Issue Management

[erictraut]

There are various philosophies for how to manage issues within an issue tracker. In my 30-year experience in the industry, I've found that it's best to be aggressive about closing out issues rather than let them linger. Some techniques I tend to use:

1. Cull older issues. If an issue is really old and has received little or no attention or interest, it's unlikely to ever be addressed. Keeping it active is probably not in the best interest of the project, so close it. It can always be reopened later.
2. Use reactions and responses to inform interest. If an issue is interesting, it will receive "thumbs up" reactions and message responses. If little interest has been shown in an issue, it's reasonable to close it as "won't fix".
3. Use GitHub's "discussion" feature to move general questions out of the issue tracker. It's easy to move posts between discussions and issues in either direction.
4. Don't suffer fools. There are a small number of people who regularly post bugs and feature requests that are poorly-motivated and would have almost no impact if addressed. These consume a lot of time for the mypy maintainers — time that could be better spent on high-impact issues, documentation, etc. Establish expectations that these low-quality issues are not appreciated and will be summarily closed if they persist.