Legitimate Interest Assessment

What is processed

When someone visits a Sifa profile page for an Atmosphere user who has not claimed their profile, Sifa fetches basic identity data from the public Atmosphere network: handle, display name, and avatar. This data is cached in memory for up to 5 minutes. It is never written to our database. No activity stats, professional history, or other enriched data is shown. The profile page displays a notice that the profile is unclaimed, with links to the privacy policy and the removal request form.

Legitimate interest

Sifa's purpose is to provide a professional identity layer for the Atmosphere ecosystem. Displaying basic identity pages makes Sifa discoverable as a directory and benefits profile visitors (recruiters, collaborators, peers) who want to look up any Atmosphere user's professional presence. Atmosphere users also benefit from discoverability: an unclaimed profile page serves as a prompt to claim and enrich their professional presence.

Necessity

To display a profile page, Sifa must fetch and render identity data. There is no way to show a useful profile page without the handle, display name, and avatar. Showing only opted-in users would eliminate directory value entirely. Only three fields are processed (handle, display name, avatar), which is the minimum set needed for a recognizable profile page.

Balancing test

Conclusion

The data subjects' interests and rights do not override the legitimate interest. The processing is minimal (three public fields, no storage, short cache), the data is already publicly available by the user's own choice on a protocol designed for public access, and a self-service removal mechanism ensures any user who objects can stop processing immediately.

What is processed

When a visitor views a claimed Sifa profile, Sifa reads recent activity content (posts, code commits, articles, etc.) live from the profile owner's data server or the relevant app. This content is cached in memory for 5 to 15 minutes depending on content type. It is never written to our database. Aggregate activity statistics (which are stored) are covered by our separate Data Protection Impact Assessment.

Legitimate interest

Displaying activity content is essential to Sifa's value proposition as a portable professional identity platform. Users who claim their Sifa profile actively seek to showcase their professional activity. Profile visitors gain access to authentic, externally-sourced professional activity rather than unverifiable self-reported claims.

Necessity

Caching is necessary for performance. Without caching, every profile page view would require multiple live API calls to the user's data server, creating unacceptable latency and placing disproportionate load on data server infrastructure. The current 5 to 15 minute TTL balances performance with freshness.

Balancing test

Conclusion

Legitimate interest is appropriate. Direct relationship, aligned expectations, minimal processing (volatile cache only), comprehensive opt-out controls.

What is processed

When a user imports their LinkedIn profile data through Sifa, we log when the import happened and how many items were imported per category (for example, 12 positions, 8 skills). We do not store the imported content itself. This metadata is stored in our database and retained until account deletion.

Legitimate interest

Import metadata enables Sifa to understand product usage patterns, identify import issues, and improve the import experience. It also provides an audit trail if a user reports problems with their import.

Necessity

Without import metadata, Sifa cannot diagnose import failures, understand which LinkedIn categories users most commonly import, or provide support when users report issues. The metadata is the minimum set needed: timestamps and counts, not content.

Balancing test

Conclusion

Legitimate interest is clearly appropriate. Minimal data, direct relationship, standard operational practice.

What is processed

Sifa uses Umami for website analytics. Umami is open-source, cookieless, and self-hosted on our own infrastructure in the Netherlands. It collects anonymous, aggregate data: page views, referrer URLs, browser type, operating system, screen size, and approximate country. No data is sent to third parties. Umami cannot link analytics data back to individual users or Sifa accounts.

Legitimate interest

Understanding how visitors use the site is necessary for product development, identifying technical issues, and making informed decisions about features and content. This is a standard and expected practice for any website operator.

Necessity

Without analytics, Sifa cannot understand which features are used, where users encounter friction, or how the site is discovered. Server-log-only analytics would be less accurate (bots, caching distortion). Third-party analytics (Google Analytics, etc.) would be more intrusive. Umami self-hosted is the most privacy-respecting option available.

Balancing test

Conclusion

Legitimate interest is clearly appropriate. The processing is anonymous by design, uses the least intrusive analytics tool available, and aligns with universal expectations.

What is processed

Sifa's servers log request metadata for each HTTP request: timestamp, HTTP method, request path, response status code, and IP address. Logs are stored on the server filesystem in Germany and rotated after a maximum of 30 days. They are not stored in the application database and are not used for analytics or profiling.

Legitimate interest

Server logs are essential for two purposes: debugging technical issues (identifying errors, diagnosing performance problems, reproducing bugs) and protecting the service against abuse, spam, and denial-of-service attacks. IP addresses are necessary for both purposes.

Necessity

Without server logs, Sifa cannot diagnose errors, identify abuse patterns, or respond to security incidents. This is a fundamental operational requirement for any internet service. Logging without IP addresses would prevent abuse identification. Shorter retention (under 30 days) would be insufficient for investigating intermittent issues.

Balancing test

Conclusion

Legitimate interest is clearly appropriate. Server logging is a fundamental operational necessity, universally expected, with proportionate retention and strong access controls.

What is processed

When an Atmosphere user requests removal of their unclaimed profile from Sifa (via sifa.id/privacy/removal), Sifa stores their DID (a permanent unique identifier assigned by the Atmosphere) and the date of the request on a suppression list. This list is stored in the application database and retained indefinitely.

Legitimate interest

The suppression list is the mechanism by which Sifa honours removal requests permanently. Without it, a removed profile could reappear after a cache flush, server restart, or any event that causes Sifa to re-encounter the identifier on the Atmosphere network. The suppression list directly serves the data subject's expressed wish not to appear on Sifa.

Necessity

The Atmosphere is a live, federated network. Sifa encounters identifiers through the event stream and direct lookups. Without a persistent record of which identifiers have requested removal, Sifa cannot distinguish between users it has never seen and users it previously removed. The DID is the minimum identifier needed (handles can change; DIDs are permanent). Time-limited suppression would eventually violate the user's removal request.

Balancing test

Conclusion

Legitimate interest is clearly appropriate. The processing serves the data subject's own expressed interest and uses the minimum data necessary to do so.