Note: now there's a QUACK page with details on some other ideas that were never brought to fruition.
An order of these beautiful QERM t-shirts was made in April 2008 and fulfilled in May. Those who did not take part in the order and still want a t-shirt have a few options (none of which are as good or cheap as being part of the original order). And if you're James Murphy, then you better make sure you don't get crossed up in the world of flying squirrels and salmon, because who knows what might happen to you for spreading heretical ideas.
The PNG files that were used to create the designs below are available through this wiki:
- Squirrel_salmon_black.png ,
- Squirrel_salmon_white.png ,
- Squirrel_salmon_black.jpg ,
- Equation_black.png (says "Invalid thumbnail parameters", huh?),
- Equation_white.png (says "Invalid thumbnail parameters", huh?),
- Equation_black_small.png ,
- QERM_rock-n-roll_black.png ,
You can upload these to CafePress.com or other online printers to make your own t-shirts or other stuff. For example, here's a mug ($10.99) available here. Note, the JPG included above was used in this case because the PNG file didn't show up on CafePress (but only the one image, not the others).
The order tally for the green shirt with the equation on the back and the awesome squirrel-salmon logo on the front was at 37. This puts the cost per shirt at the low low price of $10.77.
The black, organic cotton shirt will have a more exclusive ownership, with 20 orders. It was described by one responder as reminiscent of the "skanky bands from the late 70s/early 80s." I believe this indicates that the design was a success based on the goals set by the designers. What strange circumstances led their math addled minds to cook up these goals is a separate question. The smaller number of orders and the more expensive fabric put the cost at $14.43 per shirt.
For some reason, BlueCotton.com doesn't like to provide their formula up front: you have to enter a quantity to get a price. However, it appears from the following analysis that the relationship is pretty simple. For the grass green short-sleeve shirt shown above, the price is based on a fixed cost plus a reasonable price per shirt. However, if you go below 6 shirts, this model breaks down and you pay a lot more, so for small orders it would be better to use CafePress or other print on demand companies.
Here's the code used to create the image above.
# modeling t-shirt pricing at BlueCotton.com png <- FALSE if(png) png('C:/Documents and Settings/tayloria/My Documents/QUACK/t-shirt_prices.png',width=6,height=6,res=300,units='in') x=c(6,10,15,20,30);y=c(134,173,221,269,365) # data based on multiple inputs for quantities to BlueCotton website lm1 <- lm(y~x) fun <- function(x) lm1$coeff/x+lm1$coeff plot(x,y/x,xlim=c(6,30),axes=F,main='Price of grass green shirts at BlueCotton.com', xlab='number of shirts purchased',ylab='cost per shirt',) grid() lines(6:30,fun(6:30),lwd=3) points(x,y/x,col=2,pch=16,cex=1.5) axis(1,at=c(6,seq(10,30,5))) axis(2,las=1,at=pretty(range(y/x)),lab=paste('$',pretty(range(y/x)),sep='')) legend('topright',pch=c(16,NA),pt.cex=c(1.5,NA),lty=c(NA,1),lwd=c(NA,3),col=c(2,1),bg='white',box.col='white', legend=c('Observed',paste('Expected = $',round(lm1$coef,2),' + $',round(lm1$coef,1),'/n',sep=''))) box() if(png) dev.off()