[[“lab”, “stack-logs”, “vibe-coding”, “writing”, “blueprint”, “ai-writing”, “content-strategy”], [“H1] Title Options (pick one)\n\n- The 20-Minute Blog: An Archaeological Update (2023 \u2192 2026)\n- Three Years Later: Revisiting My "20-Minute AI Blog”, “Experiment\n- From 33 Minutes to 5: What I Got Wrong About AI Writing (2023\u20132026)\n\nWhy this title: Contains original hook (20-minute blog) + timeframe + evolution signal. SEO captures both historical and current search intent.\n\n—\n\n### [H2] The Artifact (2023)\n\nWhat to include:\n\n- [ ] Original URL: http://aihackers.net/2023/04/04/the-20-minute-blog/\n- [ ] Original date: April 4, 2023\n- [ ] Recovery status: (circle one) Fully recovered / Partially recovered / Memory only\n\nContent to recover or reconstruct from memory:\n\n\n[IF YOU HAVE WAYBACK/ARCHIVE: paste original text here]\n\n[IF WRITING FROM MEMORY, answer:]\n- What was the exact premise? (\"Create blog from nothing in 20 min...", "What tool did you use? (GPT-3.5, GPT-4, Jasper, etc.)\n- What was the actual time? (33 minutes)\n- What was the specific workflow? (prompt \u2192 draft \u2192 edit \u2192 publish)\n- What was the topic you wrote about?\n- Where did you share it? (HN, Reddit, Twitter) \n- What was the engagement? (78 upvotes? 12 comments?)\n- What surprised you about the result?\n\n\nWhy include this: Establishes you were doing this before it was mainstream. The 33-minute reality vs 20-minute goal is the authenticating detail.\n\n—\n\n### [H2] The 2026 Stack Comparison\n\nFill in this table with your actual current workflow:\n\n| Aspect | 2023 Stack | 2026 Stack | What Changed |\n|——–|———–|————|————–|\n| Primary Model | [e.g., GPT-3.5] | [e.g., Claude Sonnet 4.5] | [capability jump] |\n| Workflow Step 1 | [describe] | [describe] | [delta] |\n| Workflow Step 2 | [describe] | [describe] | [delta] |\n| Workflow Step 3 | [describe] | [describe] | [delta] |\n| Total Time | 33 minutes | [your actual time] | [what sped up, what got added] |\n| Quality Gate | [manual edit] | [describe current verification] | [how rigor changed] |\n| Publishing | [manual] | [AI-native CMS / automated] | [friction reduction] |\n\nWhy this table: Concrete before/after is more credible than abstract claims. Shows you’re not just hand-waving about "AI getting better.”, “–”, [“H2] The Real Shift: [Your Insight Here]\n\nPrompt to fill out:\n\n> 2023, I thought the question was: ________________ \n> 2026, I realize the question is: ________________\n\nExample direction:\n- 2023: "Can AI help me write faster?"\n- 2026: "Can I articulate what I want clearly enough?”, “Alternative angles to consider:\n- Speed vs. clarity\n- Generation vs. verification \n- Drafting vs. thinking\n- Production vs. intention\n\nWhy this matters:** This is your thesis. Everything else supports this insight.\n\n—\n\n### [H2] Why I’m Not Just Rewriting the 2023 Post\n\nAddress these points (your own prose):\n\n- [ ] Why recover/recreate when you could just rewrite?\n- [ ] What’s lost by not having the original?\n- [ ] What’s gained by documenting the evolution instead?\n- [ ] What does this say about tool churn in AI?\n\nMeta-note to include:”, “This post exists because I found the URL in my browser history but couldn’t access the content. Rather than recreate something outdated, I’m documenting what changed.”, “Why include this:** Shows intellectual honesty. Establishes the pattern for the other lost posts.\n\n—\n\n### [H2] Related Links (Interlinking)\n\nMust link to:\n- Article 2 in this series (the other lost posts)\n- Current stack/tools you use (your /tools/ or /implement/ pages)\n- Security considerations you’ve learned (link to /risks/ or /verify/)\n\nWhy interlink: Distributes link equity. Shows this isn’t a one-off but part of a larger body of work.\n\n—\n\n## Article 2: The Lost Posts of April 2023\n\nURL: /posts/vibe-coding-april-2023-lost-posts/ \nPurpose: Show scope of early experimentation + what survived vs. what didn’t \nRedirects: \n- /2023/04/04/the-fortune-telling-cyborg* \u2192 this article\n- /2023/04/04/the-top-10-ai-technologies* \u2192 this article \n- /2023/04/05/10-game-changing-ai-chatbot-plugins* \u2192 this article\n\n—\n\n### [H1] Title Options\n\n- The Fortune-Telling Cyborg and Other Lost Experiments (April 2023)\n- Digital Archaeology: My Deleted AI Blogging Experiments\n- What I Wrote (and Deleted) During the First Wave of GPT-4\n\n—\n\n### [H2] The April 2023 Batch\n\nList the four posts from the screenshot:\n\n| # | Title (from URL/sitemap) | Date | What I Remember | Recovery Status |\n|—|—————————|——|—————–|—————–|\n| 1 | The 20-Minute Blog | Apr 4 | [fill in] | [recovered/lost] |\n| 2 | The Fortune-Telling Cyborg: How AI is Revolutionizing Divination | Apr 4 | [fill in] | [recovered/lost] |\n| 3 | The Top 10 AI Technologies to Watch in 2023 | Apr 4 | [fill in] | [recovered/lost] |\n| 4 | 10 Game-Changing AI Chatbot Plugins That Will Change Your Life Forever | Apr 5 | [fill in] | [recovered/lost] |\n\nFor each post, answer:\n- What was the premise?\n- Why that topic?\n- How was it received?\n- Why did you delete it?\n- What would you do differently now?\n\nWhy this table: Creates the archaeological record. Shows breadth of experimentation.\n\n—\n\n### [H2] The Fortune-Telling Cyborg: [Deep Dive if recovered / Reconstruction if not]\n\nThis is the weird one. Worth exploring because:\n- It shows you were experimenting with non-obvious AI uses\n- It’s embarrassing enough to be authentic\n- It connects to current concerns about AI "hallucination”, “and anthropomorphism\n\nIf you have the content:\n- [ ] Paste original text in a blockquote\n- [ ] Annotate what you now realize was wrong/problematic\n- [ ] Connect to current understanding of AI "reasoning”, “If reconstructing from memory:\n- [ ] What was the thesis? (AI as oracle? Pattern matching as divination?)\n- [ ] What examples did you use?\n- [ ] How did you treat hallucinations? (As intuition? As noise?)\n- [ ] What would this post look like written in 2026?\n\nWhy this post matters:** The 2023 view: "AI generates convincing predictions" \u2192 2026 view:”, “AI generates convincing but ungrounded predictions, which is a security/psychology problem.”, “–”, [“H2] The Listicles: What They Reveal About 2023\n\nBoth "Top 10 AI Technologies" and "10 Game-Changing Plugins”, “are listicles.\n\nAnalyze them as artifacts:\n\n- [ ] What was on the 2023 list? (LangChain? Auto-GPT? Specific wrappers?)\n- [ ] How many are still relevant in 2026?\n- [ ] What does their obsolescence teach us?\n- [ ] What replaced them?\n\nWhy analyze listicles: Shows you’re willing to interrogate your own low-effort content. Demonstrates learning.\n\n—\n\n### [H2] Why I Deleted Them (And Why I’m Writing About Them Now)\n\nAddress the contradiction:\n\n- [ ] What felt wrong about them in 2023?\n- [ ] What felt wrong about them in 2024/2025?\n- [ ] Why revisit them now instead of letting them stay buried?\n- [ ] What does this say about content longevity in the AI era?\n\nWhy this section: Intellectual honesty about embarrassment. Shows growth trajectory.\n\n—\n\n### [H2] What Survived from April 2023?\n\nCounterpoint section: Not everything was deleted.\n\n- [ ] What principles from those experiments are still in your work?\n- [ ] What workflows survived?\n- [ ] What intuitions turned out to be correct?\n\nWhy include this: Balance. Not just "everything was wrong" but "some threads led here.”, “–”, [“H2] Related Links\n\n- Article 1 (20-Minute Blog deep dive)\n- Article 3 (lessons synthesis)\n- Current /tools/ comparisons (what replaced those 2023 tools)\n- /value/ pages (cost evolution)\n\n—\n\n## Article 3: Lessons from the Vibe Coding Trenches (2023\u20132026)\n\nURL: /posts/vibe-coding-trenches-lessons-2023-2026/ \nPurpose: Synthesis piece. Extract timeless lessons from specific experiences. \nNo redirects (this is new content, not replacing old URLs)\n\n—\n\n### [H1] Title Options\n\n- Three Years in the Vibe Coding Trenches: What I Learned\n- From 20-Minute Blog to Security Analysis: An AI Practitioner Evolves\n- Vibe Coding Grew Up: Lessons from 2023’s Naivety to 2026’s Rigour\n\n—\n\n### [H2] What "Vibe Coding”, “Meant in 2023\n\nDefine the term as you understood it then:\n\n- [ ] Describe the workflow: prompt \u2192 generate \u2192 light edit \u2192 publish\n- [ ] What felt revolutionary about it?\n- [ ] What problems did it seem to solve?\n- [ ] What was the vibe? (Optimism? Speed? Novelty?)\n\nWhy define historically: Establishes you’re not projecting current understanding backward.\n\n—\n\n### [H2] Four Things I Got Wrong\n\nFor each, use this structure:\n\n#### 1. [Mistake Name]\n\nWhat I thought in 2023: [belief]\n\nWhat happened: [consequence/reality]\n\nWhat I now understand: [revised belief]\n\nSpecific example from your work: [concrete incident]\n\nSuggested mistakes to consider:\n- Hallucination as creative feature vs. fundamental alignment problem\n- Speed as the metric vs. clarity as the bottleneck \n- Wrapper tools as sustainable vs. absorbed by foundation models\n- Prompt engineering as skill vs. asking clearly\n\n—\n\n### [H2] What Actually Worked\n\nCounterbalance to the mistakes:\n\n- [ ] Natural language as interface (still valid)\n- [ ] Rapid prototyping (still valid, but with verification)\n- [ ] AI as thought partner (still valid, but with human convergence)\n\nFor each:\n- What was the 2023 version?\n- What is the 2026 version?\n- What changed in the implementation?\n\n—\n\n### [H2] The Evolution: 2023 vs 2026\n\nFill in comparison table:\n\n| Dimension | 2023 | 2026 | The Shift |\n|———–|——|——|———–|\n| Primary question | [ ] | [ ] | [ ] |\n| Time to publish | [ ] | [ ] | [ ] |\n| Verification step | [ ] | [ ] | [ ] |\n| Security consideration | [ ] | [ ] | [ ] |\n| Source requirements | [ ] | [ ] | [ ] |\n| Hallucination handling | [ ] | [ ] | [ ] |\n| Tool stack | [ ] | [ ] | [ ] |\n\nWhy this table: Visual summary of the transformation. Reference point for readers.\n\n—\n\n### [H2] How I’d Do It Today: The 2026 Workflow\n\nDetailed workflow breakdown:\n\n| Phase | Time | Activity | AI Role | Human Role |\n|——-|——|———-|———|————|\n| Research | [ ] min | [ ] | [ ] | [ ] |\n| Verification | [ ] min | [ ] | [ ] | [ ] |\n| Structure | [ ] min | [ ] | [ ] | [ ] |\n| Drafting | [ ] min | [ ] | [ ] | [ ] |\n| Review | [ ] min | [ ] | [ ] | [ ] |\n| Security check | [ ] min | [ ] | [ ] | [ ] |\n| Publishing | [ ] min | [ ] | [ ] | [ ] |\n| Total | [ ] hours | | | |\n\nKey insight to articulate: AI doesn’t make you write faster; it makes you [what?]. The time saved on [task] gets reinvested in [task].\n\n—\n\n### [H2] On Speed (Addressing the 20-Minute Promise)\n\nDirectly confront the original premise:\n\n- [ ] Why 20\u219233 minutes was the wrong optimization\n- [ ] What speed actually matters (reader time vs. writer time)\n- [ ] Where speed helps vs. where it hurts\n- [ ] The 2026 definition of "fast”, “first draft? verified draft? published?)\n\n—\n\n### [H2] Vibe Coding Didn’t Die; It Grew Up\n\nSynthesis argument:\n\n- [ ] The vibes are still there (natural language, rapid iteration, AI assistance)\n- [ ] What’s different: verification, security, source grounding\n- [ ] Analogy:”, “We’re still vibe coding, just wearing safety equipment now”, “–”, [“H2] Lessons for the Next Three Years\n\nForward-looking section:\n\n- [ ] Verification beats generation (why this will matter more)\n- [ ] Safety is architectural (not a feature you add later)\n- [ ] The wrapper problem persists (how to build durable value)\n- [ ] Speed matters differently (reader time vs. writer time)\n\nFor each:\n- Why this lesson matters\n- What you expect to see (2026\u20132029)\n- How you’re applying it now\n\n—\n\n### [H2] Related Links\n\n- Article 1 and 2 (the specific stories)\n- /risks/openclaw/privileged-access/ (security realization)\n- /verify/ methodology (verification practice)\n- /implement/ guides (current workflows)\n- /compare/ (tool evolution)\n\n—\n\n## Meta: Research & Recovery Instructions\n\n### Before Writing: Recovery Checklist\n\nFor each of the 4 lost posts:\n\n- [ ] Wayback Machine: https://web.archive.org/web/*/aihackers.net/2023/04/04/[post-slug]/\n- [ ] Google Cache: Search cache:aihackers.net/2023/04/04/[post-slug]/\n- [ ] Browser history: Check your own browser history for the URL\n- [ ] Email archives: Search email for the post title or URL\n- [ ] Social media: Search your Twitter/Reddit/HN for links to the posts\n- [ ] Old backups: Check any WordPress/database exports from 2023\n\nFor each source, document:\n- What you found\n- Recovery percentage (full text / partial / memory only)\n- URL of recovered content\n- Date of recovery attempt\n\n### If Fully Recovered:\n\n1. Quote original in blockquote\n2. Annotate with 2026 commentary\n3. Note what you got right/wrong\n4. Archive the recovered version (see AGENT TODO)\n\n### If Partially Recovered:\n\n1. Quote what you have\n2. Reconstruct remainder from memory\n3. Clearly label reconstructed sections\n4. Note confidence level\n\n### If Memory Only:\n\n1. Write as "from memory" reconstruction\n2. Be explicit about uncertainty\n3. Focus on principles over specifics\n4. Use the "archaeological”, “framing\n\n—\n\n## Publishing Sequence\n\nRecommended order:\n\n1. Article 1 first (the 20-Minute Blog) - establishes origin story\n2. Article 3 second (synthesis) - shows you’re serious about lessons\n3. Article 2 third (other lost posts) - demonstrates breadth\n\nWhy this order: Lead with the strongest specific story, follow with synthesis to show depth, finish with breadth to show scope.\n\n—\n\n## AGENT TODO: Post-Publication Archive Task\n\nStatus: \u23f0 PENDING - To be completed after author finalizes and publishes\n\nTask: Archive final published articles via archive.org\n\nWhen to do this:\n- [ ] After author has completed their rewrite\n- [ ] After all 3 articles are published\n- [ ] Before sharing widely on social media\n\nSteps for agent:\n\n1. Verify all 3 articles are live:\n - [ ] /posts/20-minute-blog-archaeological-update/\n - [ ] /posts/vibe-coding-april-2023-lost-posts/\n - [ ] /posts/vibe-coding-trenches-lessons-2023-2026/\n\n2. Archive each via archive.org:\n bash\n curl -s \"https://web.archive.org/save/https://aihackers.net/posts/20-minute-blog-archaeological-update/\"\n curl -s \"https://web.archive.org/save/https://aihackers.net/posts/vibe-coding-april-2023-lost-posts/\"\n curl -s \"https://web.archive.org/save/https://aihackers.net/posts/vibe-coding-trenches-lessons-2023-2026/\"\n \n\n3. Record archive URLs in:\n - Article frontmatter: archived_url:\n - Source inventory file: /verification/sources-archive-2026.md\n\n4. Update this blueprint: Mark AGENT TODO as \u2705 COMPLETED\n\nWhy archive: Establishes permanence. Protects against future migrations. Shows commitment to source verification (consistent with site methodology).\n\n—\n\n## Voice & Tone Reminders\n\nThroughout all three articles:\n\n- Use "I" not "we”, “personal experience, not generic advice)\n- Include specific numbers (33 minutes, 78 upvotes, $20 burned)\n- Admit uncertainty (”, “I don’t remember exactly, but…”, “Show work (don’t just claim evolution, demonstrate it)\n- Link generously (to your own content, to sources, to archives)\n- Timestamp everything (April 2023, February 2026)\n\nAvoid:\n- Generic "AI is changing everything”, “statements\n- Humble-bragging about being early\n- Pretending you knew then what you know now\n- Wall-of-text without tables/lists/visual breaks\n\n—\n\n## Quick Reference: What Makes This Content Non-Average\n\nCheck each article against:\n\n- [ ] Contains specific version numbers, dates, or pricing\n- [ ] Includes a real incident/failure (not hypothetical)\n- [ ] Has concrete workflow/numbers (not "best practices”, [“Shows what broke and why\n- [ ] Admits what’s still unsolved\n- [ ] Compares specific time periods (2023 vs 2026)\n- [ ] Would help someone debugging at 2am\n- [ ] Could only be written by someone who lived it\n\n—\n\nBlueprint created: 2026-02-04 \nFor: Author’s personal use in writing retrospective series \nStatus: FRAME - fill in your own prose”]]]]]]]