The creator’s real content backlog isn’t in Canva — it’s in the notes app you never publish
Every social media operator I know has the same dirty secret: the best thinking lives in a private notes app, not on the feed. The carousel that would have crushed it is a half-finished outline in Obsidian. The LinkedIn post that would have started a genuine conversation is a three-line fragment you were too embarrassed to ship. Meanwhile the public-facing “content strategy” is a frantic Sunday-night scramble to reverse-engineer something publishable from whatever you can remember.
That gap — between the volume of private thinking and the trickle of public output — is the actual bottleneck in creator workflows, and it’s the problem Caldera is trying to attack. It’s a free, open-source Obsidian plugin that turns a private vault into a self-refreshing public “digital garden,” with cross-garden backlinks and signed authorship. It is not a social scheduling tool. But the mechanic it’s betting on — publish the rough draft, label its maturity, let the network link to it — is one social media teams should steal regardless of whether they ever install the plugin.
What Caldera actually is, and the problem it’s really solving
The maker, Michael Jones (who also builds Toonstar), frames the origin honestly: he keeps roughly 4,000 notes in Obsidian, and two things bugged him. His notes “never turned into anything other people could see,” and the public digital gardens that do exist are “islands” — you can’t link from your garden into a friend’s. Caldera 0.5 addresses both. It reads your vault on-device and publishes a living profile — daily journal, activity, timeline, projects — that refreshes itself every few hours. Notes dropped into a Garden folder become individual public pages, and [[@handle/Note]] links connect gardens with backlinks on both sides. Every note is hashed into a signed manifest so authorship travels with it. The plugin is MIT-licensed, the hub is AGPL, and setup is a single prompt you paste into Claude Code, Codex, or Cursor.
The problem it solves is not “how do I post more.” It’s “how do I make the thinking I already do legible to other people without turning it into a content-production project.” For a social media manager, that’s a meaningful reframe. Most of us treat publishing as a manufacturing step — draft, design, schedule, distribute. Caldera treats publishing as a state change: a note goes from private to public by being moved into a folder or flagged caldera: garden. That’s a much lower activation energy than “write a post,” and lower activation energy is the whole game when the constraint is your own avoidance.
Why the “seedling / budding / evergreen” labels matter more than the tech
The detail that generated the most genuine enthusiasm in the launch comments wasn’t the signed manifests — it was the maturity labels. One commenter, Jaewon Seo, put it plainly: “mostly it’s knowing nobody’s gonna judge the half finished ones, i’ve got way more unfinished notes than polished ones.” Another, Shubham Nair, said the labels “hit home” because his vault is “full of notes I keep meaning to revisit but never quite finish.” The maker’s reply to Seo is the thesis in one line: “Seedlings are meant to be rough. Plant the half-finished ones.”
This is the most transferable idea in the whole launch, and it has nothing to do with Obsidian. Social platforms have spent a decade training creators to treat every publish as a final, polished artifact — which is exactly why so much good thinking never ships. The digital-garden convention of labeling work by maturity (seedling → budding → evergreen) is a psychological unlock. I’ve watched teams adopt a version of this on internal wikis and Notion dashboards and ship noticeably more, simply because “draft” stopped feeling like a failure state. If you take one thing from Caldera, take the labeling convention and apply it to your content calendar.
How it differs from the tools you’re probably already using
The honest comparison set here isn’t Buffer, Hootsuite, or Later — those are distribution and scheduling layers, and Caldera doesn’t touch them. The closer analogues are Obsidian Publish (Obsidian’s own paid publishing add-on), Notion public pages, and the broader “digital garden” ecosystem around tools like Quartz and Foam. Against Obsidian Publish specifically, the differentiators are: it’s free and open source, the hub is self-hostable (a commenter, Raaed Muggo, flagged self-hosting as “a big plus” because he doesn’t want his notes “depending on yet another service staying online”), and it adds cross-garden backlinks and signed authorship that the first-party option doesn’t offer.
Against Notion public pages, the difference is philosophical. Notion is a database you happen to share; Caldera is a graph you publish, where the links between notes — and between people — are the product. The signed manifest is the genuinely novel bit: every note is hashed, so a public note carries a verifiable author and date without requiring an account. As commenter Gene Dai noted, that’s “a thoughtful touch, since it gives a public note a verifiable author and date without needing an account.” For creators who’ve been burned by content scrapers or AI training crawlers lifting their work, provenance-by-default is a real feature, not a gimmick.
Where the social-media parallel actually breaks down
Here’s where I’ll be the skeptic. Caldera is a publishing tool, not a distribution tool. It has no scheduling, no analytics, no UTM handling, no cross-posting to Instagram, TikTok, YouTube, X, LinkedIn, Facebook, Threads, or Pinterest. It won’t tell you watch time, engagement rate, or which hook drove saves. If your job is to grow a following on the major platforms, Caldera does nothing for that directly. What it does is feed the top of your content funnel — the ideation and drafting layer — and give you a public home base to point social traffic toward. That’s a legitimate role, but it’s a different role than the one a growth marketer is hired to fill.
What creators and social teams can borrow from it
The most useful thing about studying a tool like this is the workflow it implies, because you can replicate most of it with tools you already pay for. Here’s what I’d actually lift.
Treat your private notes as a content reserve, not a graveyard. The maker’s whole complaint — 4,000 notes that “never turned into anything other people could see” — is the default state of every creator’s second brain. I’ve fallen into this myself: a vault full of hooks, half-written threads, and research that never left the app. The fix isn’t a new tool; it’s a weekly ritual where you scan for “seedlings” and promote two or three into drafts. I’d bet most operators could double their publishing cadence in a month just by mining their own backlog instead of staring at a blank calendar.
Adopt maturity labels in your content pipeline. Whether you use a Notion board, an Airtable base, or a Trello list, add a status field that distinguishes “rough idea” from “ready to schedule” from “evergreen / republish candidate.” This does two things: it makes rough work feel permissible (the Seo/Nair problem), and it gives you a repurposing queue. Evergreen posts are the ones you recycle across CapCut-edited shorts, carousels, and newsletter sections months later — labeling them makes that obvious.
Use signed provenance as a trust signal. You don’t need a signed manifest to borrow the idea. If you publish original research, frameworks, or data, timestamp and attribute your work publicly — a canonical URL, a visible publish date, a byline that travels with the content. In an era where AI scrapers and content farms lift creator work constantly, making authorship legible is a small defensive moat and a credibility signal for your audience.
Point social traffic at a home base you own. The strategic reason to publish a garden is the same reason smart creators push email lists: platform risk. If your entire presence lives inside TikTok’s algorithm or LinkedIn’s feed ranking, a single distribution change can wipe your reach overnight. A public, self-hosted home base — whether that’s Caldera, Obsidian Publish, or a simple Ghost or Substack site — is where you send the people you convert on social so the relationship survives the next algorithm shift.
Why TikTok creators should care more than LinkedIn ones
Counterintuitively, the creator most likely to benefit from a tool like this isn’t the LinkedIn thought-leader — it’s the TikTok or YouTube creator whose content is heavily researched. Short-form video lives and dies on hooks and scripts, and those scripts are written somewhere before they’re filmed. A garden of research notes, hooks that tested, and audience questions becomes a script bank. The LinkedIn crowd already has a natural publishing surface (the feed tolerates long text), so the marginal value of a public garden is lower. The video creator, whose drafting happens in private and whose output is expensive to produce, gets more leverage from making that private drafting searchable and linkable. That’s my take, not the maker’s — but it follows from where the production cost actually sits.
Where my judgment says it falls short
I’ll be direct about the limitations, because a launch page won’t be.
The privacy edge cases are unresolved. A commenter, Neeraj Nathany, asked the sharpest question on the page: if you later make a linked note private, does the other side keep a stub or does the link just die? The maker’s answer was refreshingly honest: “today the link just goes dead, there’s no stub. A small ‘this note was made private’ stub is a good idea and it’s on the list.” That’s a real gap for anyone publishing work they might later retract. Similarly, Manuel Lorenzo Bouzada asked what happens to copies already shared — and the answer is the honest, unsurprising one: unpublishing removes the page from the hub on the next publish, but anything already copied or saved can’t be pulled back, “same as any website.” If your content strategy depends on being able to fully retract published material, understand that no publishing tool can promise that.
Version-pinning doesn’t exist yet. Gene Dai asked whether a cross-garden link follows the latest version of a note or the hash you saw — and whether you can pin it “the way a citation does.” The maker confirmed links follow the latest version “like a URL,” and that pinning to a specific signed version is “very doable” and “on the list” as an option. Until it ships, cross-garden links are living references, not stable citations. For academic-adjacent creators or anyone citing specific data points, that matters.
Setup assumes an AI coding tool. The one-prompt install via Claude Code, Codex, or Cursor is elegant if you already live in those tools and a genuine barrier if you don’t. A social media manager who’s never opened a terminal is not the target user here, and the maker doesn’t pretend otherwise. This is a tool for the Obsidian-native, developer-adjacent creator — not the Canva-first social coordinator.
It is not a growth tool, and the maker isn’t claiming it is. No analytics, no scheduling, no platform integrations. If you came looking for a Buffer or Later replacement, this isn’t it. The team’s claims are modest — a “living profile that refreshes itself every few hours” — and I’d treat any implied reach benefits as unproven. There’s no disclosed user count, no revenue figures, no performance data. That’s not a knock; it’s just the honest state of a young open-source project.
What I’d watch / test next
If you run social for a living, here’s what I’d actually do this week — none of it requires installing Caldera, though trying it is a reasonable side quest if you’re Obsidian-native.
First, audit your private notes for “seedlings.” Spend 30 minutes scanning your notes app, docs, or drafts folder and flag five fragments you could promote into a post, a script, or a carousel this month. That’s your content reserve, and it’s probably bigger than your calendar suggests.
Second, add a maturity field to your content pipeline — even a simple three-state status — and start labeling rough ideas as rough. Watch whether it changes how much you ship.
Third, pick one home base you own and make sure every social profile links to it. If you want to test the digital-garden approach specifically, spin up Caldera on a spare vault (the maker’s one-prompt install makes this low-risk) and see whether publishing rough notes publicly feels liberating or terrifying — that reaction tells you a lot about your own publishing psychology.
What I’d watch on the product itself: whether the “made private” stub and version-pinning ship, because those two features determine whether cross-garden links become trustworthy citations or just fragile URLs. And I’d watch whether the open-source, self-hostable model attracts the kind of community that keeps a small tool alive past launch week — because that, more than any feature, is what determines whether your public garden is still standing in two years.
