I've been using this line a great deal over the last couple of years in presentations: "Curate the content, not the container." It resonates with people who want to get at the meat of a book or article or video and not just the clever wrappings that entice you to consume content. It stems from a paper I wrote in graduate school in the late '90s, wherein I analyzed the ability of a searcher to find a known piece of content in a library containing anthologies and other such collections of small works.

Let's say you were trying to find a poem that had meaning for you. It was popular enough to be included in various collections of "The Greatest Sentimental Poems of the 19th Century," but not quite popular enough to be the shining star of such a collection. Current library practices (continuing this day) dictate that the collection be curated with it's collection title, editor, dates and so forth, and only a select few of the individual poem's titles get cataloged. It simply isn't efficient, nor does the system 'allow' for more. When you go to the library's catalog, you search on what you know - the title of poem you are looking for, and you will likely not have any results. Frustrating. You ask the librarian, but s/he cannot recall seeing it in any books in their collection. You take a few random books off the shelf and scan the contents to no avail. Then you give up and hope that Google can help you. How much more annoying to learn that the poem you wanted is in a book on the shelf in your library, but you couldn't find it in the library's catalog? Technology is no longer a limiting factor. Publisher's have sophisticated content management systems that should be configured to pass along this metadata. So why are we still failing to use that metadata?

The key part is "curate the content." The average consumer does not demand the information inside the book be curated. Business analysts have been doing so of their research portals - I know from having worked at Dow Jones that Factiva puts a great deal of effort into tagging people, companies and subjects in articles. Why aren't more content publishers doing the same? Do students want their textbooks to contain back of the book indexes only? Does a hobbyist not want to be able to find examples of a challenge they face in their learning?

It means to index not just the title, creator, publisher, dates and a few key words of an object as we do in libraries, but to tag all of the core concepts - the nouns, the verbs. I can't help but think of Schoolhouse Rock, "a noun is a person place or thing." (But don't tag all of the conjunctions, as much fun as "Conjunction Junction" was, it would be overkill for 99% of digital content!)

I've also modified the line a bit. I didn't want to imply that we shouldn't curate the container! During my librarian days I had many a patron who asked for something "which had a green cover." I appreciate the value that level of cataloging brings.

There are many activities in commercial and academic research organizations attempting to address this problem. Social tagging, browser plugins, more granular CMS systems, new SEO methods, NLP methods. They are all providing some value. We each need to find the tools right for us, and do our part in tagging our own content to bring greater value to the whole. So I ask you:

Curate the content AND the container.