the idea of “active documents” leaves me cold.

Regular folk are going to need a word that refers to a particular collection of words that are gathered to a particular purpose, and word is “Document”. Don’t take that away just because applications are changing.

And, what is changing? The file and data formats. So, say “file and data formats”.

The whole mistake was assuming that the terms “document” and “document file” was synonymous. There not.

So, our data formats for documents are changing and will surely end up as some distributed collection of dynamic content. What you’re looking for is a new word for a collection that represents such a distributed dynamic document.

So, these “DynaDocs” or “DistDocs” or “CollectoDocs” are just a new data format. Let’s not have another Buzz-Word Bandwagon that looks down upon those of us who continue to use words according to their traditional meaning.

We produce lots of “documentation”, but little of it in discrete “documents”.

Different users and organizations have different requirements in this space. Those who place very high priority on such tracking will inevitably continue to use static document formats into the foreseeable future, just as legal folks today will send around fully rasterised PDFs of important documents in the belief that those are more reliable and trustworthy.

While it’s use of the term is inspired by Lotus Notes (which itself produces the “post-documents” that you are writing about here) I think the term is quite helpful in the technical context. In distinguishing between what tools like CouchDB, Notes, and Google Docs manage, and the normalized data patterns we deal with in the relational database world, the notion of a document as a relatively stand-alone piece of data holds some water.

As you mention – documents, like web pages, are getting more intertwined with dynamic data sources, but they are still much closer to their archaic paper and PDF counterparts, than the rows and tables of a SQL store. The singular feature we’ve picked out, in writing the book, is that documents are meant to (more or less) make sense on their own. Documents carry at least some of their own context. This is why documents (like web pages) can travel well. Taking them away from their original location does not strip the meaning from them. This does not hold for a SQL row.