Tagged databases
- 2 lists of 30 test names from the same Claude model share a name in up to 98% of pairs. Faker does it in 1.29%
- My README said H2 cannot reproduce my Postgres numbers. Over TCP it landed next to local Postgres
- The index I said would buy nothing: 2.8 ms against 8.3
- Page 2 repeated a row from page 1 and nothing was broken
- A uuid primary key: 172 seconds against 59