How our random question generator works

How a question is selected

1. You define the eligible pool. “All categories” uses the full library. A category filter limits the pool to that topic. On the wheel, the first draw chooses a category and the second chooses a prompt from it.
2. The browser chooses an unbiased index. In supported browsers, RQG uses crypto.getRandomValues() and rejection sampling. That avoids the small bias that can occur when an arbitrary random number is mapped directly onto a list.
3. The chosen prompt is marked as seen. During the current visit, each pool remembers its displayed prompts. RQG does not reuse one until every eligible prompt in that pool has appeared; the pool then resets.

The random source comes from the browser, not from a prearranged wheel sequence. The Web Crypto standard describes getRandomValues() as producing cryptographically strong random values. You can read the primary W3C specification or the practical MDN browser documentation. Older environments without this API use a Math.random() compatibility fallback; the generator is for conversation and games, not security-sensitive decisions.

How the question library is reviewed

RQG selects from a maintained library rather than composing new text on demand. Before publication, prompts are reviewed for clarity, variety, suitability, and fit with their assigned category. We revise or remove prompts that are confusing, overly repetitive, or unlikely to produce a useful conversation.

• A prompt should be understandable without extra instructions.
• Its wording should match the purpose and tone of its category.
• The Kids & Family collection is separated so visitors can choose explicitly family-oriented material.
• Trivia, General Knowledge, MCQ, and Exam prompts include a checked answer, a short explanation, a topic, and a difficulty label.
• MCQ prompts keep answer choices separate from the question and require exactly one option to match the checked answer.
• Time-sensitive or easily disputed facts are dated, qualified, replaced, or linked to an authoritative reference.

Context still matters. A prompt suitable for close friends may not fit a classroom or workplace, so hosts should choose the category that matches their group and skip anything that does not feel right.

Difficulty labels are editorial guides, not standardized grade levels. Exam questions are general practice prompts rather than curriculum-aligned assessment items. If you find an unclear answer, stale fact, or broken source, send a correction; we review reported items before the next publication.

How quiz rounds and scores work

On the Trivia, General Knowledge, MCQ, and Exam pages, choose a topic and difficulty before starting a round. Each round draws different questions from the matching pool. If fewer than five or ten match, the tool explains that the round will be shorter.

MCQ rounds check your selected option automatically. Revealing or skipping an MCQ does not earn a point. For open-ended questions, reveal the suggested answer and mark your own response as correct or needing review. Those results are labelled self-assessed. At the end, review missed or skipped questions and their explanations. Filters stay fixed during a round so its question set stays consistent.

What stays in your browser

Question selection and quiz scoring run locally after the page loads. Quiz progress lasts for the current page visit and resets when you reload. Saved favorites are stored in your browser’s local storage and do not require an RQG account. Optional traffic analytics load only after consent; the full details and controls are in our Privacy Policy.

What the wheel animation means

The wheel’s visual travel and duration vary slightly so each spin feels natural. The result is selected before the animation finishes, then the wheel is rotated to the matching segment. Visitors who request reduced motion receive a shorter animation while keeping the same selection method.