We all know what this debate is (we see it argued every other day it seems) so let's digress to the positions themselves.
First position:
type is a static, genetically-coded construct. Type is defined as the arrangement of personality functions in a definite sequential order of information processing (Model-A), with each function possessing determined by the individual's biological characteristics. Type is considered a structured concept. This is the view espoused by Aushra Augusta, founder of socionics. It is also expressed in the psy, when Jung notes that raitional functions must always follow irrational functions, and vice versa. (Jung's functions correspond to Augusta's information elements) Jung also posits solidly that type does not change in people of physiologically sound mental health.
Second position:
type is prioritization of thought processes, and is measured by the frequency of focus on specific aspects of information. A person who primarily considers logic is a logical type, a person who primarily focuses on feelings is a feeling type. From this point there exist two subpositions: those who consider the plausibility of static type as a means of seeing different aspects of the primary mental focus, and those who reject it out of distrust for biological constancy (specifically, the concept of order in general), and fear that type may entail a limitation of theirs or society's ability to adapt to changing phenomena. In support of this position it is necessary for them to reject Augusta's theory of information elements, because the theory explicitly postulates that the capacity exists for the socion to respond effectively to all observable phenomenon due to the specializations inherent to static type.
The former position is solidly in favor of nature; the latter for nurture. Therefore, the type question is another manifestation of the nature vs. nurture debate. It is notable that the former position does not exclude the latter as regards the question of content consideration; indeed, there is substantial evidence it is the various aspects of base function content that one most often considers. However, the extreme form of the position, taken to its logical conclusion, denies all constancy of type in normal individuals, and is inconsistent with all empirical research.