what does research look like?

I was reviewing the notes from the last conference call in preparation for today, and saw that Mary had posed a question about what it would look like if students did open-ended bioinformatics research. To me, that means they would ask a question that could be addressed using bioinformatics (bascially sequence comparisons), determine what data (sequences) they need to collect, analyze the data (alignments, trees, etc.) and use the results to answer their question. The first step, posing a question, seems to be especially hard for students - so providing some sample question and some constrained data sets will probably make them more successful.