One event. Two headlines. Side by side.

Same event. Different headlines.

Matched pairs from RSS feeds across the spectrum. Similarity scores are mechanical — your judgment is not.

What this is — and what it isn’t

Not a “balanced” newsroom rating bias, and not both-sides-as-equal-trust either. A comparison lab for readers who notice when coverage is written to close your options down to two.

• Headlines are choices. We line up left- and right-leaning sources on the same beat — the adjectives, the villains, the framing.
• The gap is the story. When both feeds agree on a villain, the real story is what neither led with: funding, incentives, boring paperwork, the human on the ground.
• You read the sources. Every match links to the original reporting. Similarity scores are mechanical — your judgment is not.

Why not just Ground News? Bias ratings score each outlet separately. The Janus match puts the two headlines in one frame and asks what both left out — the thing a per-outlet score can never show you.