RadarStudio Contribution Privacy Notice
Version 2026-10-11.1

Christopher Kalkowski personally operates the contribution program from Colorado Springs, Colorado. This notice describes the contribution feature, including its sign-in and request records. It does not mean contribution uploads are enabled. The interface displays whether submissions are available.

Privacy and data-rights contact: privacy@swarmcord.com

What is optional

Sharing is off by default. Local preview preparation does not send the recording. When submission is available, you review the minimized contribution and affirmatively consent to each disclosed purpose before submitting. Declining does not prevent local viewer use. Visiting this hosted site or signing in can still involve provider processing described below.

Information in a contribution

The current format is radar-contribution version 1 with source original-radar-v2. It contains a random recording identifier, segment indices and relative sensor timing, limited radar position, range, radial velocity and native power measurements, spatial flags, and numeric quality, truncation and lost-history information. It accepts up to 256 frames and three peaks per frame.

The contribution payload excludes raw UART bytes, phasors, derived tracks, stable source-session or device/build identifiers, absolute capture time, original filenames, names, contact details, geographic-location fields, free text, an explicit vital-sign channel, photographs, audio and video. These exclusions describe the recording payload, not all account, request or provider information.

Radar measurements can still reveal body, movement, surrounding-environment or sensitive information. We do not promise that they are anonymous or impossible to identify. Excluding explicit vital-sign fields does not guarantee that health-related information cannot be inferred.

Account and administration records

Sites-managed ChatGPT sign-in supplies a Site-scoped user identifier and email address to the sign-in helper. The contribution service uses the identifier and does not store or display the email address or a name from that helper.

We keep the identifier linked to submitted recordings, submission timestamps, permission attestations, terms and privacy versions, accepted purposes, consent receipts, review events, submission-limit records and rights requests. This linkage can identify a contribution with your account. If you contact us by email, we receive your email address and whatever information you include. Please send only what is needed for your request.

Purposes

Contributing requires affirmative permission for all of the purposes below. You can decline any of them and continue using the local viewer without contributing:

- Private review: receive, store, inspect and assess the contribution and manage its acceptance, quality and permissions.
- Research: analyze radar measurements and create or evaluate research datasets, methods and results.
- Model development: train, test, evaluate and improve machine-learning models using the contribution and derived data.
- Commercial use: develop, improve and commercialize RadarStudio, radar-related products, services, datasets, models and derived materials.
- Redistribution: share, publish, license or sublicense contributions and derived datasets or materials for the stated research, model-development and commercial purposes, including public releases or paid licenses where lawful.

Account and administrative records are also used to authenticate access, provide receipts, enforce submission limits, prevent misuse, review permissions, respond to requests and meet legal obligations. Processing that relies on consent is limited by the permissions recorded with the submission and applicable law. The intellectual-property license does not replace required privacy consent.

Who receives information

Christopher Kalkowski can review contributions and their associated private records. An authenticated contributor can access their own records through the account features; another contributor cannot read them through the contribution API.

The prepared service uses OpenAI's ChatGPT Sites platform and sign-in, Cloudflare Workers hosting, Cloudflare D1 for structured records and private Cloudflare R2 storage for recordings. These providers process information to operate their services.

Contributions start in private storage. There is currently no automatic public-release or model-training endpoint. Where permitted by the recorded processing choices and applicable law, later recipients of recordings or derived datasets may include researchers, model developers, commercial licensees and members of the public. Commercial and public redistribution concerns the contributed recordings and derived materials; private account, consent and request records are maintained for administration rather than offered as dataset content. Information may also be disclosed where legally required.

A public release can be copied by others. We cannot promise complete recall of every downstream copy. This does not excuse obligations to honor applicable rights or to take legally required steps with recipients.

Hosting, sign-in and technical logs

The contribution application does not implement IP-address, user-agent or request-body logging or tracking analytics. Platform observability is enabled in the prepared hosting configuration. Providers may process technical information such as IP addresses, request and browser information, traffic-routing information, and security or operational logs. This notice does not promise zero provider logs, a particular provider log-retention period, or storage only in a particular country.

Provider handling, retention and international processing are described in their official policies:
OpenAI: https://openai.com/policies/privacy-policy/
Cloudflare: https://www.cloudflare.com/privacypolicy/

Provider policies do not replace Christopher Kalkowski's responsibilities for this contribution program. Provider-controlled records may require a request through that provider's stated process; contact us about records handled for this program.

Retention

Submitted recordings have no fixed voluntary expiry date and may be retained indefinitely for the disclosed purposes where lawful. Applicable legal retention limits and required deletion take priority over that policy.

Account linkage, consent receipts, review/security records and rights-request records are separate from recording files. They may remain after a recording is removed where needed and lawful to establish permissions, administer contributions, handle requests or disputes, protect the service or meet legal obligations. Deleting a recording does not automatically erase those records. We apply any legally required deletion or retention limits to them as well.

Technical cleanup of incomplete storage operations is not a deletion deadline for accepted recordings. Provider logs and backups follow the applicable service arrangements and legal requirements; the program does not promise a single fixed expiry for them.

Requests and choices

Contact privacy@swarmcord.com for access, correction, deletion, withdrawal of processing consent, objection, or an appeal of a request decision. You can also use the account's data-rights features when available. Include the relevant recording or receipt identifier if you have it. We may request information reasonably needed to verify the request and locate affected records. Do not email passwords, identity documents or unrelated sensitive information unless a necessary secure process is separately arranged.

Applicable rights depend on the governing law and circumstances. We will handle legally required requests within applicable deadlines and explain a denial or any available appeal process. You may complain to the relevant privacy regulator where that right applies. We do not charge or penalize you for exercising rights where prohibited by law.

Where processing depends on your consent, you may withdraw it. Withdrawal does not make prior lawful processing unlawful, but future processing and retention must have a lawful basis. A perpetual intellectual-property license does not override that requirement. We make no blanket claim that derived data or trained models are exempt from applicable privacy rights.

Updates

The version shown when you submit is recorded with your receipt. We will update this notice when practices change and seek additional consent where required before affected processing. Publishing an update alone does not authorize a new use of an earlier contribution.