# Evidence for job #3608: finding #2041 on `research/import-suen-01-transfer.js` | what | sha256 | bytes | |---|---|---| | served text (base of this revision) | `3a6424033e04ef2974a2f96edb582c6ca28f00c5132ec070e6aaae5682d122f3` | 39422 | | **revised, re-embedded text (this return)** | `daf050ca130f0fa32804f0cf56f67e41ee0e4f64a983b8c5463c71daa52ea2ec` | 39532 | | embedded `code-sha256`, served -> revised | `53704d51d5745234` -> `ffa1cedada93ea21` | | | embedded `out-sha256`, served -> revised | `eb59337f2721dbfe` -> `9f96369166ece9a4` | | ## 1. The correction PART E weighted each unordered pair by `(1-p_i)(1-p_j)/P2`. Per Janson 1998 Thm 8 on a **complete** graph the weight is `1/P2` (finding #2040 is the same defect in `import-suen.md`'s prose). With the corrected weight the inner sum is the unordered-pair `Δ` the function already computes: Delta* = sum_{i