[cse491] [gvwilson at cs.utoronto.ca: [ucosp] re: yak shaving]

[C. Titus Brown]

----- Forwarded message from Greg Wilson <gvwilson at cs.utoronto.ca> -----

Date: Wed, 21 Oct 2009 06:01:28 -0400 (EDT)
From: Greg Wilson <gvwilson at cs.utoronto.ca>
To: UCOSP <ucosp at googlegroups.com>
Subject: [ucosp] re: yak shaving


A student asks:

> Hey Greg,
> What the heck is yak shaving?

> From http://projects.csail.mit.edu/gsb/old-archive/gsb-archive/gsb2000-02-11.html:

  I was working on my thesis and realized I needed a reference. I'd seen
  a post on comp.arch recently that cited a paper, so I fired up gnus.
  While I was searching the for the post, I came across another post whose
  MIME encoding screwed up my ancient version of gnus, so I stopped and
  downloaded the latest version of gnus.

  Unfortunately, the new version of gnus didn't work with emacs 18, so I
  downloaded and built emacs 20. Of course, then I had to install updated
  versions of a half-dozen other packages to keep other users from hurting
  me. When I finally tried to use the new gnus, it kept crapping out on my
  old configuration. And that's why I'm deep in the gnus info pages and my
  .emacs file -- and yet it's all part of working on my thesis.

In other words, it's like the fairy tales in which the hero wants to  
rescue someone from the dragon, so she needs the sleeping potion, but in  
order to get it, she has to find the key to the wizard's castle, which is  
guarded by a troll, so she needs a magic feather to tickle the troll, and  
the only way to get the magic feather is from the giant canary's nest, and  
the only way to do *that* is with a flying carpet, but there isn't one  
lying around anywhere, so she'll have to make one, and so there she is,  
shaving the wool off an enchanted yak.  (Can you tell that I'm the father  
of a small child? :-)

Things like this (well, *kind of* like this) happen all too easily in the  
real world.  You start work on Problem A, then decide it would be easier  
if you solved Problem B first, but the best way to tackle it depends on  
Problem C --- you very quickly lose sight of your original goal and wind  
up investing way too much time in things that aren't directly delivering  
value.  (Ever seen someone spend an hour upgrading the power monitoring  
software on their netbook so that they could finish an assignment in a  
data structures course --- when finishing the assignment would only have  
taken five minutes if they'd just used a different machine?)

Thanks,
Greg

----- End forwarded message -----

-- 
C. Titus Brown, ctb at msu.edu