Recommendation Technology Policy

This Policy is published pursuant to Article 10.2-2 of Federal Law No. 149-FZ of July 27, 2006, On Information, Information Technologies and Information Protection. It describes what information the Service collects and analyzes to select materials and how it applies these technologies.

Email address for legal notices: legal@multator.org.

Last updated: 24 августа 2026 г.

1. Where recommendation technologies are used

Automated content selection and ordering are used on the home page, in the New, Popular, Following and Sandbox feeds, category and tag sections, similar works on animation pages, search results and the creator leaderboard.

2. Types and sources of user preference information

The sole source of information is the Service itself: data generated by the User’s actions and public metrics of published works. Information from third parties, advertising networks and external services is neither obtained nor used. Results are generated using:

• information about works: publication date, category, tags, title and description, and the creator’s selected visibility and section (catalog or Sandbox);
• public work metrics: likes (including during the selected period), views, community ratings and comments;
• the User’s explicit actions: follows, favorites, search queries and selected categories, tags or sorting periods;
• paid placement information: a work placed by its creator in the Featured spot appears there for the paid period.

The following are not used to generate results: not used:

• preference profiles based on viewing history, time spent interacting with content or other User behavior;
• information from external sources, advertising networks or other third parties;
• special categories of personal data, health, religious beliefs, political views, race or ethnicity;
• information obtained in violation of Russian Federation law.

3. Processes and methods

Information is collected automatically when a User acts (follows, favorites, searches or selects a category, tag or period) and when other users interact with a work (like, rate, view or comment). Organization consists of storing these actions in the Service’s database and counting them for each work. Analysis is limited to sorting works by publication date, accumulated actions during the selected period, matching categories and tags, or textual relevance to the search query. No other methods, including machine learning or behavioral profiling, are used. The method used in each section is described below.

• New — all catalog works in reverse chronological publication order. There is no personalization; everyone sees the same feed.
• Popular — catalog works ranked by likes during the User’s selected period, then by total likes and views. Everyone selecting the same period sees the same feed.
• Following — works by creators the User explicitly follows, in reverse chronological order. The feed depends only on the User’s explicit choices.
• Sandbox — unfinished and practice works, in reverse chronological order, filtered by the selected section.
• Categories and tags — works assigned the relevant category or tag by their creator, in reverse chronological order.
• Similar works — works from the same category or with matching tags. Selection is based on the work’s properties, not User data.
• Search — catalog works matching the search query, ordered by textual relevance.
• Creator leaderboard — creators ordered by public measures of their work: likes, average ratings and rank. The method is the same for everyone and does not depend on who views the leaderboard.
• Featured spot — paid placement in a dedicated block on the home page. The block is visually identified and separate from regular feeds.

4. Application and User rights

• The Service informs Users about recommendation technologies through this Policy, linked in the footer of every page.
• Users can avoid algorithmically ranked results entirely by using the chronological New feed, Following, specific search queries, categories and tags.
• The Service does not create a separate preference profile, so a way to delete or disable one is unnecessary; Users can change their follows and favorites at any time.
• Recommendation technologies are not used to violate the rights and legitimate interests of individuals or organizations. A User who believes results infringe their rights may contact legal@multator.org; a response will be sent within 30 calendar days.
• Personal data processing during Service operation is described in the Privacy Policy, and the general terms of use are in the Terms of Service.

5. Changes to this Policy

The Administration updates this Policy when content selection algorithms change. The current version is published on this page with its date.