The Open Hyperdocument System (OHS): A Living Knowledge Base Framework Introduced in 1998 by Douglas Engelbart and His Colleagues


The Open Hyperdocument System: A Blueprint for Living Knowledge

Following Up on Store, Relate, Trust, Retrieve

By Jean Louis

Hello friend,

In my previous article, I introduced four simple principles for building a knowledge base that actually works: Store, Relate, Trust, Retrieve. Today, I want to show you where those principles came from — and why they matter more than ever.

The principles are not new. They were not invented by me, and they were certainly not invented by the LLM-Wiki hype train. They were written down in 1998 by Douglas Engelbart and his colleagues in a document called the Technology Template Project — OHS Framework.

I have been building my Dynamic Knowledge Repository for 23 years. I have 95,211 hyperdocuments, 245,377 people records, and 412,048 embeddings. And every time I look at Engelbart’s framework, I realize: he already described what I was trying to build.

Let me walk you through it.

What Is the Open Hyperdocument System?

Engelbart defined the OHS as a “hierarchy of characteristics” for designing and evaluating knowledge tools. Two words matter most:

“Hyperdocument” means flexible linkages to any object in any multimedia file. Not just text. Not just markdown. Anything.

“Open” means vendor-independent access across work groups, platforms, and applications. No lock-in. No proprietary graveyards.

The framework ranges from low-level content elements all the way up to complete end-user systems. It is not a product. It is a template — a living structure that can be added to, refined, and discussed by a community.

The Four Pillars in Engelbart’s Own Words

Let me show you how my four principles map directly to Engelbart’s framework.

1. Store — With Integrity

Engelbart called for elementary objects — basic content packets of an arbitrary, user-extensible nature. Text, graphics, equations, tables, spreadsheets, images, video, sound, code elements. Anything you can imagine storing.

He required Object ID-Time Stamps — every creation or modification automatically gets a stamp with date, time, and user identification. No exceptions.

He demanded Personal Signature Encryption — users can affix cryptographic signatures to documents or specific objects. You can verify that nothing has been altered since it was signed.

What this means for you: When you store something, store it with integrity. Know who created it, when, and whether it has been tampered with. Your database should enforce this automatically.

2. Relate — With Precision

Engelbart called for explicitly structured documents — objects arranged in hierarchies or more complex cross-linked structures. Compound objects can be explicitly addressed.

He demanded back-links — information about other objects pointing to a specific object must be available. If A links to B, B must know that A exists.

He required shared documents — user-defined knowledge packages that are shareable across the entire global domain, subject to access provisions.

What this means for you: Do not settle for simple [[wikilinks]] that can break. Build typed relationships. Ensure every link is bidirectional. Use foreign keys. Your knowledge network should be a web, not a pile of threads.

3. Trust — With Provenance

Engelbart called for access control down to the object level — based on individual identity or organizational role. Read, write, or even knowledge about existence can be restricted.

He required a Journal system — an integrated library where every document gets a unique, permanent catalog number. Access is guaranteed. Back-links let you see what cites the document. Supersession notifications tell you when something has been replaced.

He specified External Document Control (XDoc) — even documents outside the system can be cataloged with back-links, so you know what is being said about them.

What this means for you: Trust is not optional. You must know where knowledge came from, who can see it, and whether it is still current. Your system must enforce this, not hope for it.

4. Retrieve — With Speed and Precision

Engelbart called for global, human-understandable object addresses — every object that someone might need to cite should have an unambiguous address, readable and interpretable by humans.

He demanded precision browsing — jumping directly to any object in any file with on-the-fly custom views. Not “load and scroll” — precision.

He required view control — objects can be displayed differently depending on user control, with filtering, truncation, or algorithmic views that show structure and content in new ways.

What this means for you: Retrieval is not grep. It is not guessing. It is deterministic, multi-dimensional, and fast. You should be able to find what you need by any property — name, content, metadata, relationships, embeddings, dates, people, projects.

The Paradigm Shift Engelbart Described

Look at the table in section 2d1e1 of the framework. Engelbart lists what we are moving from and what we are moving to:

From To
Tool-centric system Document-centric system
Function-oriented tool system Integrated end-to-end knowledge management environment
Authoring and publishing Developing, integrating and applying knowledge online
Isolated passive libraries Active “living” libraries integrated within work processes
One user class: “easy to use” Many classes of user: pedestrian to high performance
File level addressability Object-level addressability
“Load & scroll browsing” Precision browsing — jumping directly to any object
Windows tied to a file and application Windows as portals into a file repository
Each application has unique file design Applications use common file design
“Interoperability” means font styles retained Means my structured hyperknowledge interoperates in your environment

This was written in 1998. It is still ahead of most systems today.

The LLM-Wiki pattern? It is stuck in the left column. File-level addressability. One user class. Load and scroll. Tool-centric.

Engelbart was pointing to the right column. That is where a real knowledge base lives.

What This Means for You

You do not need to build everything from scratch. Engelbart gave us the blueprint. The OHS Framework specifies:

My Hyperscope implements most of these. 113 elementary object types. 76 columns of metadata. Foreign keys for relationships. Version control triggers. GPG signatures. Full-text and embedding indexes. SQL queries for precision retrieval.

You can do the same. Start with a database. Define your object types. Add foreign keys. Track versions. Set permissions. Build indexes. Use LLMs as tools, not as the engine.

The CODIAK Connection

Engelbart’s CODIAK framework — Concurrent Development, Integration, and Application of Knowledge — describes how organizations can achieve high performance. The OHS is the enabling infrastructure.

The four pillars support CODIAK directly:

Without all four, the CODIAK process collapses. Without all four, you are not augmenting human intellect. You are just moving files around.

My Invitation to You

Read the OHS Framework. It is short. It is clear. It is online at the Doug Engelbart Institute.

Then look at your own knowledge base. Ask yourself:

If you answer “no” to any of these, you have work to do. But the blueprint is there. Engelbart wrote it for you.

In my next article, I will show you how I implemented these principles in Hyperscope — the tables, the relationships, the version control, the permissions, the search. Not as theory, but as running code.

Until then, keep your hands on the wheel. The LLM is a tool. You are the curator.

And Engelbart is still ahead of us.

References

Engelbart, D. C., Lehtman, H., & Engelbart, C. (1998). Technology Template Project - OHS Framework. Doug Engelbart Institute.
https://www.dougengelbart.org/content/view/110/460/

Engelbart, D. C. (1992). The CODIAK Process Cluster: Best Strategic Application Candidate. Doug Engelbart Institute.
https://www.dougengelbart.org/content/view/116/

Engelbart, D. C. (1962). Augmenting Human Intellect: A Conceptual Framework. Stanford Research Institute.

The future is not markdown files. The future is living hyperdocuments with integrity, precision, provenance, and speed. Engelbart showed us the way. Now we build. 🧙🐘