Trust should be inspectable

How Plentropy decides what to show—and what not to claim.

The site changes the explanation, never the underlying scientific fact. It recommends a depth softly, dates changing claims and keeps personal learning data on your device.

01

One fact, several doors.

Story, everyday and rigorous views share a canonical concept. Age and comfort change examples, pacing and notation—not accuracy.

02

Models expose assumptions.

Every entropy estimate depends on a source and attacker model. Interactive outputs identify themselves as illustrations, toy models or sourced editorial claims.

03

Plenty has a reason.

The Plenty Line starts with a threat and recognized security target. It also shows the point where algorithms, implementation or key storage become the tighter limit.

Local means local

Your path belongs to your browser.

No account. No name, email or birth date is requested.

No trackers. The v1 site contains no behavioral analytics or advertising scripts.

No human-input storage. Keyboard and mouse experiment samples live only in component memory and clear on reset, navigation or refresh.

Your control. Inspect, change or erase the saved learner profile from the start page.

Editorial evidence

Stable standards and dated snapshots.

Changing claims about computing and hardware display an “as of” date, review date and evidence class. They are reviewed quarterly rather than fetched from an unexamined live API.

Explicit boundaries

What Plentropy is not.

Access is part of accuracy

Every exhibit needs more than color and motion.

Interfaces target WCAG 2.2 AA, work from the keyboard, honor reduced motion, provide visible focus, use text labels beside color and pair visual charts with readable summaries or tables.

Inspect my local profile →