Interactive Type Dichotomy Tutorial

[Sol]

The end goal of this project is a synthesis

You are using non-classical hypothesis like it has the trust close to classical part. It's not acceptable. Reinin's traits are not even close to Jung's dichotomies in the sense of basis. They are like other typology, as there is nothing good to think those traits are linked to Jung's types like it's claimed. Even their descriptions differ between different authors, even own Augustinavichiute's text has contradictions, unclear and strange parts (look Lytov's article). It's low quality hypothesis, while you mix it with core Socionics theory like equal.

[Lao Tzunami]

You are using non-classical hypothesis like it has the trust close to classical part. It's not acceptable. Reinin's traits are not even close to Jung's dichotomies in the sense of basis. They are like other typology, as there is nothing good to think those traits are linked to Jung's types like it's claimed. Even their descriptions differ between different authors, even own Augustinavichiute's text has contradictions, unclear and strange parts (look Lytov's article). It's low quality hypothesis, while you mix it with core Socionics theory like equal.

I'm in the process of reading the article (which may take a while) and I understand that some empirical definitions do not fit the model A descriptions, but let me ask you this. If you consider quadra to be "classical", then should quadra depend on the four Jungian dichotomies, or can any type be in any of the four quadras (effectively increasing the number of types to 64)? If quadra does depend on type, then that is a testable hypothesis. You can have a test for quadra, and a test for type, and they have to statistically coincide, or one or both tests are wrong.

I also understand that synthesis is not the same as integration. There are lots of bad ideas in the socionics community that should not be used. What I want to do is rank them based on how well they work with other dependent combinations. Sort of like Cronbach's Alpha, it would be a measure of internal consistency. You could then iteratively try different combinations until you found the best combination.

EDIT: I finished skimming though the article. The author seems to care very much about the names of the traits rather than the actual empirical items. If it would make people happy, we could refer to all reinin dichotomies with their types code, like I could say I am <ET+, N+, ENP-, ETP+>, but no one would know what I am talking about. We could also change the names every times someone came up with a bright idea, and have as many ways of referring to the dichotomies as we have referring to the types (should I say I am IEI, INFp, Yesenin, the lyricist, a romantic, NiFe, NiFi, NiFx, Ni-F, TE, TR, , ?) Personally, I am extremely annoyed I have to learn every person's individual jargon, and want to keep and use the current reinin dichotomy names until we get everyone on the same page.

That being said, I don't have the time right now to analyze everything that is said in the article, but I can say that I have been pleasantly surprised when I have tried to actually apply the reinin dichotomies to people I am trying to type. Based on the descriptions, not the names in the 2003 reinin dichotomy study, this is my experience so far:

Great:
Irrational / Rational
Logic / Ethics
Intuitive / Sensory
Merry/ Serious
Yielding / Obstinate
Process / Results

Good:
Extrovert / Introvert
Carefree / Farsighted
Judicious / Decisive
Constructivist / Emotivist

Iffy:
Democratic / Aristocratic
Positivist / Negativist
Static / Dynamic

Unusable:
Tactical / Strategic
Asking / Declaring

But again, trying to evaluate all of them at once is too much to handle. Every if some of them do not work, we should go one at a time.