My first rule as a low-budget Twitter researcher is to collect interesting data first, ask (research) questions later. I have a lot of data I’ve never used, but I’d rather deal with that than a missed opportunity.