Why build a website for a research paper? Most papers exist as static PDFs. You download them, read them (or skim the abstract and figures), cite them, and move on. The PDF is a snapshot in time — frozen at submission. FDRP is different. It is a living system that evolves with every production run, every correction, every newly catalogued method. A static PDF cannot show the dashboard updating in real-time, cannot let you explore 76 vector figures at full resolution, cannot give you downloadable data to verify our claims independently. So we built a companion.
The site follows a three-tier access model. The first tier is fully public: methodology descriptions, key metrics, the timeline, this blog, and the roadmap. Anyone can see how FDRP works, how it has evolved, and where it is going. The second tier is gated for paper readers: raw data exports, the full technical report, and supplementary materials. Credentials are published in the paper footnote. The third tier is private: operational details, internal dashboards, and security-sensitive findings that should never be public. This mirrors FDRP's own progressive disclosure architecture — information is revealed at the resolution appropriate to each audience.
The most visible feature of this site is something most research projects would hide: a retraction notice. We originally published an 85.7% convergence rate. Cross-model verification — Claude Opus, GPT-5.4, and Gemini 3.1 working independently — caught an arithmetic error. Double-counted findings from overlapping expert domains inflated the number. The correct rate is 75.0%. We retracted publicly, updated the paper, and recorded the correction in git. In traditional publishing, retractions carry stigma. In FDRP, self-correction is the system working as designed. We display it prominently because it validates the methodology.
What will you find here? The paper itself, downloadable as PDF with full version history. An interactive evolution dashboard tracking 104 methods across 6 pages. Eight case studies from particle physics to cybersecurity, each with structured findings and expert analysis. Downloadable data exports so you can verify every claim we make. A public roadmap showing what we are working on and what we are exploring. And this blog, where we will write long-form notes on methodology decisions, corrections, and things we got wrong.
FDRP treats every planning decision as a manufactured artifact — traceable through its complete lifecycle, from initial proposal to final verification. This website treats every published claim the same way: traceable, verifiable, retractable. If we publish a number, you should be able to find the query that produced it. If we retract a number, you should see the retraction before you see the original claim. That is the point.