Someone has forwarded you a screenshot: a search result, or a Google AI Overview, with the word "predatory" next to our name. Perhaps from a co-author, perhaps from a head of department who signs off before your manuscript leaves the hospital.
The normal response is a paragraph of reassurance. Reassurance is worth nothing here: a publisher that was lying would write the same paragraph.
So this post hands you the queries, names us as the specimen, and prints each result in advance — including every row that goes against us. An answer that does not reproduce is a reportable error, and the last section says where to send it.
Every answer resolves off our own website
Every row below resolves at Crossref, OpenAlex, DOAJ, the US National Library of Medicine, ORCID or a commercial source list — registers we do not control. The Crossref rows are our own deposits, published as we filed them, which is why several go against us. Four rows touch our servers — the DOI resolutions, the landing-page comparison, the OAI-PMH endpoint, the board page — and all four exist to be broken. We left out the check that cannot fail: a publisher that reviews nothing can publish the same policy pages.
Which rows audit the publisher, which audit one journal
PUBLISHER rows — membership, prefix ownership, the participation report, the OAI-PMH endpoint — answer for everything we publish. JOURNAL rows — DOAJ, the NLM catalogue, source lists — run against one title only.
One row was cut rather than fudged: an ISSN Portal result differs per title, so look yours up yourself.
Last verified: 30 August 2026. Every query was re-run that day; the stated results are what came back. Rows that did not reproduce were deleted, not softened.
| Check | Entity | Query | Result we state you will get | What a different result means |
|---|---|---|---|---|
| Crossref member record | PUBLISHER | https://api.crossref.org/members/29803 | Member 29803, "Directive Publications", prefixes ["10.52338"] | No record, or another name: our claim is wrong. Ours. |
| Prefix ownership, not just membership | PUBLISHER | https://api.crossref.org/prefixes/10.52338 | Resolves to member 29803, Directive Publications | Another member owns it: someone else deposits our DOIs. Ours. |
| Registered DOI count | PUBLISHER | Member call, counts | 596 total — 521 current, 75 backfile | The split shifts as records pass two years old; a fall in the total means withdrawn deposits. Ours. |
| Resolve three DOIs you pick yourself | PUBLISHER | Take three DOIs off article pages, prefix with https://doi.org/ | Each redirects to a live page for that article. Some older deposits still point at retired PDF URLs and 404 — ours to fix. | Dead link or wrong article: a broken deposit. Ours. |
| Landing page matches the deposit | PUBLISHER | https://api.crossref.org/works/10.52338/joi.2026.6049 | Title, journal and date match the page the DOI opens. No author list on either side: the deposit carries no author field. This row fails. | Any other mismatch is a metadata defect. Ours. |
| Licence element, recent deposit | PUBLISHER | Same call, license field | creativecommons.org/licenses/by/4.0/ | Absent: the page promises what the record lacks. Ours. |
| Licence element, older backfile deposit | PUBLISHER | https://api.crossref.org/works/10.52338/apps.2023.1018 | Empty. No licence element. A gap we own. | Expected. We did not deposit a licence element with these records. Ours to fix. |
| Licence coverage across all deposits | PUBLISHER | Member call, coverage-type.all.journal-article.licenses, then the same field under current and backfile | 0.681 overall; 0.779 current, 0.0 backfile | Lower is worse than we said. Ours. |
| Similarity Check full-text links | PUBLISHER | Member call, coverage-type.all.journal-article.similarity-checking | 0.0. A field we leave unset. | A crawl link we do not deposit, not a record of screening. Ours to fix. |
| References, funders, update policies | PUBLISHER | Member call, same object | 0.0, 0.0 and 0.0. A shortfall we own. | Our deposits are thin. No reading favours us. Ours. |
| ORCID and ROR identifiers | PUBLISHER | Member call, same object, orcids and ror-ids | 0.0034 and 0.0. Near zero. | People and institutions unidentified in our records. Ours. |
| OAI-PMH endpoint answers | PUBLISHER | https://directivepublications.org/oai?verb=Identify | Identify response, repository "Directive Publications", earliest datestamp 2025-10-08 | Error or timeout: our endpoint is down. Ours. |
| OpenAlex record, by DOI only | PUBLISHER | https://api.openalex.org/works/https://doi.org/10.52338/joi.2026.6049 | A work record, type article, source the journal, oa_status hybrid | A 404 means the record has not propagated. Not an indexing verdict. |
| OpenAlex publisher entity | PUBLISHER | https://api.openalex.org/publishers?search=Directive%20Publications | count: 0. We hold no publisher entity. | Anyone citing an OpenAlex publisher page for us cites nothing. |
| DOAJ register | JOURNAL | https://doaj.org/api/search/journals/bibjson.publisher.name:%22Directive%20Publications%22 | "total": 0. Not listed. This row fails. | Correct as printed: not in DOAJ. |
| NLM catalogue, and therefore PubMed | JOURNAL | https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi?db=nlmcatalog&term=%22Directive+Publications%22%5Bpubl%5D&retmode=json | count: "0". No catalogue record. This row fails. | Correct as printed. Use [publ]: without it the search matches the words, not the publisher. The inference runs one way — no record means no PubMed, but a record does not mean PubMed carries it. |
| Commercial source lists | JOURNAL | Search the title on the Scopus source list and the Master Journal List | No matching title on either. Both fail. | Correct as printed: on neither. |
| Named preservation archive | PUBLISHER | Search the archives' participating-publisher lists | We appear on none. This row fails. | Correct as printed: enrolled in none. |
| Editorial board reality check | JOURNAL | Take three board names and look each up at https://orcid.org | Some resolve to a public record with matching affiliation, some do not. Verification is in progress, not complete. | A listed name with no trace anywhere is a real problem. Send it to us. |
What a pass does not prove: every row that passes tests infrastructure. None tests whether a manuscript was reviewed properly.
Resolve a DOI, then read the deposit behind it
A DOI is a redirect plus a metadata record, and the audit is in the second half. Pick three articles yourself. Resolve each DOI, then fetch the same DOI from the Crossref API and compare title, authors, journal and date. Run the prefix row too: membership says an organisation joined, while prefix ownership says this prefix is registered to it.
Read our participation report before our policy pages
The participation report is the coverage data inside the member record. Ours is not flattering: references, funders and update policies at zero, ORCID at 0.34 per cent, ROR at zero.
One figure needs separating. Similarity Check reads 0.0. Crossref defines it as the percentage of our records carrying a full-text link for its Similarity Check service to crawl — an obligation on members of that service, not a record of whether a manuscript was screened. Screening is run on submissions, as described in how similarity screening works.
Check the licence in the metadata, not the licence on the page
Our article pages state CC BY 4.0; reuse tools read the deposit, not the page. A recent DOI carries the licence element. An older backfile record does not, and coverage sits at 68.1 per cent of deposits overall — 77.9 per cent of current ones, none of the backfile. So do not accept "every deposit carries the licence" from anyone. The gap is ours, and not because the licence field did not exist. What the licence permits is on Open Access.
Query OpenAlex by DOI, never by journal name
Name searches return whichever entity matches the string, which for common titles can be the wrong one. Anchored on a DOI, our records return oa_status: hybrid, not gold, and we hold no publisher entity. We do not quote the mean citedness or h-index those pages carry — see what impact factor, CiteScore and h-index really mean. OpenAlex presence is plumbing, not indexing.
Four claims we cannot make
DOAJ. The API returns zero journals. We are not in the Directory of Open Access Journals.
The source lists. No NLM catalogue record, and therefore nothing in PubMed or PubMed Central. Not on the Scopus source list, not on the Master Journal List. If you have seen any claimed for us, tell us where.
Preservation. We are on no named archive's participating-publisher list, so we name none. What that gap means is in digital preservation in scholarly publishing.
The board. Our verification programme, built on ORCID, ROR and OpenAlex, is in progress, not complete. The same picture without API syntax is in where we are indexed and where we are not.
What a clean sweep still does not prove
Crossref membership is purchasable. A prefix follows from it. An OAI-PMH endpoint is a software feature. An OpenAlex record follows from a deposit. Publishers that appear on watchlists hold all four, because infrastructure has no opinion about whether a manuscript was read.
A full column of passes proves the plumbing works, not that two reviewers read the paper or that an editor read their reports — see how to identify a predatory journal.
Run this table on the next journal that emails you
Swap the prefix, member number and journal title, and this audits any publisher in twenty minutes. What a publisher says when a check goes against it tells you more than the check did.
If a row of ours returns something other than what we printed, write to [email protected] with the query and the result. That is an address, not a service commitment: no turnaround is attached. Maintained by the Directive Editorial Team.
The policies behind the rows: Open Access, Copyright & Licensing, Digital Preservation, Publication Ethics.