11-04-2003, 10:41 PM
|
#11 (permalink)
|
Fast'n'Bulbous
Location: Australia, Perth
|
Quote:
stingc said
Heh. This reminds of my math major roommate junior year. He had a crush on this girl, and proceeded to write down and solve a set of differential equations modeling when the optimal time to ask her out would be. He was semi-serious too! So he eventually asks her out, and it turned out that she already had a boyfriend - for over six months. He called it an error in his initial conditions
|
lol, this is gold.
Back in high school, when we more commonly refered to girls in numbers, quite shallow i know, a friend divised a program which would calulate the probablity of sleeping with someone based on how many drink's you've had. Pretty basic stuff, but it was good for a luagh at the time. At no point did i take it seriously though.
on a sorta of unrealted note, i sent this to a friend a while back in a measn of quantifying what's the best drink for you. I had a bit of time on my hands.....
As for drinks, it would be interesting to one day, form an analysis using
graphs and other techniques to try and objectivly measure a better drink,
rather than arbitarily saying "i like X better cause i say so".
i've done this quite a bit recently at uni, in anaysing software and
gathering metrics to determine, basically how good it is (in simpler terms).
And it has become quite applicable to anything you want to measure which
genrally has a subjective nature. A basic way of doing this would be to
measure the alchohol density per can or conatiner (eg a 5% woody at 440ml is
22 units of alcohol)
(excuse the maths, im doing it quite hurriedly)
measure the cost per unit of alcohol. ie a 4 pack of woodies is 4*22 = 88
units of alchol for $10 (say a 4 pack cost 10)
then its 88/10 = 8.8 units of alchol per dollar
then subjectily assign a multiplier to multiply this quantatative value.
This multiplier could be determined by adding a few subjective fields
together.
possible fields:
how much you like it (rate it 1-5 ---> 5 = love it, 1 = taste like shit)
how embarrased you'd be to be seen with it (1-5 ---> 5= not embarrased,
almost proud 1=i hope no one sees me)
the look/feel of the container(1-5 ---> 5=handles and looks great 1=my hand
hurts or poor marketing design)
useful disscussion about the particular drink with other people at the party
(1-5 ---> 5= talk of the party 1=bland/uninteresting)
when rating this multiplier obviously the subjective fields have a different
level of importance, so they're multiplied as well by universally accepted
constants. Eg the look and feel of the conatainer should be considered less
important than the taste, so you give less percent of the total value you're
multiplying with the original quantifiable value you obtained. This value
could then show which drink is better for the given criteria. The universal
constants, aren't so universal, i subjectivly assigned them for arguements
sake. ie how much you like it has a multiplier of 10 (to my subjective value
assigned), emabarrasment is less important so i'll say a 5, look feel is
less important perhaps a 1, and conversation about the drink is about 2.
This is done so if a drink tastes really bad but has a good look and feel,
its not considered as good as a drink with good taste and bad look feel.
Perhaps you didn't understand that, its quite complicated to explain so i'll
clear it up with an example:
ok i have my pack of 4 woodies with a quantifiable (ie anyone who measures
this will get the same result) 8.8 units of alchol per dollar.
now for the multiplier:
these are purely my subjective values (as for anyone rating their drink)
multiplied by my subjective universal constant (as above) to get the
subjective multiplier.
here we go---->
i kinda like woodies, so i'll give it a 3.5 out of 5 with a multiplier of 10
is equal to 35
the embarrasment of woddies isn't so much so i'll give it a 3.5 again,
multiplied by my constant of 5 is 17.5
the look feel of the can isn't particulary great, its kinda large so i'll
give it a 3, by the universal multiplier of 1 is 3
the novelty/disscussion of the drink is also not much so i'll give it a 2,
by my universal multiplier 2 is equal to 4.
now we have 4 subjective values based on this 4 criteria and there
subjective importance to the overall decision of how good the drink is. Now
we add these up 35+17.5+3+4 = 59.5 now this becomes our overall subjective
multiplier.
59.5*8.8 = 3540.25 goodness per dollar in an alcoholic drink
so i could graph woodies as a 3540 goodness per dollar ( i know goodness is
not really a word, but hopefully you get what i mean) and then compare it to
other drinks i evaluated to determine the best drink for my dollar. Of
course you could subjectivly measure the best overall (where money becomes
irrelevant), but i feel you need some objective criteria to even things out
more, in the grand scheme of things.
|
|
|