This is an excerpt from our research paper, which will publish in concert with the StartWithXML Forum on January 13th at the McGraw-Hill Auditorium in New York. Early bird discounting for BISG members is ending soon!
A good taxonomy is the backbone of your business — it’s how you sort your content. It allows for effective merchandising, effective marketing — you can aim your content with the precision of a pool cue. It allows for inventorying your content — so you know what you have … and what you need. With your content tagged and organized, you know where everything is and how to deploy it.
Taxonomies are contextually sensitive and rigorous — and in establishing your own, it helps to look at what other industries are doing. Wiley has adopted accounting and cooking and psychology taxonomies from those industries to organize information in its professional development titles. Educational publishers are increasingly arranging their textbooks around “learning objects” — taxonomized pedagogical goals developed by educators themselves. Even the BISAC codes — which are part of the ONIX system of organizing book information and therefore an XML-based taxonomy — are developed very carefully and consensually among book industry professionals in monthly meetings.
An important aspect of taxonomy development is scope notes. Terms need definition and clarity around how they’re going to be used. Documenting your taxonomy — what you mean when you say “porcelain” (collectible china, dental work, household fixtures?), parent-child relationships between categories, and why you choose certain terms over others — is important for the long term. Future editors and authors will need to know why your taxonomy has developed as it has.
Consistency in application is also crucial. Drop-down menus (as opposed to free-text fields) enforce structure and ensure that users don’t come up with their own terms that pollute your taxonomy with duplicates or irrelevancies (or misspellings).
An advantage to using XML is that you don’t have to accomplish everything at once, perfectly, from the outset. You will not be able to tag your documents thoroughly right off the bat — who can know everything in advance? The act of tagging is recursive, and depends on market and company needs. XML allows for this flexibility. Depending on how you envision chunking and re-use, you’ll tag your documents differently with each iteration. Unlike the “fire and forget” model, iterative tagging means that your books are living documents.