We all think of 'crazy' experiments we'd like to try 'just for the fun of it'. They are crazy either because the technique is something we are not too familiar with, or the chance of getting a definitive result is pretty slim or simply because we don't know what to expect. The other day, I was reminded of this when reading Andre Geim's interview. He talks of 'Friday evening experiments', wherein he tries things that aren't part of his main line of investigation - things that are fun but unlikely to produce any new data most of the time. As he states it, it takes courage - to just go for it - and long hours - just so you don't take time away from your main projects.
I am not that free-spirited to try things that are too far removed from my main area of study, but I do want to try techniques I've never tried before and phenomenon I've not tried to look at in my main projects. Starting today afternoon. If not every saturday afternoon, atleast every once in a while. If not anything else, I'll atleast learn something.