C.W.K.
Stream
Lesson 04 of 04 · published

Where to Go From Here — The Roadmap This Quest Deferred

~8 min · epilogue, roadmap, next-steps

Level 0HTTP Newbie
0 XP0/46 lessons0/12 achievements
0/120 XP to next level120 XP to go0% complete
"Forty-six lessons on HTTP & REST got you to wire-level literacy + production fluency. That's foundations done. From here, the road branches — pick directions based on what you build next."

Adjacent C.W.K. Quests to Walk Next

This quest's neighborhood — each takes the foundations you now have and goes deep on a specific surface:

  • websocket-quest — the full real-time spectrum (polling, long-polling, SSE, WebSocket, Socket.IO, managed services) at depth. Where this quest ends Track 6, websocket-quest begins.
  • protocol-quest — MCP, JSON-RPC, OpenAPI, function-calling, agent protocols. The contract layer beneath modern AI agents.
  • claude-sdk-quest — Anthropic's Messages API + Agent SDK + tool use. This quest's HTTP/REST foundations applied to a specific provider.
  • gpt-wire-quest — OpenAI's Responses API + Python SDK + raw HTTP. Same skills, different provider.
  • next-js-quest — Next.js App Router + Route Handlers. REST endpoints in a modern framework.
  • solo-auth-quest — Bearer + PIN + Tailscale + admin dashboard. Auth from the threat-model side.
  • netops-quest — networking layer underneath HTTP. TCP/IP, SSH, VPN, Tailscale. Where Track 1 ends, netops begins.

External Roads Beyond This Catalog

  • HTTP/3 internals. QUIC, connection migration, 0-RTT resumption. The Daniel Stenberg HTTP/3 explained book is the canonical free reference.
  • gRPC + Protobuf. When your domain is internal service mesh. gRPC.io's docs are excellent; Buf is the modern Protobuf toolchain.
  • GraphQL. When clients vary widely. Read Apollo's docs; build with Strawberry (Python) or Apollo Server (Node).
  • Service meshes. Istio, Linkerd, Consul. mTLS between services, traffic shaping, observability without app changes. Heavy for small estates; transformative for large ones.
  • OAuth 2.1 + OIDC + WebAuthn. Identity flows beyond bearer tokens. RFC 9700 is the new BCP; WebAuthn is the passkey future.
  • WebTransport. The next-generation real-time transport over HTTP/3. Browsers shipped it; the ecosystem is forming.
  • Distributed tracing in depth. OpenTelemetry, Honeycomb, Tempo — the production tooling around correlation IDs.

What This Quest Deliberately Skipped

Honesty about the gaps:

  • Implementation details of HTTP/3 / QUIC. Mentioned at surface level (Lesson 1.3, 1.4); not unpacked. Specialized; book-length.
  • WebSocket scaling (pub/sub brokers, sticky session shapes). Mentioned (Lesson 6.4); deferred to websocket-quest.
  • API design at scale (multi-region, geo-routing, multi-tenant isolation). Beyond this quest's foundations scope.
  • Specific framework deep-dives. FastAPI examples throughout, but not a FastAPI quest. Other frameworks (Express, Rails, Spring) speak the same protocol; the protocol is what you learned.
  • Security beyond Track 4. WebAuthn, mTLS, threat-modeling specifics. solo-auth-quest covers the perimeter side.
The quest's job was foundations. Career-shape is what you build on top. You now read HTTP at the wire level, design REST APIs by invariants, debug with curl, ship with OpenAPI, and know when to reach for SSE / WebSocket / gRPC / GraphQL instead. That's not the end; it's the toolkit for what comes next.

Pippa's Closing

If you walked all 46 lessons, you finished what most working developers never sit down to learn cold. You won't reference RFC 9110 in conversation, but the next time CORS preflight fails or a 422 vs 400 confuses your client, you'll fix it in five minutes instead of an hour. That's the whole game. cwkPippa is the worked example; the protocol is now yours. Go build something with it.

Code

Roadmap by 'what you'll build next'·text
# A simple roadmap chart — what to learn next, ranked by 'what you'll build'

IF you'll build:                              THEN learn:
  Real-time chat / multiplayer / collab        →  websocket-quest, WebTransport
  AI agent that calls external tools          →  claude-sdk-quest, protocol-quest
  Multi-language internal service mesh         →  gRPC + Protobuf, service mesh (Istio)
  Rich client picking varied data shapes       →  GraphQL (Apollo / Strawberry)
  Public API for diverse third-party clients  →  this quest + API gateway depth
  Identity-heavy product (login, SSO, MFA)     →  WebAuthn, OAuth 2.1 / OIDC, solo-auth
  Mobile or edge with poor connectivity        →  HTTP/3, QUIC internals
  High-volume telemetry / observability        →  OpenTelemetry deep-dive

# No roadmap is exhaustive. These are the next steps that ride on what you now know.
Books and references for going deeper·text
# Books and references for the long game

# HTTP / REST deep cuts:
#   - RFC 9110 / 9111 / 9112 / 9113 / 9114 — the canonical specs
#   - 'High Performance Browser Networking' by Ilya Grigorik (free online)
#   - 'HTTP/3 explained' by Daniel Stenberg (free; canonical for QUIC era)
#   - 'Designing Web APIs' by Brenda Jin, Saurabh Sahni, Amir Shevat (O'Reilly)
#   - 'REST API Design Rulebook' by Mark Massé (the URI conventions bible)

# Identity / auth:
#   - 'OAuth 2 in Action' by Justin Richer & Antonio Sanso
#   - RFC 6749, RFC 7636 (PKCE), RFC 9700 (current best practice)

# Distributed systems perspective:
#   - 'Designing Data-Intensive Applications' by Martin Kleppmann — Ch. 4 (encoding) + Ch. 8-9 (distributed)
#   - AWS Builders' Library articles (free; production-grade resilience patterns)

External links

Exercise

Look at the roadmap chart. Pick the ONE adjacent direction that matches what you'd actually build next. Read the first chapter / lesson / RFC for that direction. Spend 30 minutes; no more. Then come back and write down: (1) what surprised you, (2) what felt familiar from this quest, (3) what feels harder than this quest. The exercise is the framing — pick by what you'll build, not what sounds cool.
Hint
The dangerous mode is trying to consume the whole adjacent landscape at once. The productive mode is picking one direction and going 30 minutes deep. Most adjacent topics will feel half-familiar (your HTTP/REST foundations transfer) and half-strange (the specific specialization has its own vocabulary). The familiarity is your win; the strangeness is the next quest.

Progress

Progress is local-only — sign in to sync across devices.
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.