Privacy Policy

The short version

SanskritKatha Feedback is a research platform. We collect only what we need to run the platform and conduct our research. We do not sell your data, show you ads, or share your personal information with third parties. Your review scores may appear in an anonymized research dataset. That is the whole story.

Who runs this

This platform is operated by Mahesh Ramakrishnan, an independent researcher based in Bengaluru, India. It is part of the SanskritKatha research project, which studies AI-generated Sanskrit stories. This is not a commercial product. There is no company behind it, no investors, no business model.

What data we collect

When you use the platform, we collect:

• Account information: your email address, name, affiliation (e.g. university or school), and self-assessed Sanskrit proficiency level. We need these to understand who our reviewers are and to acknowledge contributors in publications.
• Review data: your scores (1-5) on six quality dimensions for each story, an optional overall impression score, optional free-text comments, and the time you spent on each review. This is the core research data.
• Session data: a session cookie (a random identifier) so we know you are logged in. This is purely functional.
• Audit data: your IP address and timestamps when you log in or submit reviews. This helps us detect misuse and maintain data integrity.

That is everything. No browsing history, no device fingerprinting, no location tracking beyond IP address.

What we do NOT collect

• No third-party analytics (no Google Analytics, no Mixpanel, nothing)
• No tracking cookies or advertising cookies
• No third-party scripts of any kind
• No passwords (we use magic link authentication)

How we use your data

• Research: Your review scores are aggregated and analyzed to validate AI-generated quality scores for Sanskrit stories. This research targets publication at academic venues (ACL/EMNLP 2026).
• Quality control: We compute trust scores based on review patterns (consistency with other reviewers, performance on calibration stories, time spent). This helps us identify careless or bad-faith reviews and maintain data quality.
• Communication: We use your email solely to send magic login links and, occasionally, invite notifications related to the platform. No newsletters, no marketing.
• Acknowledgment: Reviewers with a trust score above 0.5 are acknowledged by name in research publications, unless you opt out.

Anonymized research data

Review scores may be released as part of an anonymized open research dataset alongside our academic paper. In that dataset, your personal information (name, email, affiliation) is stripped. You would appear only as a reviewer ID (e.g. "reviewer_042") with your proficiency level and review scores. There is no practical way to re-identify you from the anonymized data.

Where your data is stored

All data is stored in a SQLite database on a server hosted by Hetzner Online GmbH in Germany. The server is managed via Coolify (a self-hosted deployment platform). Data is not replicated to any cloud database service. Backups are stored on the same Hetzner infrastructure.

Cookies

We use exactly one cookie: a session cookie that stores a random session identifier. It exists solely to keep you logged in. It is not used for tracking, analytics, or advertising. There is no cookie banner because there are no optional cookies to consent to.

Email

We send emails via AWS Simple Email Service (SES) for two purposes only:

• Magic link login emails (when you sign in)
• Invite notifications (when you are invited to the platform)

We do not send marketing emails, newsletters, or promotional content of any kind.

Data sharing

We do not sell, rent, or share your personal data with anyone. Period.

The only form of data sharing is the anonymized research dataset described above, which contains no personal information.

Our email provider (AWS SES) processes your email address to deliver login links. They are a data processor, not a data controller, and are bound by their own privacy commitments.

Data retention

Your account data and review data are retained for the duration of the SanskritKatha research project. We expect this to be 2-3 years from the date you create your account.

The anonymized research dataset, once published, may persist indefinitely as part of the academic record. It contains no personal information.

Your rights

You can:

• Access your data: Your profile page shows your account information. Your review history shows all your submitted reviews.
• Request deletion: Email [email protected] and we will delete your account and all associated personal data. Note: anonymized review scores that have already been included in a published dataset cannot be retroactively removed, as they contain no personal information.
• Withdraw: You can stop using the platform at any time. Simply do not log in again. If you want your data deleted, let us know.

India's Digital Personal Data Protection Act, 2023

This platform processes personal data of individuals in India. We are committed to complying with the Digital Personal Data Protection (DPDP) Act, 2023, as its provisions come into effect. In particular:

• We collect data only for a clear, stated purpose (research evaluation of Sanskrit stories)
• We collect the minimum data necessary for that purpose
• We provide mechanisms to access and delete your personal data
• We do not process children's data (this platform is intended for adult reviewers)
• We store data securely and do not share it with unauthorized parties

As the DPDP Act's rules are finalized, we will update this policy as needed to remain in compliance.

Changes to this policy

If we make material changes to this policy, we will update the "Last updated" date at the top. For significant changes, we may also notify active reviewers by email.

Questions?

Email [email protected] with any questions about this privacy policy or how your data is handled.