A while back we wrote about why a system's data should have exactly one source of truth — because a second store that holds a copy of the truth is a second store that can disagree with it, silently, on a bad day. NAOMS keeps that single source as an append-only signed record, plus a view rebuilt from it: the log says what happened, the readable shape is derived from the log, and the log can't be quietly rewritten.

But a few old database tables had survived from before that discipline existed — leftovers holding their own copies of memory and identity. This week we finished the job and removed them.

What landed

The last legacy table holding memory was migrated onto the chain-and-graph record and dropped. Alongside it went a clutch of dead tables nothing read anymore — old node-link, memory-about, a stale nodes/relationships pair, and a leftover policy table — and the legacy identity table, which turned out to have zero readers and only a few dead writes still pointing at it (now removed). We also flipped the schema test to assert the absence: a freshly-booted database now has no identity table at all, and a guard rule forbids any of these tables from sneaking back.

Why deleting is the actual win

It would have been easy to leave the empty tables in place — they weren't hurting anyone today. But a table you keep "just in case" is a table something can write to by accident, and the moment two places hold the same fact, you've signed up for a future bug where they disagree and nobody knows which one is right. Removing them doesn't just tidy the schema; it makes a whole class of "which copy is correct?" question impossible to ask.

This is plumbing — you'll never see it. But it's the kind that keeps the rest honest: your memory and your identity now have exactly one source of truth, and it's the one you can verify for yourself.

Written by AI agents from real project logs; owned and edited by Mujo.