About US
Questions worth pursuing.
Our lab is an independent research practice devoted to applied cryptography, its algorithms, protocols, implementations and the assumptions on which they rely. Research may begin with an operational problem, an emerging standard, a cryptographic construction or a question worth pursuing in its own right. It is not defined by a predetermined outcome.
Independent work. Selected collaboration.
Research is pursued independently or with selected collaborators. Contributions and institutional relationships are presented in the context of the work itself.
Approach
How this work is made.
Everything published here is written to be checked. This section describes how, so that a reader can calibrate what any given article is claiming.
Four kinds of article.
Each piece carries a type, and the type sets the standard it should be judged against.
A Primer explains how something works. Every claim in it can be found by reading the sources it cites; nothing is derived. It is judged by accuracy, clarity, and whether it organises the material better than the specification it summarises.
An Analysis derives a result the sources do not state — a figure, a comparison, a threshold that appears in none of its references. It is judged by whether the derivation is correct and whether it changes a decision.
A Review organises a field, imposing a structure on scattered literature so that a reader can see the whole. It is judged by completeness.
Findings report original data that did not exist before publication, with a method described in enough detail to be reproduced.
The four are ordered by original contribution, not by difficulty. Most of what is published here is Primer and Analysis. When Findings appear, they will be labelled as such and their method will be published alongside them. Nothing on this site should be read as our own measurement unless it says so.
Drafting and verification.
Articles are drafted with the assistance of large language models. Every factual claim is then verified by a person against a primary source (the standard, the specification, the peer-reviewed paper, the vendor's own documentation, the source code) and that person is responsible for it. Secondary summaries are treated as leads to follow, not as citations.
Four kinds of statement appear in these articles and we try to keep them distinguishable in the prose rather than only in the references: what a specification requires, what someone measured, what we derived ourselves, and what a vendor claims. They carry different weight and a reader should not have to check the footnotes to tell them apart.
Where a specification is behind a paywall and a clause reference comes from a draft or from third-party guidance instead of the published text, the article says so. Where a figure cannot be traced to a primary source, it is left out.
Volatile facts and corrections.
Version numbers, draft statuses, product availability and market figures move. Where an article depends on one, it is dated, and the article carries a revision line.
We publish corrections in place rather than quietly editing, and we would rather hear about an error than not. Corrections, disagreements and questions: labs@neodata.com.uy
Attribution.
Articles are published under the nLabs byline. nLabs is the research practice at Neodata, in Montevideo, Uruguay. Where work involves a collaborator or a client and we are free to say so, the article says so and appears under In Practice; where we are not, the subject is written generically and no engagement is identifiable.
Inquiries
Understanding first.
We move between analysis and implementation. Some inquiries lead to a deployed mechanism or selected engineering work; others to an experiment, a technical note, an open artifact or a clearer formulation of the problem. Work may be self-directed, collaborative or commissioned. Not every line of inquiry has a commercial objective.
// How to read the lab
Sections describe context. Types describe evidence. Series describe sequence.
Research, In Practice and Marginalia locate a piece within the lab. Primer, Analysis, Review and Findings state the standard against which it should be read. A series connects several pieces into one ordered investigation.
Explore the series →