Personal Knowledge Management
Personal knowledge management (PKM) is a systematic process by which individuals manage and organize their own knowledge and information.[^c1] The field grew out of the recognition, dating to Peter Drucker's identification of "knowledge workers" in the 1960s, that intellectual capital is a decisive resource,[^c2] and it evolved from organizational knowledge-management systems into digital "second brains"—external systems that capture, organize, and retrieve an individual's ideas, notes, and tasks to alleviate cognitive overload.[^c3] The concept of the second brain was popularized most influentially through Tiago Forte's Building a Second Brain methodology and has become a standard metaphor for personal knowledge practice.
The discipline is organized around methods and tools. The Zettelkasten method, invented by the German sociologist Niklas Luhmann in the 1950s, structures notes as atomic, interconnected cards and underpins much of modern note-taking practice.[^c4] The PARA method organizes information by actionability into Projects, Areas, Resources, and Archives.[^c5] The software ecosystem spans local-first Markdown applications such as Obsidian, which stores notes as plain files the user controls,[^c6] cloud workspaces such as Notion, outliners such as Roam Research and Logseq, the legacy tool Evernote, privacy-focused alternatives such as Anytype, and reading-to-knowledge pipelines such as Readwise, which exports highlights into other tools.[^c15] By 2026 the tool landscape had divided into two camps in how people build second-brain tools—connected databases with AI summaries that live in the cloud, versus local plain files under the user's full control—and storage architecture became the axis along which trust, portability, and exit strategy are judged.[^c10] Local-first systems illustrate the spectrum directly: one stores notes as plain Markdown files on the user's disk, one as encrypted objects that sync peer-to-peer, and one in a polished cloud the user never directly touches.[^c11]
The most recent development is the integration of artificial intelligence into the whole workflow. This includes AI-native productivity agents such as Tencent's WorkBuddy, which turns a natural-language prompt into completed deliverables,[^c14] agent-memory frameworks that automatically collect context from email, code repositories, and news feeds into a local wiki,[^c12] persistent agent memory systems such as GBrain, which turns Markdown notes into a self-wiring knowledge graph,[^c13] and systems in which a large language model continuously compiles a user's notes into a structured, interlinked wiki; its originator argues that building a personal knowledge base with a language model is more valuable than using it to write code.[^c8]
The history of the field reaches back to pre-digital practices such as the commonplace book and the card index, and forward through Vannevar Bush's vision of a personal knowledge machine—an idea that goes back 80 years,[^c7] Douglas Engelbart's demonstration of networked computing, and the contemporary tool boom. In 2026 that vision re-entered the field directly: the AI-maintained wiki was framed as the long-delayed realization of Bush's Memex, the machine that had been stuck for eighty years suddenly starting to turn.[^c9] Ongoing debates concern whether tools or methods matter more, whether collecting information substitutes for creating knowledge, the risks of over-engineering systems, and the implications of privacy and artificial intelligence for how individuals manage knowledge.