QUOTE(loc2k @ Dec 29, 2006, 12:15 AM)

The point here is that mathematics is, although formed from the real world around us, no longer just descriptive
Exactly, although I'm more spiritually inclined to ponder on the beginning and how it must have started from a simple process. This is why im intrigued by phi, Fibonacci and the regular n-gons.
QUOTE(loc2k @ Dec 29, 2006, 12:15 AM)

As far as spirituality, if you equate that with passion, you are sadly mistaken. I am passionate in all my mathematical research, none of which I have remotely revealed to you. If you mean spirituality in the sense in which it is typically used, that is an unfair charge. I believe math draws conclusions from physical/temporal space, and not that math is put in place by something or due to some spiritual reason.
Granted passion and spirituality may be perceivably different to you, but I believe you may be confusing religion and spirituality. To see a possible truth and be passionate about its attainment is the belief in yourself and all that mathematics stands for. To believe God made you do it, is a completely separate notion. I believe you are spiritually motivated, you defend your belief with conviction.
QUOTE(loc2k @ Dec 29, 2006, 12:15 AM)

For your information, the reason I brought this topic back to light is I believe I have found an answer for your original question, or at least an arrow in the right direction. It reminds me of a question Flex asked me a couple months ago about the Fibonacci sequence converging to certain constants. If you have ever read about probabilistic ways to derive pi (the famous needle and parallel lines experiment), you know that there are methods for the geometric convergence of constants. I believe such is the case for the golden ratio with respect to n-grams. There is a regular pattern in the series of n-grams. It is important to note that there may be more than one way to form any given n-gram (there are up to 2, at least for small n). For example, the 7-gram can be formed by vertices at k*2pi/7, k in K = {1, 2, ... , 7} (K congruent mod 7), with each line connecting vertices at a*2pi/7 to those at (a+2)*2pi/7, and it can also be formed by connecting vertices at a*2pi/7 to those at (a+3)*2pi/7, for a in K. It is easy to see the reason for this (isomorphism to cyclic groups). I point this out in case you have not covered the second (or perhaps higher) cases for each n-gram. Now the direction I'm suggesting is that we derive a convergence to phi based on this scheme. You noted that the 5-gram has special properties involving phi. Since the circle in which all n-grams are inscribed relates deeply to pi, the strategy is to find a n-gram sequence that converges to phi (using the fact that pi and phi relate). If this is accomplished, we have the "why" answer for the relationship between phi and pi through a strictly geometric, and thus physical, model. We may be able to make a separate argument using polar coordinates to avoid non-elementary functions (then apply a transformation). I'll post when/if I think of something.
Nice work! Are you saying that if you connect the first vertex to the third and fourth, then continue to apply that logic, that you end up recreating the original ngon? I think this is what you are saying, although I'm rusty with my maths - just starting a University course to brush up. I drew a lot of shapes manually with my daughter, to physically check phi and other ngons. I noticed the ngon did repeat, but almost always lines would be drawn through the new ngon.
QUOTE(loc2k @ Dec 29, 2006, 12:15 AM)

At this point, I take back what I said about 5 being arbitrary. This is very speculative, but since I mentioned groups above, the intimately related Galois theory may apply to your hypothesis about the 5-gram. Abel's impossibility theorem proves there is no closed-form general equation for 5th degree polynomials. (In fact, the quartic equation is the highest degree in which there is a closed-form general solution.)
I'm glad you are attempting to disprove what I originally proposed through your speciality and some new passion has been fostered. I merely spoke philosophically and am unable to say with confidence whether 5th degree polynomials have anything to do with what I spoke of. I will probably have difficultly appreciating what you put forward, but I am working on that.
QUOTE(loc2k @ Dec 29, 2006, 12:15 AM)

It is probably the case that we have coincidental 5's, but I am not yet ready to reject the 5-gram as special with regard to phi. This is interesting because if indeed a convergence to phi can be derived by the aforementioned method from n-grams, the 5-gram case should not be unique (although this is hard to prove). In other words, my feeling is that the method above contradicts your hypothesis, but at this point it's ambitious to prove rigorously.
Its clear that the ngons prior to 5 have no relationship to phi. After 5 its a different story completely. But what I said is that IF this was a mathematical symptom of the force passing information forwards, the force behind DNA, and IF it is indeed infinite, then it would forever reappear after the point of its first appearance. However you are probably not the one to have this conversation with.
I'd be interested to know any unusual fact or fiction about Phi you have learned on your path, as its quite a grey area to research, as well as any findings on this matter.