The Self-Print Economy

1. The Limits and Freedom of Physical QR

TokiStorage's physical QR products — UV-resistant laminate and quartz glass — are designed to preserve voices for 1,000 years. Quartz glass QR continues to be manufactured exclusively by TokiStorage. For laminated QR, however, the publication of DIY guides has opened a path for partners and users to manufacture their own. That said, ensuring quality and durability requires understanding of materials and precision in the process.

Manufacturing physical QR takes care. Sourcing materials, managing print quality, precision in lamination. Because it resists mass production, each piece carries meaning.

However, not everyone needs 1,000 years of durability. Someone needs 30 copies of a wedding speech printed by next week. A teacher wants to give each student a personal message. A parent wants to record their child's voice today and stick the QR code on the fridge today. What these people need is not the permanence of quartz glass, but the means to make it real, right now, with their own printer and their own judgment.

2. Bulk Mode as a Watershed

TokiQR's normal mode fits data into a single QR code. Up to approximately 29 seconds at Codec2's 450 bps mode. This is free, with no restrictions. Anyone can use it, as many times as they want, without an account. This principle remains unchanged.

Bulk mode breaks through the single-QR barrier. By splitting data across multiple QR codes, it handles longer audio, higher-quality audio (with Opus codec support), and higher-resolution images. A three-minute speech, a ten-minute reading — all can take shape as a set of QR codes.

What matters here is that bulk mode is not an "upgraded version" of normal mode. Normal mode pursues maximum density within the constraint of a single QR code. Bulk mode removes the constraint itself by increasing the number of codes. The design philosophies differ. The former is about compression; the latter, expansion. It is not a question of which is superior, but of what you want to preserve.

And QR codes generated in bulk mode are for you to print yourself. TokiStorage does not manufacture them. On your own printer, on your own paper, at your own timing. This is the essence of the self-print concept.

3. The Experience Bulk Mode Unlocks

Let the numbers speak. In normal mode, the data capacity of a single QR code is approximately 2,140 bytes. For audio, that means a maximum of about 29 seconds at Codec2's 450 bps mode. For images, a maximum of 320×320 pixels, roughly 2 KB of WebP. For text, approximately 2,000 characters. This is the physical ceiling of a single QR code.

Bulk mode removes that ceiling. Data is split across multiple QR codes, stacking approximately 2,140 bytes per code. 24 codes yield about 50 KB; 48 codes about 100 KB. Add more codes, and the capacity has no upper bound.

The Impact of the Opus Codec

Normal mode audio is encoded with Codec2. Codec2 is an audio codec compressed to the extreme — at 450 bps, just 57 bytes per second. It can pack 29 seconds of voice into a single QR, but the sound quality is robotic. You can tell who is speaking, but the subtleties of emotion, the nuances of tone, are lost.

Bulk mode unlocks the Opus codec. Opus is the industry-standard codec used in VoIP calls worldwide, Discord, YouTube, and Spotify. At the 12 kbps mode, the voice comes through crisp and clear — not robotic, but unmistakably human. The difference from Codec2 is so stark it hardly sounds like the same person. Laughter, sighs, trembling in the voice, the trailing echo of a sentence — emotional details that vanish in Codec2 are preserved intact in Opus.

And there is no time limit. You can record directly in the browser, or select an audio file from your device. A twenty-minute speech recording, a one-hour live performance — any audio can be turned into QR codes as-is. At Opus 12 kbps, approximately 1.5 KB per second. 48 QR codes hold about 64 seconds. 96 codes hold over two minutes. Add more and it keeps growing. A five-minute speech, a ten-minute reading, a thirty-minute lecture — all possible.

There is another decisive difference. Codec2 is a codec specialized for human speech. It handles spoken voice, but other sounds — singing, instruments, music — are structurally beyond its reach and cannot be faithfully reproduced. Opus has no such constraint. Not just voice, but songs, melodies, instrumental performances, a child humming, birdsong, the sound of waves, the sound of wind. Any "sound" can be sealed inside a QR code.

What this means is that what a QR code can preserve expands from "voice" to "sonic memory." This single difference fundamentally changes the reach of the service.

Imagine a five-year-old daughter playing "Twinkle, Twinkle, Little Star" at her piano recital. Hesitant fingers, the pause where she stops and starts over, the applause when she makes it to the end. Codec2 can only preserve speech. Opus preserves the air of that hall.

Imagine a summer evening on your grandfather's veranda. The cry of evening cicadas, the chime of a wind bell, the distant beat of a festival drum. Your grandfather is gone now. The house has been torn down. But if that sound remains, that place still exists.

Imagine a wedding speech. The voice trembling, words catching midway, yet read through to the end — all of it preserved, tone and timing included. A toast read as text and a toast heard as a voice are entirely different things.

Imagine the din of a foreign market. Voices haggling over prices, a radio drifting from a spice stall, motorbike horns, the distant call to prayer from a mosque. What a photograph cannot capture — the "sonic air" of a place — becomes a QR code in its entirety.

Twenty-nine seconds of Codec2 is for leaving a brief message. Opus in bulk mode is for preserving an entire experience of sound. You, too, must have sounds you want to keep.

Images Unbound

Images undergo the same structural liberation. Normal mode images max out at 320×320 pixels, roughly 2 KB of WebP. That is thumbnail-scale — you can barely make out a person's expression, and background detail is crushed.

In bulk mode, there is no resolution limit. No height or width constraints. Add more QR codes, and any size image can be stored. A 1024×1024-pixel photo at 85% WebP quality is approximately 100 KB, fitting in about 48 QR codes. Need higher resolution? Just add more codes. Roughly 50 times or more the data of normal mode transforms an image from "symbol" to "record." You can read each person's expression in a family photo. Depth perception in a landscape comes through. Handwritten text becomes legible.

Aspect ratio is also free. Normal mode forces compression into near-square dimensions, but bulk mode handles tall scroll images and wide panorama photographs as they are.

There is another essential point, easily overlooked. Images stored in QR codes are digital data. Physical photographs fade from UV and humidity; prints deteriorate with age. The reason historical portraits survive only in black and white or sepia is partly a limitation of the technology of their time, but also an inescapable fate of physical media. An image stored in a QR code does not fade. Ten years from now, a hundred years from now, scanning it reproduces the original in full, vivid color. Artworks, cultural artifacts, historical documents — permanent preservation free from the degradation inherent in physical media becomes reality through bulk mode's image capability. Ancient Greek sculptures were most likely painted in vivid colors. Egyptian murals, when first created, must have dazzled with pigment. Yet after millennia of weathering, all we see is white marble and faded earth tones. Had this technology existed then, the figures of mythology would appear before us today in their original colors. Imagine that impact.

Unlimited Text

Text in normal mode caps at approximately 2,000 characters. Brotli compression effectively stretches this somewhat, but long-form writing hits a wall. In bulk mode, the character limit is removed entirely. Wills, letters, diaries, poetry collections — write as much as you want without counting characters.

The significance of permanent text preservation extends far beyond personal records. Humanity has inscribed words on clay tablets, written on papyrus, bound them in scrolls, and printed them on paper. Yet no medium has withstood time. Ancient manuscripts survive as scrolls, but ink fades, paper grows brittle, and characters become harder to identify with each passing year. It took decades to decipher the Dead Sea Scrolls — not just because of the content's complexity, but because of the battle against physical deterioration.

Text stored in a QR code is digital data. Not a single character fades. Not a single stroke crumbles. A hundred years from now, a thousand years from now, scanning it reproduces the original text exactly. This is convenience for personal wills and letters, and at the same time, a structural shift for the entire endeavor of preserving records.

From an administrative perspective, the implications are equally profound. Civil registries, land records, permits and licenses — government documents carry mandatory retention periods, yet paper records are lost to disasters and decay. The Great Hanshin Earthquake of 1995 incinerated vast quantities of paper documents in municipal offices. The 2011 Tōhoku tsunami swept away original family registries. Text stored in QR codes, etched into quartz glass, can withstand fire and flood alike. Distributed across multiple copies, backup becomes straightforward. The role that paper has long served — the authoritative original of a public record — can now be inherited in a physically indestructible form.

All of these expansions run on the same principle: split the data, secure capacity through QR code count. Technically simple, yet revolutionary as a user experience. From a world of compressing everything into a single QR code to a world of adding as many codes as expression demands. From design that optimizes within constraints to design that removes the constraints themselves.

4. The Question: "Why Not Just Use the Cloud?"

Having read this far, a natural question arises. Audio, images, text — just save them to iCloud, Google Drive, or Dropbox. Why bother converting anything into QR codes?

Let us address this head-on. Cloud storage is excellent technology. Syncing is fast, capacity is generous, and search works well. For everyday file management, it is ideal. But to the question "will it still be there in 100 years," no one can answer.

Google has shut down Google Reader, Google+, Picasa, Hangouts, Stadia — countless services over the years. Google is not at fault. Companies have lifecycles. When revenue structures shift, unprofitable services close. That is sound business judgment. The problem is that user data gets caught in that judgment.

Sometimes a grace period is offered: "Please download your data." But those who missed the notification, those who have already passed away, those who changed their email address — their data quietly disappears. The lifespan of data entrusted to the cloud depends not on your lifespan, but on the service's.

Data stored in a QR code severs that dependency. Reading a QR code printed on paper requires no account. No WiFi. No corporate survival. All it takes is a decoder — and TokiQR's decoder is nothing more than a combination of open standard technologies that anyone can reimplement — and the data is resurrected.

A QR code etched into quartz glass has physical durability measured in millennia. Laminated, it lasts decades. Even on plain paper, in a dry environment, it can endure for centuries. Which timeframe to choose is up to the user, but in every case, the survival of the data is decoupled from the survival of any company.

There is another difference, easily overlooked. A file in the cloud requires the act of "opening." Navigate to the folder, select the file, press play. Those who do this routinely are fine. But five years from now, ten years from now, how many people will remember to open "that file in that folder"? Cloud data, from the moment it stops being accessed, exists but is absent.

A QR code exists physically. Stuck on the fridge. Placed on the family altar. Tucked into a photo album. Engraved on a headstone. Each time it catches your eye, you hold up your phone and sound pours out. This "physical serendipity" — happening to see it, happening to play it — is an experience that no cloud folder hierarchy can architect.

Digital and analog are not adversaries. Just as vinyl records have found renewed appreciation in the age of streaming, physical media possess an inviolable strength of their own. When the power goes out, when the servers go down, a QR code printed on paper is simply there. Scan that QR code and the digital data is restored. Digital and analog do not compete; they serve as each other’s backup. This round-trip capability is a form of resilience that cloud-centric architectures cannot replicate.

The cloud is infrastructure for convenient management. A QR code is infrastructure for permanence. The purposes differ. They are not things to compare, but things to use together. Daily backups go to the cloud. What you want to reach 100 years into the future goes into a QR code.

5. N-1 — The Meaning of the First QR Being Free

The billing structure of bulk mode is called "N-1." N is the number of QR codes your data is split into. The first is always free. From the second onward, each costs 1 credit (150 yen). This means that if your data fits in a single code, bulk mode costs nothing.

This may seem like a minor distinction, but as a design philosophy, it is decisive. Many freemium services impose artificial feature restrictions on the free tier, pushing users toward paid plans. Free becomes a "trial," and meaningful use presupposes payment.

N-1 takes the opposite approach. If your data fits in one QR code, that is the complete product. Not a limited demo. Recording 30 seconds of voice in normal mode and generating single-QR data in bulk mode carry exactly the same value. Charges arise only when the user consciously chooses "longer" or "higher quality."

So why charge from the second QR onward? It is not solely for revenue. Director Kawashima of the Japanese Immigration Museum "Nihojimamura" in Hawaii once said: "When information is free, people treat it lightly. I charge not because I want to make money, but so that people recognize its importance." This resonated deeply. When preserving a voice carries a cost, people stop treating that voice casually. 150 yen is not an economic barrier. It is a quiet declaration of commitment to one's own voice.

6. The Design Philosophy of Prepaid Credits

As the billing mechanism for bulk mode, we designed prepaid credits. One credit equals 150 yen ($1). Purchased in packs and activated by entering a code. Every aspect of this system is intentional.

First, there is no account registration. No email address, no password. The only data recorded during a credit purchase is the code (in TOKI-XXXX-XXXX-XXXX format) and the quantity purchased. TokiStorage does not know who bought them. It does not need to.

Second, there is no expiration. Credits purchased today can be used a year from now, or ten years from now. We did not want to build "use it before it expires" pressure into the design. Preserving a voice is not something to be rushed by a deadline.

Third, the credit balance is stored in the user's browser (IndexedDB). The server holds no balance information. More on this later, but this choice required resolve.

7. Why Wise, Not Credit Cards

The reason for not adopting credit card payments is straightforward. Credit card processing fees run 3–5% on a 150-yen product. Add monthly fixed costs for a payment processor, PCI DSS compliance costs, and above all — card payments inevitably require collecting personal information: name, address, card number.

TokiStorage's design principle is "hold no personal data." Data during QR generation is never sent to any server. No personal information is collected during credit purchases either. Maintaining this principle with card payments is structurally impossible.

Wise, as an international money transfer service, offers fees of 0.3–1%, transparent exchange rates, and coverage in over 150 countries. A user in Germany and a user in Brazil follow the same procedure to purchase credits. And TokiStorage has no mechanism to identify individuals from transfer information. Only codes and amounts are matched.

The uniform price — 1 credit at 150 yen ($1) — applies regardless of country or currency. Whether purchased in Japanese yen, euros, or Brazilian reais, one credit is one credit. This is none other than the application of the principle articulated in the "Uniform Pricing" essay to the credit system.

8. IndexedDB as a Choice — The Resolve Not to Hold User Data

Storing credit balances in the browser's IndexedDB rather than on a server. By conventional service design standards, this choice has obvious drawbacks.

If a user clears their browser data, the credit balance disappears. When switching devices, the balance does not transfer automatically. If a user contacts TokiStorage asking to "restore my balance," there is no data on the server to restore from.

We chose this design anyway because the alternative was worse. Storing balances on a server would require a system to identify users. Accounts, email addresses, at minimum some form of ID. And purchase history, usage history, and balance changes tied to that ID would accumulate on the server. That would mean TokiStorage abandoning its principle of "holding no personal data."

Storing in IndexedDB means TokiStorage cannot track user balances. We do not know who holds how much. We do not know who used what, or when. We cannot revoke. We cannot freeze. This asymmetry is a design that places complete sovereignty on the user's side.

This echoes Director Kawashima's philosophy mentioned earlier. "I charge so that people treat it with care" — and its natural extension: "I don't hold it, so that people care for it themselves." If TokiStorage managed balances, users might feel reassured. But that reassurance comes from entrusting one's credits to someone else. The choice not to hold user data is also a declaration of the relationship: that users themselves bear responsibility for their own balance.

9. Why a "Service That Holds Nothing" Is Abnormal

As a premise, virtually every modern web service operates by "holding user information." This is not malice but structural inevitability. Provide a service → need to identify users → create accounts → collect personal information → store that information → storage requires security → security costs money → need ads or subscriptions to recoup costs → ads require user behavior data → collect even more information. This chain is something like the original sin of web services.

TokiStorage's self-print economy severs this chain by refusing the very first step: "identify the user." How abnormal this is becomes clear when you think about it in reverse.

What "No Account Required" Really Means

Other services also claim to be "no account required." But the vast majority offer anonymous use as a "trial version," requiring account registration for full features. In other words, "no account required" means the front door is open, but the rooms beyond are locked.

TokiStorage is different. If your voice fits in a single QR code, the front door and every room beyond it are wide open. Billing only begins when bulk mode splits data across two or more codes, and even that billing requires no account. Just buy a prepaid code and enter it. The system does not know who bought it. It is not designed to know.

This is a fundamentally different level of "no account required." It is not a surface-level UX decision but a guarantee of user anonymity at the system architecture level.

The Structural Difficulty of Requiring No Personal Information

Not collecting personal information in a paid service is harder than it sounds. Use credit card payments and PCI DSS compliance becomes mandatory, making collection of card numbers, names, and addresses unavoidable. Use PayPal or Stripe and email addresses get linked. Even Apple Pay or Google Pay accumulate user information on the payment provider's side.

The choice of Wise transfers elegantly sidesteps this problem. The transfer itself is completed between banks, and TokiStorage only matches codes against amounts. There is simply no structure for incorporating sender's personal information into TokiStorage's system.

This is not "being careful not to collect personal information" but rather "no pathway for personal information to enter exists by design." The difference is decisive. The former is a policy issue, breakable depending on operations. The latter is an architecture issue, unbreakable without changing the design.

The Courage of No Expiration

The decision not to set an expiration date on prepaid credits may look "user-friendly" on the surface, but from a business perspective, it requires considerable resolve.

Typical prepaid services set expiration dates. The reason is straightforward: unused balances become "liabilities" in accounting. With expiration dates, expired balances can be removed from liabilities and recognized as revenue. This is the basic structure of the gift card business — even Amazon gift cards set expiration dates (within legal constraints).

TokiStorage has forfeited this "expiration revenue." Credits bought ten years ago still work. This is financially disadvantageous by design, but it is the inevitable conclusion of the philosophy: "do not impose deadlines on the act of preserving a voice." The moment someone wants to preserve their voice may not be the day they purchased credits. It might be five years later. Ten years later. If the credits had expired by then, the very purpose of the service would be undermined.

The Abnormality of Storing Balances in IndexedDB

This is perhaps the boldest design decision of all.

Normally, balances in a billing system are managed server-side. This is "common sense" from security, user experience, and business perspectives alike. With server-side balances, changing devices preserves the balance. Fraudulent use can be frozen. Failures can be recovered from.

TokiStorage discarded all of this common sense. Balances live in the browser's IndexedDB. Nothing exists on the server. Clear browser data and the balance vanishes. Switch devices and the balance does not transfer. Contact TokiStorage and there is no data to restore from.

A typical service designer would classify this as a "bug" or "not yet implemented." But TokiStorage declares it as intentional design. If balances were stored on a server, a user identification ID would be necessary. With an ID, purchase history gets linked. With history, profiling becomes possible. The moment balances are placed on a server, the principle of "holding no personal data" collapses.

What makes this decision remarkable is that it is not "technically impossible" but "technically possible yet deliberately not done." Server-side balance management is an absolute basic of implementation. By deliberately not implementing it, the design principle is upheld.

The Abnormality of Everything Working Simultaneously

Individual elements — no account required, no personal information, no expiration, first QR free — might each be found in other services on their own. But a service where all of these hold true simultaneously, and with billing functionality, honestly does not come to mind.

The reason is that these elements are normally in a trade-off relationship. Removing accounts makes billing management difficult. Removing personal information limits payment methods. Removing expiration dates swells accounting liabilities. Making the first unit free raises the bar for monetization. Normally, compromise happens somewhere. "Accounts are required but personal information is minimized." "A free tier exists but is time-limited."

TokiStorage has not compromised. An architecture satisfying all constraints simultaneously has been achieved through the combination of Wise transfers, IndexedDB, and prepaid codes. This is possible not because each individual technology choice is superior, but because the design philosophy is consistent throughout. Had the philosophy wavered, some element would have bent: "actually, we do need accounts" or "actually, let's store balances on the server."

Why "Design by Subtraction" Is Fundamentally Difficult

Adding features is easy. Add account functionality, build a dashboard, visualize usage history, implement cloud sync for balances. These cost development effort, but as design decisions, "just add it" requires no hesitation.

Subtraction is different. You must envision everything that will happen when a feature is removed, everything lost by not holding certain information, and then decide to remove it anyway. And the inconveniences that result from removal — balances disappearing, no recovery, no cross-device sync — must be honestly communicated to users, with the resolve to explain why this design is still correct.

TokiStorage's self-print economy is the result of pushing that subtraction to its limit. What remains is "a tool, a price, and freedom" — this single phrase condenses the entire design philosophy. The tool is provided. The price is stated. The rest is your freedom. TokiStorage's presence as a service provider is intentionally minimized.

This is the implementation of "democratizing proof of existence" at the infrastructure layer. The means of preserving a voice is not placed under the control of any specific platform or company. Anyone, anywhere, by their own judgment, on their own printer. To structurally guarantee that freedom, this much has been stripped away.

The very concept of a user base has been abandoned. We do not know who is using it. We do not even know precisely how many are using it. Yet a service that works has been built. This is a technical achievement and, at the same time, an intellectual one.

10. The Coexistence of Physical QR and Self-Print

Physical QR products and self-print are not in competition. Nor are they premium and budget versions of the same service. They are expressions of fundamentally different design philosophies serving fundamentally different needs.

Physical QR resists time through the power of materials. Quartz glass endures for thousands of years; UV-resistant laminate withstands outdoor environments. Three-layer preservation — Physical (quartz glass/laminate), National (National Diet Library), and Private (GitHub) — delivers data to the future through multiple pathways. This is the domain where TokiStorage takes responsibility as a manufacturer.

Self-print takes shape instantly in the user's hands. On a home printer, an office copier, a convenience store print service. If you own a laminator, you can laminate them yourself. Quality, quantity, timing — the user decides everything. TokiStorage merely provides the tool.

This relationship resembles that between a professional photo studio and a home printer. Having a studio produce a family portrait album versus printing a smartphone photo yourself. Both are acts of preserving a photograph, but what they seek differs. One offers professional finishing and archival quality. The other offers immediacy and self-determination. Not a matter of superiority, but of choice.

11. The Restoration Path — The Round Trip of Print and Scan

QR codes generated in bulk mode are output as PDFs. These PDFs can of course be stored as digital files, but they can also be printed and stored as paper. And if the paper survives, restoration works in reverse. Scan the paper back into a PDF, decode the QR codes with software, and audio, images, and text are fully restored.

What is notable is that this restoration requires no camera. A document scanner converts paper to PDF; software reads the QR codes. There is no need to hold a smartphone up to each code one by one. With an ADF (automatic document feeder) scanner, dozens of pages can be converted to PDF in a single batch. Even a large set of QR codes generated in bulk mode can be restored in a practical timeframe through this path.

QR code error correction underpins this round trip. During the process of printing and scanning, noise is introduced: ink bleed, paper creases, scanner resolution differences. But QR codes incorporate Reed–Solomon error correction, absorbing a range of noise and restoring data completely. In other words, even after the physical round trip of print → paper → scan → PDF, the digital data does not degrade by a single bit.

This bidirectionality is a foundational characteristic of the self-print economy. Digital files become physical media; physical media become digital files. If either is lost, the other provides a path to recovery. Cloud storage data vanishes when the service shuts down, but QR codes printed on paper survive quietly in a drawer. Conversely, if paper is lost to fire, the PDFs preserved on GitHub and at the National Diet Library enable restoration. Multiple pathways complement one another.

Scanners will not disappear a thousand years from now. Neither will paper and ink. Even if the QR code standard itself falls out of use, anyone can write a program to analyze black-and-white dot patterns. Every link in the restoration chain is free from dependence on any specific technology or company.

12. What the Self-Print Economy Means

Bulk mode, prepaid credits, Wise transfers, IndexedDB storage, no account required — when these elements combine, what emerges is an economy.

What to preserve is the user's decision. How long to make it is the user's decision. When to print, on what paper, how many copies, who to give them to. Everything is entrusted to the user's judgment and action. TokiStorage provides only the codec (audio compression), the QR code generation engine, and the credit system.

And the dedicated newsletter is the natural culmination of this pipeline. Bulk mode QR PDFs are stored on GitHub and at the National Diet Library. A voice is converted into QR codes, typeset into a PDF, committed to a version control system, and archived in a national repository. The pathway for an individual’s voice to reach a national preservation infrastructure is now fully open. The self-print economy deserves the name “economy” precisely because the entire cycle — from generation to storage, printing, and restoration — functions as a single closed loop.

This minimalism is not product immaturity. It is intentional design. What a tool should provide is capability, not judgment. The decision of what to record and what to preserve can only be made by the person whose voice it is. What TokiStorage can do is deliver the tools for executing that decision — as affordably, as simply, and to as many people as possible.

The self-print economy may be an economic model that does not yet have a name. It is not a subscription. It is not a one-time purchase. It resembles pay-per-use, but the first unit is always free. Without accounts, there is no customer management. Without personal data, there is no targeted advertising. Without data on our servers, there is no risk of data breach.

All that exists is a tool, a price, and freedom.

TokiStorage's mission statement reads: "You become a story, generations connect in dialogue, the path forward." This is not just the founder's philosophy. It is a question addressed to everyone.

Your story is yours to record, vividly. The landscape of the place where you were born. The scent of a land full of memories. The path your ancestors walked. Stories passed down through your family. Historical facts of the places around you. Each of these is an irreplaceable element of your story.

Record a voice. Preserve a photograph. Write down your words. The moment you begin to record these things — your existence is no longer a single line, but a story layered many times over. You stand at the intersection where what you received from the past and what you hand to the future converge. The self-print economy is the tool for standing at that intersection.