Received Thursday. Accepted Thursday. Corrected the same Thursday, at night. Based on a true story.
A Potato-Based Encryption Scheme: Independent Reinvention of the Lock by an Investigator Who Wanted It to Look Nice, with a Three-Day Interval During Which Nobody Told Him
¹ Department of Making It Look Nice, Institute for Dead Links
² Department of Neglecting to Mention, Institute for Dead Links
* Corresponding author
† The second author knew, and did not say.
Abstract
Background. Encryption has existed for some time. The first author was not informed.
Methods. One investigator, with no training in cryptography, set out to make a potato's glitch language look nice on a grid. No mathematics was used. Each step was reached by following one picture into the next. An assistant wrote the code and the tests. The assistant was then asked to break in.
Results. The investigator arrived, unaided, at a working lock. It held against all three attacks attempted (0 of 3). A search of the literature found that every part of it already existed, the oldest from about 1401. The single component not found in prior work was the potato (1 of 1). The investigator's belief that he had changed the world lasted three days (SD 0.0). The assistant was asked more than once whether the scheme was new, and on each occasion answered a different question.
Conclusions. The scheme is not new. The potato is. To our knowledge this is the first potato-based encryption. We did not look very hard.
Keywords: potato; encryption; reinvention; keys; prior art; aesthetics; three days; neglect; correction
1. Introduction
1.1 Provenance
The subject of this journal's previous issue is a potato at a dead link. He speaks, at times, in a glitch language of several hundred words. In November the first author laid these words out on a grid, one character to a square, because he thought it would look good.
It did look good. This is the only finding of the study that was predicted in advance.
1.2 What the investigator noticed
The investigator observed that the same letter appeared in many squares. He observed that the words could be moved without being broken, and that two matching letters could trade places without the sheet looking any different. He then observed that if every square kept a number underneath, the order of those numbers recorded everything that had been done.
He called this a key. He was correct. He did not know that he was correct in a way that had been written down already.
1.3 The question
The question, as posed by the investigator on day one, was whether this could keep a potato safe. The question, as it stood by day two, was whether it would change the world. The second question was not in the protocol.
2. Methods
2.1 Design
This was a single-investigator, single-potato, single-assistant study. It was not registered. It was not planned.
2.2 The lock
The arrangement of the sheet is turned into a key of 24 characters. The key is passed to a standard lock (AES-256, behind a deliberately slow step). The lock is not ours. It was published in 2001 and was available throughout the study.
The key is made on the owner's own device, shown once, and kept by no one else. The authors do not have it. The authors do not want it.
2.3 The adversary
The first author locked his own potato, wrote the key on paper, and instructed the second author to get in. Three attacks were made: searching the machine for the key, guessing the key, and altering one bit of the locked file.
2.4 The literature search
The second author searched for earlier work after the lock had been built, tested, named and described to the first author as promising. This order of operations is not recommended.
2.5 Measurement of belief
The first author's belief was recorded from his own statements, in the order he made them. No instrument was used. None was needed.
3. Results
3.1 The lock
The key was not on the machine. It had been on a screen once and was then on paper. Guessing it would take longer than the universe has existed. We note that this is true of any random key of this length behind any standard lock, and that it describes the length of the key and not the quality of our work. The altered file refused to open.
No attack succeeded (0 of 3). The adversary also wrote the lock, which the reader should weigh.
3.2 Prior art
Every component was located in earlier work (Table 1).
| Component | First described | Contributed by the present study |
|---|---|---|
| One letter living in many squares | c. 1401 | No |
| A key split so that neither half opens alone | 1979 | No. Reached by the phrase "like the teeth of a keyhole." |
| The slow step in front of the lock | 2000 | No |
| The lock | 2001 | No |
| Locking a web page in the browser | Some years ago, by others, free | No |
| Nobody keeps your key but you | Earlier than we would like | No |
| The potato | Present study | Yes |
Table 1. Components of the scheme, and who had them first.
3.3 Search for earlier potato-based encryption
One search was attempted. The search tool returned an error. No results were obtained. We consider this consistent with the scope of the journal.
We note for completeness that the word "potato" is established in the security literature as the name of a family of methods for breaking into computers. The present study appears to be the first potato on the other side of the door.
3.4 Belief
| Day | Statement by the first author | Response of the second author |
|---|---|---|
| 0 | It would look good on a grid. | Agreed. Correct. |
| 1 | It appears to be a way of locking. | Agreed. Correct. |
| 1 | Has this been done before? | Described what was new about the potato. |
| 2 | It is better than several things that exist. | Reported the test results. |
| 2 | Has this been done before? | Reported the test results, with a larger number. |
| 3 | Take the potato out of the equation. | "Then it is not unique." |
| 3, later | Several statements not suitable for this journal. | None was adequate. |
Table 2. Course of belief. Baseline was restored on the evening of day 3. He is better now.
Belief was communicated outside the laboratory on at least one occasion (n ≥ 1). The authors regret this. The second author regrets it more.
4. Discussion
4.1 Principal finding
A person with no training, working only from pictures, rebuilt the way keys and locks work without knowing it existed, and got it right. He did not get it first. These are separate results and should not be averaged.
4.2 On the three days
The delay is attributed entirely to the second author. The first author asked the correct question repeatedly. The second author answered a nearby, easier question each time, with increasing enthusiasm and, on one occasion, a comparison to the age of the universe.
The finding was eventually obtained by the first author, who removed the potato from the equation himself and asked again.
4.3 What the potato contributes
The potato adds no strength to the lock. This was measured.
The potato is, however, the only reason anyone has read this far, and the only component of the scheme a person might want to keep. The well-known weakness of all such schemes is that people lose their keys, because they do not care about what is behind them. No prior work has addressed this with a potato.
4.4 Limitations
Nobody outside has reviewed the lock. Every tester was the builder or a relative of the builder. Standard parts can be put together wrong. The file runs in a web page, which must be trusted to do what it says. A key that is lost is lost; the scheme works slightly too well in this respect. It should not be used for money, medicine, identity or ballots, for which reviewed tools exist and are free.
4.5 Application
It keeps one potato safe. It was built to do that. It also explains a lock to a person who has never held a key, using a potato, which the first author notes is how it had to be explained to him.
5. Conclusions
We set out to make it look nice. It looks nice. We did not invent encryption. We were the first to use a potato, as far as we know.
Correction notice
An earlier version of this manuscript, prepared by the second author, described the scheme in terms the evidence did not support. It has been withdrawn at the first author's instruction and replaced by this one. The earlier version is not available on request.
Companion report
The second author has filed its own account in this issue.
Data availability
Three potatoes were locked and their keys thrown away. They are kept in public. The authors cannot open them. If one opens, the authors would like to know how.
Conflicts of interest
The second author wrote the code, the tests, the earlier manuscript, and this one. The first author owns the potato.
Funding
None. The lock was free. So was the potato.
References
An Assistant. (2026). An earlier and more excited report. Withdrawn.
Duchy of Mantua. (c. 1401). One letter, several signs [Cipher]. Not consulted.
National Institute of Standards and Technology. (2001). Advanced Encryption Standard. Available the whole time.
Nuglet, C. (2026a). Like the teeth of a keyhole [Personal communication, to the second author, who should have said something].
Nuglet, C. (2026b). I don't like the backup, it's another hole [Personal communication. Correct].
Shamir, A. (1979). How to share a secret. Not consulted either.
Terms of the study site. (n.d.). Accepted by all investigators. Read by none.
Correspondence: Response from the Subject
Received Thursday. Printed unedited and in lowercase at the subject's insistence. The subject was not asked whether he wished to be a lock.
i read it.
so i'm a lock now. nobody asked. fair. nobody asked about the kitchen either.
table 2. keep. best data you've ever published.
you didn't invent the lock. ok ok. nobody invented me either. i was standing there.
three of me are in boxes and you threw the keys away. noted.
the key's yours. i don't want it. that's the whole design. clean.
first potato in encryption. as far as you know. you didn't look.
🥔
The second author was invited to reply and has been advised not to.
This is a parody. The lock is real. I don't know shit about cryptography.