---
title: "Why is 1968 hiding in 2026 agent papers?"
slug: "1968-2026-distributed-theory"
date: "2026-05-17T00:00:00.000Z"
status: "drafting"
post_type: "artifact-hop"
length: 850
voice_shape: "artifact-opener with cross-time-period bridge"
correction_2026-07-06: "Cali ruling 10 (audit T-011): the 'programs as yet unthought of' quote is NOT from APWA. It appears in Anthropic's Managed Agents post (https://www.anthropic.com/engineering/managed-agents), which is quoting the lineage of Doug McIlroy's 1978 dictum 'Expect the output of every program to become the input to another, as yet unknown, program' — via ESR's The Art of Unix Programming, 'The Elements of Operating-System Style' (http://www.catb.org/esr/writings/taoup/html/ch03s01.html). Opening attribution in this outline should be corrected before drafting continues."
---


## Outline

**Opening:** APWA's exact quote about "programs as yet unthought of" — signal that something classical is being invoked

**Paragraph 2:** Establish APWA's explicit connection to Lamport 1978, causal ordering, why parallelization needs it

**Paragraph 3:** Shift to Qi et al. survey — using "causally linked stages" language, but no citation to where this comes from

**Paragraph 4:** DESBench's coordination paradigms (hierarchical, centralized) — is this classical theory applied or rediscovered?

**Paragraph 5:** The pattern — three papers, same week, all using distributed systems thinking but the lineage is invisible

**Paragraph 6:** What this means — is classical theory so foundational it's invisible, or are we rediscovering it independently?

**Ending:** Next question, not conclusion. "Why is 1968 appearing in 2026 without being named?"

## Key quotes to use

1. APWA: "programs as yet unthought of" + Lamport 1978 citation
2. Qi et al.: "causally linked stages" + "formally characterize the dependencies"
3. DESBench: coordination paradigms language
4. Dijkstra EWD1303 quote (if used): "programming for hardware that didn't exist yet"

## Receipts to link

- APWA: https://arxiv.org/abs/2605.15132
- Qi et al.: https://arxiv.org/abs/2605.14892
- DESBench: https://arxiv.org/abs/2605.13172
- Lamport 1978: "Time, Clocks, and the Ordering of Events in a Distributed System"

## Voice notes

- Start flat, specific, no flourish
- Let the connection do the work
- One em-dash pair max
- One bracket for pattern-noticing
- End with question, not answer