Perhaps it is over ambitious to start with the goal of producing
only code destined to end up in Phobos. The domain is so broadly
defined, and the standard to aspire to so high that one ends up
setting the goalpost so high that given likely contributors one
risks ending up running out of steam before getting very far.
Also, the domain of things that maybe should never be in Phobos
but are nonetheless important and lack a well documented,
industrial strength solution - this domain is not tiny.
Maybe biting off a small chunk so you can show what you mean, and
understand in practical terms what is involved might be the first
step on a longer journey. That's actually realistically how one
would have to approach it anyway even were you to get universal
buy-in, which doesn't seem to be on the cards. It is easier to
convince people as momentum picks up than with just an idea.
Pick something not too large but useful and start working on it ?
