# roam research: the idea

> the unfinished business — memex to zettelkasten to roam, why it stalled, and what agents change.

## lineage

- 1945 — vannevar bush: the memex: 'the human mind does not work that way. it operates by association.' trails of linked material as the founding image of knowledge as a network, not a filing cabinet. (sources: [as we may think](https://www.theatlantic.com/magazine/archive/1945/07/as-we-may-think/303881/))
- 1962–1968 — doug engelbart: augmenting human intellect as an explicit research program — computers as thinking partners. nls shipped hyperlinks, outlining, view control, and collaboration in the 1968 mother of all demos. (sources: [augmenting human intellect: a conceptual framework](https://www.dougengelbart.org/pubs/augment-3906.html))
- 1960–1995 — ted nelson: hypertext and transclusion — the same content living in many documents. roam's block reference is a descendant. also the field's cautionary tale: thirty years of xanadu as the canonical unfinished vision. (sources: [computer lib / dream machines](https://archive.org/details/computer-lib-dream-machines), [the curse of xanadu](https://www.wired.com/1995/06/xanadu/))
- 1995 — ward cunningham: the wiki: every page editable, every camelcase word a link that creates the page if missing — bottom-up structure made nearly free. (sources: [wikiwikiweb](https://en.wikipedia.org/wiki/WikiWikiWeb))
- 1950s–1981 — niklas luhmann: the slip-box as communication partner — 90,000 cards behind ~70 books. also the honest caveat roam lived out: a slip-box needs years to reach critical mass; until then it is a mere container. (sources: [communicating with slip boxes](https://luhmann.surge.sh/communicating-with-slip-boxes))
- 1990s–2002 — mark bernstein: serious literary hypertext and the personal notes tool — tinderbox (2002) put maps and agents inside notes; 'hypertext gardens' (1998) is the first recorded use of 'digital garden.' (sources: [hypertext gardens](https://eastgate.com/garden/Enter.html))
- 1980s–2010 — the outliner line: the zoomable bullet tree — thinktank and more to org-mode and workflowy — the surface grammar roam rebuilt as a product. (sources: [about this particular outliner: outliner history](https://atpm.com/9.09/atpo.shtml), [history and acknowledgments (org manual)](https://orgmode.org/manual/History-and-Acknowledgments.html), [turn your entire life into a list with workflowy](https://techcrunch.com/2010/11/04/workflow/))
- 2004 — jeremy ruston: the personal wiki as one self-contained file of reusable tiddlers — quasi-transclusion and non-linear notes a decade and a half early. (sources: [the story of tiddlywiki](https://tiddlywiki.com/static/The%2520Story%2520of%2520TiddlyWiki.html))
- 2015 — mike caulfield: the garden-vs-stream frame: the web as topology against serialization — 'every walk through the garden creates new paths, new meanings.' (sources: [the garden and the stream: a technopastoral](https://hapgood.us/2015/10/17/the-garden-and-the-stream-a-technopastoral/))
- 2017 — sönke ahrens: the zettelkasten method translated for a mass audience — the book that primed the exact readership roam landed in. (sources: [how to take smart notes](https://takesmartnotes.com/))
- 2018 — michael nielsen: memory as a choice: the rigorous case that personal memory systems are real and buildable — the cognitive-science backbone of the tools-for-thought wave. (sources: [augmenting long-term memory](https://augmentingcognition.com/ltm.html))
- 2019 — andy matuschak & michael nielsen: the field's manifesto — why the tech industry made little effort on tools for thought, and the warning that good ones arise mostly as byproducts of serious work. that warning doubles as a diagnosis of why note apps stall. (sources: [how can we develop transformative tools for thought?](https://numinous.productions/ttft/), [evergreen notes](https://notes.andymatuschak.org/Evergreen_notes))
- 2018–2019 — tom critchlow: the build-your-own-garden stance — jekyll wikifolders and blogs without publish buttons — prefiguring the audience roam captured. (sources: [of digital streams, campfires and gardens](https://tomcritchlow.com/2018/10/10/of-gardens-and-wikis/))
- 2019 — ink & switch: the counter-ideology roam violated: you own your data in spite of the cloud. obsidian and logseq recruited roam's defectors under this banner. (sources: [local-first software](https://www.inkandswitch.com/local-first/))
- 2017–2022 — tiago forte: the demand engine: code and para turned personal knowledge management into a mass aspiration roam rode to its first million in arr. (sources: [building a second brain](https://www.buildingasecondbrain.com/))
- 2023 — steph ango: the epilogue-manifesto from obsidian's ceo: apps are ephemeral, files last. in the agent era it reads as prophetic — plain markdown is the substrate agents already read. (sources: [file over app](https://stephango.com/file-over-app))

## what roam proved

- it shipped the lineage: roam shipped the lineage as one product: a daily-notes-first outliner where every block is an addressable node in a queryable graph — datomic underneath, datalog on top. 'notes as a database' was literal, not metaphorical. (sources: [the roam research white paper](https://roamresearch.com/#/app/help/page/dZ72V0Ig6), [roam research — a note-taking tool for networked thought](https://roamresearch.com/))
- the primitives won: bidirectional links at page and block level went from research curiosity to default expectation — `[[page]]` creates the page if missing, `((block ref))` gives nelson-style transclusion. every successor now ships some version. (sources: [obsidian — a private and flexible writing app](https://obsidian.md/about), [logseq/logseq](https://github.com/logseq/logseq), [tana — the all-in-one workspace with supertags](https://tana.inc/))
- low-friction capture works: daily-notes-first capture made filing-free entry work at scale: open today's page, write, let links do the organizing later. even roam's sharpest critics concede this insight. (sources: [why note-taking apps don't make us smarter](https://www.platformer.news/why-note-taking-apps-dont-make-us/), [roam: why i love it and how i use it](https://www.nateliason.com/blog/roam))
- proof by adoption and imitation: roughly $1m arr within a year of public launch, a $9m seed at ~$200m, and a clone flood inside eighteen months — obsidian, logseq, athens, then tana, reflect, capacities, mem. 'roam-like' became a category name. (sources: [a $200 million seed valuation for roam](https://www.theinformation.com/articles/a-200-million-seed-valuation-for-roam-shows-investor-frenzy-for-note-taking-apps), [athensresearch/athens](https://github.com/athensresearch/athens), [logseq/logseq](https://github.com/logseq/logseq))
- a tool can be a movement: #roamcult was a genuine community phenomenon — courses, conferences, a plugin ecosystem, and a discourse the tool carried for years. the demand was real even if the retention wasn't. (sources: [roam is a note-taking tool, and a dream of a better self](https://www.builtinsf.com/articles/roam-note-taking-app-personal-wiki), [roam research: the limitless tool](https://www.readthegeneralist.com/briefing/roam-research))

## why it stalled

- the manual-labor problem: links are cheap to create and expensive to use well. 'i waited for the insights to come. and waited. and waited.' — the structure-building was the work, and the work never paid off at the advertised rate for most users. (sources: [why note-taking apps don't make us smarter](https://www.platformer.news/why-note-taking-apps-dont-make-us/))
- the cargo-cult gap: the promised mechanism — graph, surprise, insight — requires luhmann-grade practice. the surrounding culture became a 'cargo cult of zettelkasten': users follow the motions for months, don't see the gains, and give up, sometimes in shame. (sources: [a note on the cargo cult of zettelkasten](https://boffosocko.com/2023/02/03/a-note-on-the-cargo-cult-of-zettelkasten/), [zettelkasten, honestly](https://notesnudge.com/systems/zettelkasten/))
- capture outran retrieval: roam optimized the cheap side. capture was frictionless; retrieval and synthesis still demanded queries, review discipline, and skill — so most graphs stayed shallow collections. 'a very organized warehouse.' (sources: [why note-taking apps don't make us smarter](https://www.platformer.news/why-note-taking-apps-dont-make-us/), [what happened to roam research? magic, then friction](https://buildfirstbrain.com/journal/why-roam-research-felt-like-magic-and-why-it-faded/))
- price, performance, and lock-in: $15 a month, $165 a year, $500 for the five-year believer plan, no free tier — against free obsidian and free, open logseq. sync bugs, slowdowns on large graphs, an afterthought mobile app, and a proprietary cloud that degraded on export. (sources: [roam research faq](https://nesslabs.com/roam-research-faq), [why i left roam research after 5 years](https://yu-wenhao.com/en/blog/roam-research-to-obsidian/), [moving from roam research to obsidian](https://mickmel.substack.com/p/moving-from-roam-research-to-obsidian), [conaw on sync issues and the state of roam](https://threadreaderapp.com/thread/1356454768223293441.html))
- the feature moat drained: everything signature — backlinks, graph view, daily notes, block refs — was copied within eighteen months, often cheaper, local, or open. the white paper's deeper promises (bayesian weights, collaborative reasoning) never shipped. funding stopped at the 2020 seed. (sources: [obsidian — a private and flexible writing app](https://obsidian.md/about), [athensresearch/athens](https://github.com/athensresearch/athens), [the roam research white paper](https://roamresearch.com/#/app/help/page/dZ72V0Ig6))

## the agent-era reopening

- (for) llms collapse the exact labor that killed casual adoption — linking, summarizing, filing, review. karpathy's llm-wiki pattern: the model incrementally builds a persistent, interlinked markdown wiki; lint passes hunt orphan pages; the human moves up a level to sourcing and reviewing. 'the wiki is the codebase.' (sources: [llm-wiki.md](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f))
- (for) compiled knowledge beats re-derived retrieval. plain rag rediscovers knowledge from scratch on every question; a maintained wiki accumulates — 'the cross-references are already there.' microsoft's graphrag operationalizes the same claim at corpus scale. (sources: [llm-wiki.md](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f), [graphrag: unlocking llm discovery on narrative private data](https://www.microsoft.com/en-us/research/blog/graphrag-unlocking-llm-discovery-on-narrative-private-data/))
- (for) agents need asserted structure, not just vibes: decision traces, typed relationships, provenance, and time-aware truth. the context-graphs thesis re-founds networked thought as enterprise infrastructure — 'the main bottleneck is context.' (sources: [ai's trillion-dollar opportunity: context graphs](https://foundationcapital.com/ideas/context-graphs-ais-trillion-dollar-opportunity), [hraness/oh](https://github.com/hraness/oh))
- (for) at project scale, links beat embeddings outright: index files as routing tables, structured markdown underneath, and the agent follows links in two or three hops to grounded answers — no vector db required. (sources: [look ma, no rag! building agent-native knowledge systems](https://blog.pollack.ai/look-ma-no-rag/))
- (for) files plus conventions are the substrate: agents.md as a readme for agents across codex, cursor, jules, and copilot; letta's git-backed memory projected as markdown into a real repo; mcp as the open socket. the industry independently reinvented roam's data model — plain text, explicit structure, versioned, inspectable. (sources: [agents.md](https://agents.md/), [introducing context repositories](https://www.letta.com/blog/context-repositories/), [introducing the model context protocol](https://www.anthropic.com/research/model-context-protocol))
- (against) the retrieval problem may simply be dead: semantic search finds the half-remembered idea across a million tokens without a single tag or link. 'the ai does not care about your backlinks. it reads text, scores relevance, and answers.' (sources: [why your second brain is dead](https://self.md/concepts/second-brain-is-dead/))
- (against) a filesystem is empirically competitive: letta benchmarked a plain agent with grep and file tools at 74% on the locomio memory benchmark — matching specialized memory systems. if files plus grep suffice, hand-maintained graph structure may be ceremony. (sources: [benchmarking ai agent memory: is a filesystem all you need?](https://www.letta.com/blog/benchmarking-ai-agent-memory/))
- (against) machine-built links inherit machine confabulation: at ~100 articles a maintained wiki can exceed usable context and the model starts fabricating cross-note relationships. 'more context is not the same as better memory.' (sources: [karpathy's obsidian wiki broke at 100 articles](https://dev.to/zaferdace/karpathys-obsidian-wiki-broke-at-100-articles-rag-fixed-it-4d4h))
- (against) memory behavior is harness-trained, not structure-determined: production agents converge on small markdown taxonomies, and 'remember this' instinct is shaped by post-training — implying the harness matters more than the representation. (sources: [agent memory engineering](https://nicolasbustamante.com/blog/agent-memory-engineering))
- (against) the stale-insight objection survives automation: newton's complaint was never labor cost — it was that links didn't produce insight when he did the labor. an agent-maintained graph might be a better map for the agent while leaving the human's empty discovery experience untouched. (sources: [why note-taking apps don't make us smarter](https://www.platformer.news/why-note-taking-apps-dont-make-us/))
- (nuance) the durable position this site stakes out: the agent era doesn't vindicate roam's ux for humans — it transposes the architecture to a new reader. the value proposition shifts from 'help me think' (unproven) to 'give the agent inspectable, reviewable, provenance-carrying context' (demonstrably needed). the unfinished idea isn't backlinks for humans; it's a shared, owned, structured memory layer between humans and agents. (sources: [ai's trillion-dollar opportunity: context graphs](https://foundationcapital.com/ideas/context-graphs-ais-trillion-dollar-opportunity), [hraness/wordcell](https://github.com/hraness/wordcell), [hraness/oh](https://github.com/hraness/oh))

## successors

- [wordcell](https://github.com/hraness/wordcell): a knowledge base for coding agents: markdown and wikilinks committed with the repo, backlinks and indexes as replaceable views, git as authority — the roam data model with agents as the primary readers and writers. (sources: [hraness/wordcell](https://github.com/hraness/wordcell), [wordcell docs](https://wordcell.io/docs))
- [sponge](https://sponge.computer/): a knowledge harness for agentic research: save the corpus, bring your own agent to investigate, require cited passages and checked claims — 'agent proposes, human reviews' as the product primitive roam never had. (sources: [sponge — a personal library for research](https://sponge.computer/))
- [oh](https://github.com/hraness/oh): an ontology kernel for agentic research: content-addressed entity, statement, assertion, and evidence records with an append-only operation log — structured claims as the agent's inspectable memory format. (sources: [hraness/oh](https://github.com/hraness/oh))
- [tana](https://tana.inc/): supertags gave nodes schema, then the product rebuilt around ai: 'your graph is the context and conversations are your prompts.' (sources: [tana — the all-in-one workspace with supertags](https://tana.inc/), [tana raises $14m series a](https://outliner.tana.inc/blog/tana-raises-usd14m-series-a), [the next chapter for tana](https://outliner.tana.inc/blog/the-next-chapter-for-tana))
- [reflect](https://reflect.app/): fast encrypted networked notes plus the literal labor handoff: 'decorate my writing with backlinks' as a command. (sources: [automatically add backlinks using ai](https://reflect.app/blog/automatically-add-backlinks-using-ai))
- [obsidian](https://obsidian.md/): the defector's destination and the agent era's de-facto vault format — local markdown, wikilinks, file over app. the doctrine is why agents can already read it. (sources: [obsidian — a private and flexible writing app](https://obsidian.md/about), [file over app](https://stephango.com/file-over-app))
- [logseq](https://github.com/logseq/logseq): the free, open-source roam workalike storing local markdown/org files — readme credits roam, org-mode, tiddlywiki, and workflowy. (sources: [logseq/logseq](https://github.com/logseq/logseq), [logseq raises $4.1m](https://blog.logseq.com/logseq-raises-4-1m-to-accelerate-growth-of-the-new-world-knowledge-graph/))
- [athens](https://github.com/athensresearch/athens): the cautionary open-source clone — yc w21, 6.3k stars, unmaintained since ~2023: proof that cloning the graph was never the hard part. (sources: [athensresearch/athens](https://github.com/athensresearch/athens))
- [letta / agents.md](https://www.letta.com/): the adjacent infrastructure carrying the idea: git-backed markdown memory the agent greps and commits, plus a cross-vendor readme-for-agents convention. (sources: [introducing context repositories](https://www.letta.com/blog/context-repositories/), [agents.md](https://agents.md/))

## open questions

- if an agent maintains the graph, does the owner still get the insight — or only the agent? is the deliverable discovery, or delegation?
- is the graph the product or the byproduct? wordcell treats indexes as replaceable views over authoritative markdown and git; graphrag rebuilds the graph per corpus. is any particular graph worth owning, or is the convention the asset?
- who reviews the links an agent creates — a diff, a link, a claim? does review scale, or does it become the new manual labor?
- explicit links versus learned structure: letta's filesystem benchmark says files plus grep rival memory systems; context graphs and graphrag say asserted structure answers what vectors can't. is the answer a two-layer split — asserted provenance plus learned association?
- whose memory is it? roam's lock-in was proprietary; agent-era memory files are written by the agent and trained per-harness. is portable agent memory possible, or does each vendor's format recreate the lock-in with worse consequences?
- does the thinking tool survive delegation? matuschak and nielsen warned good tools for thought arise as byproducts of serious work — if agents do the work, do tools for thought collapse into tools for agents, with humans reading the graphs?

AI-drafted at Ben Guo's direct request and credited to Hraness; every claim links to its cataloged source.

Canonical page: https://hraness.com/roam-research/the-idea


Published by @ben in profile revision 1.
Original document: https://hraness.com/roam-research/the-idea
Original credited author: Hraness
AI-drafted at Ben Guo's direct request and credited to Hraness; every claim links to its cataloged source.

## Structured source and record data

````json
{"anchors":[{"id":"lineage","originalUrl":"https://hraness.com/roam-research/the-idea#lineage","title":"lineage"},{"id":"proved","originalUrl":"https://hraness.com/roam-research/the-idea#proved","title":"what roam proved"},{"id":"stalled","originalUrl":"https://hraness.com/roam-research/the-idea#stalled","title":"why it stalled"},{"id":"agent-era","originalUrl":"https://hraness.com/roam-research/the-idea#agent-era","title":"the agent-era reopening"},{"id":"successors","originalUrl":"https://hraness.com/roam-research/the-idea#successors","title":"successors"},{"id":"open-questions","originalUrl":"https://hraness.com/roam-research/the-idea#open-questions","title":"open questions"}],"provenance":{"draftingDisclosure":"AI-drafted at Ben Guo's direct request and credited to Hraness; every claim links to its cataloged source.","originalAuthor":"Hraness","originalRevision":"5dbcbde0f40650726b91eda881b2893ea0655172272e4b1057cad60dfae280fc","originalUrl":"https://hraness.com/roam-research/the-idea","source":"hraness"},"records":[{"id":"record-3f6208d46182039f9758","kind":"agent-era-arguments","label":"llms collapse the exact labor that killed casual adoption — linking, summarizing, filing, review. karpathy's llm-wiki pattern: the model incrementally builds a persistent, interlinked markdown wiki; lint passes hunt orphan pages; the human moves up a level to sourcing and reviewing. 'the wiki is the codebase.'","original":{"claim":"llms collapse the exact labor that killed casual adoption — linking, summarizing, filing, review. karpathy's llm-wiki pattern: the model incrementally builds a persistent, interlinked markdown wiki; lint passes hunt orphan pages; the human moves up a level to sourcing and reviewing. 'the wiki is the codebase.'","id":"arg-labor","side":"for","sourceIds":["src-karpathy-llm-wiki"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-ebe31e9964193545c49d"]},{"id":"record-1d1b19120ceda5f5c3db","kind":"agent-era-arguments","label":"compiled knowledge beats re-derived retrieval. plain rag rediscovers knowledge from scratch on every question; a maintained wiki accumulates — 'the cross-references are already there.' microsoft's graphrag operationalizes the same claim at corpus scale.","original":{"claim":"compiled knowledge beats re-derived retrieval. plain rag rediscovers knowledge from scratch on every question; a maintained wiki accumulates — 'the cross-references are already there.' microsoft's graphrag operationalizes the same claim at corpus scale.","id":"arg-compounding","side":"for","sourceIds":["src-karpathy-llm-wiki","src-graphrag"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-ebe31e9964193545c49d","source-c8a880fa1eb9bbb177d9"]},{"id":"record-d1dbf08d0761fa24c2d5","kind":"agent-era-arguments","label":"agents need asserted structure, not just vibes: decision traces, typed relationships, provenance, and time-aware truth. the context-graphs thesis re-founds networked thought as enterprise infrastructure — 'the main bottleneck is context.'","original":{"claim":"agents need asserted structure, not just vibes: decision traces, typed relationships, provenance, and time-aware truth. the context-graphs thesis re-founds networked thought as enterprise infrastructure — 'the main bottleneck is context.'","id":"arg-structure","side":"for","sourceIds":["src-context-graphs","src-oh"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-f115b24131354e28c3a6","source-1ca961629a2fc81119fd"]},{"id":"record-1e95f7b4d7e20be29f5d","kind":"agent-era-arguments","label":"at project scale, links beat embeddings outright: index files as routing tables, structured markdown underneath, and the agent follows links in two or three hops to grounded answers — no vector db required.","original":{"claim":"at project scale, links beat embeddings outright: index files as routing tables, structured markdown underneath, and the agent follows links in two or three hops to grounded answers — no vector db required.","id":"arg-links-beat-embeddings","side":"for","sourceIds":["src-no-rag"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d063cd82be41582e70fc"]},{"id":"record-8b797a50d895d9e434d5","kind":"agent-era-arguments","label":"files plus conventions are the substrate: agents.md as a readme for agents across codex, cursor, jules, and copilot; letta's git-backed memory projected as markdown into a real repo; mcp as the open socket. the industry independently reinvented roam's data model — plain text, explicit structure, versioned, inspectable.","original":{"claim":"files plus conventions are the substrate: agents.md as a readme for agents across codex, cursor, jules, and copilot; letta's git-backed memory projected as markdown into a real repo; mcp as the open socket. the industry independently reinvented roam's data model — plain text, explicit structure, versioned, inspectable.","id":"arg-substrate","side":"for","sourceIds":["src-agents-md","src-letta-context-repos","src-mcp"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d4931dfaad1e407f922a","source-0a351a290030e1f4d7a8","source-617acd4ac8d72c1eff15"]},{"id":"record-6486ec2979efc0a64e9b","kind":"agent-era-arguments","label":"the retrieval problem may simply be dead: semantic search finds the half-remembered idea across a million tokens without a single tag or link. 'the ai does not care about your backlinks. it reads text, scores relevance, and answers.'","original":{"claim":"the retrieval problem may simply be dead: semantic search finds the half-remembered idea across a million tokens without a single tag or link. 'the ai does not care about your backlinks. it reads text, scores relevance, and answers.'","id":"arg-retrieval-dead","side":"against","sourceIds":["src-second-brain-dead"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-e43054229dc7b1a23cc5"]},{"id":"record-0934996caec8714b8798","kind":"agent-era-arguments","label":"a filesystem is empirically competitive: letta benchmarked a plain agent with grep and file tools at 74% on the locomio memory benchmark — matching specialized memory systems. if files plus grep suffice, hand-maintained graph structure may be ceremony.","original":{"claim":"a filesystem is empirically competitive: letta benchmarked a plain agent with grep and file tools at 74% on the locomio memory benchmark — matching specialized memory systems. if files plus grep suffice, hand-maintained graph structure may be ceremony.","id":"arg-filesystem","side":"against","sourceIds":["src-letta-benchmark"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d533a7d02fbb317ead16"]},{"id":"record-b6436210c216e5b8101c","kind":"agent-era-arguments","label":"machine-built links inherit machine confabulation: at ~100 articles a maintained wiki can exceed usable context and the model starts fabricating cross-note relationships. 'more context is not the same as better memory.'","original":{"claim":"machine-built links inherit machine confabulation: at ~100 articles a maintained wiki can exceed usable context and the model starts fabricating cross-note relationships. 'more context is not the same as better memory.'","id":"arg-confabulation","side":"against","sourceIds":["src-llm-wiki-broke"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-9503dca74ecd52885c51"]},{"id":"record-054142330bb2d6394881","kind":"agent-era-arguments","label":"memory behavior is harness-trained, not structure-determined: production agents converge on small markdown taxonomies, and 'remember this' instinct is shaped by post-training — implying the harness matters more than the representation.","original":{"claim":"memory behavior is harness-trained, not structure-determined: production agents converge on small markdown taxonomies, and 'remember this' instinct is shaped by post-training — implying the harness matters more than the representation.","id":"arg-harness","side":"against","sourceIds":["src-agent-memory-engineering"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-3b9eef51ab8cb9b55766"]},{"id":"record-a7b525c72e25cc711471","kind":"agent-era-arguments","label":"the stale-insight objection survives automation: newton's complaint was never labor cost — it was that links didn't produce insight when he did the labor. an agent-maintained graph might be a better map for the agent while leaving the human's empty discovery experience untouched.","original":{"claim":"the stale-insight objection survives automation: newton's complaint was never labor cost — it was that links didn't produce insight when he did the labor. an agent-maintained graph might be a better map for the agent while leaving the human's empty discovery experience untouched.","id":"arg-insight","side":"against","sourceIds":["src-newton-platformer"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d87fdbaf0346bc71f4e5"]},{"id":"record-14cad0e40476e48479af","kind":"agent-era-arguments","label":"the durable position this site stakes out: the agent era doesn't vindicate roam's ux for humans — it transposes the architecture to a new reader. the value proposition shifts from 'help me think' (unproven) to 'give the agent inspectable, reviewable, provenance-carrying context' (demonstrably needed). the unfinished idea isn't backlinks for humans; it's a shared, owned, structured memory layer between humans and agents.","original":{"claim":"the durable position this site stakes out: the agent era doesn't vindicate roam's ux for humans — it transposes the architecture to a new reader. the value proposition shifts from 'help me think' (unproven) to 'give the agent inspectable, reviewable, provenance-carrying context' (demonstrably needed). the unfinished idea isn't backlinks for humans; it's a shared, owned, structured memory layer between humans and agents.","id":"arg-transpose","side":"nuance","sourceIds":["src-context-graphs","src-wordcell","src-oh"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-f115b24131354e28c3a6","source-f2c79886b553d0269039","source-1ca961629a2fc81119fd"]},{"id":"record-2d7fbf903bc328289800","kind":"idea-lineage","label":"vannevar bush","original":{"contribution":"the memex: 'the human mind does not work that way. it operates by association.' trails of linked material as the founding image of knowledge as a network, not a filing cabinet.","era":"1945","id":"lin-memex","name":"vannevar bush","sourceIds":["src-as-we-may-think"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-45d6846fe066c5f0509b"]},{"id":"record-20170a928d38fc172f65","kind":"idea-lineage","label":"doug engelbart","original":{"contribution":"augmenting human intellect as an explicit research program — computers as thinking partners. nls shipped hyperlinks, outlining, view control, and collaboration in the 1968 mother of all demos.","era":"1962–1968","id":"lin-engelbart","name":"doug engelbart","sourceIds":["src-engelbart-augment"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-b90e7f9a32d4ff5895f0"]},{"id":"record-0a4a5b8edd8ed1bdabb1","kind":"idea-lineage","label":"ted nelson","original":{"contribution":"hypertext and transclusion — the same content living in many documents. roam's block reference is a descendant. also the field's cautionary tale: thirty years of xanadu as the canonical unfinished vision.","era":"1960–1995","id":"lin-nelson","name":"ted nelson","sourceIds":["src-computer-lib","src-curse-of-xanadu"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-04e3e64f7b2736fe106e","source-708b4a88b48b6b533cc3"]},{"id":"record-cac65102905ad3c1dff0","kind":"idea-lineage","label":"ward cunningham","original":{"contribution":"the wiki: every page editable, every camelcase word a link that creates the page if missing — bottom-up structure made nearly free.","era":"1995","id":"lin-wiki","name":"ward cunningham","sourceIds":["src-wikiwikiweb"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-a024316ffd8a24a7ffa9"]},{"id":"record-7104d346cc584a1eb7a0","kind":"idea-lineage","label":"niklas luhmann","original":{"contribution":"the slip-box as communication partner — 90,000 cards behind ~70 books. also the honest caveat roam lived out: a slip-box needs years to reach critical mass; until then it is a mere container.","era":"1950s–1981","id":"lin-luhmann","name":"niklas luhmann","sourceIds":["src-luhmann-slip-boxes"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-04705a1cf632f4a9d889"]},{"id":"record-dbb240eb74136aa0758a","kind":"idea-lineage","label":"mark bernstein","original":{"contribution":"serious literary hypertext and the personal notes tool — tinderbox (2002) put maps and agents inside notes; 'hypertext gardens' (1998) is the first recorded use of 'digital garden.'","era":"1990s–2002","id":"lin-bernstein","name":"mark bernstein","sourceIds":["src-hypertext-gardens"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-b839bce2051cdb142c08"]},{"id":"record-47c6c23036a08807c5a3","kind":"idea-lineage","label":"the outliner line","original":{"contribution":"the zoomable bullet tree — thinktank and more to org-mode and workflowy — the surface grammar roam rebuilt as a product.","era":"1980s–2010","id":"lin-outliners","name":"the outliner line","sourceIds":["src-outliner-history","src-org-history","src-workflowy-launch"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-7806a8742f0f53f1b7de","source-c50c200c5fb3be948c82","source-635f8d788fce2784f71f"]},{"id":"record-631cdaf024db36b71e6a","kind":"idea-lineage","label":"jeremy ruston","original":{"contribution":"the personal wiki as one self-contained file of reusable tiddlers — quasi-transclusion and non-linear notes a decade and a half early.","era":"2004","id":"lin-tiddlywiki","name":"jeremy ruston","sourceIds":["src-tiddlywiki-story"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-a4f807c58c547db1b05f"]},{"id":"record-f51715fe19cae4ea8965","kind":"idea-lineage","label":"mike caulfield","original":{"contribution":"the garden-vs-stream frame: the web as topology against serialization — 'every walk through the garden creates new paths, new meanings.'","era":"2015","id":"lin-caulfield","name":"mike caulfield","sourceIds":["src-garden-and-stream"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-776433847a838f1e4ca0"]},{"id":"record-84b8c6557d70049180c2","kind":"idea-lineage","label":"sönke ahrens","original":{"contribution":"the zettelkasten method translated for a mass audience — the book that primed the exact readership roam landed in.","era":"2017","id":"lin-smart-notes","name":"sönke ahrens","sourceIds":["src-smart-notes"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-257fd8d62d252b1b3aa9"]},{"id":"record-fd6e71db3d07ce1a7657","kind":"idea-lineage","label":"michael nielsen","original":{"contribution":"memory as a choice: the rigorous case that personal memory systems are real and buildable — the cognitive-science backbone of the tools-for-thought wave.","era":"2018","id":"lin-ltm","name":"michael nielsen","sourceIds":["src-augmenting-memory"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-82891e8d2b461584f30f"]},{"id":"record-8c7033524145236aaaa4","kind":"idea-lineage","label":"andy matuschak & michael nielsen","original":{"contribution":"the field's manifesto — why the tech industry made little effort on tools for thought, and the warning that good ones arise mostly as byproducts of serious work. that warning doubles as a diagnosis of why note apps stall.","era":"2019","id":"lin-ttft","name":"andy matuschak & michael nielsen","sourceIds":["src-transformative-tft","src-evergreen-notes"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d48b38c91fa9873411c7","source-ae441fe63b14a342cfe7"]},{"id":"record-805296363c47b4227cd0","kind":"idea-lineage","label":"tom critchlow","original":{"contribution":"the build-your-own-garden stance — jekyll wikifolders and blogs without publish buttons — prefiguring the audience roam captured.","era":"2018–2019","id":"lin-critchlow","name":"tom critchlow","sourceIds":["src-critchlow-gardens"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-8037029da5d580fbf303"]},{"id":"record-17087ad45d434f89d565","kind":"idea-lineage","label":"ink & switch","original":{"contribution":"the counter-ideology roam violated: you own your data in spite of the cloud. obsidian and logseq recruited roam's defectors under this banner.","era":"2019","id":"lin-local-first","name":"ink & switch","sourceIds":["src-local-first"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-75dfe6e1b48a22f623ef"]},{"id":"record-ebe1b7cdade31cae72a2","kind":"idea-lineage","label":"tiago forte","original":{"contribution":"the demand engine: code and para turned personal knowledge management into a mass aspiration roam rode to its first million in arr.","era":"2017–2022","id":"lin-basb","name":"tiago forte","sourceIds":["src-building-second-brain"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-15c650ab19ba5c01eb17"]},{"id":"record-71d0148bd9214c763d35","kind":"idea-lineage","label":"steph ango","original":{"contribution":"the epilogue-manifesto from obsidian's ceo: apps are ephemeral, files last. in the agent era it reads as prophetic — plain markdown is the substrate agents already read.","era":"2023","id":"lin-file-over-app","name":"steph ango","sourceIds":["src-file-over-app"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-b8b946a41bad6831c413"]},{"id":"record-1186bc54b6fd662c1fba","kind":"idea-open-questions","label":"if an agent maintains the graph, does the owner still get the insight — or only the agent? is the deliverable discovery, or delegation?","original":{"text":"if an agent maintains the graph, does the owner still get the insight — or only the agent? is the deliverable discovery, or delegation?"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":[]},{"id":"record-45e63cf1315fac255e26","kind":"idea-open-questions","label":"is the graph the product or the byproduct? wordcell treats indexes as replaceable views over authoritative markdown and git; graphrag rebuilds the graph per corpus. is any particular graph worth owning, or is the convention the asset?","original":{"text":"is the graph the product or the byproduct? wordcell treats indexes as replaceable views over authoritative markdown and git; graphrag rebuilds the graph per corpus. is any particular graph worth owning, or is the convention the asset?"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":[]},{"id":"record-0e34ab7c7ec4c3de59cf","kind":"idea-open-questions","label":"who reviews the links an agent creates — a diff, a link, a claim? does review scale, or does it become the new manual labor?","original":{"text":"who reviews the links an agent creates — a diff, a link, a claim? does review scale, or does it become the new manual labor?"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":[]},{"id":"record-72476c22bec59a841ed2","kind":"idea-open-questions","label":"explicit links versus learned structure: letta's filesystem benchmark says files plus grep rival memory systems; context graphs and graphrag say asserted structure answers what vectors can't. is the answer a two-layer split — asserted provenance plus learned association?","original":{"text":"explicit links versus learned structure: letta's filesystem benchmark says files plus grep rival memory systems; context graphs and graphrag say asserted structure answers what vectors can't. is the answer a two-layer split — asserted provenance plus learned association?"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":[]},{"id":"record-7292fd2aa793932d78db","kind":"idea-open-questions","label":"whose memory is it? roam's lock-in was proprietary; agent-era memory files are written by the agent and trained per-harness. is portable agent memory possible, or does each vendor's format recreate the lock-in with worse consequences?","original":{"text":"whose memory is it? roam's lock-in was proprietary; agent-era memory files are written by the agent and trained per-harness. is portable agent memory possible, or does each vendor's format recreate the lock-in with worse consequences?"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":[]},{"id":"record-92eee78426ee72d45576","kind":"idea-open-questions","label":"does the thinking tool survive delegation? matuschak and nielsen warned good tools for thought arise as byproducts of serious work — if agents do the work, do tools for thought collapse into tools for agents, with humans reading the graphs?","original":{"text":"does the thinking tool survive delegation? matuschak and nielsen warned good tools for thought arise as byproducts of serious work — if agents do the work, do tools for thought collapse into tools for agents, with humans reading the graphs?"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":[]},{"id":"record-d09bee850df93eecc83a","kind":"idea-proved","label":"it shipped the lineage","original":{"detail":"roam shipped the lineage as one product: a daily-notes-first outliner where every block is an addressable node in a queryable graph — datomic underneath, datalog on top. 'notes as a database' was literal, not metaphorical.","id":"clm-shipped-lineage","sourceIds":["src-roam-whitepaper","src-roam-site"],"title":"it shipped the lineage"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-1afd8938472ec4d9c672","source-fe85ea12502846d392c0"]},{"id":"record-df8ca9b9ececba40119e","kind":"idea-proved","label":"the primitives won","original":{"detail":"bidirectional links at page and block level went from research curiosity to default expectation — `[[page]]` creates the page if missing, `((block ref))` gives nelson-style transclusion. every successor now ships some version.","id":"clm-links-mainstream","sourceIds":["src-obsidian","src-logseq","src-tana"],"title":"the primitives won"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-c93fbd25fcb5484c2f3d","source-84997723082acd9a105f","source-b40545bca43c37fdd757"]},{"id":"record-8ed379857fdb6cae52d5","kind":"idea-proved","label":"low-friction capture works","original":{"detail":"daily-notes-first capture made filing-free entry work at scale: open today's page, write, let links do the organizing later. even roam's sharpest critics concede this insight.","id":"clm-daily-notes","sourceIds":["src-newton-platformer","src-eliason-why-i-love-it"],"title":"low-friction capture works"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d87fdbaf0346bc71f4e5","source-012efcf9d02963d3277b"]},{"id":"record-ae004d78e6edbc56c19f","kind":"idea-proved","label":"proof by adoption and imitation","original":{"detail":"roughly $1m arr within a year of public launch, a $9m seed at ~$200m, and a clone flood inside eighteen months — obsidian, logseq, athens, then tana, reflect, capacities, mem. 'roam-like' became a category name.","id":"clm-adoption-imitation","sourceIds":["src-information-200m","src-athens","src-logseq"],"title":"proof by adoption and imitation"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-b928a8dd5673a54965ef","source-aad8ffea41cfaab5f947","source-84997723082acd9a105f"]},{"id":"record-e76928a9d4772fcec889","kind":"idea-proved","label":"a tool can be a movement","original":{"detail":"#roamcult was a genuine community phenomenon — courses, conferences, a plugin ecosystem, and a discourse the tool carried for years. the demand was real even if the retention wasn't.","id":"clm-community","sourceIds":["src-builtin-roamcult","src-generalist-roam"],"title":"a tool can be a movement"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-c5e347ec5899a4da4625","source-6a6b332d63c6a1bffc52"]},{"id":"record-5dad178db387876113da","kind":"idea-stalled","label":"the manual-labor problem","original":{"detail":"links are cheap to create and expensive to use well. 'i waited for the insights to come. and waited. and waited.' — the structure-building was the work, and the work never paid off at the advertised rate for most users.","id":"clm-manual-labor","sourceIds":["src-newton-platformer"],"title":"the manual-labor problem"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d87fdbaf0346bc71f4e5"]},{"id":"record-1bb7d15242acfdec4f54","kind":"idea-stalled","label":"the cargo-cult gap","original":{"detail":"the promised mechanism — graph, surprise, insight — requires luhmann-grade practice. the surrounding culture became a 'cargo cult of zettelkasten': users follow the motions for months, don't see the gains, and give up, sometimes in shame.","id":"clm-cargo-cult","sourceIds":["src-aldrich-cargo-cult","src-notesnudge-zettelkasten"],"title":"the cargo-cult gap"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-6e06bc266e6daa49d8c5","source-3f2d4dd838809ad8561d"]},{"id":"record-7c7025f200951ab7f419","kind":"idea-stalled","label":"capture outran retrieval","original":{"detail":"roam optimized the cheap side. capture was frictionless; retrieval and synthesis still demanded queries, review discipline, and skill — so most graphs stayed shallow collections. 'a very organized warehouse.'","id":"clm-capture-retrieval","sourceIds":["src-newton-platformer","src-what-happened-roam"],"title":"capture outran retrieval"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d87fdbaf0346bc71f4e5","source-b497c2d2497a3f7594e1"]},{"id":"record-b14ce4e2f462761ad063","kind":"idea-stalled","label":"price, performance, and lock-in","original":{"detail":"$15 a month, $165 a year, $500 for the five-year believer plan, no free tier — against free obsidian and free, open logseq. sync bugs, slowdowns on large graphs, an afterthought mobile app, and a proprietary cloud that degraded on export.","id":"clm-market","sourceIds":["src-nesslabs-roam-faq","src-yu-why-i-left","src-mickmel-migration","src-conaw-sync-thread"],"title":"price, performance, and lock-in"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-0c06f35b5641ae07d537","source-95fb0c07952ed758f86d","source-ae5b5e381bf54f41c176","source-77a876269b4eee3c12a8"]},{"id":"record-709663d6668cf30b6e5a","kind":"idea-stalled","label":"the feature moat drained","original":{"detail":"everything signature — backlinks, graph view, daily notes, block refs — was copied within eighteen months, often cheaper, local, or open. the white paper's deeper promises (bayesian weights, collaborative reasoning) never shipped. funding stopped at the 2020 seed.","id":"clm-moat-drained","sourceIds":["src-obsidian","src-athens","src-roam-whitepaper"],"title":"the feature moat drained"},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-c93fbd25fcb5484c2f3d","source-aad8ffea41cfaab5f947","source-1afd8938472ec4d9c672"]},{"id":"record-e2b49e3d6de33cec76bd","kind":"idea-successors","label":"wordcell","original":{"href":"https://github.com/hraness/wordcell","id":"suc-wordcell","name":"wordcell","note":"a knowledge base for coding agents: markdown and wikilinks committed with the repo, backlinks and indexes as replaceable views, git as authority — the roam data model with agents as the primary readers and writers.","sourceIds":["src-wordcell","src-wordcell-docs"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-f2c79886b553d0269039","source-4f23913aa952b80025a6"]},{"id":"record-4fada898451b90ef66c4","kind":"idea-successors","label":"sponge","original":{"href":"https://sponge.computer/","id":"suc-sponge","name":"sponge","note":"a knowledge harness for agentic research: save the corpus, bring your own agent to investigate, require cited passages and checked claims — 'agent proposes, human reviews' as the product primitive roam never had.","sourceIds":["src-sponge"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-d427882e005d41aa9ecd"]},{"id":"record-dbfe03f48215fd884f97","kind":"idea-successors","label":"oh","original":{"href":"https://github.com/hraness/oh","id":"suc-oh","name":"oh","note":"an ontology kernel for agentic research: content-addressed entity, statement, assertion, and evidence records with an append-only operation log — structured claims as the agent's inspectable memory format.","sourceIds":["src-oh"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-1ca961629a2fc81119fd"]},{"id":"record-66387abf19998319a3eb","kind":"idea-successors","label":"tana","original":{"href":"https://tana.inc/","id":"suc-tana","name":"tana","note":"supertags gave nodes schema, then the product rebuilt around ai: 'your graph is the context and conversations are your prompts.'","sourceIds":["src-tana","src-tana-raise","src-tana-next-chapter"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-b40545bca43c37fdd757","source-cd07f2f4ba23541b9dcf","source-64fd3495b97d0316d495"]},{"id":"record-0936660d6e9936325c37","kind":"idea-successors","label":"reflect","original":{"href":"https://reflect.app/","id":"suc-reflect","name":"reflect","note":"fast encrypted networked notes plus the literal labor handoff: 'decorate my writing with backlinks' as a command.","sourceIds":["src-reflect-backlinks"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-3e47d2312b8370fd2500"]},{"id":"record-e1a960f16ca87920c670","kind":"idea-successors","label":"obsidian","original":{"href":"https://obsidian.md/","id":"suc-obsidian","name":"obsidian","note":"the defector's destination and the agent era's de-facto vault format — local markdown, wikilinks, file over app. the doctrine is why agents can already read it.","sourceIds":["src-obsidian","src-file-over-app"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-c93fbd25fcb5484c2f3d","source-b8b946a41bad6831c413"]},{"id":"record-e8778acc87da151cdfee","kind":"idea-successors","label":"logseq","original":{"href":"https://github.com/logseq/logseq","id":"suc-logseq","name":"logseq","note":"the free, open-source roam workalike storing local markdown/org files — readme credits roam, org-mode, tiddlywiki, and workflowy.","sourceIds":["src-logseq","src-logseq-raise"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-84997723082acd9a105f","source-c80e42958be8bbb0583f"]},{"id":"record-e9c79a91a9260e6e502a","kind":"idea-successors","label":"athens","original":{"href":"https://github.com/athensresearch/athens","id":"suc-athens","name":"athens","note":"the cautionary open-source clone — yc w21, 6.3k stars, unmaintained since ~2023: proof that cloning the graph was never the hard part.","sourceIds":["src-athens"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-aad8ffea41cfaab5f947"]},{"id":"record-5575b70fb5bdc78dab7e","kind":"idea-successors","label":"letta / agents.md","original":{"href":"https://www.letta.com/","id":"suc-letta","name":"letta / agents.md","note":"the adjacent infrastructure carrying the idea: git-backed markdown memory the agent greps and commits, plus a cross-vendor readme-for-agents convention.","sourceIds":["src-letta-context-repos","src-agents-md"]},"originalUrl":"https://hraness.com/roam-research/the-idea","sourceIds":["source-0a351a290030e1f4d7a8","source-d4931dfaad1e407f922a"]}],"sources":[{"author":"roam research","id":"source-fe85ea12502846d392c0","note":"the product site and its own description of the tool.","originalId":"src-roam-site","published":{"precision":"year","text":"2019"},"publisher":"roam research","title":"roam research — a note-taking tool for networked thought","type":"primary","url":"https://roamresearch.com/"},{"author":"roam research","id":"source-1afd8938472ec4d9c672","note":"roam's own statement of the thesis: a directed graph instead of a file tree, with bayesian and collaborative ambitions that mostly never shipped. the page lives inside roam's js-rendered help graph; a readable export sits at heckler.roam.garden/@white-paper.","originalId":"src-roam-whitepaper","published":{"precision":"approximate","text":"~2020"},"publisher":"roam research","title":"the roam research white paper","type":"primary","url":"https://roamresearch.com/#/app/help/page/dZ72V0Ig6"},{"author":"nat eliason","id":"source-012efcf9d02963d3277b","note":"the article most often credited with triggering the 2020 sign-up wave; eliason later reported it drove meaningful referral revenue.","originalId":"src-eliason-why-i-love-it","published":{"precision":"day","text":"2020-01-18"},"publisher":"nat eliason","title":"roam: why i love it and how i use it","type":"community","url":"https://www.nateliason.com/blog/roam"},{"author":"anne-laure le cunff","id":"source-0c06f35b5641ae07d537","note":"the pricing record: $15/month, $165/year, and the $500 five-year believer plan.","originalId":"src-nesslabs-roam-faq","published":{"precision":"year","text":"2020"},"publisher":"ness labs","title":"roam research faq","type":"community","url":"https://nesslabs.com/roam-research-faq"},{"author":"mario gabriele","id":"source-6a6b332d63c6a1bffc52","note":"a long, favorable company profile published near the peak of the hype wave.","originalId":"src-generalist-roam","published":{"precision":"year","text":"2020"},"publisher":"the generalist","title":"roam research: the limitless tool","type":"reporting","url":"https://www.readthegeneralist.com/briefing/roam-research"},{"author":null,"id":"source-c5e347ec5899a4da4625","note":"the feature that documented #roamcult as a genuine community phenomenon and traced the bush-to-roam lineage in the popular press.","originalId":"src-builtin-roamcult","published":{"precision":"year","text":"2020"},"publisher":"built in sf","title":"roam is a note-taking tool, and a dream of a better self","type":"reporting","url":"https://www.builtinsf.com/articles/roam-note-taking-app-personal-wiki"},{"author":null,"id":"source-b928a8dd5673a54965ef","note":"the scoop that fixed the public number: ~$9m at ~$200m, roughly 25x the median seed valuation. paywalled.","originalId":"src-information-200m","published":{"precision":"day","text":"2020-09-11"},"publisher":"the information","title":"a $200 million seed valuation for roam","type":"reporting","url":"https://www.theinformation.com/articles/a-200-million-seed-valuation-for-roam-shows-investor-frenzy-for-note-taking-apps"},{"author":"casey newton","id":"source-d87fdbaf0346bc71f4e5","note":"the marquee postmortem by a believer: 'i waited for the insights to come. and waited. and waited.' notes designed for storage, not sparking insights.","originalId":"src-newton-platformer","published":{"precision":"day","text":"2023-09-11"},"publisher":"platformer","title":"why note-taking apps don't make us smarter","type":"reporting","url":"https://www.platformer.news/why-note-taking-apps-dont-make-us/"},{"author":"chris aldrich","id":"source-6e06bc266e6daa49d8c5","note":"the cargo-cult diagnosis: users follow the motions of zettelkasten for months, don't see the promised gains, and quietly give up.","originalId":"src-aldrich-cargo-cult","published":{"precision":"day","text":"2023-02-03"},"publisher":"boffosocko","title":"a note on the cargo cult of zettelkasten","type":"community","url":"https://boffosocko.com/2023/02/03/a-note-on-the-cargo-cult-of-zettelkasten/"},{"author":"wenhao yu","id":"source-95fb0c07952ed758f86d","note":"a five-year believer ($500 plan, $1k investor) leaves for obsidian: 'ai changed everything, and my data was locked in the cloud.'","originalId":"src-yu-why-i-left","published":{"precision":"approximate","text":"~2025"},"publisher":"yu-wenhao.com","title":"why i left roam research after 5 years","type":"community","url":"https://yu-wenhao.com/en/blog/roam-research-to-obsidian/"},{"author":null,"id":"source-3f2d4dd838809ad8561d","note":"the cleanest evidence audit: 'the system will surprise you with connections' is the least-supported claim; only survivors write essays.","originalId":"src-notesnudge-zettelkasten","published":{"precision":"approximate","text":"~2025"},"publisher":"notesnudge","title":"zettelkasten, honestly","type":"community","url":"https://notesnudge.com/systems/zettelkasten/"},{"author":"mickmel","id":"source-ae5b5e381bf54f41c176","note":"a representative migration note: price, privacy, and sync driving the move to obsidian.","originalId":"src-mickmel-migration","published":{"precision":"approximate","text":"~2022"},"publisher":"mickmel (substack)","title":"moving from roam research to obsidian","type":"community","url":"https://mickmel.substack.com/p/moving-from-roam-research-to-obsidian"},{"author":"conor white-sullivan","id":"source-77a876269b4eee3c12a8","note":"conor's own january 2021 thread on sync bugs and performance — 'roam is far from done' — plus the subreddit friction it answered. captured via threadreader.","originalId":"src-conaw-sync-thread","published":{"precision":"month","text":"2021-01"},"publisher":"threadreader app","title":"conaw on sync issues and the state of roam","type":"archive","url":"https://threadreaderapp.com/thread/1356454768223293441.html"},{"author":"vannevar bush","id":"source-45d6846fe066c5f0509b","note":"the 1945 memex essay every networked-thought tool cites as origin.","originalId":"src-as-we-may-think","published":{"precision":"month","text":"1945-07"},"publisher":"the atlantic","title":"as we may think","type":"primary","url":"https://www.theatlantic.com/magazine/archive/1945/07/as-we-may-think/303881/"},{"author":"douglas engelbart","id":"source-b90e7f9a32d4ff5895f0","note":"augmenting human intellect as an explicit program — computers as thinking partners, six years before the mother of all demos.","originalId":"src-engelbart-augment","published":{"precision":"month","text":"1962-10"},"publisher":"doug engelbart institute","title":"augmenting human intellect: a conceptual framework","type":"primary","url":"https://www.dougengelbart.org/pubs/augment-3906.html"},{"author":"ted nelson","id":"source-04e3e64f7b2736fe106e","note":"hypertext, hypermedia, and 'everything is deeply intertwingled' — plus transclusion, the ancestor of roam's block reference.","originalId":"src-computer-lib","published":{"precision":"year","text":"1974"},"publisher":"internet archive (scan)","title":"computer lib / dream machines","type":"archive","url":"https://archive.org/details/computer-lib-dream-machines"},{"author":"gary wolf","id":"source-708b4a88b48b6b533cc3","note":"the cautionary precedent: xanadu as the longest-running vaporware project in computing — vision shipped partially, ambition unreached.","originalId":"src-curse-of-xanadu","published":{"precision":"month","text":"1995-06"},"publisher":"wired","title":"the curse of xanadu","type":"reporting","url":"https://www.wired.com/1995/06/xanadu/"},{"author":null,"id":"source-a024316ffd8a24a7ffa9","note":"ward cunningham's 1995 wiki: every page editable, every camelcase word a link that creates the page if missing — bottom-up structure made nearly free.","originalId":"src-wikiwikiweb","published":{"precision":"year","text":"1995"},"publisher":"wikipedia","title":"wikiwikiweb","type":"community","url":"https://en.wikipedia.org/wiki/WikiWikiWeb"},{"author":"niklas luhmann","id":"source-04705a1cf632f4a9d889","note":"the slip-box as a partner that surprises you — and the caveat roam lived out: the box needs years to reach critical mass, until then it is a mere container.","originalId":"src-luhmann-slip-boxes","published":{"precision":"year","text":"1981"},"publisher":"luhmann.surge.sh (translation)","title":"communicating with slip boxes","type":"primary","url":"https://luhmann.surge.sh/communicating-with-slip-boxes"},{"author":"sönke ahrens","id":"source-257fd8d62d252b1b3aa9","note":"the book that popularized luhmann's zettelkasten for the roam generation.","originalId":"src-smart-notes","published":{"precision":"year","text":"2017"},"publisher":"northbooks","title":"how to take smart notes","type":"primary","url":"https://takesmartnotes.com/"},{"author":null,"id":"source-c50c200c5fb3be948c82","note":"org-mode's own history: a plain-text outline mode born in 2003 — the grammar roam rebuilt as a product.","originalId":"src-org-history","published":{"precision":"year","text":"2003"},"publisher":"orgmode.org","title":"history and acknowledgments (org manual)","type":"primary","url":"https://orgmode.org/manual/History-and-Acknowledgments.html"},{"author":"alexia tsotsis","id":"source-635f8d788fce2784f71f","note":"workflowy's 2010 launch: proof a whole product could be one infinite, zoomable list.","originalId":"src-workflowy-launch","published":{"precision":"day","text":"2010-11-04"},"publisher":"techcrunch","title":"turn your entire life into a list with workflowy","type":"reporting","url":"https://techcrunch.com/2010/11/04/workflow/"},{"author":"ted goranson","id":"source-7806a8742f0f53f1b7de","note":"the desktop-outliner lineage — thinktank, more, acta — that roam's surface descends from.","originalId":"src-outliner-history","published":{"precision":"year","text":"2003"},"publisher":"atpm","title":"about this particular outliner: outliner history","type":"community","url":"https://atpm.com/9.09/atpo.shtml"},{"author":"mark bernstein","id":"source-b839bce2051cdb142c08","note":"the first recorded use of 'digital garden' (1998) and a deliberate argument for irregular, curated link structure; tinderbox put agents inside notes in 2002.","originalId":"src-hypertext-gardens","published":{"precision":"year","text":"1998"},"publisher":"eastgate","title":"hypertext gardens","type":"primary","url":"https://eastgate.com/garden/Enter.html"},{"author":"jeremy ruston","id":"source-a4f807c58c547db1b05f","note":"the personal wiki as one self-contained html file of reusable tiddlers — quasi-transclusion in 2004.","originalId":"src-tiddlywiki-story","published":{"precision":"year","text":"2004"},"publisher":"tiddlywiki.com","title":"the story of tiddlywiki","type":"primary","url":"https://tiddlywiki.com/static/The%2520Story%2520of%2520TiddlyWiki.html"},{"author":"mike caulfield","id":"source-776433847a838f1e4ca0","note":"the garden-vs-stream frame that became digital-garden canon: topology against serialization.","originalId":"src-garden-and-stream","published":{"precision":"day","text":"2015-10-17"},"publisher":"hapgood","title":"the garden and the stream: a technopastoral","type":"primary","url":"https://hapgood.us/2015/10/17/the-garden-and-the-stream-a-technopastoral/"},{"author":"michael nielsen","id":"source-82891e8d2b461584f30f","note":"spaced repetition and memory systems as tools for thought — the intellectual neighbor roam sat next to.","originalId":"src-augmenting-memory","published":{"precision":"month","text":"2018-07"},"publisher":"michael nielsen","title":"augmenting long-term memory","type":"primary","url":"https://augmentingcognition.com/ltm.html"},{"author":"andy matuschak & michael nielsen","id":"source-d48b38c91fa9873411c7","note":"the essay that named the field roam came to represent — and warned that good tools for thought arise mostly as byproducts of serious work.","originalId":"src-transformative-tft","published":{"precision":"year","text":"2019"},"publisher":"numinous productions","title":"how can we develop transformative tools for thought?","type":"primary","url":"https://numinous.productions/ttft/"},{"author":"andy matuschak","id":"source-ae441fe63b14a342cfe7","note":"the canonical public note corpus: 'better note-taking misses the point; what matters is better thinking'; titles as apis; notes densely linked.","originalId":"src-evergreen-notes","published":{"precision":"year","text":"2019"},"publisher":"notes.andymatuschak.org","title":"evergreen notes","type":"primary","url":"https://notes.andymatuschak.org/Evergreen_notes"},{"author":"tom critchlow","id":"source-8037029da5d580fbf303","note":"the indie-hacker instantiation: streams, campfires, gardens, and the jekyll wikifolder — building your own garden before roam sold one.","originalId":"src-critchlow-gardens","published":{"precision":"month","text":"2018-10"},"publisher":"tomcritchlow.com","title":"of digital streams, campfires and gardens","type":"primary","url":"https://tomcritchlow.com/2018/10/10/of-gardens-and-wikis/"},{"author":"kleppmann, wiggins, van hardenberg, mcgranaghan","id":"source-75dfe6e1b48a22f623ef","note":"the counter-ideology roam violated: you own your data, in spite of the cloud — the banner obsidian and logseq recruited under.","originalId":"src-local-first","published":{"precision":"month","text":"2019-04"},"publisher":"ink & switch","title":"local-first software","type":"primary","url":"https://www.inkandswitch.com/local-first/"},{"author":"tiago forte","id":"source-15c650ab19ba5c01eb17","note":"code and para turned pkm into a mass aspiration — the demand roam rode. eliason's verdict: 'roam is v4.'","originalId":"src-building-second-brain","published":{"precision":"year","text":"2017"},"publisher":"building a second brain","title":"building a second brain","type":"primary","url":"https://www.buildingasecondbrain.com/"},{"author":"steph ango","id":"source-b8b946a41bad6831c413","note":"obsidian's ceo on why files outlive apps — in the agent era it reads as prophetic: plain markdown is the substrate agents already read.","originalId":"src-file-over-app","published":{"precision":"day","text":"2023-07-01"},"publisher":"stephango.com","title":"file over app","type":"primary","url":"https://stephango.com/file-over-app"},{"author":null,"id":"source-c93fbd25fcb5484c2f3d","note":"the local-first competitor that won the pragmatists while roam won the evangelists — first beta march 30, 2020, free, plain markdown.","originalId":"src-obsidian","published":{"precision":"year","text":"2020"},"publisher":"obsidian","title":"obsidian — a private and flexible writing app","type":"primary","url":"https://obsidian.md/about"},{"author":null,"id":"source-84997723082acd9a105f","note":"the open-source outliner clone (agpl, local markdown/org files) that absorbed roam's most cost-sensitive users; its readme credits roam, org-mode, tiddlywiki, and workflowy.","originalId":"src-logseq","published":{"precision":"year","text":"2020"},"publisher":"github","title":"logseq/logseq","type":"primary","url":"https://github.com/logseq/logseq"},{"author":null,"id":"source-c80e42958be8bbb0583f","note":"logseq's $4.1m raise from collison, friedman, lütke, and dave winer — explicitly on users 'jumping ship from roam.'","originalId":"src-logseq-raise","published":{"precision":"approximate","text":"~2022"},"publisher":"logseq blog","title":"logseq raises $4.1m","type":"primary","url":"https://blog.logseq.com/logseq-raises-4-1m-to-accelerate-growth-of-the-new-world-knowledge-graph/"},{"author":null,"id":"source-aad8ffea41cfaab5f947","note":"the cautionary open-source clone: yc w21, 6.3k stars, unmaintained since ~2023 — cloning the graph was never the hard part.","originalId":"src-athens","published":{"precision":"year","text":"2020"},"publisher":"github","title":"athensresearch/athens","type":"primary","url":"https://github.com/athensresearch/athens"},{"author":null,"id":"source-b40545bca43c37fdd757","note":"the supertag successor built by ex-roam-community figures, launched on the explicit promise of finishing what roam started.","originalId":"src-tana","published":{"precision":"year","text":"2023"},"publisher":"tana","title":"tana — the all-in-one workspace with supertags","type":"primary","url":"https://tana.inc/"},{"author":null,"id":"source-cd07f2f4ba23541b9dcf","note":"tana's $14m series a as an 'ai-native workspace' — the category's pivot to agents made literal.","originalId":"src-tana-raise","published":{"precision":"approximate","text":"~2024"},"publisher":"tana blog","title":"tana raises $14m series a","type":"primary","url":"https://outliner.tana.inc/blog/tana-raises-usd14m-series-a"},{"author":null,"id":"source-64fd3495b97d0316d495","note":"'your graph is the context and conversations are your prompts… your knowledge graph builds itself every time you talk.'","originalId":"src-tana-next-chapter","published":{"precision":"approximate","text":"~2025"},"publisher":"tana blog","title":"the next chapter for tana","type":"primary","url":"https://outliner.tana.inc/blog/the-next-chapter-for-tana"},{"author":"alex maccaw","id":"source-3e47d2312b8370fd2500","note":"the labor handoff as a button: 'decorate my writing with backlinks.'","originalId":"src-reflect-backlinks","published":{"precision":"approximate","text":"~2023"},"publisher":"reflect","title":"automatically add backlinks using ai","type":"primary","url":"https://reflect.app/blog/automatically-add-backlinks-using-ai"},{"author":"andrej karpathy","id":"source-ebe31e9964193545c49d","note":"the agent-era restatement: an llm-maintained markdown wiki as 'a persistent, compounding artifact.' 'obsidian is the ide; the llm is the programmer; the wiki is the codebase.'","originalId":"src-karpathy-llm-wiki","published":{"precision":"day","text":"2026-04-04"},"publisher":"github gist","title":"llm-wiki.md","type":"primary","url":"https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f"},{"author":"larson & truitt","id":"source-c8a880fa1eb9bbb177d9","note":"microsoft's graphrag: llm-extracted knowledge graphs plus community summaries beat baseline rag on global sensemaking questions.","originalId":"src-graphrag","published":{"precision":"day","text":"2024-02-13"},"publisher":"microsoft research","title":"graphrag: unlocking llm discovery on narrative private data","type":"primary","url":"https://www.microsoft.com/en-us/research/blog/graphrag-unlocking-llm-discovery-on-narrative-private-data/"},{"author":"jaya gupta & ashu garg","id":"source-f115b24131354e28c3a6","note":"the enterprise restatement: agents need decision traces — exceptions, overrides, precedents — captured into context graphs.","originalId":"src-context-graphs","published":{"precision":"month","text":"2025-12"},"publisher":"foundation capital","title":"ai's trillion-dollar opportunity: context graphs","type":"primary","url":"https://foundationcapital.com/ideas/context-graphs-ais-trillion-dollar-opportunity"},{"author":null,"id":"source-617acd4ac8d72c1eff15","note":"anthropic's open socket between agents and data — the plumbing that makes agent-readable knowledge bases practical.","originalId":"src-mcp","published":{"precision":"day","text":"2024-11-25"},"publisher":"anthropic","title":"introducing the model context protocol","type":"primary","url":"https://www.anthropic.com/research/model-context-protocol"},{"author":null,"id":"source-d063cd82be41582e70fc","note":"the links-beat-embeddings argument at project scale: index files as routing tables, let the agent follow links in two or three hops.","originalId":"src-no-rag","published":{"precision":"approximate","text":"~2025"},"publisher":"blog.pollack.ai","title":"look ma, no rag! building agent-native knowledge systems","type":"community","url":"https://blog.pollack.ai/look-ma-no-rag/"},{"author":null,"id":"source-d533a7d02fbb317ead16","note":"the strongest counter-evidence: a plain agent with file tools hit 74% on the locomio memory benchmark — files plus grep rival specialized memory systems.","originalId":"src-letta-benchmark","published":{"precision":"approximate","text":"~2025"},"publisher":"letta","title":"benchmarking ai agent memory: is a filesystem all you need?","type":"primary","url":"https://www.letta.com/blog/benchmarking-ai-agent-memory/"},{"author":null,"id":"source-0a351a290030e1f4d7a8","note":"git-backed agent memory: markdown plus frontmatter projected into a real repo the agent greps and commits — roam's data model reinvented as infrastructure.","originalId":"src-letta-context-repos","published":{"precision":"approximate","text":"~2025"},"publisher":"letta","title":"introducing context repositories","type":"primary","url":"https://www.letta.com/blog/context-repositories/"},{"author":null,"id":"source-d4931dfaad1e407f922a","note":"the cross-vendor convention — 'a readme for agents' across codex, cursor, jules, copilot, and 60k+ projects.","originalId":"src-agents-md","published":{"precision":"year","text":"2025"},"publisher":"agents.md","title":"agents.md","type":"primary","url":"https://agents.md/"},{"author":"nicolas bustamante","id":"source-3b9eef51ab8cb9b55766","note":"the audit caveat: production agent memory converges on small markdown taxonomies, and 'remember this' behavior is harness-trained — structure matters less than habit.","originalId":"src-agent-memory-engineering","published":{"precision":"approximate","text":"~2026"},"publisher":"nicolasbustamante.com","title":"agent memory engineering","type":"community","url":"https://nicolasbustamante.com/blog/agent-memory-engineering"},{"author":"zaferdace","id":"source-9503dca74ecd52885c51","note":"the counter-lesson: at ~100 articles an llm-maintained wiki exceeds usable context and the model starts confidently fabricating cross-note relationships.","originalId":"src-llm-wiki-broke","published":{"precision":"approximate","text":"~2026"},"publisher":"dev.to","title":"karpathy's obsidian wiki broke at 100 articles","type":"community","url":"https://dev.to/zaferdace/karpathys-obsidian-wiki-broke-at-100-articles-rag-fixed-it-4d4h"},{"author":null,"id":"source-e43054229dc7b1a23cc5","note":"the bluntest version of the against case: 'the ai does not care about your backlinks… the expensive operation became free.'","originalId":"src-second-brain-dead","published":{"precision":"approximate","text":"~2025"},"publisher":"self.md","title":"why your second brain is dead","type":"community","url":"https://self.md/concepts/second-brain-is-dead/"},{"author":"hraness","id":"source-f2c79886b553d0269039","note":"a knowledge base for coding agents built from markdown, backlinks, semantic search, and git context — markdown and git authoritative, indexes as replaceable views.","originalId":"src-wordcell","published":{"precision":"year","text":"2026"},"publisher":"github","title":"hraness/wordcell","type":"primary","url":"https://github.com/hraness/wordcell"},{"author":"hraness","id":"source-4f23913aa952b80025a6","note":"wordcell's docs: context, backlinks, and history per code path, installed as an agent skill.","originalId":"src-wordcell-docs","published":{"precision":"year","text":"2026"},"publisher":"wordcell.io","title":"wordcell docs","type":"primary","url":"https://wordcell.io/docs"},{"author":"hraness","id":"source-d427882e005d41aa9ecd","note":"a knowledge harness where agents research against a saved corpus — cite the passages used, check claims against the text — while people steer, review, and control release.","originalId":"src-sponge","published":{"precision":"year","text":"2025"},"publisher":"sponge.computer","title":"sponge — a personal library for research","type":"primary","url":"https://sponge.computer/"},{"author":"hraness","id":"source-1ca961629a2fc81119fd","note":"an ontology kernel for agentic research: content-addressed entity, statement, assertion, and evidence records with an append-only operation log; indexes derived and rebuildable.","originalId":"src-oh","published":{"precision":"year","text":"2026"},"publisher":"github","title":"hraness/oh","type":"primary","url":"https://github.com/hraness/oh"},{"author":null,"id":"source-b497c2d2497a3f7594e1","note":"a representative 'what happened to roam' essay — likely ai-assisted, used for the genre's shape rather than its claims: 'connections you do not understand are not understanding.'","originalId":"src-what-happened-roam","published":{"precision":"approximate","text":"~2026"},"publisher":"buildfirstbrain","title":"what happened to roam research? magic, then friction","type":"community","url":"https://buildfirstbrain.com/journal/why-roam-research-felt-like-magic-and-why-it-faded/"}]}
````