Algorithm visualization is an educational tool that is already used for many years to show the progress of an algorithm by updating a visual representation of a data structure. They usually cover CS1 and CS2 algorithms. However, algorithm animations and high level data structure visualizations are not yet widely used in algorithm learning.

The demonstration presents Algovision (www.algovision.org) that tries to overcome the problems of the existing algorithm visualizations by using the following design principles: Algovision is implemented in JavaScript to be immediately available at any device with a web browser; it does not offer simple visualizations, but full visual algorithm lectures; it is based on feedback of several hundreds of students and many teachers that have already used it in courses; it also incorporates some rather advanced algorithms; it visualizes not only the action of the code, but also the underlying algorithmic idea; it offers an embedded guide with optional synthetic voice output.

The most important and innovative feature are ``playgrounds'': interactive scenes, where learners can prepare their own inputs (trees, graphs, $ldots$) and play with operations that are later used in algorithms (e.g., rotations of binary tree edges, color flips in red-black trees, node joins and splits in B-trees).

Playgrounds make instruction much more efficient and highly motivate learners. Advantageously, playgrounds are also used at home or in cooperating student groups.