Privacy

On this page: who is responsible · reading the site · private responses · machine access · tools · children · security · where information is processed · how long information is kept · why information is processed · your rights · changes

Who is responsible

The Goodwill Ledger is a personal publication and collection of software tools operated by Veemal Gungadin, based in Mauritius.

For anything about privacy or information held about you, write through the private response form.

Reading the site

There are no user accounts, advertising, behavioural analytics or tracking cookies on the publication itself. Its typefaces and normal page assets are served from this domain.

Reader preferences in the header, such as colour mode, accent, density, width and font size, are kept in your own browser's local storage. They are not sent to the site, and clearing your browser data removes them.

The site is hosted on Cloudflare. Like any web infrastructure provider, Cloudflare processes technical information required to deliver and protect web requests, which can include IP addresses and request metadata. The site's own application code does not store visitors' IP addresses, and application request logging is switched off.

The RSS feed and machine-readable index contain published content only.

Private responses

The response form asks for your name, email address, optionally the page or essay you are responding to, and your message.

Submitting it sends the message to the author as email. The site's application does not keep a separate database copy.

The message is then held as ordinary correspondence and is used to read and respond to what you sent. It is not published without your permission.

If you ask for correspondence to be deleted, it will be removed from the author's active mailbox, subject to any temporary backup or recovery periods operated by the underlying email service.

Two anti-abuse measures apply. A hidden field that ordinary visitors do not fill in causes obvious automated submissions to be discarded. Submissions are also rate-limited per IP address. The resulting counter is temporary, exists only to prevent abuse, is not written to the site's database and expires automatically.

Machine access

AI systems and other software can read published content through the site's read-only Model Context Protocol server. It provides access only to information that is already public. It does not expose drafts, private responses or unpublished material.

The service does not create profiles of people who connect to it or keep persistent histories of their requests. Requests may be rate-limited using temporary technical identifiers such as an IP address. Those identifiers are used only for service protection and expire automatically.

If you connect an AI product or another third-party application to this service, information you provide to that product is also subject to that provider's own privacy practices.

Tools

The tools section contains software created and published by The Goodwill Ledger. These may include demonstrations, utilities, calculators, educational tools, research tools, experiments, prototypes, AI-powered applications or other forms of interactive software.

Different tools may work in different ways. Some operate entirely within your browser and send nothing you enter to a server. Others may require server-side processing, temporary storage, browser permissions, external APIs or other technical services in order to perform their function.

The guiding principle is to collect and retain the minimum information reasonably necessary for the tool to work.

Where a tool asks you to provide information or grant access to something on your device, the tool should make clear what is happening at the point where it matters. Depending on the tool, this may include explaining:

• what information is being used;
• why it is needed;
• whether it remains on your device or is transmitted elsewhere;
• whether it is stored;
• how long it is kept;
• whether another service provider processes it;
• whether a browser permission such as camera, microphone or location is being requested; and
• whether an AI or automated system is involved.

Information collected for the operation of one tool is not used for an unrelated purpose simply because it is technically available. The tools do not sell personal information, use it for advertising, build advertising profiles or engage in cross-site behavioural tracking.

Identifiers and sessions

Some tools may need to distinguish one browser, participant or session from another in order to function. Where necessary, they may use randomly generated identifiers, first-party cookies or browser storage for this purpose.

These identifiers are designed to provide the functionality required by the tool rather than to establish someone's real-world identity. Where an identifier can still distinguish one user or session from another, it should be treated as pseudonymous information until it is deleted or irreversibly aggregated.

The tools do not intentionally use device fingerprinting techniques to identify people across sessions unless a specific tool clearly explains otherwise before doing so.

Information you enter

A tool may ask you to enter information as part of its operation. Where possible, processing is performed locally in your browser so that the information does not need to leave your device. If information must be transmitted to a server or external service for the tool to work, that should be made clear within the tool.

Sensitive information such as passwords, authentication codes, payment-card details or financial account credentials should never be entered into a Goodwill Ledger tool unless the tool explicitly states that it is designed to handle that kind of information and explains how it will be protected.

Browser permissions

Some tools may demonstrate or make use of browser capabilities such as the camera, microphone, location, notifications or access to files. Your browser or device remains responsible for asking you whether to grant those permissions.

Granting a browser permission does not necessarily mean that the underlying information is transmitted or stored. Where a tool uses a permission, it should process only what is necessary for the stated function and explain any storage or transmission that takes place.

You can revoke browser permissions through your browser or device settings.

AI and external services

Some tools may use an external API, cloud service or artificial-intelligence model to perform part of their function. Where information you provide is sent to such a service, the tool should tell you that this is happening and identify the purpose of the transfer. Only information needed for that function should be sent.

External providers process information according to their own infrastructure and contractual terms. Where their processing is material to your privacy, the relevant tool should make that clear before the information is sent.

Organisers and shared sessions

Some tools may be designed for use with a group, classroom, organisation, event or facilitated session. In those cases, an organiser may be able to see information about the session.

Where possible, shared views are designed around aggregate results rather than identifying individual participants. Any access to private presenter, administrative or moderation functions should be protected separately from public participant access. If a particular tool allows an organiser to view participant-level information, that should be clearly disclosed within that tool.

Children

The publication is written primarily for adults, but reading the site does not require someone to identify themselves. Some tools may also be used in schools or with younger audiences.

Tools intended for children should, wherever reasonably possible, be designed so that they do not need to collect personal information. Where a tool processes personal data relating to a child, the appropriate consent and safeguards required by applicable law must be in place. Additional protections may be applied for younger audiences, including disabling particular browser permissions, media capture or other functionality.

Security

Reasonable technical measures are used to protect the site and its tools against unauthorised access, abuse and accidental disclosure. These can include encrypted connections, access controls, short-lived credentials, rate limiting, separation between public and administrative interfaces, and automatic deletion of temporary data.

No internet service can guarantee absolute security, but tools are designed to minimise both the information collected and the period for which it exists.

Where information is processed

The publication and its tools principally use Cloudflare for web delivery and application infrastructure. Cloudflare operates internationally, so technical processing associated with delivering the site or operating a tool may take place outside Mauritius.

A tool may use another service provider where its function requires it. Where that involves material processing of information you provide, the tool should identify that processing when it occurs.

Personal information is not provided to third parties merely because it might be commercially useful to them.

How long information is kept

Information is kept only for as long as reasonably necessary for the purpose for which it was collected.

Many tools may keep no user-provided information at all. Others may keep temporary session information only while the tool is being used, or for a limited period afterwards where this is necessary for the functionality being provided. Where a tool retains personal or participant-level information, the applicable retention period should be disclosed within the tool.

Information may also be transformed into aggregate statistics that no longer identify or distinguish individual users. Those statistics may be retained for longer.

Infrastructure providers may maintain temporary backups or recovery copies after information has been removed from the active application. Those copies expire through the provider's normal recovery cycle and are not used as active application data.

Private correspondence is retained for as long as reasonably needed or until deletion is requested, subject to any temporary backup or recovery periods.

Why information is processed

Where personal data is processed, it is used only for a defined purpose and under an appropriate lawful basis under applicable data-protection law, including the Data Protection Act 2017 of Mauritius.

Depending on the circumstances, this may include your consent, providing a service or function you requested, protecting the security and integrity of the service, complying with a legal obligation, or another legitimate and lawful purpose. Where processing relies on consent, that consent can be withdrawn.

Your rights

Under the Data Protection Act 2017 of Mauritius, you have rights in relation to personal data concerning you. Depending on the circumstances, these may include asking whether information is held about you, obtaining access to it, correcting inaccurate information, requesting deletion or restriction, objecting to certain processing, and withdrawing consent where processing is based on consent.

Because many of the tools deliberately avoid collecting identifying information, there may sometimes be no reliable way to determine which pseudonymous record belongs to a particular person. Information may also already have been automatically deleted or converted into aggregate information by the time a request is received.

For private correspondence, provide the name and email address you used. For a tool, provide whatever non-sensitive information is reasonably available to help locate the record, such as the tool used, approximate date and time, or a session identifier if one was provided. Write through the private response form.

You may also make a complaint to the Data Protection Office of Mauritius.

Changes

This page may change as the publication, its infrastructure and the kinds of tools published here evolve. A tool may also provide additional privacy information where its particular functionality requires it. The date below records the most recent revision of this page.

Last updated 18 September 2026.