Public dossier section · @ben
the idea
This is @ben's profile revision 1. The original date and editorial confidence below are attributed to its publisher, not independent verification.
Original credited author: Hraness
AI-drafted at Ben Guo's direct request and credited to Hraness; every claim links to its cataloged source.
Back to the full profile · Markdown version · Original document
Published text
tldraw: the idea
canvas-as-infrastructure — the lineage tldraw inherited, what it proved, and what agents do to it.
lineage
- 1963 — sketchpad: sutherland's sketchpad — direct manipulation of geometry as computation, the ancestor every canvas interface cites. tldraw inherits the claim that drawing on a surface is a way of programming. (sources: sutherland 1963 — lineage anchor)
- 1970s — the dynabook: kay's dynabook — the personal dynamic medium. the canvas as a place where a person builds their own tools rather than consuming someone else's documents. (sources: kays dynabook — lineage anchor)
- 1987 — hypercard: hypercard — end-user authoring with a real object model underneath. the direct ancestor of 'a canvas is a thing you build with': the surface is friendly, the substrate is programmable. (sources: hypercard 1987 — lineage anchor)
- 2019–2020 — spatial software: john palmer's spatial interfaces and spatial software essays — 'a world within its interface' — the 2019–2020 genre that made the spatial canvas a legible product category again before the tools caught up. (sources: john palmer, spatial interfaces (reprint of darkblueheaven), john palmer, spatial software — a world within its interface)
- 2019–2023 — muse: muse — the thinking-tool canvas that wound down in fall 2023. adam wiggins wrote the post-mortem and sits on tldraw's cap table: the predecessor's failure is literally inside the successor's investor list. (sources: wiggins muse post-mortem — canvas thinking-tool business failure, tldraw.dev)
- 2020– — excalidraw: excalidraw — the mit-licensed hand-drawn canvas that adopted perfect-freehand. the open alternative that defined the aesthetic and now defines the license comparison tldraw keeps losing. (sources: worlds fair talk, name origin (tl, excalidraw-vs-tldraw evergreen comparison)
- 2024– — the canvas revival: the zoomable-ui re-emergence — artifacts, chatgpt canvas, mistral's le chat: the frontier labs made 'canvas' an ai-interface paradigm while tldraw was already there. the category arrived around the company. (sources: zui/infinite-canvas re-emergence in genai apps, canvas-as-ai-interface paradigm: artifacts→chatgpt canvas→mistral)
what tldraw proved
- infrastructure beats the app: canvas-as-infrastructure survives where canvas-as-app struggles — ~$1m of first-year license revenue, a named roster (clickup whiteboards, padlet, autodesk, replit, shopify, luma, runway, blackrock), and a $10m series a on the strength of it. muse wound down; tldraw compounded. (sources: $10m series a (lux + definition), clickup whiteboards-on-tldraw case study with pm quote, wiggins muse post-mortem — canvas thinking-tool business failure, ruiz talk: make real quaint in 2026)
- the dom bet: the html-over-webgl bet paid — 'i could probably just make normal website stuff just as fast… then i would be able to put anything onto that canvas.' a dom canvas can host iframes, which is what made make real possible at all; ruiz argues miro 'could not' have built it. (sources: html-vs-rendering bet quotes, full make real story talk, full make real chronology, how tldraw built)
- demos as r&d: demos are the growth engine — make real, draw fast, computer, fairies each generated the largest coverage moments in the record, and each funnels into sdk awareness. ruiz calls the program 'marketing, r&d, and fun'; the hn numbers back the marketing part. (sources: biggest tldraw hn thread — 553 pts, 120 comments, full make real chronology, product hunt: #2 of the day, 418 upvotes, launch post: $25 one-time, december-only timed experiment)
- permanence as trust: permanence is a feature — v1 parked at old.tldraw.com 'until the end of time' and released under mit is the trust primitive that makes the license argument survivable. the company keeps an older self alive on purpose. (sources: apr 4 2023 switchover plan, original 2021-2022 version, released under mit, v2 exits beta apr 4 2023)
- the agent surface: the canvas is a real agent surface — carry on, teach, computer, the agent starter kit, fairies, and mcp access all ran on the same substrate. 'if it's busy, you just go somewhere else' is a multi-agent argument a chat window can't make. (sources: agent starter kit docs — cursor-style agent on canvas, github, canvas-vs-chat multi-agent quotes, aiewf talk: agent-canvas category (paper, replit agent 4, stitch), offline desktop launch)
what stayed unfinished
- the app is a showroom: the app never became the business — tldraw.com stayed free, the sync demo server wipes data after 24 hours, and offline is a free desktop app. the product-shaped canvases are showrooms for the sdk, not revenue. (sources: sync architecture: rooms, tlsocketroom, cloudflare template, demo server, offline desktop launch, origins quote, $2)
- the spent goodwill: the open-source goodwill was spent — 'open source is a spectrum,' the readme correction, the formerly-foss listings, the license-key gate. the community that made the 2021 launch warm now writes the comparisons that recolor tldraw as closed. (sources: bespoke license w/ sdk 2, its not foss… changed in december 2023, quietly stopped being a fair open source pick… late 2025, most contested thread — 98 pts, 50 comments)
- the sunset pattern: demos sunset faster than they convert — draw fast went down in january 2024, computer stayed closed source, fairies ran one december. the attention keeps arriving; the products it pointed at keep expiring. (sources: shutdown notice date, team: computer stays closed-source, fairies landing page, tool review noting demo-vs-production caveats for make real)
- the pricing opacity: pricing stays opaque — the protest issue, the 100-day trial, hobby and commercial paths that still route through sales. 'just ask for money' is easier to say than to publish a price for. (sources: protest issue: get a license for production are you joking?, license-key mechanics: trial/commercial/hobby, client-side validation, adobe turndown on a friday, most contested thread — 98 pts, 50 comments)
- the cadence nobody watches: release news stopped being news — sdk 4.0's thread was a fight, sdk 5.0's was a shrug (1 point). hn attention now arrives only for demos and controversies, not for the product cadence. (sources: tldraw sdk 5, most contested thread — 98 pts, 50 comments, sdk 5)
the agent-era turn
- (for) the canvas is the right agent surface — spatial, parallel, inspectable. 'if it's busy, you just go somewhere else' generalizes to agents: many workers on one visible substrate, which is what the fairy experiments were actually testing. (sources: canvas-vs-chat multi-agent quotes, multi-agent coordination lessons, agent starter kit docs — cursor-style agent on canvas, show hn agent-drawing demo on tldraw, 53 pts)
- (for) infrastructure position compounds — the sdk already lives inside clickup, replit, luma, and google's own stitch experiments, so agent features ride existing deployments rather than fighting for new surface area. (sources: ruiz talk: make real quaint in 2026, $10m series a (lux + definition), aiewf talk: agent-canvas category (paper, replit agent 4, stitch))
- (against) the non-determinism problem is real — natural-language node graphs fail unpredictably, the computer's own hn thread collected failure reports, and the agents are 'essentially working blind.' the canvas can host agents; that doesn't make them reliable. (sources: essay on natural-language node graphs non-determinism, biggest tldraw hn thread — 553 pts, 120 comments, multi-agent coordination lessons)
- (against) the license gate collides with agent-era distribution — a keyed production sdk can't ride inside downstream open source, and the community that builds agent demos is exactly the one priced out. the ai-slop flood is the same ecosystem arriving uninvited. (sources: most contested thread — 98 pts, 50 comments, contributions policy — auto-closing external prs over ai slop, quietly stopped being a fair open source pick… late 2025)
- (nuance) the paradigm is commoditizing — artifacts, chatgpt canvas, le chat, google's own canvases. tldraw's edge is the substrate (dom, sync, licensing), not the concept; if 'canvas' becomes a feature every chatbot ships, the sdk is either the picks-and-shovels winner or a commodity input. (sources: canvas-as-ai-interface paradigm: artifacts→chatgpt canvas→mistral, zui/infinite-canvas re-emergence in genai apps, aiewf talk: agent-canvas category (paper, replit agent 4, stitch))
- (nuance) gating may be the honest trade — 'fate and capital both demand that tldraw be a sustainable project.' if the sdk is infrastructure, someone has to pay for it, and the mit alternative (excalidraw) exists for everyone the license excludes. (sources: cli, 5 starter kits, license-key enforcement, wcag 2, excalidraw-vs-tldraw evergreen comparison, company page correcting llm misinformation about the license)
successors
- excalidraw: the mit control group — the canvas that adopted perfect-freehand and now anchors every 'fair open-source pick' comparison tldraw loses. (sources: excalidraw-vs-tldraw evergreen comparison, self-hosting lens: excalidraw mit vs tldraw non-commercial, quietly stopped being a fair open source pick… late 2025)
- draw-a-ui: the community seed — hood's mit-licensed sketch-to-html demo that tldraw forked into make real. the ecosystem makes the demos; the company packages them. (sources: original draw-a-ui repo, ~13, sawyer hoods show hn — 43 pts, 5 comments, full make real chronology)
- muse: the predecessor's ghost — the thinking-tool canvas whose post-mortem is the thesis's control case, and whose founder backs tldraw. (sources: wiggins muse post-mortem — canvas thinking-tool business failure, tldraw.dev)
- the canvas cohort: kosmik and the app-shaped canvases that kept launching and sunsetting around the sdk — the field the infrastructure bet was made against. (sources: kosmik profile naming the infinite-canvas cohort, kosmik sunset announcement)
- the agent-canvas category: paper, replit agent 4, google stitch — the aiewf 'spatial harness' talk already treats the agent canvas as a category with tldraw inside it, not above it. (sources: aiewf talk: agent-canvas category (paper, replit agent 4, stitch), agent starter kit docs — cursor-style agent on canvas, google gemini api showcase of tldraw computer (via vuink mirror))
- the lab canvases: artifacts, chatgpt canvas, mistral's le chat — the frontier-lab versions of the same surface, owned by the model providers tldraw demos against. (sources: canvas-as-ai-interface paradigm: artifacts→chatgpt canvas→mistral, zui/infinite-canvas re-emergence in genai apps)
open questions
- what is tldraw's exact public launch day — the hn thread is november 18, 2021, but the reports leave the precise opening-day sequence unresolved.
- does license revenue compound past ~$1m a year — and does the $10m series a price a platform or a services-adjacent sdk?
- is the linkedin-aggregated '$24.1m total funding' figure a phantom row, or does a real unannounced round sit between the extension and the series a?
- can a production-keyed sdk participate in the agent era it demos for, when the agent ecosystem builds downstream open source the license excludes?
- do the demos ever become products — or is 'marketing, r&d, and fun' the permanent terminal state of make real, computer, and fairies?
- does 'source available' hold as the settled self-description, or does the spectrum framing return each time an llm misstates the license?
- who owns the agent-canvas category — the sdk that hosts the surface, or the labs whose models fill it?
- does tldraw itself become 'a vibe coded alternative to tldraw' — ruiz's own joke about where the tests-to-closed-source bit was actually pointing?
AI-drafted at Ben Guo's direct request and credited to Hraness; every claim links to its cataloged source.
Canonical page: https://hraness.com/tldraw/the-idea
Original source occurrences
- company page correcting llm misinformation about the license · tldraw.dev · 2025 (year) · original ID: src-ci-20-things
company page correcting llm misinformation about the license
- agent starter kit docs — cursor-style agent on canvas · tldraw.dev · undated (unknown) · original ID: src-ci-agent-starter
agent starter kit docs — cursor-style agent on canvas
- ruiz talk: make real quaint in 2026 · ai.engineer · 2026 (year) · original ID: src-ci-ai-engineer-talk
ruiz talk: make real "quaint in 2026"; stitch "not built with our canvas"
- essay on natural-language node graphs non-determinism · atharvaraykar.com · 2024-12 (month) · original ID: src-ci-atharva-computer
essay on natural-language node graphs' non-determinism
- clickup whiteboards-on-tldraw case study with pm quote · tldraw.dev · undated (unknown) · original ID: src-ci-clickup-case
clickup whiteboards-on-tldraw case study with pm quote
- original draw-a-ui repo, ~13 · github · 2023-11-07 (day) · original ID: src-ci-draw-a-ui-repo
original draw-a-ui repo, ~13.5k stars, mit
- multi-agent coordination lessons · linkedin · 2025-12-19 (day) · original ID: src-ci-fairies-coord
multi-agent coordination lessons; agents "working blind"
- launch post: $25 one-time, december-only timed experiment · linkedin · 2025-12-01 (day) · original ID: src-ci-fairies-linkedin
launch post: $25 one-time, december-only timed experiment
- google gemini api showcase of tldraw computer (via vuink mirror) · vuink.com · 2024-12-19 (day) · original ID: src-ci-gemini-showcase
google gemini api showcase of tldraw computer (via vuink mirror)
- show hn agent-drawing demo on tldraw, 53 pts · hacker news · 2026-07-06 (day) · original ID: src-ci-hn-agent-draw
show hn agent-drawing demo on tldraw, 53 pts
- biggest tldraw hn thread — 553 pts, 120 comments · hacker news · 2024-12-20 (day) · original ID: src-ci-hn-computer
biggest tldraw hn thread — 553 pts, 120 comments; ruiz "marketing, r&d, and fun" quote
- sawyer hoods show hn — 43 pts, 5 comments · hacker news · 2023-11-08 (day) · original ID: src-ci-hn-draw-a-ui
sawyer hood's show hn — 43 pts, 5 comments
- most contested thread — 98 pts, 50 comments · hacker news · 2025-09-18 (day) · original ID: src-ci-hn-sdk4
most contested thread — 98 pts, 50 comments; pricing/centralization criticism; ruiz engagement
- tldraw sdk 5 · hacker news · 2026-05-08 (day) · original ID: src-ci-hn-sdk5
"tldraw sdk 5.0" — 1 pt
- zui/infinite-canvas re-emergence in genai apps · interfacestudies.substack.com · undated (unknown) · original ID: src-ci-interfacestudies-zui
zui/infinite-canvas re-emergence in genai apps
- its not foss… changed in december 2023 · isitreallyfoss.com · undated (unknown) · original ID: src-ci-isitreallyfoss
"it's not foss… changed in december 2023"
- contributions policy — auto-closing external prs over ai slop · github · 2026-01-15 (day) · original ID: src-ci-issue-7695
contributions policy — auto-closing external prs over ai slop
- protest issue: get a license for production are you joking? · github · undated (unknown) · original ID: src-ci-issue-8577
protest issue: "get a license for production are you joking?"
- kosmik sunset announcement · launchpad.kosmik.app · undated (unknown) · original ID: src-ci-kosmik-sunset
kosmik sunset announcement
- license-key mechanics: trial/commercial/hobby, client-side validation · tldraw.dev · undated (unknown) · original ID: src-ci-license-key-docs
license-key mechanics: trial/commercial/hobby, client-side validation
- quietly stopped being a fair open source pick… late 2025 · meetrix.io · undated (unknown) · original ID: src-ci-meetrix-oss
"quietly stopped being a fair 'open source' pick… late 2025"
- canvas-as-ai-interface paradigm: artifacts→chatgpt canvas→mistral · patmcguinness.substack.com · 2024-11 (month) · original ID: src-ci-mistral-canvas-essay
canvas-as-ai-interface paradigm: artifacts→chatgpt canvas→mistral
- wiggins muse post-mortem — canvas thinking-tool business failure · adamwiggins.com · 2024-10 (month) · original ID: src-ci-muse-retro
wiggins' muse post-mortem — canvas thinking-tool business failure; fall-2023 wind-down
- excalidraw-vs-tldraw evergreen comparison · ossalt.com · 2026 (year) · original ID: src-ci-ossalt-compare
excalidraw-vs-tldraw evergreen comparison
- product hunt: #2 of the day, 418 upvotes · webmail.hunted.space · 2023-11-27 (day) · original ID: src-ci-ph-drawfast
product hunt: #2 of the day, 418 upvotes
- v2 exits beta apr 4 2023 · tldraw.dev · 2023-03-20 (day) · original ID: src-ci-relaunch-blog
v2 exits beta apr 4 2023; v1 parked under mit
- tool review noting demo-vs-production caveats for make real · the rundown · undated (unknown) · original ID: src-ci-rundown
tool review noting demo-vs-production caveats for make real
- self-hosting lens: excalidraw mit vs tldraw non-commercial · selfhosting.sh · undated (unknown) · original ID: src-ci-selfhosting-compare
self-hosting lens: excalidraw mit vs tldraw non-commercial
- aiewf talk: agent-canvas category (paper, replit agent 4, stitch) · nextjs-conf-scheduler.sentry.dev · 2025 (year) · original ID: src-ci-spatial-harness
aiewf talk: agent-canvas category (paper, replit agent 4, stitch)
- john palmer, spatial interfaces (reprint of darkblueheaven) · balajis.com · 2019-08-30 (day) · original ID: src-ci-spatial-interfaces
john palmer, "spatial interfaces" (reprint of darkblueheaven)
- john palmer, spatial software — a world within its interface · balajis.com · 2020-04-09 (day) · original ID: src-ci-spatial-software
john palmer, "spatial software" — "a world within its interface"
- kosmik profile naming the infinite-canvas cohort · techcrunch · 2023-12-21 (day) · original ID: src-ci-tc-kosmik
kosmik profile naming the infinite-canvas cohort
- kays dynabook — lineage anchor · wikipedia · undated (unknown) · original ID: src-ci-wiki-dynabook
kay's dynabook — lineage anchor
- hypercard 1987 — lineage anchor · wikipedia · undated (unknown) · original ID: src-ci-wiki-hypercard
hypercard 1987 — lineage anchor
- sutherland 1963 — lineage anchor · wikipedia · undated (unknown) · original ID: src-ci-wiki-sketchpad
sutherland 1963 — lineage anchor
- worlds fair talk · ai.engineer · 2025-07-21 (day) · original ID: src-hp-aiengineer-computer-talk
world's fair talk; excalidraw ink origin; computer internals; google collab story; liquid layers/observable shoutouts
- team: computer stays closed-source · github · 2024-12-15 (day) · original ID: src-hp-computer-issue
team: computer stays closed-source; workflow starter kit teased
- shutdown notice date · drawfast.tldraw.com · 2024-01-14 (day) · original ID: src-hp-drawfast-site
shutdown notice date
- github · github · 2025 (year) · original ID: src-hp-fairies-pr
internal fairy agent: @tldraw/fairy-shared, cloudflare fairy-worker, settool/removetool apis
- fairies landing page · fairies.tldraw.com · 2025-12 (month) · original ID: src-hp-fairies-site
fairies landing page
- full make real story talk · gitnation.com · 2024 (year) · original ID: src-hp-gitnation-makereal
full make real story talk
- name origin (tl · hacker news · 2022-12-01 (day) · original ID: src-hp-hn-seed
name origin ("tl;dr aw"); ruiz on excalidraw/perfect-freehand, globs origin
- bespoke license w/ sdk 2 · tldraw.dev · 2023-12-20 (day) · original ID: src-hp-license-update
bespoke license w/ sdk 2.0 beta; non-commercial default; v1 mit forever; 2.x pre-releases apache-2.0
- full make real chronology · tldraw.substack.com · 2023-11-18 (day) · original ID: src-hp-makereal-story
full make real chronology; sawyer hood fork; drawmyui.com; starter template fri; quotes
- offline desktop launch · tldraw.dev · 2026-07-16 (day) · original ID: src-hp-offline-blog
offline desktop launch; local .tldraw files; scripting/agents
- apr 4 2023 switchover plan · tldraw.substack.com · 2023-03-20 (day) · original ID: src-hp-relaunch-post
apr 4 2023 switchover plan; old.tldraw.com; v1 repo move; footer "tldraw gb ltd"
- cli, 5 starter kits, license-key enforcement, wcag 2 · tldraw.dev · 2025-09-18 (day) · original ID: src-hp-sdk40
cli, 5 starter kits, license-key enforcement, wcag 2.2 aa
- sdk 5 · tldraw.dev · 2026-05-07 (day) · original ID: src-hp-sdk50
sdk 5.0 "all about features"; shipped-the-backlog quote
- tldraw.dev · tldraw.dev · 2022-11-30 (day) · original ID: src-hp-seed-blog
$2.7m seed investor list; "new tldraw" beta; team roster; early sdk adopters
- $10m series a (lux + definition) · tldraw.dev · 2025-04-09 (day) · original ID: src-hp-series-a
$10m series a (lux + definition); ~$1m first-year license revenue; customers; team ~20; digital-ink origin
- sync architecture: rooms, tlsocketroom, cloudflare template, demo server · tldraw.dev · 2026-09 (month) · original ID: src-hp-sync-docs
sync architecture: rooms, tlsocketroom, cloudflare template, demo server
- how tldraw built · syntax.fm · 2022 (year) · original ID: src-hp-syntax-505
how tldraw built; no-canvas-element rendering; globs/excalidraw lineage
- origins quote, $2 · techcrunch · 2022-12-16 (day) · original ID: src-hp-techcrunch
origins quote, $2.7m seed, five fte, sponsors-only revenue
- html-vs-rendering bet quotes · twosetai.com · 2026-03-22 (day) · original ID: src-hp-twosetai-canvas
html-vs-rendering bet quotes
- canvas-vs-chat multi-agent quotes · twosetai.com · 2026-03-28 (day) · original ID: src-hp-twosetai-chat
canvas-vs-chat multi-agent quotes
- adobe turndown on a friday · uxtools.co · 2026-03-16 (day) · original ID: src-hp-uxtools
adobe turndown "on a friday"; product-maker-fit; "just ask for money"
- original 2021-2022 version, released under mit · github · undated (unknown) · original ID: src-hp-v1-repo
"original 2021-2022 version, released under mit"
Structured research records
the canvas is the right agent surface — spatial, parallel, inspectable. 'if it's busy, you just go somewhere else' generalizes to agents: many workers on one visible substrate, which is what the fairy experiments were actually testing. · tldraw-agent-era-arguments
Original location · source-744df855d18c40017423 · source-95c36b914fed84f78930 · source-61e14e50eec664a4e09c · source-fefb08ce6bb1e6e4e1c9
Original structured fields
{ "claim": "the canvas is the right agent surface — spatial, parallel, inspectable. 'if it's busy, you just go somewhere else' generalizes to agents: many workers on one visible substrate, which is what the fairy experiments were actually testing.", "id": "arg-canvas-agents", "side": "for", "sourceIds": [ "src-hp-twosetai-chat", "src-ci-fairies-coord", "src-ci-agent-starter", "src-ci-hn-agent-draw" ] }infrastructure position compounds — the sdk already lives inside clickup, replit, luma, and google's own stitch experiments, so agent features ride existing deployments rather than fighting for new surface area. · tldraw-agent-era-arguments
Original location · source-f488236bcad5bba908e4 · source-7a834ad66795691be4ed · source-81e53b450674497ce864
Original structured fields
{ "claim": "infrastructure position compounds — the sdk already lives inside clickup, replit, luma, and google's own stitch experiments, so agent features ride existing deployments rather than fighting for new surface area.", "id": "arg-distribution", "side": "for", "sourceIds": [ "src-ci-ai-engineer-talk", "src-hp-series-a", "src-ci-spatial-harness" ] }the non-determinism problem is real — natural-language node graphs fail unpredictably, the computer's own hn thread collected failure reports, and the agents are 'essentially working blind.' the canvas can host agents; that doesn't make them reliable. · tldraw-agent-era-arguments
Original location · source-1d7bed610ba4b974aaab · source-8a22e22c0bae60d06c25 · source-95c36b914fed84f78930
Original structured fields
{ "claim": "the non-determinism problem is real — natural-language node graphs fail unpredictably, the computer's own hn thread collected failure reports, and the agents are 'essentially working blind.' the canvas can host agents; that doesn't make them reliable.", "id": "arg-nondeterminism", "side": "against", "sourceIds": [ "src-ci-atharva-computer", "src-ci-hn-computer", "src-ci-fairies-coord" ] }the license gate collides with agent-era distribution — a keyed production sdk can't ride inside downstream open source, and the community that builds agent demos is exactly the one priced out. the ai-slop flood is the same ecosystem arriving uninvited. · tldraw-agent-era-arguments
Original location · source-cf1c4c60f1205546d41f · source-abfd1080d08b791c1845 · source-178858b0cba70c16d626
Original structured fields
{ "claim": "the license gate collides with agent-era distribution — a keyed production sdk can't ride inside downstream open source, and the community that builds agent demos is exactly the one priced out. the ai-slop flood is the same ecosystem arriving uninvited.", "id": "arg-gate-friction", "side": "against", "sourceIds": [ "src-ci-hn-sdk4", "src-ci-issue-7695", "src-ci-meetrix-oss" ] }the paradigm is commoditizing — artifacts, chatgpt canvas, le chat, google's own canvases. tldraw's edge is the substrate (dom, sync, licensing), not the concept; if 'canvas' becomes a feature every chatbot ships, the sdk is either the picks-and-shovels winner or a commodity input. · tldraw-agent-era-arguments
Original location · source-f19ba1929c1c527876ec · source-765cc6f7a1ed080fc177 · source-81e53b450674497ce864
Original structured fields
{ "claim": "the paradigm is commoditizing — artifacts, chatgpt canvas, le chat, google's own canvases. tldraw's edge is the substrate (dom, sync, licensing), not the concept; if 'canvas' becomes a feature every chatbot ships, the sdk is either the picks-and-shovels winner or a commodity input.", "id": "arg-commodity", "side": "nuance", "sourceIds": [ "src-ci-mistral-canvas-essay", "src-ci-interfacestudies-zui", "src-ci-spatial-harness" ] }gating may be the honest trade — 'fate and capital both demand that tldraw be a sustainable project.' if the sdk is infrastructure, someone has to pay for it, and the mit alternative (excalidraw) exists for everyone the license excludes. · tldraw-agent-era-arguments
Original location · source-7c437e0f7a7a9080ceda · source-d5c46b24b087a89c7547 · source-62b10619cfc1eb6f0cc9
Original structured fields
{ "claim": "gating may be the honest trade — 'fate and capital both demand that tldraw be a sustainable project.' if the sdk is infrastructure, someone has to pay for it, and the mit alternative (excalidraw) exists for everyone the license excludes.", "id": "arg-sustainability", "side": "nuance", "sourceIds": [ "src-hp-sdk40", "src-ci-ossalt-compare", "src-ci-20-things" ] }sketchpad · tldraw-idea-lineage
Original location · source-1bb0b40cc88528148f88
Original structured fields
{ "contribution": "sutherland's sketchpad — direct manipulation of geometry as computation, the ancestor every canvas interface cites. tldraw inherits the claim that drawing on a surface is a way of programming.", "era": "1963", "id": "lin-sketchpad", "name": "sketchpad", "sourceIds": [ "src-ci-wiki-sketchpad" ] }the dynabook · tldraw-idea-lineage
Original location · source-0b478b11ce9a1e9b2084
Original structured fields
{ "contribution": "kay's dynabook — the personal dynamic medium. the canvas as a place where a person builds their own tools rather than consuming someone else's documents.", "era": "1970s", "id": "lin-dynabook", "name": "the dynabook", "sourceIds": [ "src-ci-wiki-dynabook" ] }hypercard · tldraw-idea-lineage
Original location · source-b6758bc5cb0908acdd2d
Original structured fields
{ "contribution": "hypercard — end-user authoring with a real object model underneath. the direct ancestor of 'a canvas is a thing you build with': the surface is friendly, the substrate is programmable.", "era": "1987", "id": "lin-hypercard", "name": "hypercard", "sourceIds": [ "src-ci-wiki-hypercard" ] }spatial software · tldraw-idea-lineage
Original location · source-c585e0049a16a05e0d85 · source-93f714c9c48ae8330204
Original structured fields
{ "contribution": "john palmer's spatial interfaces and spatial software essays — 'a world within its interface' — the 2019–2020 genre that made the spatial canvas a legible product category again before the tools caught up.", "era": "2019–2020", "id": "lin-spatial-software", "name": "spatial software", "sourceIds": [ "src-ci-spatial-interfaces", "src-ci-spatial-software" ] }muse · tldraw-idea-lineage
Original location · source-679234f10732200f4e19 · source-3a1f4235a952006eb88b
Original structured fields
{ "contribution": "muse — the thinking-tool canvas that wound down in fall 2023. adam wiggins wrote the post-mortem and sits on tldraw's cap table: the predecessor's failure is literally inside the successor's investor list.", "era": "2019–2023", "id": "lin-muse", "name": "muse", "sourceIds": [ "src-ci-muse-retro", "src-hp-seed-blog" ] }excalidraw · tldraw-idea-lineage
Original location · source-e3adcc327eb109991d9c · source-faeb26dbf94c7d4c2c78 · source-d5c46b24b087a89c7547
Original structured fields
{ "contribution": "excalidraw — the mit-licensed hand-drawn canvas that adopted perfect-freehand. the open alternative that defined the aesthetic and now defines the license comparison tldraw keeps losing.", "era": "2020–", "id": "lin-excalidraw", "name": "excalidraw", "sourceIds": [ "src-hp-aiengineer-computer-talk", "src-hp-hn-seed", "src-ci-ossalt-compare" ] }the canvas revival · tldraw-idea-lineage
Original location · source-765cc6f7a1ed080fc177 · source-f19ba1929c1c527876ec
Original structured fields
{ "contribution": "the zoomable-ui re-emergence — artifacts, chatgpt canvas, mistral's le chat: the frontier labs made 'canvas' an ai-interface paradigm while tldraw was already there. the category arrived around the company.", "era": "2024–", "id": "lin-zui-revival", "name": "the canvas revival", "sourceIds": [ "src-ci-interfacestudies-zui", "src-ci-mistral-canvas-essay" ] }what is tldraw's exact public launch day — the hn thread is november 18, 2021, but the reports leave the precise opening-day sequence unresolved. · tldraw-idea-open-questions
Original structured fields
{ "text": "what is tldraw's exact public launch day — the hn thread is november 18, 2021, but the reports leave the precise opening-day sequence unresolved." }does license revenue compound past ~$1m a year — and does the $10m series a price a platform or a services-adjacent sdk? · tldraw-idea-open-questions
Original structured fields
{ "text": "does license revenue compound past ~$1m a year — and does the $10m series a price a platform or a services-adjacent sdk?" }is the linkedin-aggregated '$24.1m total funding' figure a phantom row, or does a real unannounced round sit between the extension and the series a? · tldraw-idea-open-questions
Original structured fields
{ "text": "is the linkedin-aggregated '$24.1m total funding' figure a phantom row, or does a real unannounced round sit between the extension and the series a?" }can a production-keyed sdk participate in the agent era it demos for, when the agent ecosystem builds downstream open source the license excludes? · tldraw-idea-open-questions
Original structured fields
{ "text": "can a production-keyed sdk participate in the agent era it demos for, when the agent ecosystem builds downstream open source the license excludes?" }do the demos ever become products — or is 'marketing, r&d, and fun' the permanent terminal state of make real, computer, and fairies? · tldraw-idea-open-questions
Original structured fields
{ "text": "do the demos ever become products — or is 'marketing, r&d, and fun' the permanent terminal state of make real, computer, and fairies?" }does 'source available' hold as the settled self-description, or does the spectrum framing return each time an llm misstates the license? · tldraw-idea-open-questions
Original structured fields
{ "text": "does 'source available' hold as the settled self-description, or does the spectrum framing return each time an llm misstates the license?" }who owns the agent-canvas category — the sdk that hosts the surface, or the labs whose models fill it? · tldraw-idea-open-questions
Original structured fields
{ "text": "who owns the agent-canvas category — the sdk that hosts the surface, or the labs whose models fill it?" }does tldraw itself become 'a vibe coded alternative to tldraw' — ruiz's own joke about where the tests-to-closed-source bit was actually pointing? · tldraw-idea-open-questions
Original structured fields
{ "text": "does tldraw itself become 'a vibe coded alternative to tldraw' — ruiz's own joke about where the tests-to-closed-source bit was actually pointing?" }infrastructure beats the app · tldraw-idea-proved
Original location · source-7a834ad66795691be4ed · source-57d19aafab1825165370 · source-679234f10732200f4e19 · source-f488236bcad5bba908e4
Original structured fields
{ "detail": "canvas-as-infrastructure survives where canvas-as-app struggles — ~$1m of first-year license revenue, a named roster (clickup whiteboards, padlet, autodesk, replit, shopify, luma, runway, blackrock), and a $10m series a on the strength of it. muse wound down; tldraw compounded.", "id": "clm-proved-infrastructure", "sourceIds": [ "src-hp-series-a", "src-ci-clickup-case", "src-ci-muse-retro", "src-ci-ai-engineer-talk" ], "title": "infrastructure beats the app" }the dom bet · tldraw-idea-proved
Original location · source-673c5e9a5d658db8ef66 · source-6bcf37076af9abf7b07f · source-79ab2993b3b272875b42 · source-1bee7d2f3a71a94dfbd6
Original structured fields
{ "detail": "the html-over-webgl bet paid — 'i could probably just make normal website stuff just as fast… then i would be able to put anything onto that canvas.' a dom canvas can host iframes, which is what made make real possible at all; ruiz argues miro 'could not' have built it.", "id": "clm-proved-dom", "sourceIds": [ "src-hp-twosetai-canvas", "src-hp-gitnation-makereal", "src-hp-makereal-story", "src-hp-syntax-505" ], "title": "the dom bet" }demos as r&d · tldraw-idea-proved
Original location · source-8a22e22c0bae60d06c25 · source-79ab2993b3b272875b42 · source-e43d12bae5478dd37078 · source-3ba21e1c95153d60d382
Original structured fields
{ "detail": "demos are the growth engine — make real, draw fast, computer, fairies each generated the largest coverage moments in the record, and each funnels into sdk awareness. ruiz calls the program 'marketing, r&d, and fun'; the hn numbers back the marketing part.", "id": "clm-proved-demos", "sourceIds": [ "src-ci-hn-computer", "src-hp-makereal-story", "src-ci-ph-drawfast", "src-ci-fairies-linkedin" ], "title": "demos as r&d" }permanence as trust · tldraw-idea-proved
Original location · source-a4e1605b7cddc01f5f72 · source-e384207ea8b8b759591f · source-eae4a38e1e6b16cf519c
Original structured fields
{ "detail": "permanence is a feature — v1 parked at old.tldraw.com 'until the end of time' and released under mit is the trust primitive that makes the license argument survivable. the company keeps an older self alive on purpose.", "id": "clm-proved-permanence", "sourceIds": [ "src-hp-relaunch-post", "src-hp-v1-repo", "src-ci-relaunch-blog" ], "title": "permanence as trust" }the agent surface · tldraw-idea-proved
Original location · source-61e14e50eec664a4e09c · source-245c21085ba98313880b · source-744df855d18c40017423 · source-81e53b450674497ce864 · source-235e61c6c27be6651285
Original structured fields
{ "detail": "the canvas is a real agent surface — carry on, teach, computer, the agent starter kit, fairies, and mcp access all ran on the same substrate. 'if it's busy, you just go somewhere else' is a multi-agent argument a chat window can't make.", "id": "clm-proved-agents", "sourceIds": [ "src-ci-agent-starter", "src-hp-fairies-pr", "src-hp-twosetai-chat", "src-ci-spatial-harness", "src-hp-offline-blog" ], "title": "the agent surface" }the app is a showroom · tldraw-idea-stalled
Original location · source-91a32c7e59a6982b05a7 · source-235e61c6c27be6651285 · source-f9b5159929dc5994dfd4
Original structured fields
{ "detail": "the app never became the business — tldraw.com stayed free, the sync demo server wipes data after 24 hours, and offline is a free desktop app. the product-shaped canvases are showrooms for the sdk, not revenue.", "id": "clm-stalled-app", "sourceIds": [ "src-hp-sync-docs", "src-hp-offline-blog", "src-hp-techcrunch" ], "title": "the app is a showroom" }the spent goodwill · tldraw-idea-stalled
Original location · source-90b048d01304e82e6fb1 · source-98df06b35a0fcf01f7a4 · source-178858b0cba70c16d626 · source-cf1c4c60f1205546d41f
Original structured fields
{ "detail": "the open-source goodwill was spent — 'open source is a spectrum,' the readme correction, the formerly-foss listings, the license-key gate. the community that made the 2021 launch warm now writes the comparisons that recolor tldraw as closed.", "id": "clm-stalled-goodwill", "sourceIds": [ "src-hp-license-update", "src-ci-isitreallyfoss", "src-ci-meetrix-oss", "src-ci-hn-sdk4" ], "title": "the spent goodwill" }the sunset pattern · tldraw-idea-stalled
Original location · source-ec1484dccade15e61d87 · source-75bc3cd74980ab366df3 · source-9bf4b496cfebb13fc5a1 · source-b0eb9ce12f605c2075e4
Original structured fields
{ "detail": "demos sunset faster than they convert — draw fast went down in january 2024, computer stayed closed source, fairies ran one december. the attention keeps arriving; the products it pointed at keep expiring.", "id": "clm-stalled-demos", "sourceIds": [ "src-hp-drawfast-site", "src-hp-computer-issue", "src-hp-fairies-site", "src-ci-rundown" ], "title": "the sunset pattern" }the pricing opacity · tldraw-idea-stalled
Original location · source-5dfed536365c48de7a0b · source-0180c84b437596224223 · source-09a1160301b136414bb9 · source-cf1c4c60f1205546d41f
Original structured fields
{ "detail": "pricing stays opaque — the protest issue, the 100-day trial, hobby and commercial paths that still route through sales. 'just ask for money' is easier to say than to publish a price for.", "id": "clm-stalled-pricing", "sourceIds": [ "src-ci-issue-8577", "src-ci-license-key-docs", "src-hp-uxtools", "src-ci-hn-sdk4" ], "title": "the pricing opacity" }the cadence nobody watches · tldraw-idea-stalled
Original location · source-780c11f9238f69650cbc · source-cf1c4c60f1205546d41f · source-d4909fc3d2be5a14264d
Original structured fields
{ "detail": "release news stopped being news — sdk 4.0's thread was a fight, sdk 5.0's was a shrug (1 point). hn attention now arrives only for demos and controversies, not for the product cadence.", "id": "clm-stalled-cadence", "sourceIds": [ "src-ci-hn-sdk5", "src-ci-hn-sdk4", "src-hp-sdk50" ], "title": "the cadence nobody watches" }excalidraw · tldraw-idea-successors
Original location · source-d5c46b24b087a89c7547 · source-cb61a660542cd54746f0 · source-178858b0cba70c16d626
Original structured fields
{ "href": "https://excalidraw.com", "id": "suc-excalidraw", "name": "excalidraw", "note": "the mit control group — the canvas that adopted perfect-freehand and now anchors every 'fair open-source pick' comparison tldraw loses.", "sourceIds": [ "src-ci-ossalt-compare", "src-ci-selfhosting-compare", "src-ci-meetrix-oss" ] }draw-a-ui · tldraw-idea-successors
Original location · source-2b81dfac8cf182dac144 · source-0676868dcb4aa6b3aad9 · source-79ab2993b3b272875b42
Original structured fields
{ "href": "https://github.com/SawyerHood/draw-a-ui", "id": "suc-draw-a-ui", "name": "draw-a-ui", "note": "the community seed — hood's mit-licensed sketch-to-html demo that tldraw forked into make real. the ecosystem makes the demos; the company packages them.", "sourceIds": [ "src-ci-draw-a-ui-repo", "src-ci-hn-draw-a-ui", "src-hp-makereal-story" ] }muse · tldraw-idea-successors
Original location · source-679234f10732200f4e19 · source-3a1f4235a952006eb88b
Original structured fields
{ "href": "https://museapp.com", "id": "suc-muse", "name": "muse", "note": "the predecessor's ghost — the thinking-tool canvas whose post-mortem is the thesis's control case, and whose founder backs tldraw.", "sourceIds": [ "src-ci-muse-retro", "src-hp-seed-blog" ] }the canvas cohort · tldraw-idea-successors
Original location · source-b8a5228d06ccd96bc34d · source-9d884d07151e468c8773
Original structured fields
{ "href": "https://launchpad.kosmik.app", "id": "suc-kosmik", "name": "the canvas cohort", "note": "kosmik and the app-shaped canvases that kept launching and sunsetting around the sdk — the field the infrastructure bet was made against.", "sourceIds": [ "src-ci-tc-kosmik", "src-ci-kosmik-sunset" ] }the agent-canvas category · tldraw-idea-successors
Original location · source-81e53b450674497ce864 · source-61e14e50eec664a4e09c · source-a4203fdba76346513679
Original structured fields
{ "href": "https://tldraw.dev/starter-kits/agent", "id": "suc-agent-canvas", "name": "the agent-canvas category", "note": "paper, replit agent 4, google stitch — the aiewf 'spatial harness' talk already treats the agent canvas as a category with tldraw inside it, not above it.", "sourceIds": [ "src-ci-spatial-harness", "src-ci-agent-starter", "src-ci-gemini-showcase" ] }the lab canvases · tldraw-idea-successors
Original location · source-f19ba1929c1c527876ec · source-765cc6f7a1ed080fc177
Original structured fields
{ "href": "https://patmcguinness.substack.com/p/mistrals-new-le-chat-and-the-canvas", "id": "suc-lab-canvases", "name": "the lab canvases", "note": "artifacts, chatgpt canvas, mistral's le chat — the frontier-lab versions of the same surface, owned by the model providers tldraw demos against.", "sourceIds": [ "src-ci-mistral-canvas-essay", "src-ci-interfacestudies-zui" ] }