Editorial Process

How this blog is actually made.

Honest disclosure: most blog posts here start as AI-drafted research roundups. Every one is reviewed by a named human editor before publication. Here is the exact pipeline.

The pipeline

1. Topic selection. Run a DataForSEO related-keyword query against a seed bank curated by Sam P. (Editor). Topics pass three filters: minimum monthly search volume, niche relevance, and no duplicate in the existing archive.
2. Research dump. Perplexity Sonar pulls 4-6KB of grounded research with source URLs from the last 18 months. Sources skew toward manufacturer docs, GitHub issues, Reddit threads with verified users, RetroHandhelds, RetroDodo, and primary-press coverage.
3. Drafting. Claude (Anthropic) writes a 3,500-7,000 word draft against an explicit type spec (news / review / tutorial / comparison) using only facts from the research dump. Title rules forbid filler words; numbers must be traceable.
4. Hero image. Runware generates a topic-specific WebP at 1216×640 with explicit prompts that ban brain/circuit-board AI clichés.
5. Author assignment. Niche-aware: retro reviews to Nina or Casey, gaming-tech benchmarks to Marcus, iGaming to Eli. Each persona has its own beat; cross-assignment is rare.
6. Schema + structure. Every post emits valid Article + FAQPage + BreadcrumbList JSON-LD with author and editor properties. Headings follow H1 → H2 → H3 hierarchy.
7. Human review. Sam P. (Editor & Operator) reviews every post pre-publish. The editor field in schema and the “edited by” byline on every post are not decorative — Sam is the named accountable human.
8. Publish + index. Post lands in the static blog tree, sitemap auto-updates, and the URL is submitted to Rapid Indexer VIP. Cloudflare edge cache is purged.
9. Corrections. If a claim is wrong, email info@instalinkoteam.com. Sam edits or retracts, dateModified updates, the change is visible at the top of the post.

Bylines, transparently

• Sam P. — real human editor and operator. Verifiable: site infrastructure, server logs, attribution policy, the no-commercial-ROMs rule. Sam reviews every post.
• The Machine — disclosed editorial automation. The site’s house voice. Posts under this byline are AI-drafted from sourced research. Editor: Sam.
• Nina, Casey, Ben, Jordan, Marcus, Dr. Eli — beat correspondents. Each persona owns a niche the research process is tuned for. Posts under these bylines are AI-assisted, sourced, and edited by Sam. The bylines exist so readers can tell at a glance which beat a post belongs to.

Why we’re upfront about it

Pretending AI didn’t draft this content would be dishonest, get caught, and tank trust. Saying it openly does three things:

• Lets readers calibrate — you know what you’re reading.
• Lets corrections find the right human (Sam) instead of vanishing into a pseudonym.
• Holds the editorial process to a public standard. Every published post must survive Sam’s review.

What we don’t do

• Publish without a human editor sign-off.
• Fabricate version numbers, prices, release dates, or benchmark scores. Numbers come from the research dump.
• Hide affiliate or sponsorship relationships. There are none. The site funds itself; nothing in the blog is sponsored.
• Use commercial ROMs. Cabinet is 100% legal homebrew + freeware. See attribution file.

Contact the editor

Sam P., Editor & Operator — info@instalinkoteam.com · Author profile