Interesting breakdown! I usually work across these sections in a descending ratio of effort to likelihood of elimination:

Fast:

- Tone detection: Not definitive, but sets expectations.

- Logical fallacies: Most things fail here.

- Clarity: Similar to fallacies, people like making overly broad statements that don't mean anything.

- Check claims: This is often easy these days.

Slow:

- Context: If I don't understand it, I won't be able to evaluate it further.

- Arguments: The real meat, and where most of the time will be spent (if if hasn't failed already)

- Bias detection: Again not definitive, but can be easy to evaluate once you've made it this far.

- Assumptions: The second heaviest step.

- Source: Is almost an afterthought. "Reputable" sources make bad arguments all of the time. "Seedy" sources make good arguments more often than you expect. Source is often more related to bias than correctness. nostr:note1yck6mfk69tfj5lxusgqk2slkvxwh0a0uteej32r626gdtt7qslmsnusejk