Tagging arose from library classification systems. The key thinker was Shiyali Ranganathan (1872-1972). Peter Memes was the first to introduce tagging on the web.

The usefulness of tagging depends upon context.
In a bookmarking service the user is effectively building a relational database where the tags represent both field headings and the query criteria.
Unfortunately most users have no experience of databases or the process of normalisation. Consequently their tag cloud billows in size and they end up adding more and more tags in the hope of finding something again, a process that becomes self defeating.

A great many people also have great difficulty in grasping the concept of tagging. Hence so many 'cool' 'cute' & 'funny' tags applied to just everything. Only of limited use and increasingly less so when overused. (More to look through.)

Imagine a library where people could put books back where they expected to find them. In some cases it would lead to a more intuitively logical order. In other cases it would lead to books becoming unfindable by anyone other than the person who placed it on a shelf. (Unless you had enough copies of each book for everyone.)

Interestingly I now pay more attention to tracking down people I surmise think like me (Friends & Fans in bookmarking) rather than individual tags