I'll throw in an utterly pointless post to counter the previous one written by lemons.
Socionics sucks when attempting to figure out the individual.
That is only because it has not been naturalized and humanized toward individuals of your specific culture. For me it is quite easy to understand the patterns within socionics to attribute them to self growth. I see more distinction among people with socionics than I do with MBTI. It might take someone like me who understands the material to naturalize it to thousands upon thousands of people, and that will come in the form of reorganizing the theory and giving it the impression of a fresh approach while keeping all the same principles. To someone like me, all of this has already been materialized.
Whatever you choose to believe that I-S-T-P or I-S-T-j represents is meaningless since both MBTI and Socionics can only hook a reader into their way of thinking if the reader only learns by using basic dichotomies.
Although this idea is quite valid, there are still many in whom I have encountered several dichotomy switches, and it is because neither the functions or dichotomies are exactly the same. Even seeming obvious to you that they are the same, you have to take a fresh approach to the theory and not make any orientation to MBTI. Not everyone is going to have both Tis or both Ses. For that matter I am glad they called it Ti in socioinics and not something else, because now people will easily start to contrast the two, which is what should happen. If they called Ti, "Structural Logic," I am sure there would be more intention to say it is the same thing as Ti in MBTI. But socionics is not some watered down version of MBTI. It is a new level of self-thought and relationships. Unfortunately it has some unofficial extremities in form of obstacles in the way, being formed into hundreds of new theories. It is a good thing, but many of the theories need to stop making branch upon branch. People need to learn to grow their own trees.