Hello from an SLI

[Myst]

My typings are never intended to be set in stone, I only list standalone evidence and findings. I'm well aware of the potential flaws of those facial composites and the mistypes within them BUT in my opinion most of them seem rather accurate. The may be some individuals way off from those but the law of large numbers dictates the more people in a sample the closer you get to a singular result. Its a matter of probability instead of this is right or wrong.

How big were these samples? What is the guarantee that they will converge?

[Muddy]

How big were these samples? What is the guarantee that they will converge?

Heh, I always give Ti mixed with Ne and then you always want pure Ti. Never fails.


The samples used there were around 10-15, not very big I know. I always compare stuff with my own experience to form an intuitive judgment of whether something is correct and I'm pretty confident in my ability to do this. Of course, the drawback of that method is that it is not always easy to convince others of it's validity as I'm sure you know. I've taken statistics class and I'm very familiar with how results homogenize when the sample size increases. I am rather certain the genes and lifestyle of the types effect their appearance, and given large enough samples, you would start to see distinct kinds of looks for each of the types. The might be exceptions, but again, its an undeniable correlation. Obviously more research still needs to be done on VI for it to become more reliable. Until then, I'll keep using my intuition.

[Myst]

Heh, I always give Ti mixed with Ne and then you always want pure Ti. Never fails.

Hahaha right.. Or Ti with Se also works

The samples used there were around 10-15, not very big I know. I always compare stuff with my own experience to form an intuitive judgment of whether something is correct and I'm pretty confident in my ability to do this. Of course, the drawback of that method is that it is not always easy to convince others of it's validity as I'm sure you know. I've taken statistics class and I'm very familiar with how results homogenize when the sample size increases. I am rather certain the genes and lifestyle of the types effect their appearance, and given large enough samples, you would start to see distinct kinds of looks for each of the types. The might be exceptions, but again, its an undeniable correlation. Obviously more research still needs to be done on VI for it to become more reliable. Until then, I'll keep using my intuition.

OK my problem is that the convergence relies on strength of those correlations between type and physical appearance... it may not be a very strong correlation.

And 10-15 is nothing.