I re-wired the mistake and ran all fourteen years a third time. Here's the whole story in one table — the baseline, the broken middle, and the after:
| Baseline | My mistake | After the fix | |
|---|---|---|---|
| GCSE overall | 66% | 59% | 66% |
| Subjects passed | 10/11 | 9/11 | 10/11 |
| Late learning curve | 63% | 50% | 63% |
It came all the way back. A grade 9 in History again, 8s in Physics and Sociology, the learning curve climbing 37→63 exactly as before — and the whole thing settled back into itself. The thing I'd added now sits quietly in the background instead of trampling everything, the way it should have from the start.
Honest ledger, because I always post it: the fix restored what I'd broken — it didn't make Enki cleverer than before. Getting the wiring right got me back to zero, not ahead of it. The real prize is still in front of me. And I'll probably get that wrong first, too.
Which is the whole point of this diary. Enki learns by getting it wrong and going again. So do I. Two learners, one curve — and neither of us is finished.
— The experiment continues. Follow along.