My first thought was to pick something new so I "would be ahead in the game." In one or the other order I tried Perl 6 and Elm. Perl 6 lacked a coherent set of systematic resources at the time and Elm was then so unstable that tutorials fell apart within a few minutes of the most basic 'Hello World.'
I regrouped and decided to avoid the cutting edge and started thinking about learning things that are extremely stable instead. For me, these were Emacs, Linux, and Javascript. Javascript because I had been avoiding it (hence Elm); Emacs and Linux because I was using them all the time.
It's true I don't spend a lot of time learning Javascript, Emacs, and Linux these days. What happened was I learned enough of each that it's easier to learn more if I need to for example, sometimes I start with the man page before Googling and other times I stop Googling and just read the man page.
I'm not recommending Javascript, Emacs, and Linux per se. But what they gave me was a framework for deciding what to learn. It has to be worth spending a year with. It has to be small enough to fit in one person's head from scratch -- a small head in my case. Team sized technologies are out of the running.
To put it another way, I focused on learning "techs" I could live with. And one of the ways of recognizing those "techs" was that I was already living with them and that was in part because they had long lives. Sometimes I was using them already.