Note apps come and go, but plain text files from 1980 still open today. Why plain text is the safest bet for notes you want to keep for decades.
· Privacy & Ownership · 7 min read
There is a file format that has survived every platform shift of the last fifty years. It opens on Windows, macOS, Linux, iOS, Android, and a Raspberry Pi. It needs no license, no account, and no particular vendor to stay alive. Files written in it during the 1980s open today exactly as they were written — and will open in 2076 too, on machines that have not been invented yet.
That format is plain text, and if you write notes you intend to keep for decades, it is very likely the only storage decision that actually matters. Apps are temporary. The average note-taking tool has a lifespan measured in years; your notes, if they are worth taking at all, have a lifespan measured in decades. Whatever app you use today, the honest question is not "is this the best app?" but "what happens to my notes when this app is gone?" Plain text is the answer that keeps working after every other answer expires.
Plain text is the simplest possible representation of writing: a sequence of characters, one after another, with nothing hidden. No embedded styling engine, no binary container, no schema that only one program understands. When people say a note app "stores notes as plain text," they mean you could open the file in any editor ever written and read your own words.
Two clarifications, because the term gets muddied:
#, **, []()) rather than invisible codes. That is precisely why Markdown became the default dialect for note-taking: it adds structure without giving up readability in a bare editor.The strongest case for plain text is not theoretical — it is the graveyard of formats that seemed permanent.
WordPerfect files from the early 1990s need conversion software to open today. Works documents, ClarisWorks, AppleWorks: effectively dead. Evernote's .enex format is XML, but stuffed with a private markup dialect that no other tool renders faithfully. OneNote sections are binary blobs. Every one of these was, at its peak, the "obvious" place to put your writing.
The pattern repeats because it is structural, not accidental. A proprietary format is a moat: it makes leaving expensive, which is good for the vendor and bad for you. When the vendor thrives, the moat locks you in; when the vendor dies, the moat becomes a grave. Plain text is the one format with no vendor to die.
Notes make this worse than other documents, because notes compound. A contract from 2009 you probably never need again. But a personal knowledge base is an asset whose value comes from accumulation — the note you wrote eight years ago is exactly the one you want your search to find today. A second brain that resets every time you switch apps is not a second brain; it is a series of abandoned diaries.
Every tool works on it. Search your notes with the operating system's built-in search, or grep, or any indexing tool. Version them with Git. Sync them with anything that moves files. Script against them. Feed them to an AI assistant. No plugin ecosystem required — the entire history of computing tooling is your plugin ecosystem.
Migration becomes trivial. Switching between plain-text-based apps is a copy-paste, or nothing at all if they read the same files. The switching cost that traps people in mediocre tools simply disappears, which — usefully — also keeps your current tool honest.
Corruption is survivable. When a binary database corrupts, you lose everything after the last backup. When a text file corrupts, you lose a few characters and keep the rest. Simple formats fail gracefully.
It is diff-able and mergeable. Two versions of a text file can be compared line by line and reconciled. This is why plain text pairs so naturally with local-first apps that sync between your own devices: merging is tractable when the data is transparent.
It is future-proof by construction, not by promise. No company has to stay in business, no standard body has to maintain a spec, no converter has to be written. The format is so simple that supporting it is unavoidable.
Plain text is not free of trade-offs, and pretending otherwise undersells the argument.
You give up rich embedded content: images, PDFs, and audio can only be referenced from a text file, not stored inside it, so attachments need a folder convention sitting next to your notes. Real-time collaborative editing is harder — plain text has no built-in change tracking, which is why shared team documents still tend to live in cloud editors. And some structure is genuinely awkward to express: a large spreadsheet-like table wants to be a spreadsheet, not a Markdown grid.
The practical resolution is scope. For the writing that constitutes your actual thinking — meeting notes, journals, project logs, ideas, reference notes — plain text covers essentially everything, and the exceptions (a few images, an occasional table) attach around the edges without compromising the core.
One subtlety worth untangling: format and storage location are independent axes, and you need both right.
Plain text files on someone else's server are still hostage to an account. A proprietary database on your own laptop is still hostage to one program. The durable combination is open format and storage you control — files or a local database on your device, with backups going to places you own.
This is also where a fair test has to cut against apps you like, not just apps you left. Indenta makes a useful example precisely because it sits in the middle. On the format axis it passes cleanly: the canonical copy of every note is the Markdown outline text, stored verbatim in the browser's local database and exported as that same text — there was never a private encoding to escape from. But the storage axis deserves honesty: a browser database is not a folder of files. You cannot point grep, Git, or a file-sync tool at it, and it lives inside a single browser profile that the browser itself manages. So the earlier warning — an export you must remember to run is a promise, not a property — applies in part here too. The export button, and the optional backup to your own cloud drive, are not courtesies in this design; they are the step that turns plain text in a database into plain text you hold. Run them on a schedule, rather than trusting any database — local or not — to be eternal.
What fails the test is the opposite arrangement: notes held in a private encoding, with "export to text" as a lossy afterthought that drops your links, tags, or structure on the way out. Run the export today, open the result in a bare text editor, and see whether your notes are still your notes. That five-minute test tells you more about an app's respect for your data than any feature list.
A reasonable way to make note-app decisions is to imagine reading your notes at eighty. Not the app — the notes. Which of today's choices still matter at that distance?
Features won't: every feature you love will be commonplace or obsolete. The company won't: it is unlikely to exist in its current form. What survives the fifty-year horizon is exactly two things — whether you built the habit of writing things down, and whether what you wrote is in a format that can still be read.
The first is up to you. The second is a decision you can make once, today, and never revisit: keep your notes in plain text, in storage you control, and let the apps come and go around them. The words are the asset. The format is what decides whether you get to keep it.
Try it in practice: Indenta is a free, offline-first outliner — nested notes, backlinks, tags, and peer-to-peer sync, with no account required. Start writing in your browser, or read the User Guide first.