Ask HN: When would you fork instead of waiting for proposal/PR acceptance?

I love Go's templating system for its simplicity and frankly being part of the stdlib. I am however in need of a parser extensions. I created a proposal for this (https://github.com/golang/go/issues/59453), but like many large open source projects, the proposal queue is substantial with uncertain review timelines. I'm considering forking just this part of the stdlib to implement my needed changes in a backward-compatible way.

This raises a broader question: At what point do you decide to fork and maintain your own version versus waiting for upstream acceptance? What has been your experience with similar situations, especially when dealing with core libraries or language features?

I guess one sure reason to fork it is if the proposal is completely shot down and for some reason the proposer absolutely cannot live without it - but let us not go that far!