Bad interview

Today I did what was probably my poorest interview so far. I was carried away from my main subjects and didn't have the power to stir him back in.

The interviewee -- being a schooled journalist himself -- made sure to take a long time and add a lot of detail to the inoffensive questions and not so much in the ones I was really interested in, which, in turn, made the interview longer than necessary, very dull and not very useful.

Also he didn't tell me what he really thought, I think. I have the impression that either I failed in making him comfortable with the confidentiality with which I would treat the data, or he does believe in the success of the operation for the sake of his job. After all, waking up in the morning for something you think is doomed mustn't be easy.

What could I have done better? Interrupt him in the dull parts was out of question. Contesting what he said was impossible, because he was giving his opinion about things.

So my question is: what kind of science is this that ethnography does that is based on people's opinions? Is this what we are trying to capture: detailed accounts of what people think of ongoing processes? How much objectivity needs to be put in an interview to make it valid?