Wikipedia co-founder: I no longer trust the website I created

But for non-political subjects, wiki is still great IMO.

https://larrysanger.org/2021/06/wikipedia-is-more-one-sided-...

He says this seemingly as a "liberty loving conservative". I don't find the arguments about these particular topics very compelling. And these article are naturally politically charged...

I'd be more interested in a discussion of bias in less politically charged articles, because I'm sure this also exists to some degree.

It's not that the sources that en.wikipedia uses to arrive at the opposite conclusion are even banned or neglected, or the reliable source policy is different at all - the difference is largely achieved by structure, prioritisation and editorialising, such as immediately offsetting every statement at odds with the narrative end with a healthy dose of FUD (e.g. the Unit 731 article's mention of https://en.wikipedia.org/wiki/Yoshio_Shinozuka is a four-line paragraph of which the last two lines are to the effect of "however, his statement has been cast into doubt because POWs in China have been known to be brainwashed into self-criticism, and moreover Unit 731 did not have a Youth Brigade at the time") while letting the ones in agreement stand uncritically. Looking at the talk page, it is clear that the articles tend to be maintained in that state by a number of opinionated power users who have far more devotion and domain knowledge than any drive-by editor in disagreement can bring to the table, which is basically the same thing as the English Wikipedia is being frequently accused of (though perhaps more stark here because they look to be so obviously in the wrong to the Anglo-American observer).

I do wonder how this problem could be solved in a systematic way. It does seem to me that the effect of powers users who feel ownership of an article or topic is basically universally negative the moment the article becomes a subject of political controversy (as the power users tend to use their power to defend the undue weight given to their preferred viewpoint, at the expense of the NPOV ideal), and this perhaps would suggest that a sort of no-questions-asked https://en.wikipedia.org/wiki/Ostracism mechanism might be good (make it so any editor that becomes too prominent can and will be easily banned from editing an article; if they are actually right, then surely others will take their place in defending their position), but it is also easy to imagine that this rule would be gamed to detrimental effect by off-site canvassers (say, a real-world club of homeopaths could over time muster enough warm bodies to outright take over some article on medicine).