This thread is an attempt to get some sort of clear discussion going between two different approaches to Socionics.
CONTENTISTS
On one hand, you have what I call the contentists. These people point to descriptions of typical interests associated with the quadras as being the most reliable typing method. For example, in their view, being interested in intellectual discussion "just for the fun of it" basically means one is Alpha. Interest in drama and artistic expression would be Beta. And so on. Generally, they don't put a lot of stock into temperaments, as these often contradict typings based on content/interest-based quadra descriptions.
The contentists often point out that the quadra descriptions are an official part of Socionics, seem to be connected with common-sense definitions of the functions, and provide a clear and compelling approach for typing. The chief weakness is that content-based typings often don't fit with temperament; for example, a person may have "Alpha interests" but be N and have Ip temperament. Also, content interest isn't always stable; a person may be interested in many things, which would appear to cause someone to be in more than one quadra.
STRUCTURALISTS
Then there are the structuralists. These people believe that people have a certain style of thinking and expressing themselves that's more fundamental than any particular content interest. They see content interest as something that may be quite flexible. For example, in their view, a person may be interested in what the quadra descriptions would say are Alpha sorts of things, but still not be Alpha.
The structuralists often point out that type is something that is fundamental and ingrained, and not dependent on whatever subject matter a person is focusing on at any given time, or even on changes in a person's belief system or philosophy (e.g., changing between a materialist and anti-materialist point of view). The chief weakness of the structuralist view is that it doesn't seem to either explain the contentist viewpoint or state clearly what's wrong with it. Hence, contentism is never fully defeated.
ATTEMPTS TO MERGE THE VIEWS
Then there's Tcaud's whole exertion thing, which I don't quite understand, but I think it's just basically embracing the contentist and structualist approaches and therefore giving a person two types. The challenge here is in fleshing out the details of what it means, for example, for someone to be "INTp-INTj," and proving that such types exist in real life, and aren't just the result of misunderstanding.
One possible related approach is to think of one's structural type as sort of like one's basic hardware or operating system, whereas content types may be more like various software programs that ride above it. It isn't clear that there has to have just one predominant content type; it seems that one may have several.