DocEng 2026

Computational creation and processing of documents as structured, multifaceted information objects — text, images, equations, tables, diagrams, and musical scores. Not "documents" as in files; documents as in typed, transformable, queryable structure.

This site is an org-mode source tree with six independent projections off a single parser (search index, feed, sitemap, header lint, verdict annotations, provenance map). That is a document-engineering architecture, argued in exactly DocEng's terms: one structured representation, many derived artifacts, and the parser — not the renderer — as the contract.

Specific overlaps to look for in the program:

• Round-tripping and lossy transformation between markup representations
• Provenance and change tracking inside documents (drawers/annotations here)
• Tables, equations, and diagrams as first-class structure rather than images
• The sustainability theme — incremental re-export and cache invalidation is the same problem this repo's org-publish timestamp cache solves badly

• [ ] General and program chairs — not on the landing page
• [ ] Call for Papers deadlines (submission, notification, camera-ready)
• [ ] Accepted-paper list and program (early August)
• [ ] Whether proceedings are open-access this year

Documents-as-data here

pocket-es (index built off the org parser) · spec · controlled vocabulary v1.

Transformation and reversibility

reversible pipeline transforms.

Retrieval counterpart

SIGIR 2026.

Symposium site

doceng.org/doceng2026

Series

ACM DL DocEng · doceng.org