A data scientist makes the following statement:

I cannot see how hierarchical clustering can be of any use to data science. It may work for small, "toy" problems, but it is not appropriate for the massive data sets that we work with every day. Whenever I face a real problem that requires cluster analysis, I use k-means partitioning.

Do you agree or disagree with this data scientist? Explain your thinking. Provide examples from your reading, research, or work experience to support your answer.