Privacy & Data Protection
SigmaCV is built on data minimization. It reads only public research metadata, matches your work by identifier (ORCID / OpenAlex ID) — never by name — and logs nothing about your choices unless you explicitly opt in. This notice explains what we process and the rights you have under the EU GDPR and Japan's APPI.
Data controller
SigmaCV is operated by Basile Chrétien (PharmD, MSc, MPH) as an independent personal project — not by, or on behalf of, any university or employer — and he is the sole data controller for the personal data SigmaCV processes. For any privacy question or request, contact him at privacy@sigmacv.org.
What we process
Account data from your sign-in provider (ORCID iD, or email / Google account); the public research metadata used to build your CV (publications, grants, affiliations and metrics from OpenAlex, ORCID, Crossref, DataCite, OpenAIRE, DBLP, Wikidata, Open Editors Plus, NIH iCite, OpenCitations, Sciety and Software Heritage, plus — matched by name and organisation, and shown only as review candidates — UKRI, NIH RePORTER, NSF, ClinicalTrials.gov, EU CTIS, WHO ICTRP and the European Patent Office); the CV content and display choices you create; and an optional profile photo you upload. Your CV is private unless you choose to publish a public page. If you choose to, you can also record optional self-declared career context (career breaks, part-time periods, clinical or caring duties, a first-publication year); it is stored with your CV, shown only if you enable it, never used to compute any score, and included in your account data export. If you freeze a version of your CV, that frozen copy (without your private notes) is kept with your account, is public only if you make it so, and is exported and deleted with the rest of your data; if you minted a DOI for a version, the DOI record held by DataCite (the public data on that version's page: your name, ORCID iD, affiliation, the DOIs of the works shown and the funder / award identifiers of the grants shown) is withdrawn — hidden and pointed at a “withdrawn” page — rather than deleted.
Supervision and mentoring records may include the names of the people you supervised — third-party personal data that you, as the author of your CV, choose to publish. Add a person's name only with their agreement; you can instead show a degree-level description (for example “PhD student”) in place of the name on the public page and in every export. Those names are deleted with your account, like the rest of your CV.
Purpose & legal basis
We process this data to provide the service you requested — generating and exporting your CV (GDPR Art. 6(1)(b), performance of a contract). Publishing a public CV page happens only on your instruction. We do not sell your data or use it for advertising.
Who receives your data, and where it is stored
We never sell your data or use it for advertising. We share it only to run the service: with our hosting provider in the European Union (Germany), which stores it on our behalf under a data-processing agreement; and, when you sign in or sync, with the public research services we query by your identifier to build your CV (OpenAlex, ORCID, Crossref, DataCite, OpenAIRE, DBLP, ROR and similar), which receive your ORCID iD or name to return your public records and act as independent controllers of that public data. Your account and CV are stored in the EU. Backup copies are also held by the controller in Japan on encrypted storage, and are not shared with anyone else; the controller is based in Japan, and transfers between Japan and the EU are covered by the 2019 Japan–EU mutual adequacy decision. When you opt into email digests or confirm a contact email, we deliver those messages through SMTP2GO, an email-delivery processor based in New Zealand, which receives only your email address and the message content (for digests, the titles of your own public works) to send the email on our behalf; New Zealand is recognised by the EU as ensuring an adequate level of data protection. If you choose to mint a DOI for a frozen version of your CV, the DOI record — the public data on that version's page: your name, ORCID iD, affiliation, the DOIs of the works shown and the funder / award identifiers of the grants shown — is registered with DataCite, which holds it as an independent controller under its own persistent-identifier policy; when you delete your account the record is not erased but withdrawn: hidden from DataCite's search and pointed at a “withdrawn by its owner” page, so the DOI keeps resolving without leading to your data. When you allow search-engine indexing of your public page, its metadata — the public record of your CV and the works it lists, as Dublin Core — is also available to OAI-PMH harvesters (open repositories, institutional CRIS systems and aggregators) through our OAI-PMH endpoint; they act as independent controllers of what they collect. Being listed under your current institution's ROR set for such harvesters is a separate, opt-in choice, off by default. Appearing on your institution's public page on SigmaCV (/i/<ROR id>) is a further separate opt-in, pinned to the institutions you tick among your current positions: your name, ORCID iD, current position and the works your public page lists are then shown there and counted in its figures; it is off by default, is asked again rather than moved when your affiliation changes, and can be withdrawn at any time with immediate effect. When you are listed, the institution's page on SigmaCV (/i/<ror>) shows how many researchers are listed under it — a count only, never who. The organisation view (/i/compare) sets side by side OpenAlex's public record of two or three organisations — counts of works and of open copies, never people: nothing about you beyond the fact that at least one researcher is listed under an organisation (which only decides whether it can be chosen), and none of the figures of the researchers listed on an institution's page, is used, shown or compared there. If you also choose to share your reconciliation rows, a public CSV/JSON export under /i/<ror>/reconciliation lists, from the frozen version you designate, your ORCID iD, the works your public page lists (DOI, year, type, licence and the open-access state found) and the institutions you ticked; it is off by default and stops with the next request once you untick it.
Public ORCID preview
Anyone can enter an ORCID iD to see a CV assembled from public research metadata, with no account and no sign-in. The preview is built on request from the public sources named under “What we process”. Most are queried by the iD itself; the national funder, clinical-trial and patent registries that hold no ORCID iD (UKRI, NIH RePORTER, NSF, ClinicalTrials.gov, EU CTIS, WHO ICTRP and the European Patent Office) are queried by the researcher's name and organisation, and anything they return is shown only as a flagged candidate for the researcher to confirm, never as an established fact. The preview is generated when requested, held in a short-lived cache for a few minutes and then discarded; it is not stored as a profile, and our analytics do not record which iD was looked up. The preview never shows citation figures, indicators or an h-index for a researcher who has not signed in and chosen them; those appear only on a page its owner published. The name search at /search asks OpenAlex for researchers with an ORCID iD and shows the names and affiliations OpenAlex already publishes; results are held only in a short-lived memory cache for a few minutes, nothing about the query or the people listed is written to disk, no figure is shown, and each result leads to that preview. Because the person shown may never have used SigmaCV, this notice is the information we owe them (GDPR Art. 14(5)(b)); the legal basis is our legitimate interest in the accurate display of already-public bibliographic records (Art. 6(1)(f), with Art. 5(1)(d), accuracy), and under the APPI the purpose of use of this acquired data is displaying public research records on request. If the iD entered belongs to a SigmaCV account holder, we additionally read that person's own disambiguation corrections and apply them: works they marked “not mine” are removed, and the “may not be yours” flag is cleared on works they checked and kept. We read nothing else from their account — not their display choices, private notes, saved layouts or contact details — and this can only ever remove a work or remove a warning, never add anything a visitor would otherwise not see. If you would rather your corrections were not used this way, or do not want an automatic preview of your record shown at all, email privacy@sigmacv.org and we will exclude your iD.
Research use (opt-in only)
SigmaCV is also a research project on responsible research assessment and author-name disambiguation. Logging of your curation choices for research is OFF by default and happens only with your explicit, separate consent (GDPR Art. 6(1)(a)). You can withdraw at any time from your account; withdrawal stops future logging and erases the research records already collected. Confirmatory analyses are pre-registered and conducted under an ethics/IRB protocol. If the scope of research logging materially changes, we ask for your consent again.
AI drafting assistant (optional)
If you use the optional AI drafting assistant, you supply your OWN AI provider's API key — it is held only in your browser's memory for the session (never saved to your browser or on our servers). When you request a draft, a short, public-only summary of your visible research outputs (the module, output counts and a few titles — no contact details or identifiers) is relayed from our server to the provider YOU chose, under YOUR key, and nothing about the request is stored or logged. Where your data goes and how it is handled is governed by the provider and terms YOU pick, so choose one you're comfortable with (for example an EU provider such as Mistral AI if EU processing matters to you). The feature is off by default, you must explicitly request each draft and accept the on-screen notice, and the result is never added to your CV automatically. If you paste the text of a call you are answering, it is sent with that summary to your provider and remembered only in your browser.
Retention
We keep your account and CV data until you delete your account. Deleting your account removes your profile, CV, and any research records from the live service immediately. Encrypted backups are kept on a rolling schedule and overwritten within 14 days, so a deleted account disappears from them within that window; we do not restore deleted accounts from backup. Published public pages stop being served immediately when you unpublish or delete. For 30 days after a record stops being harvestable over OAI-PMH (you unpublish your page, turn indexing off, leave an institution set, take a work off the page, or delete your account), the endpoint keeps answering for that record's identifier with “deleted”, the date and, in a harvest of an institution set it has left, that set, so the repositories that harvested it can delete their copy; the identifier contains your page's address, which includes your name.
Security & cookies
We protect your data with HTTPS, access controls, and database-backed sessions (sign-in provider access tokens are dropped, not stored); the code has undergone independent security audits, and we minimise what we hold. We use a single strictly-necessary cookie to keep you signed in, and no advertising or tracking cookies. For aggregate usage statistics we run privacy-friendly, cookieless, self-hosted analytics — first-party only, with no third parties and no cross-site tracking — so no cookie banner is needed. No online service can be perfectly secure, but we take reasonable, audited measures.
Your rights
You can access and export all your data (a one-click JSON export), correct it in the editor, withdraw research consent, and delete your account and all associated data at any time — directly from the app. Under the GDPR and APPI you also have the right to lodge a complaint with your data-protection authority.
If you have never used SigmaCV, your public research records may still be shown in the on-request preview described under “Public ORCID preview”. You can object at any time (GDPR Art. 21): verify your ORCID iD once at /object (no account is created, only a keyed hash of the iD is kept), or email privacy@sigmacv.org with your iD and we will honour it within 30 days. Either way the automatic preview of your record is answered as if no public record existed, without asking for a reason, and we do not tell visitors that you objected. You also have the rights of access, rectification and erasure over any data we hold about you, and the right to complain to a data-protection authority.
Contact
To ask a question or exercise your rights, email the data controller at privacy@sigmacv.org or use the contact page. We aim to reply within 30 days.
This notice is provided in several languages for convenience; if the versions differ, the English text prevails.
We may update this notice; material changes to research logging require renewed consent.