The Boring Infrastructure Nobody Wants to Talk About
Cross-border sellers obsess over the wrong stack. We spend weeks comparing Shopify themes and Amazon PPC bids, then run our entire operation on a pile of half-synced notes scattered across Slack, Notion, and someone’s personal Google Drive. The listing copy that took three rounds of compliance review, the supplier spec sheet, the returns policy that varies by marketplace — all of it lives in proprietary silos that break the moment a VA leaves or a subscription lapses. So when a small, deliberately unglamorous tool shows up on Product Hunt arguing that your files should just live in your own folders, in plain text, forever, I pay attention. That tool is hibi, a free, open-source Markdown/LaTeX/Typst editor from maker Ryana May Que, and the reason it matters to operators has almost nothing to do with writing prose.
What hibi actually solves — and why it’s an ops problem, not a writing problem
Strip away the launch-page framing and hibi is a local-first text editor. Files stay in folders you control, on your machine; you open an existing directory and start editing. The maker describes it as free and open-source, supporting Markdown, LaTeX, Typst “and more,” and — importantly — as “somewhat in a stable alpha state,” with an explicit warning to back up your notes first. That honesty is refreshing and also a flag I’ll come back to.
For a cross-border operator, the interesting property isn’t the editor. It’s the file format and the storage model. Markdown in a local folder is the closest thing we have to a vendor-neutral data layer. Your product research doc, your supplier contact sheet, your per-marketplace compliance checklist, your TikTok Shop content calendar — written as plain text, they survive tool migrations, they diff cleanly in Git, they’re greppable across thousands of files, and they can be read by any AI agent you point at the folder. That last point is the one most sellers are sleeping on: an LLM can ingest a directory of Markdown far more reliably than it can scrape a proprietary database behind a login.
Why Amazon sellers should care more than Shopify ones
Shopify-first DTC brands tend to centralize in one admin and one theme repo, so their documentation pain is lower. Amazon FBA sellers are the opposite: they juggle Amazon Seller Central listing copy, A+ content drafts, brand-registry assets, per-country compliance docs, and a separate Helium 10 research trail — often across multiple marketplaces and multiple team members. That’s a documentation surface area begging for plain-text, folder-based organization. A shared, version-controlled folder of Markdown specs is worth more to a five-marketplace FBA operation than any note app’s fancy graph view.
How it differs from the incumbents you’re probably already paying for
The obvious comparison is Obsidian, and the Product Hunt thread makes it explicit: a commenter asks whether they can open an existing Obsidian vault or must convert first, and the maker answers “you can just open it outright.” That’s a meaningful compatibility signal — Obsidian vaults are already just folders of Markdown, so hibi slots into an existing local-first workflow rather than demanding a migration.
Against Notion, the contrast is philosophical. Notion is a hosted database with a beautiful interface and a hard dependency on Notion’s servers and pricing. hibi is the inverse: ugly-duckling local files, zero lock-in. For a seller whose Notion workspace holds supplier terms and margin math, the lock-in risk is real — export fidelity from database-style tools is notoriously lossy. Against a plain editor like VS Code, hibi’s bet is that writers and documentarians want a focused editor rather than an IDE, and its LaTeX/Typst support targets people producing formatted output (spec sheets, printable labels, PDFs) rather than code.
The genuinely differentiated bit, per the thread, is breadth of format: a commenter notes that going “beyond markdown to Typst and LaTeX” is “the gap I feel with most note apps.” For an operator generating print-ready supplier POs or formatted compliance PDFs from source text, that’s a real workflow, not a novelty.
What cross-border sellers can borrow from this launch
Three transferable ideas, regardless of whether you ever install hibi:
1. Treat your operational knowledge as portable files, not app state. The single highest-leverage migration any seller can do this quarter is moving critical SOPs — supplier onboarding, listing QA checklists, returns triage trees — out of chat threads and into a versioned folder of Markdown. You don’t need hibi to do it; you need the discipline. The tool just makes the discipline pleasant.
2. Quick capture beats perfect organization. One commenter nails the real failure mode: “i lose half my notes cause opening the app and finding the right folder takes too many clicks before i even start typing.” Every operator knows this pain — the supplier quote that arrives in a WeChat voice note, the ad-hoc idea during a factory call. Whatever stack you pick, optimize for time-to-capture, because a note you didn’t write down is worth zero.
3. Own your AI-ready data layer. As more sellers wire LLMs into listing generation and review analysis, the winners will be the ones whose institutional knowledge is machine-readable. A folder of clean Markdown is the cheapest RAG corpus you’ll ever build.
Where my judgment says it falls short
I’ll be blunt: this is a hobby project in alpha, and the maker says so. The download page is a nightly build tagged “nightly-green,” with an explicit “back up your notes first” warning. For a seller whose supplier contracts and margin models live in these files, “back up first” is not a production-grade posture. There’s no disclosed pricing beyond “free,” no disclosed team beyond the maker, and no disclosed roadmap or SLA — because there isn’t one. The community-addon question in the thread (“Can the community build addons today, or is the API still settling?”) went unanswered in the scrape, which tells you the extension ecosystem is immature.
Where the math breaks
The economics of “free and open-source” cut both ways. Zero license cost is great for a bootstrapped seller; zero vendor accountability is not. If hibi stalls, your files survive — that’s the whole point of local-first — but your editing workflow doesn’t. Compare that to paying for a hosted tool where the tradeoff is lock-in for continuity. Neither is free; you’re just choosing which risk to carry. For a solo operator, hibi’s risk profile is acceptable. For a ten-person brand with VAs in three time zones, I’d want a paid support path that doesn’t exist yet.
The other honest gap: hibi is an editor, not an operations platform. It won’t sync across your warehouse team, it won’t integrate with Klaviyo or your 3PL, and it won’t replace the marketplace-native tools where listing data actually has to live. It’s a layer beneath those, and sellers who mistake it for a Notion replacement will be disappointed.
What I’d watch / test next
This week, do a one-hour audit: pick your three most business-critical documents — say, your supplier terms, your per-marketplace compliance checklist, and your returns policy — and confirm each exists as a portable file you could hand to a new hire or an AI agent tomorrow. If they’re trapped in a proprietary tool, export them to Markdown and store them in a folder you control. Then, if you want to try hibi, install a nightly build in a sandbox directory with throwaway files, not your real SOPs, and test two things: how fast quick-capture actually is from a cold start, and whether it opens an existing Obsidian-style vault without corruption. Watch the GitHub repo for whether the addon/API question gets answered — that’s the signal for whether this becomes a platform or stays a personal project. And keep your real operational knowledge in plain files either way, because that decision pays off no matter which editor wins.
