<p>
I still strongly disagree with this premise, which is not to say I think there is a “best” route.</p>
<p>Obviously, a person who goes the tech route should do so in a lab that allows them to have an independent project, and in a lab that does work they find interesting – in short, they should treat the year(s) as a tech as an extended graduate school rotation. In addition, they should try to pick up all the techniques and skills they can, the merit of doing which, I hope, is obvious. (Importantly, though, the students who came to my program after a year or two of tech work did not uniformly have these opportunities. They still got into my program. Overall, the students who did tech work come from weaker schools and with less impressive accomplishments than the students who came straight from undergrad, indicating that, yes, tech work has some sort of “PhD virtue” to it.)</p>
<p>And quickly, sooner, whatever. There’s no advantage to rushing your training period, particularly because hurrying at one stage can significantly set you back at another.</p>