Radar Perene / Archive / science

What a DOI is and why every Radar Perene study has one

◦ Index methodology v2.2 (working papers with DOI). See the methodology.

Science

The internet forgets at a speed that would embarrass any library: studies of the phenomenon estimate that an enormous fraction of the links cited in academic work stops working within a few years — the address changes, the site goes down, the page gets reorganized. For a social media conversation, this is irrelevant. For science, it is fatal: an argument that cites sources which disappear becomes an unverifiable argument.

A DOI (Digital Object Identifier) is a permanent code that identifies a digital document — in the format 10.XXXX/name — and keeps leading to the document even when the page's address changes. Registered in an international infrastructure maintained by a consortium, it is the difference between citing a study and citing a link.

The distinction sounds bureaucratic until the day one tries to reconstruct, five years later, what exactly a piece of research claimed — and in which version.

How the code works

The DOI is a deliberate layer of indirection. Whoever cites does not point to the file's physical address, but to the code; the resolver system — reachable at doi.org — maintains the table that translates the code into the current address. If the document moves to another server, its owner updates the table, and every citation in the world keeps working without anyone touching it.

Two properties complete the mechanism. First: a DOI comes with registered metadata — author, title, date, version — which makes the document findable and citable by machines, not just by readers. Second: serious repositories keep versions — each revision of a study can receive its own code, and the old version remains accessible. The history of an argument becomes auditable: what was claimed first, what was corrected later, all dated.

It is the registry office's logic, transplanted into scientific literature: no one is forced to register, but what is unregistered counts for less in a dispute.

Using the code in daily life is trivial, and worth describing because almost no one does: faced with any DOI — in a bibliography, a footnote, a study page — prefixing it with `doi.org/` in the browser leads to the document. The reader who acquires this habit gains an instant seriousness test: the citation that offers a DOI invites verification; the one that offers only a vague title and a website asks for trust on credit.

Why an independent study needs this even more

A university lends its researchers its institutional permanence: the repository, the library, the archive. The independent researcher has no such backing — the credibility of what they publish depends entirely on the verifiability of what they have published. Without permanent registration, an independent house's output is a collection of pages that can vanish; with it, it is citable, dated, immutable literature on the same terms as university output.

That was the reason for the house's choice of Zenodo, the scientific repository operated by CERN: every deposited study receives a DOI, every version is preserved, and none of it depends on the house's website continuing to exist. In numbers: six working papers of the RPWP series are public with their own DOIs — the Perene Risk Index study (10.5281/zenodo.21325743), the structural Ânima Index study (10.5281/zenodo.21327595), the tactical Ânima Index study (10.5281/zenodo.21327608), the Real Estate Liquidity Index study (10.5281/zenodo.21327629), the Brazilian macro regime score study (10.5281/zenodo.21402939) and the intermarket relationships study (10.5281/zenodo.21327663).

The data package behind the first two studies is deposited as well, with checksums that let any reader confirm the downloaded file is the published file. The whole — study with DOI, data with DOI, versions preserved — is what turns "trust us" into "check for yourself".

What a DOI is not

The code concerns permanence, never quality. A weak study with a DOI remains weak — merely verifiably weak, which is already progress. Judging merit belongs to other mechanisms: robustness tests, public criticism, peer review. The DOI guarantees only that the object under evaluation is findable, intact and dated — the minimum condition for any honest evaluation to begin.

Frequently asked questions

Is registering a DOI expensive?

It depends on the route. Publishers and registration agencies charge; open repositories such as Zenodo assign DOIs free of charge to any public deposit. For independent researchers, that free route changed the game.

Is a DOI a mark of scientific quality?

No. It is a mark of permanence and traceability. Any document can receive one — quality is judged by the content, the tests, and the scrutiny the text has survived.

What is the difference between DOI, ISBN and ISSN?

The ISBN identifies books; the ISSN, serial publications (journals, magazines); the DOI, individual digital objects — an article, a dataset, a specific version of a study. For citing standalone research on the internet, the DOI is the right instrument.

If a study is updated, does the DOI change?

In versioned repositories, each version can have its own code, and a concept code always points to the latest one. The cited version remains accessible forever — that is what makes the history auditable.

---

Registering a study forever only has value if someone can redo it — and that is the trail's next question: what reproducibility in finance means in practice.

House reading: what these registrations support in practice is in today's reading, in the Diário.

Studies of this kind, applied to a specific case, are the house's bench work — the registered output is collected at /pesquisa.