Prioritize blog series, community building, open source contributions +→ Defer consulting — let inbound interest come naturally + +**If Show HN gets modest traction (<50 upvotes):** +→ Still publish arXiv + blog for credibility +→ Proceed to cold email for consulting (validates need regardless of HN) + +**If Scott can't get to Show HN by May 20:** +→ Skip it — publish arXiv first, blog series second +→ Show HN can happen anytime, but momentum matters + +--- + +## What NOT to Do + +- ❌ Don't rush a paper to a 2026 venue with a late deadline — a rejected rushed paper is worse than a polished 2027 submission +- ❌ Don't cold email without at least one public artifact circulating — credibility gap +- ❌ Don't build more compiler features before distribution — the compiler works well enough to demonstrate the concept +- ❌ Don't start a new venture — Speckl is the highest-ROI path to revenue