@jprint714: I’m not an advocate of “the ‘one thought - one notecard’ school (I believe this is from the Bill de Ville school!), which involves selecting a paragraph for annotation, and therein a separate note annotation with its own tags describing just that clip of information”–although I sometimes use such tools.
Most of my annotations use the Annotation template, which will likely contain multiple notes about a document and may link to other notes and documents as well. Very often my notes form a kind of network via links.
As to tags, I got my baptism of fire using keywords/tags to do information searches back in the 1960s-70s, when I was director of a university computer information center with the mission of disseminating federally sponsored R&D relevant to environmental issues. Full text searches were not available back then on computer tapes that contained a couple of million references to documents. When we received a query for information, we had to translate it into keywords that were likely descriptive of the topic, and then (via punched cards) into a search query. At the other end, the federal agencies that created the computer tapes used keywords to convey the information content of each document.
If we were lucky, my staff had used keywords that were the same as those used by staff at a federal agency, and the search pulled documents that were highly relevant to our client’s request for information about a topic.
Why did I use the term “lucky”? The two major problems with using keywords/tags to define the information content of a document are: very often topics that are present in a document are missing because the reviewer ignored or failed to recognize a topic, and people also tend to be inconsistent in applying keywords/tags to a topic. These are very serious problems. They have been discussed extensively in the literature of information science.
Although the federal agencies spent a lot of effort to train the persons who assigned keywords/tags to documents, the problem of “missing topics” simply cannot be solved without spending a lot of time and effort to identify all potentially important topics in a document. In practice, it’s simply too expensive to do a comprehensive job. As a result, information that might be important in a search will not be found.
In spite of personnel training and the provision of “dictionaries” of keywords/tags, different persons are likely to use different terms to define a topic, and the same person exhibits that problem at different times. This makes it more difficult to design a search query, to take into account probable inconsistencies in application of keywords/tags.
Given that experience, I use tags very sparingly. I don’t assign them at the time new content is added. I learned that to assign comprehensive tags to each of the more than 30,000 documents in my main research database would take years of time and effort. And I learned that such assignments would turn out to have inconsistencies. The purpose of my database work isn’t to fight the problems of comprehensive and consistent tagging. Instead, the purpose of my database work is to do things with its information content that meet my needs and interests. I consider tagging a tool, but a limited one that is best used when it saves time and effort rather than burning time and effort without serving my real purpose.