One of the “knocks” against this multi-step approach is that there is a danger of “propagation of error”. That is, each step contains a level of uncertainty, so after several steps the uncertainty accumulates until finally you have no confidence whatsoever in your process. In principal, I agree with that. As a matter of fact, I would say that the situation is even more distinct here: unlike propagation of error for a continuous physical quantity, my steps are binary in nature. That is, either they are the “right thing” to do, or they are wrong. The best one can hope for is to maintain confidence via the “breadcrumbs” I have mentioned. If any one of my previous steps is wrong, then my breadcrumbs are weak, and my current process is nothing but a crazy random walk in the park. Is that what this is? You can find plenty of crazy theories for solving K4 on the internet. Is mine just another one of them? Well, let’s take some time to identify some of the common qualities of those crazy theories:
-
(1)Steps are taken without any discernible motivation other than the results that they lead to. In other words, random things are tried until something “neat” seems to spring out.
-
(2)The “neat” things that spring out often involve one sifting through large arrays of text hunting for groups of neighboring letters, or patterns, that suggest words. This is an “alphabet soup” methodology, and the problem with it is that humans are very good at finding patterns in large arrays. This is especially true if they have the freedom of attempting numerous random steps provided by common quality (1) above.
Now, let’s compare my theory with respect to these qualities above:
-
(1)Every single one of my steps has been motivated by quotes, clues, or both. Often, the steps were the most simple interpretation possible under the circumstances. Example: once having observed the “Doorway”, instructions (1) and (2) from K3 were applied using the most obvious interpretation available. No “reaching” for significance was performed during each step. Any oddities observed after each step are therefore independent.
-
(2)The “neat” things that spring out of my work are “in your face” obvious, with no need for alphabet soup or scrabble methods. Also, the significance that resulted from a step often fell into a highly constrained category that was consistent with the premises. (In other words, I was not “free” to invent any significance that I could come up with.) Examples: the positions of the misspelled characters conjured the image of a “doorway” as described in K3; the shape of the array after debris removal and candle insertion is right for a flat map of the Earth, as described in K2; and K2 lands on the “abscissa”, as the keyword suggests.
The question, then, is whether my breadcrumbs are compelling. In spite of the fact that Jim Sanborn admits to the use of visual clues and that the locations of the misspelled letters are important, suppose that:
-
-you are not convinced that the “doorway” is significant all by itself. (Call this event A.)
-
-you are not convinced that the “flat map shape” of the subsequent array is significant all by itself. (Call this event B.)
-
-you are not convinced that K2 aligning with the “abscissa” is significant all by itself. (Call this event C.)
That is fine, but you must note that these things are not occurring all by themselves. They are independent observations occurring simultaneously and connected only by the simple application of a set of reasonable, and highly motivated, steps.
Some reviewers have opined that my theory is a crazy random walk and that this is all one big coincidence. They assert that the uncertainties surrounding these observations should be accumulated, resulting in decreased confidence. On the contrary, in my opinion, these independent observations serve to increase confidence. The question is: what is the probability that “unlikely event A”, “unlikely event B”, and “unlikely event C” all occur simultaneously by pure chance? To compute the combined probability, one multiplies the independent probabilities (not the “confidence values”). The result is a lower probability that pure chance is the driver, which means increased confidence in the significance of my process. It is a matter of “preponderance of the evidence”, and the key of it is that I maintained independence of observations by not forcing the interpretations toward significance. My assessment at this point is that my “breadcrumbs” are strong and so it makes sense to continue down the path and see where it leads.
