Gradually general principles of classification and knowledge organizations have become more clear to me. An important insight is that any classification of a domain must be based on theories or views of that domain. I try to verify this on different domains. This is not always easy, and some of the most influential classifications seems to contradict this principle. This is, for example, the case with Linnaean classification in biology, which was developed before Darvian and modern evolutionary theory and it is the case with the influential psychiatric classification DSM IV.
Ereshefskys book on the Linnaean classification in biology represents high class scholarship in both biology and philosophy and it is a goldmine of arguments for a general philosophy of classification. Besides it is an indication that philosophy can make itself a productive part of of specific specific scientific research.