Summary: Sent (posted) an incredibly flamboyant sing/letter to a random address in Montana bearing, among other things, the message "Thank You!!!!"
Devised a rough method for quantifying thanks deserved and posted it on the outside of some kind computer room in the physics department. Also decked out in the functional nightsuit.

Details:At first, I thought of going on a roof and posting up a big thank you sign. Then I got around to asking, "What for?" Thanking masses of people, who very well may deserve the thanks, has to be done very generally (due to lack of specific things in common between people that deserve a 'thank you'). I then asked, "Why?" which led to "How?" This led me to devise a guide for how much people should be thanked. But first, the task says I can post a sign. I am on my way to send rthanks.doc (in this task's files) as post to a random address in Montana. Basically, the gist of the sign / letter is THANK YOU, unicorns, almond milk, scandalous news Kraig.

Now for that guide:
YOU MIGHT WANNA FALL ASLEEP FOR THIS SECTION. I'LL LET YOU NOW WHEN THE BORING SHIT ENDS.
My graphing calculator of 4 years is no longer with us (RIP Clif ... you were the best TI 83 a man could ask for (actually you were not in terms of processing power b/c the silver edition has more ram than you, but your aesthetic qualities were the best)). So, I used my soulless 89 titanium to generate some graphs.
I figured that a good way to judge gross amount of thanks deserved with respect to an observed person is directly proportional to the total number of positive thoughts an observer has had about the observed person (and then scaled by some dilation, I really don't know how to handle that per person w/o some unique constant per person). Also, I figured that the longer a positive friendship is established (at least for the first 60 days or so), the rate of change of positive thoughts with respect to time by the observer concerning the observed should increase exponentially and then fluctuate after a while. Since I had the idea of what that graph should look like in my head, I set out to graph it first. I got the shape of the curve right quickly, but it took me a bit longer to keep that shape and get the scale of the graph to something reasonably resembling a healthy friendship:
f'(x) = .69(.025x^2-.01x+.000008*4*x^3*sin(3*x))

where the x axis is time in days, and the y axis is the change in number of positive thoughts with respect to the observed person per day. I figured this graph could be interpreted as the positiveness of mood with respect to another person.
If you take the integral of this graph with respect to time, you get what I was originally after (because it, somehow, is proportional to gross thanks deserved) - total number of positive thoughts towards a person.
And no, I did NOT integrate this by hand. no firggin way.
f(x) = -.000000817778(3x*(3x^2-2)*cos(3x)-(9x^2-2)*sin(3x)-7031.25*x^2(x-.6))

where the x axis is time in days, and the y axis is total number of positive thoughts had for an observed person.
(It also makes sense that if you take the derivative of number of positive thoughts with respect to time, you get rate of change of number of positive thoughts, which ostensibly could tell you a lot about someone's mood towards one other person.)
It's also interesting to note that f''(x) = .000066x^3*cos(3x)+.000066x^2*sin(3x)+.0345x-.0069 and is the rate of change of the rate of change of the number of positive thoughts had about a person with respect to time. This could be interpreted as volatility of mood, but I did non make any graphical representation of this.
Let it be noted that there are many problems with this. For starters, if this was a real experiment, the care of significant figures and standard deviation analysis (on the regression) would be really important to show if this method is even halfway legitimate. Also, The equations I came up with are totally BS theoretical speculations; if this was the real deal, you'd have to collect a lot of data and then maybe run a regression on the data to get an equation (it would be incredible if it came anywhere near my speculation). Also, whereas I started with rate of change of # positive thoughts and integrated that, it would be easier to record total number of positive thoughts, run a regression on that data, and then derive the resulting equation.
Also let it be noted that this whole thing is used to calculate gross positive thoughts about a person, which is proportional to gross thanks due to to one person. Net thanks would be a more complete and useful result, which would take into account negative thoughts as well. While I feel as if the negative thoughts are somewhat accounted for in the sinusoidal fluctuation of either graph, a more thorough analysis of negative thought change over time would make this project more accurate.
Finally, though not the objective of this project, it should be noted that mood towards one person isn't the most useful mood determining factor. I really really wish I knew multivar calc, b/c that would make analysis of net mood fathomable.
OK BORING SHIT'S OVER
That whole mathy project I described I printed out:

So I set out and reached my destination: the outside of a window of a physics computer lookin lab (or something). It was a tiny bit of a climb from ground level to get there.
I assembled the package with trusty duct tape:

Put it up (trusty DT again):

enjoyed for a sec:

walked away:

headed home tired bed night.

Also, During my potted plant distribution task, I wrote "Thank you for not moving me" on the pot.