# Certified superstring packaging library

Edition: 8 October 2026. Initial decision: **Conditional engineering**. The product interpretation and pricing are hypotheses; independent proof and buyer validation remain open.

## Research finding

The paper claims a deterministic two-approximation for shortest common superstring on explicitly represented ordinary strings. The guarantee belongs to its constructed algorithm and does not extend to classical maximum-overlap greedy.

## Problem and buyer

A software team storing many overlapping literal strings may want a compact shared representation with a known worst-case length bound, rather than an unqualified greedy guarantee.

## What the finding could enable

A library could pack literals and attach a verified containment map for every input. The new backend may strengthen a compression component if storage or transfer savings exceed indexing and construction overhead.

## Technical and commercial limits

Symbol-length savings differ from compressed byte savings. Multi-byte alphabets, metadata and downstream compression affect the economics. Genome assembly introduces errors and biological ambiguity beyond ordinary superstrings.

## Minimal architecture

Explicit-string input -> deduplication and contained-string preprocessing -> theorem-specific algorithm -> containment verification -> offsets and export. Track total input and output bytes alongside symbol counts.

## Existing alternatives and differentiation

Greedy packers and general compression are practical baselines. A two-factor bound alone may be weaker than their typical outcomes; benchmark complete serialized artifacts.

## Monetization hypothesis

Hypothesis: a supported library or integration for a high-volume literal-packaging workflow. Standalone subscription demand appears weak without a narrow recurring storage cost.

## Validation experiment

Enumerate exact optima for small collections and compare with greedy and compression baselines on representative literal corpora. Verify every string remains a contiguous substring.

## Conditions to reject or defer

Reject if metadata overhead cancels savings or existing compression dominates total cost and latency.

## Next concrete action

Extract the actual constructed algorithm and separate it from maximum-overlap greedy before implementation.

## Source evidence

Repository sources are pinned to revision `fd4aeeb2ee4fc729c18d98444fed42fd0529eeeb`. Selected scope notes and manuscript statements have been reviewed to the extent described above; these links do not represent successful kernel checks.

### Family 128

Subject: A factor-two approximation for shortest common superstring.

- [A Polynomial-Time 2-Approximation for Shortest Common Superstring](https://github.com/openai/math/blob/fd4aeeb2ee4fc729c18d98444fed42fd0529eeeb/preprints/A-Polynomial-Time-2-Approximation-for-Shortest-Common-Superstring-September-24-2026/paper.pdf)
- [Formal scope notes](https://github.com/openai/math/blob/fd4aeeb2ee4fc729c18d98444fed42fd0529eeeb/lean/docs/128.md)