Comment moderation policy

Comments on this site currently run on Giscus + GitHub Discussions.

That is an intentional first step. The initial rollout prefers the lowest-maintenance and most auditable platform path. If moderation later needs finer private rules, stronger filtering, or a separate account system, the site can still move to a self-hosted backend without rewriting the public baseline first.

What is welcome

• corrections, questions, and additions that stay close to the article
• relevant examples, counterexamples, and implementation notes
• disagreement backed by reasons and facts
• feedback about layout, links, citations, or readability problems

What may be moderated

The following content is not a fit for the comment area:

• obvious ads, lead-gen drops, SEO link spam, or bulk junk
• phishing, credential theft, malware distribution, or wallet-seed/private-key bait
• threats, harassment, personal attacks, or deliberate flooding
• doxxing or exposure of personal information that should not be public
• off-topic content posted only to hijack the discussion or redirect attention elsewhere

What automation actually does

Automation is intentionally narrow.

It only acts on comments that look clearly like:

• ad or scam spam
• malicious links, credential-phishing prompts, or malware-style bait

When a comment matches that narrow rule set, GitHub may automatically minimize it and keep a platform-visible reason such as spam or abuse.

Automation does not replace human judgment for gray areas. Sharp disagreement, blunt criticism, or comments that need context are not supposed to be auto-moderated just because they are uncomfortable.

If a comment was minimized by mistake

If your comment was minimized and you believe that was wrong:

• open a GitHub issue and include the discussion link
• explain the original intent of the comment and why the minimization looks incorrect

Review will follow the public policy above instead of an undisclosed private rule set.