Living Database / LENS — at1 lens

A .lensfile holds a dataset as typed compressed columns, a Merkle root over every section, and an English self-description — it is the data, the report that reads it, and the proof it's all correct, in one file with no database and no server. Reports run on the compressed columns: a sum touches each delta-chunk once, a category mix each RLE run, a log search each grammar rule — so the work scales with the structure in your data, not its size. Every number is emitted by a named program and signed by a certificate.

Use it

# build a .lens container from a CSV (typed compressed columns + a Merkle root)
at1 lens build fleet.csv -o fleet.lens

# run the dashboard programs on the compressed columns (Tier-3) + per-number certificates
at1 lens report fleet.lens --search retry
#   Summary: latency sum=8.9M avg=447  [cert 08e6.../80ef...]
#     628 compressed ops vs ~20000 row ops (31.8x less work)

at1 lens verify fleet.lens             # Merkle-verify the container (refuses on tamper)

What it gives you

  • Tier-3 execution — compute on the rule, not the data: a sum reads chunk totals, a mix reads run lengths, a search walks the grammar. On the reference fleet, a numeric summary is about 31.8× less work than a row scan.
  • Certified numbers — every value carries a program hash, output hash and container root; re-run and the certificate is identical.
  • Refuse-on-tamper — flip one byte and verify refuses; the container won't load.

Honest scope

This is the CLI reference VM (M1): text reports, one file, local. Wins scale with the structure in your data; speed is stated as durable ops-ratios, with wall-clock only as same-substrate. “Secure” here means integrity + reproducibility, not confidentiality. Browser host, live ingest, and the compact self-carrying interpreter are the next milestones, not claimed here.

See the marketing overview at LENS, and zero-scan GenQuery for query-on-the-generator.

← CLI reference