@Neuropolarbear opined:

Now that my lab is collecting data, I think my students would be more productive if I leave and stop pestering them for 2 months.

Stop pestering and stop helping them. And maybe I really mean postdocs rather than students. But I often think this.

Trainees would in many cases be better served if the PI said “Hey guys, I’m taking off on sabbatical, see you in six months or so.”

I’ve never had the nerve to try that one out though. I can’t see my way to risking my precious projects that way. Either that or I can’t bear the thought of having a trainee really bollocks up their training stint with me to the point that it is an abject failure…and think I could have been more involved in oversight.

I worry, however, that this hovering weakens them.

So how about it, PI-tariat? How long would you let your trainees go without having any decent idea what they were up to? More than two weeks? More than a month?

….Six months????