These Rules are published under Article 10.2-2 of Russian Federal Law No. 149-FZ of 27 July 2006 "On information, information technologies and information protection". They describe where the Frolick service (the "Service") uses recommendation technologies, what information about user preferences they rely on, where that information comes from and how recommendations are formed.
Recommendation technologies are information technologies that present information based on collecting, organising and analysing information about user preferences.
The owner of the Service is Ulyana V. Stepanova (self-employed, taxpayer ID 544052568288). Email address for legally significant messages: support@frolick.ru. Other details are on the About the operator page.
2. Where recommendation technologies are used
Community catalogue (available after signing in): the games, content sets, lessons, boards and widgets sections. Materials are ordered as "Popular" by default; in the games and content sets sections you can switch to "Newest".
Popular picks on the catalogue overview and on the workspace home page: a few community materials.
Authors page: a block of popular authors.
"More from this author" on a material's page: other materials by the same author.
Smart search over content sets in the catalogue (turned on with a switch).
Content set suggestions when creating a game or lesson.
Pre-filled AI generation settings (difficulty level and age group).
The Service shows no advertising; recommendations are not used to serve ads.
3. What information is used and where it comes from
Information about the materials themselves: type, title, description, the content of a content set, subject, tags, author, creation and update dates.
Usage counters the Service keeps automatically as users act: the number of times a game was launched, a lesson was copied from the catalogue, a board or widget was used in an embed, the number of games built on a content set, and the number of uses in the last 7 days. The counters do not record who used a material.
Your current request: the search string, selected filters and sort order, the material you have open.
Information from your account — only for suggestions to you personally: the content sets you created, the sets you recently used, and the settings (difficulty level, age group) of your recent AI generation requests.
All of this comes from users' actions in the Service. The Service does not buy user data from third parties, does not use data from other services and does not share your preferences with third parties.
4. How recommendations are formed
"Popular" — materials are ordered by their total all-time number of uses: games by launches, lessons by copies from the catalogue, boards and widgets by uses in embeds. Content sets are listed by last update, and each page of results is ordered by the number of games built on them.
"Newest" — materials are ordered by date.
Popular picks — the materials with the most uses in the last 7 days, then all-time; if there are not enough, the remaining places are filled with a random selection of public materials that changes once a day and is the same for all users.
Popular authors — the authors whose games have the most launches in total.
"More from this author" — a few other public materials by the same author.
Smart search — the text of a content set (title, description, items) and your search query are converted into numeric vectors on the Service's own servers, without external services; the sets closest in meaning to the query are shown.
Content set suggestions — three lists: your sets, the sets you recently used, and recently updated public sets.
Pre-filled AI generation settings — if several of your recent AI generation requests share the same difficulty level and age group, the form proposes them by default.
No individual interest profiles are built. Catalogue results, popular picks and smart search are the same for all users with the same query, filters and sort order; only content set suggestions and pre-filled AI generation settings depend on your personal data. No individual material is promoted manually.
5. Controlling recommendations
In the games and content sets sections you can switch the order from "Popular" to "Newest". You can leave smart search off.
Content set suggestions and pre-filled AI generation settings are only suggestions: you can ignore or change them without losing access to any feature.
If new recommendation technologies appear in the Service or the way recommendations are formed changes, these Rules are updated before the change is applied.
Questions and legally significant messages about these Rules: support@frolick.ru or the Contact page.