Reading memory

A reading memory is the durable, queryable layer of what you have read — built automatically from bookmarks rather than authored as notes, organised as concept pages rather than time-ordered highlights.

Personal wiki

A personal wiki is a private, hyperlinked knowledge base maintained by an individual to record, connect, and re-find ideas they've encountered.

Second brain

The second brain is Tiago Forte's term for an external system that captures, organises, and resurfaces ideas you encounter, so your biological brain doesn't have to.

Zettelkasten

Zettelkasten (German for *slip-box*) is a knowledge-management method using small, atomic, hyperlinked notes — each on a single idea — popularised by sociologist Niklas Luhmann.

Knowledge base

A knowledge base is a structured repository of information used by individuals or organisations to consolidate what they know about a topic, product, customer, or process.

Digital garden

A digital garden is a public, evolving collection of interconnected notes — published as you think, not as a finished work.

Retrieval-augmented generation (RAG)

Retrieval-augmented generation is an AI technique that retrieves relevant documents at query time and feeds them into the language model's context, so the model answers from sources rather than from parametric memory alone.

Semantic search

Semantic search retrieves results based on meaning rather than literal keyword match, typically by computing similarity between vector embeddings of the query and indexed content.

Vector embedding

A vector embedding is a fixed-length array of numbers that represents the meaning of a piece of text (or an image, or audio), produced by a neural network so that semantically similar inputs produce numerically close vectors.

Knowledge graph

A knowledge graph is a structured representation of entities and the relationships between them, typically stored as triples (subject — predicate — object) and queryable as a graph.

Entity extraction

Entity extraction is the task of identifying named entities (people, companies, places, products, concepts) within unstructured text, typically performed by a language model or specialised NLP pipeline.

Topic modeling

Topic modeling is the unsupervised discovery of themes in a corpus of documents, classically via algorithms like LDA and increasingly via embedding-based clustering.

Model Context Protocol (MCP)

The Model Context Protocol (MCP) is an open standard, introduced by Anthropic in November 2024, that defines a uniform way for AI applications to connect to external tools, data sources, and services. It uses JSON-RPC 2.0 messaging so that any compliant AI client can discover and use any compliant server without bespoke integration code.

MCP server

An MCP server is a program that exposes data and capabilities to AI applications over the Model Context Protocol, advertising them as tools (callable functions), resources (readable data such as files or records), and prompts (reusable templates). AI clients connect to it over JSON-RPC and negotiate which of these capabilities are available.

Agent memory

Agent memory is the capacity of an AI agent to store, recall, and reuse information across interactions, typically divided into short-term memory (the working context of a single session) and long-term memory (persistent storage that survives across sessions). Long-term memory is usually held outside the model in databases or vector stores and retrieved when relevant.

Context window

A context window is the maximum amount of text, measured in tokens, that a language model can take into account at once, spanning both the input it is given and the output it generates. Anything outside this window is invisible to the model unless it is retrieved and re-supplied.

AI hallucination

An AI hallucination is output from a generative model that is fluent and plausible but factually incorrect, fabricated, or unsupported by any real source. It arises because language models predict likely text rather than retrieve verified facts.

Source grounding

Source grounding is the practice of tying an AI system's outputs to specific, identifiable source documents, so that each claim can be traced back to and verified against the material it came from. A grounded answer carries citations or references rather than relying on the model's parametric memory alone.

Tool calling

Tool calling (also called function calling) is the mechanism by which a language model invokes external functions, such as a search, a database query, or an API, by emitting a structured request that an application executes and feeds back as a result. It lets a model act beyond text generation by delegating to deterministic code and live data.

PDF extraction

PDF extraction is the process of turning a PDF — a format designed for fixed visual layout, not for data — into structured, searchable text, typically combining optical character recognition (OCR) with layout analysis to recover reading order, headings, tables, and columns.

Bookmark management

Bookmark management is the practice and tooling around saving, organising, and re-finding web pages — distinct from read-later (deferred consumption) and knowledge-management (synthesised understanding).

Read-later

Read-later is the workflow of saving an article during one session and consuming it during a later, dedicated reading session — typically supported by tools that strip ads, sync across devices, and offer offline access.

Highlight

A highlight is a marked passage within a document — selected during reading, retained as a separate object, and typically searchable, citeable, and re-surfaceable.

Passage highlight

A passage highlight is a multi-sentence selection within a document, distinguished from a phrase-level highlight by spanning a complete unit of thought (often a paragraph).

Briefing

A briefing is a concise, prepared summary delivered to someone before a meeting, decision, or action — typically distilling source material into the points relevant to that specific context.

Morning briefing

A morning briefing is a regular, time-of-day-anchored digest — typically delivered at the start of the workday — summarising what's new, what's pending, and what the day's priorities should be.

Client briefing

A client briefing is a per-account digest prepared in advance of a meeting with that client, distilling recent reading, news, and internal notes into the points relevant for the conversation.

Account research

Account research is the practice of accumulating context on a specific customer or prospect — their organisation, market, leadership, recent news, internal stakeholders — to inform sales, customer success, or consulting work.

Web clipper

A web clipper is a browser tool — usually an extension or bookmarklet — that saves the page you are reading into a knowledge tool, capturing its content, title, and source URL in one action.

Note-taking app

A note-taking app is software for capturing personal notes — typically text, sometimes with rich media — optimised for fast capture, organisation, and retrieval.

Bookmarking app

A bookmarking app is software for saving web URLs (typically with metadata: title, tags, notes) and re-finding them later — distinct from a read-later app, which emphasises deferred consumption.

RSS reader

An RSS reader is software that subscribes to RSS or Atom feeds — structured publication formats produced by blogs, news sites, and podcasts — and presents new entries to the reader in chronological or relevance-ranked order.

Knowledge tool

A knowledge tool is software designed to help individuals or teams accumulate, organise, retrieve, and act on information — distinct from a productivity tool (which optimises task completion) or a communication tool (which optimises message exchange).

Consulting knowledge management

Consulting knowledge management is the discipline and tooling around capturing, organising, and reusing the institutional knowledge generated during consulting engagements — frameworks, deliverables, client research, and reusable methodology.

Hosted MCP

Hosted MCP refers to a Model Context Protocol server that is run and managed as a remote, cloud-based service over an HTTP transport, rather than installed and run locally by each user. The provider operates the server, handles authentication and updates, and exposes it at a network endpoint that AI clients connect to.

Text-to-speech (TTS)

Text-to-speech (TTS) is the synthesis of spoken audio from written text, performed by a neural model that converts a string of words into a natural-sounding voice.