Because expert opinions are sometimes the only data available. "What will computer architecture look like in 20 years?" Clearly there's no factual content to answer that question, but I would argue that it's still an interesting question to ask an expert.