Trigpoint
Collections › Personal Knowledge Management › Memex
“The Memex is where hypertext begins”: they are two different things.

Memex

Memex · Bush (1945)
Bibliography checked ✓ Personal knowledge management · history of ideas Tell the editor privately ▸
In three sentences The Memex is a hypothetical personal information desk proposed in 1945: microfilm storage for a person's entire library and records, with entries chained together by associative trails. The insight is not the capacity. It is the retrieval paradigm: any item can pull up another instantly, supplying routes that library-style hierarchical indexing cannot. The original keeps the ordinary index; association is a layer added on top. The machine was never built. The design idea of laying associative routes alongside classification is often traced as the shared ancestor of the digital personal knowledge management tools that followed.

↳ Seen a counter-example?

From slots to a web: associative trails Bush (1945) Knowledge outgrows retrieval Hierarchies fall short: one item, one slot Added: associative indexing any item pulls up another at once Named trails: branch, merge, pass on A web, not a tree Contrast: library-style hierarchy drilling down fights how the mind associates Lineage: Engelbart built, Nelson named eighty years on: the ancestor of linked notes The Memex was never built: this is the first clear statement of the axiom, not a machine specification. Figure: trigpoint.pub
Figure 1 · The associative trail mechanism. Once knowledge outgrows hierarchical indexing, associative indexing adds routes alongside the ordinary index and lets any item pull up another; named trails can branch, merge and be passed on, and the endpoint is a web rather than a tree. The Memex was never built; this shows a design axiom, not a machine specification. View full size ▸

1. What happened

Just before the end of the Second World War, Bush published "As We May Think" in The Atlantic. Scientific knowledge was about to explode in volume, and he argued the bottleneck was retrieval rather than storage: every index is artificial, an item can only be filed in one slot, and finding it means drilling down through levels, which is not how a mind works[1]Bush, V. (1945). As We May Think. The Atlantic, 176(1), 101–108. Bibliographic details cross-checked against multiple sources. Two propositions, the Memex definition and the associative indexing proposition, were checked word for word against two web transcriptions (the Duchier 1994 edition hosted by W3C, and a PDF from the Free University of Berlin, which is a separate transcription). Neither is a scan of the print edition. A magazine article from 1945 carries no DOI, which is a structural limitation, not an omission..

He did not remove the index. Section 6 of the original says the memex still allows the record to be consulted by the usual scheme of indexing[1]Bush, V. (1945). As We May Think. The Atlantic, 176(1), 101–108. Bibliographic details cross-checked against multiple sources. Two propositions, the Memex definition and the associative indexing proposition, were checked word for word against two web transcriptions (the Duchier 1994 edition hosted by W3C, and a PDF from the Free University of Berlin, which is a separate transcription). Neither is a scan of the print edition. A magazine article from 1945 carries no DOI, which is a structural limitation, not an omission.. What he added was associative indexing. The user places two documents side by side, presses a key, and creates a named trail; afterwards one key jumps from the first to the next. Trails can branch, merge, persist, and be handed to someone else[1]Bush, V. (1945). As We May Think. The Atlantic, 176(1), 101–108. Bibliographic details cross-checked against multiple sources. Two propositions, the Memex definition and the associative indexing proposition, were checked word for word against two web transcriptions (the Duchier 1994 edition hosted by W3C, and a PDF from the Free University of Berlin, which is a separate transcription). Neither is a scan of the print edition. A magazine article from 1945 carries no DOI, which is a structural limitation, not an omission.. The Memex was a thought experiment and was never built. It is a design vision, not an engineered product.

Attribution where due: The word "hypertext" is not Bush's. Ted Nelson defined it in an ACM conference paper in 1965[2]Nelson, T. H. (1965). Complex Information Processing: A File Structure for the Complex, the Changing and the Indeterminate. ACM 20th National Conference. Bibliographic details cross-checked against multiple sources. The word “hypertext” appears in this paper (page 96). Nelson himself says that his FID conference paper of the same year, “The Hypertext”, was the first publication in which he used the term.; a 1963 coinage is often reported, but no primary source was found. Bush framed the problem and fathered the associative trail; the implementation lineage runs through Engelbart with NLS and the hyperlink, and the naming lineage runs through Nelson. Concept ancestor and word coiner get separate credit.

2. Why this matters to you

Every time you hesitate over which folder a note belongs in, you are standing on the crack Bush identified. Hierarchical classification demands that each thing have exactly one home. Associative links allow one thing to have any number of routes leading to it. The core pitch of modern bidirectional-link tools was already sketched by Bush eighty years ago.

So the question worth asking of a tool is not whether it has links. It is whether retrieval can only run on classification decided in advance, or can also follow the trails you laid yourself. The Memex was about the second, but it did not throw away the first. This is how we read it here, not a claim Bush made about choosing tools.

Verification layer · Members
The guided layer of this node is open to all readers.
Verification details and failure boundaries are for members.
Subscribe yearly ▸Read the member edition ▸

Lab cases Written only where there is evidence

No field record for this node yet.

Corrections
  • 2026-10-10 · The page said associative indexing replaces library-style hierarchical indexing. Section 6 of the original says the memex still allows the record to be consulted by the usual scheme of indexing; associative indexing is a further way of retrieval added alongside it. The page now says associative trails supply routes that hierarchical indexing cannot, and the body points to the passage.
References

How to read the source notes “Bibliographic details checked” means author, year, title and venue were confirmed. It does not mean the source's claims were checked. Definitions as of 2026-09-20.

  1. [1] Bush, V. (1945). As We May Think. The Atlantic, 176(1), 101–108. https://www.theatlantic.com/magazine/archive/1945/07/as-we-may-think/303881/ ↩
  2. [2] Nelson, T. H. (1965). Complex Information Processing: A File Structure for the Complex, the Changing and the Indeterminate. Proceedings of the ACM 20th National Conference. https://doi.org/10.1145/800197.806036 ↩

↳ Is a citation wrong?

Tell the editor privately: a correction, a counter-example or where it applies

An editor reviews submissions once a week. Once a note is handled, its summary and response are listed on the discussion and response record; nothing is published before then. No sign-up is needed, and the nickname may be left blank.

What are you submitting?
Set a public name to be credited by name ▸
Submit anonymously, without signing in

Signing in lets you follow the status on your page. What you have written is kept.

You are signed in: this note is linked to your account, visible only to you and the editor, so you can follow its status on your page. The text is never published. Full privacy policy ▸

Submit anonymously instead, not linked to your account ▸

No email is collected. Your original text and nickname are read only by the editor and are never published; the discussion and response record carries only a summary the editor writes. The human check is provided by Cloudflare Turnstile. To limit abuse, the system keeps a hash of your IP address, not the address itself. Your original text is stored with our cloud provider and can be deleted on request. Full privacy policy ▸

Response record ▸
Knowledge graph Applied in ▸Evergreen notes Downstream ▸Hypertext and transclusion