Then that's why that principle should not make it into the guidelines. It counters the long-standing principle of choosing a default interpretation of a person's behavior based on general and specific behavior over time. Mark Zuckerbergs business began with a series of crimes/deceptions, increased with even more where he betrayed his customers in various ways, he continued to lie about the value of privacy for ad revenue, and proved it wasn't just his belief by protecting his own privacy buying up all those houses near his. In business & personal life, he stayed a selfish, control freak who destroyed friend, stranger, and foes wherever for ego and esp profit. This pattern of behavior stretches back years.
So, it's rational to default on the belief that... after years of such behavior... his next move will be motivated by selfish gain (ego/control probablg) with potential deception. Also note that character assessment and source integrity ratings are standard practice of experienced journalists, police, and intelligence gathering. Example below in ratings section:
https://en.m.wikipedia.org/wiki/Intelligence_collection_mana...
Also, my brief reading of Charity link suggested it fails instantly if person talking is a con artist or habitual liar. Like Zuckerberg but Frank Abignale of Catch Me if You Can even better example. Or litany of PR people working for companies always getting caught doing devious things with cover stories. The Empirical, evidence-over-time model handles both the good and the bad while the Charity model effectively encourages people to act like sheep in front of hungry, intelligent wolves. So, Im strongly against it being part of the guidelines of any forum.
Now, I do get the concept and still thank you for the link. Ill think on it more in the near future as I can see room for self-improvement. I suggest a lightweight version of it where we try to assume a reasonable/charitable version of a comment by default then reply to that or ask clarification before slamming it. No charitable version if person, claim, or topic has a consistent meaning that's neutral or negagive. Probability applies then. Misinformation that's easily disproven by linked evidence, as I often deal with, would be example. Maybe combine this with "keep it civil, focus on claims, and no personal attacks" a la FreeBSD's Code of Conduct. Might make a nice compromise between charity principle and reality.
Just a few of my thoughts for the team's consideration. Just concerned that guideline could censor comments that need to be there while protecting subtle trolls or unreliable sources from quickly being dealt with. It was a good idea but just goes too far into idealism in that link. Especially for discussions on... the Internet. ;)