Open any popular notes app and watch what happens when you type. The words leave your keyboard, get serialized, get encrypted (hopefully), and travel across the internet to a data center you'll never see. They live there. Your notes app is really a window into a remote database with your name on it.

This is so normal we don't even notice it anymore. But it's a fairly recent default, and it costs more than most people realize.

What "the cloud" actually is

"The cloud" is just someone else's computer. When your notes live there, three things become true at the same time:

• You don't fully own them. You access them under terms of service that can change. If the company shuts down, gets acquired, or bans your account, your notes can become inaccessible overnight.
• You depend on a connection. No internet, no notes. Most apps cache locally for offline use, but the canonical copy is somewhere else.
• You inherit their security posture. Every breach of their servers is a breach of your notes. You can't audit their employees, their access controls, or their infrastructure.

The local-first principle

Local-first software flips the relationship: your device holds the canonical copy. The network, if used at all, is for sync between your own devices — not for storage on someone else's.

The data lives where you live. The app is a tool you use on it, not a gateway to access it.

This isn't a new idea. It's how every desktop app worked before the cloud era. What changed isn't that local-first stopped making sense — it's that selling cloud subscriptions makes more money than selling apps once.

What you get back

When notes are local-first:

• They work everywhere. Airplane mode, dead Wi-Fi, no signal — none of it matters.
• They can't be scanned. Server-side scanning for ads, AI training, or "safety" requires the data to be on a server in the first place.
• They don't disappear. Account bans, service shutdowns, and billing failures can't lock you out of your own writing.
• They're not part of a breach surface. The biggest breaches of the last decade were all server-side. Notes that never leave your phone are not in those headlines.

The honest tradeoff

Local-first isn't free. The main thing you give up is automatic cross-device sync. If you want your notes on your phone and your laptop simultaneously, someone has to coordinate that.

The cloud's pitch is: let us coordinate it for you, in exchange for permanent custody of your data. That's a real convenience, and for some workflows it's the right call. But for the bulk of what most people put in a notes app — daily journaling, lists, drafts, half-formed thoughts, financial scratch work, medical notes, passwords-you-shouldn't-store-here-but-do — the convenience isn't worth the custody trade.

How BetterNotes implements it

BetterNotes stores every note, checklist, RSVP reader, drawing, folder, and setting in your device's private storage. There is no account. There is no sync server. There is no cloud bucket with your username on it, because we never created one.

If you lose your phone, your notes are gone — which is the same risk you take with any local-first tool, and is the price of the guarantees above. We recommend Android's built-in backup mechanisms if you want a safety net.

This is the version of the app we wanted to use ourselves. We think it's the version more people will want, once they realize the alternative was never inevitable.

Most adults read between 200 and 300 words per minute. That number has barely budged in a century, despite vast improvements in education, lighting, and the size and clarity of the text we read. The bottleneck isn't your eyes. It's how your eyes move.

Why normal reading is slow

When you read a line of text, your eyes don't glide smoothly across it. They jump — small, fast movements called saccades — and pause briefly on roughly every other word in what are called fixations. During the saccade itself, you're effectively blind. Each pause is short (about 200–250 ms), but they add up.

Worse, you also regress: your eyes flick back to re-read words you've already passed. Studies estimate 10–15% of reading time is spent re-reading the same words, often without you noticing.

Stacked on top of all this is subvocalization — the silent inner voice that pronounces each word as you read it. Subvocalization is capped by the speed of human speech (~250 wpm), which is roughly where unaided reading plateaus.

What RSVP does differently

Rapid Serial Visual Presentation flashes words one at a time at a fixed point on the screen. Three things happen at once:

• Saccades disappear. Your eyes stay still. The words come to you.
• Regressions become impossible. You can't dart back to a word that's already gone. (Most RSVP readers let you pause or rewind manually, which is healthier than the unconscious habit.)
• Subvocalization weakens. At high enough speeds, your inner voice can't keep up. Comprehension shifts from hearing the words to recognizing them visually.

The combined effect: most people can comfortably read at 400–600 wpm with RSVP after a short adjustment period. Some go much higher.

What the research actually says

RSVP has been studied since the 1970s. The short version of the literature:

• At moderate speeds (300–500 wpm), comprehension is essentially identical to normal reading.
• At high speeds (600–800 wpm), comprehension drops on complex material but stays strong on narrative or familiar text.
• At very high speeds (1000+ wpm), comprehension degrades noticeably. The apps that advertise "read War and Peace in an afternoon" are oversold.

The honest framing: RSVP is a powerful tool for the middle range, not a magic trick that abolishes the cognitive cost of understanding ideas.

What it's good for

• Long articles, blog posts, and newsletters you'd otherwise skim.
• Reference material and documentation where you mostly know the shape of the content.
• Re-reading material you've already studied once.
• Reducing eye fatigue on long sessions — your eyes literally don't move.

What it's not good for

• Dense technical writing where you need to pause and reason about each sentence.
• Poetry, where the line breaks and rhythm are the meaning.
• Material with diagrams, equations, or footnotes — RSVP is a stream, not a page.

How BetterNotes' reader works

Paste any text into a BetterNotes RSVP reader and it plays back word-by-word at a speed you control. The reading stays entirely on your device — we don't send the text anywhere to "process" it, because there's nothing to process.

Every RSVP reader in BetterNotes is free and unlimited — create as many as you want, with or without Premium. There's no note cap to work around here.

About this passage

The text below is an original 1,000-word essay called "A Brief History of Note-Taking," written for this demo. It's deliberately a mix of narrative and reference material so you can feel where RSVP shines and where it gets challenging.

If you want to follow along, the full text is here. Most people find that after a minute at 400 wpm the words stop feeling like a strobe and start feeling like a voice.

A Short Story of Note-Taking

People have written things down for as long as we have written at all. Long before printed books, even before paper, there was the urge to take a thought out of your head and put it somewhere you could find it later. That urge is one of the oldest parts of how we think.

Many early notes were not important on their own. They were scratched into soft clay by a single writer — practice marks, fixes, half-finished ideas, things meant to come back to and never followed up on. The notes were rough, and they were honest. The first note-takers were not trying to make books. They were just trying to remember.

Later, people used flat wooden boards covered in wax. A pointed stick wrote on one side, and a flat blade on the other side smoothed the wax clean so you could start again. It was a draft surface. You could think on it, change your mind, and try again, all without using up paper, which at the time was expensive and hard to come by. Many ideas that ended up in famous books started as scratches on wax that the writer planned to erase.

Hundreds of years later, in quiet rooms where books were copied by hand, a new kind of writing showed up: small notes in the margins. A reader copying a long book would jot down a thought as they worked. A correction. A pointer to another page. Sometimes just a line about the cold weather or the bad ink. These small notes were never the main point of the page. But they were honest, and they showed what a careful reader was actually thinking. Long after, people studying the same books would read those side notes and learn more from them than from the polished text itself.

Once paper became cheap, people started keeping their own personal notebooks. Shopkeepers used them for money. Students used them to collect quotes and ideas, sorted by topic instead of by source. These books were not diaries, and they were not work logs. They were a personal list of ideas — built up slowly, over years — that a person could draw from later when they had something of their own to write.

Some of the most creative people in history kept this kind of notebook. Sketches of inventions sat next to grocery lists. Drawings of the human body shared a page with recipes for mixing paint. The notebooks were never made for the public. They were a workshop for the mind, and their value came from the fact that no one else was reading them.

In more recent centuries, the small pocket notebook became popular. Cheap, light, easy to carry. People used them everywhere — at work, on the road, on long trips into places that no one back home had ever seen. A traveler far from home would write down what they saw, without knowing which detail would matter later. They wrote it all down anyway. Many ideas we now take for granted started as a single line in a notebook that no one else read for years.

In the last hundred years, note-taking became something people studied on purpose. Some kept their notes on small cards, one idea per card, so the cards could be sorted and re-sorted and pulled together in new ways. A writer's stack of cards turned into a thinking partner. Cards pointed to other cards. Patterns showed up that no one had planned. Whole books grew out of the shape of the notes, not the other way around.

Then came computers, and everything changed at once. The earliest digital notes were simple text files. They were faster than paper, but not much better. Then came links between notes, search across all of them at once, and tags to group them. Suddenly the limits were no longer about how much paper you had, or where you could store it. You could keep more notes than any person could ever read.

But something quiet was lost. Digital notes were no longer fully yours. They lived in someone else's system, behind someone else's login, under rules someone else could change. The most personal kind of writing — the place where you put thoughts you would never say out loud — was now being scanned for ads, fed into machines that learned from it, and stored on servers in places you would never visit.

Many people did not mind. The trade-off was real. You got automatic backups, your notes on every device, smart search, and helpful tools built into the writing. For many uses, that was worth it.

But for the older purpose of a notebook — to be a private place where unfinished thoughts could exist safely until they were ready — the cloud was the wrong shape. A wax tablet did not phone home. A pocket notebook did not send anything to anyone. The whole point was that no one else was looking.

This is the older idea that BetterNotes belongs to. Not the cloud-first, account-based, sync-everywhere model that fills the app stores. The simpler one. The one where a notebook is a thing you own, kept somewhere you control, used by a mind that does not need an audience to think.

The surface has changed many times. Clay, then wax, then animal skin, then paper, then a screen. What stays the same is the relationship: one person, alone with their own thoughts, writing them down before they are gone. The tool exists to serve that relationship. When the tool stops doing that, it is time for a new tool.

That is what we are trying to be.

"We don't collect your data" is one of the most overused claims in software marketing. It's said by apps that sync everything you write to their servers, by apps that ship third-party trackers in their SDKs, and by apps that quietly upload "anonymized" usage telemetry by default. The phrase has been so diluted that it barely means anything anymore.

So let's say what it means here, with no hedging.

What BetterNotes doesn't collect

• No account. There is nothing to sign up for. We never see an email address, a phone number, or a username.
• No analytics SDKs. No Firebase Analytics, no Mixpanel, no Amplitude, no homegrown event logger.
• No third-party trackers. No Facebook Pixel, no AdMob, no advertising IDs, no fingerprinting.
• No crash telemetry from us. We don't ship a crash reporter SDK. (More on what Google Play does below.)
• No content uploads. Notes, checklists, RSVP text, folders, drawings, and settings stay on your device. There is no server they're sent to, because we don't run one.
• No biometric data. The lock feature uses Android's BiometricPrompt API. Your fingerprint or face never leaves your device's secure enclave, and the app never sees it.
• No permissions. The app requests no Android permissions. No internet access, no camera, no microphone, no contacts, no location.

What Google Play does, and why we have to mention it

Publishing on the Play Store means accepting a layer of monitoring that Google performs at the platform level — across every app on the store, automatically, with no opt-out for the developer. We don't configure it, we don't receive the data, and we don't control it. But it exists, and pretending otherwise would be dishonest.

The main pieces:

• Crash reports and ANRs — Google Play Console aggregates "Application Not Responding" events and crashes across all installs. The developer sees aggregate stability data, never per-user identity.
• App signing — Google signs each release to verify integrity.
• Play Protect scanning — Google scans installed apps for security threats.
• Play SDK scanning — Google scans the SDKs we bundle for known vulnerabilities.
• Spam, anomaly, and fairness protection — Google polices ratings, reviews, and ranking signals.

These are governed by Google's Privacy Policy, not ours. We mention them in our Privacy Policy because you deserve a complete picture, even of the parts we don't run.

Why this distinction matters

A lot of "privacy-first" notes apps claim end-to-end encryption and call it done. End-to-end encryption is excellent, but it's a defense for data that's already being transmitted. It does nothing about the fundamental question of whether transmission needs to happen at all.

The most private data is data that was never sent anywhere.

Notes are unusually personal. People keep medical symptoms, financial scratch work, journal entries about relationships, half-thoughts they'd never say out loud. The right architecture for that data isn't "encrypted in transit and at rest on someone else's server." It's "never on anyone else's server in the first place."

The tradeoff we accept

Not collecting data means we can't see what's broken until you tell us. We don't know which features are popular. We don't know which screen people get stuck on. When a user reports a bug, we genuinely don't have logs to look up — we have to reproduce it from scratch.

That's a cost we'd rather pay than ask you to pay it for us. If you find something wrong, email tachyon@tachyondev.net — that's the support channel, and the absence of telemetry is exactly why it matters.

Most of the paper you deal with in a given month was never meant to be kept forever. A receipt for something you might return. A doctor's note your employer wants on file. A whiteboard sketch from a meeting nobody typed up. A signed lease page a landlord asked for "just for our records." None of it belongs in a filing cabinet, but all of it needs to exist somewhere, in a form someone else can actually open, at least for a while.

A photo of a document isn't a document

The instinct is to just take a picture. It's fast, the camera's already in your pocket, and for a few seconds it feels like the problem is solved. It isn't, and the difference shows up the moment you try to do anything with the result.

• No fixed page order. A three-page form becomes three separate photos, sitting wherever the camera roll happened to put them — not necessarily next to each other, and with no record of which page came first.
• Perspective and lighting artifacts. Desks aren't camera stands. The page comes out keystoned, slightly rotated, with a glare streak across the text you actually needed to read.
• Nothing to hand off as a unit. You can't attach "three photos" to an email and expect the other person to reconstruct your form in the right order. You can attach one PDF.
• It competes for space with everything else. A scan of your insurance card lives in the same undifferentiated stream as vacation photos and screenshots, getting harder to find every week that passes.

Why PDF, specifically

PDF won this fight decades ago for reasons that still hold. A page rendered as PDF has a fixed size and layout that looks the same on every device and prints the same way every time — unlike an image, which just gets bigger or smaller depending on the screen. A multi-page document is one file instead of a folder of loose images, in the order you actually scanned them. And unlike a photo that gets re-compressed every time it's forwarded through a messaging app, a PDF page doesn't quietly lose quality with each hop.

None of this is exotic. It's the same reason contracts, tax forms, boarding passes, and academic papers have shipped as PDFs for thirty years — it's a format built for documents that need to look identical no matter who opens them or how.

The camera roll is not an archive

Document photos accumulate quietly and never really leave. There's no natural boundary between "receipt from March" and the hundred other photos taken that month, no folder structure, no lock. A routine "clear out my screenshots" pass can just as easily take a warranty photo with it. And when you actually need one of these — tax season, a warranty claim, a dispute with a landlord — you're scrolling through a camera roll trying to remember roughly when you took it, instead of opening a folder that was built to hold exactly this kind of thing.

How BetterNotes' Documents section supplements this

This is the gap the Documents section is built to close. The built-in scanner does automatic edge detection and perspective correction, so a page photographed at an angle on a cluttered desk still comes out cropped, straightened, and readable — not just a photo, but an actual scan. Multi-page documents assemble directly into a single PDF as you go, in the order you scanned them, instead of scattering across separate images.

If a page comes out wrong, you don't have to start over — pages can be reordered, deleted, or individually retaken after the fact. Photos or PDFs you already had before you started using the app can be imported and appended alongside scanned pages, so nothing has to be rescanned just to end up in the same place. Every document lives in the same folder structure and optional biometric lock as the rest of BetterNotes, instead of sitting exposed in a general-purpose photo library. When a document's finished, it can be downloaded straight to your device or shared as an actual PDF, in the format whatever you're sending it to already expects.

And in keeping with how the rest of the app works, none of it leaves your device to get "processed" somewhere else — scanning and page rendering both happen locally.