Moderation is important; it deserves to be done better than this

Firstly, there's the issue you identified: the ratio between effort to moderate and effort to spam/flame. If it requires all of this effort to ban someone that's _clearly_ acting in bad faith, then you create the perfect opportunity to troll. Trolling is the art of the output amplification attack: the ideal troll puts in a small amount of effort and induces others to put in large amounts of effort to combat him. Obviously if you allow this process to be short-circuited then you just go back to the original problem if that short-circuit process is abused.

Secondly, there's just the issue of the stakes. Ultimately it's a discussion forum for a programming language on the internet, and one of many. In the real world we have more rigorous decision-making processes before we convict people and sentence them for crimes, because the stakes are much higher! You can be tarnished with a conviction, fined, and even sent to prison. The required standard of proof in order to remove someone from an internet mailing list is surely lower.

Thirdly, the problem is quite different from the problems that arise in real life. It isn't usually about whether there is adequate proof of what is claimed to have happened, but whether what has happened actually contravenes community standards. What has actually been said is clear on the record: it's all right there in the mailing list/forum archives. But whether the conduct falls outside what is acceptable in the "community" (for lack of a better word) is really the question. And so it doesn't really matter what the rules say, because the rules don't usually correspond perfectly with what the community standards actually are.

If the rules are vague, that gives a lot of power to whomever interprets them. But on the other hand, if the rules are very specific, it gives more leeway to trolls that purposefully edge right up to the boundary of acceptable and unacceptable behaviour but don't cross it. If you make a list of unacceptable words, they'll find one you don't list and use that, then cry foul when you try to ban them. "It isn't on the list". But if you don't, then you allow moderators to claim that "chairman" or "master" are offensive.

There are reasons why all your rules are unacceptable in certain scenarios. For example, spam-detection can be an arms race where you lose if you tell spammers exactly why they're being banned. Security by obscurity is sadly the name of that game. Generally though, I agree that you should have to give reasons when someone is banned. However, sometimes people are banned because they're dickheads and that's all you need to say. This depends on the scale and stakes of the "community" though. Given that this is the official Python discussion forum we're talking about, it probably should lean towards more formalism than it has and more transparent decision-making. But it would be unreasonable to expect Jon Corbet to follow formal processes like these when banning people for obnoxious comments on this website. It's a private comment section on the website of a business.