Context: You are collecting opinions about a soft-science discipline. You are using your judgement to decide which opinions you believe in.
Problem: How do you efficiently, and as objectively as possible, collect opinions.
Forces: Since you can't prove or disprove soft-science opinions, you must use your judgement about whether you like them. Analyzing an opinion and thinking about it is too much like trying to prove it; a better, and less time consuming way to judge it is just to see if you still believe in it as time goes by.
Solution: Collect soft science ideas and opinions without even judging them. Occasionally summarize your current beliefs. The summary should contain the ideas and opinions you currently judge useful.
Resulting context: This approach lets you separate the generation and collection of ideas from the judgement of them, and then later uses the test of time instead of reasoning to judge the ideas. It also lets ideas grow or fade gracefully; just because an idea or opinion is not in the current summary does not mean it is completely discarded - it may still be "rising in the charts" or "falling in the charts".
See also: UseJudgementNotProofs, CollectAndOrganize
It's not so much whether you like them, as whether you think they are significant. Insignificant stuff gets left out of the summary. You cannot avoid making this kind of judgement, but you can certainly include ideas that you dislike or don't believe in the summary. -- DaveHarris
For example, I don't believe in BigDesignUpFront ... but I admit it's a valid and popular theory. ;-> --RonJeffries